NHCE CSE Compendium of R&D Projects

LAUNCH The UNITYSat project is designed to work in a polar orbit of 575 km. It will be put into this orbit using ISROs Polar Satellite Launch Vehicle. The satellite is set to be launched on February 28th, 2021, from ISROs launch port at the Satish Dhawan Space Centre at Sriharikota, Andhra Pradesh, India. The satellite will be launched aboard the PSLV C-51, ISROs 53rd launch mission. The PSLV C51 is planned to be launched at 04:53 (UTC) / 10:23 (IST) on 28 February 2021 with the main payload from Brazil, INPE’sAmazônia-1and20otherride- sharingsmallsatellites.This includes three of the UNITYSats. Once in the right altitude the satellites will be injected into orbit using ISROs 1U cube sat deployer where it will separate each other using the on-board spring plungers. The satellites are stacked in the deployer separated by Spring Plungers used in between TWO of the Satellites. During the deployment sequence the force exerted by the primary spring of the deployer and simultaneous opening of the door of the deployer will pave the way for deployment of 1st satellite along with cumulative effect of Spring Plungers mounted in between preceding satellite. Following this, the remaining two satellites will follow because of the force from the deployer ‟spring and separate because of the spring plungers. Fig. 3 UNITYsat Team has Innovated CanSats Systems and Subsystems for Global Markets 196

Mr. Sainath Vamshi, Core Team Member of UNITYsat is working on SETI@home Project of The University of California, since 28 September 2018 and during this period he has contributed 17,528 Cobblestones of Computation (1541 Quadrillion floating-point operations) to SETI@home Search for Extraterrestrial intelligence (SETI) Life! On 08 Nov 2019, the above Certificate of Computation has been issued to him as an Appreciation to his tireless efforts! SETI@home is an Internet-based public volunteer computing project employing the BOINC software platform created by the Berkeley SETI Research Center and is hosted by the Space Sciences Laboratory, at the University of California, Berkeley. Fig.4 Development Process and Testing 197

Consultancy Project Period: 2017-2018 Ref: 17-18-CP/CSE/001 Project Name: Online Evaluation Tools Funding Agency: EDUMERGE Development Leads: Dr. Beena B M, Ms. Soja Rani, Mr. Vinay Khande, Mr. Vishwas INTRODUCTION EDUMERGE is a platform where the School/College and the Parents interact for the student’s betterment. EDUMERGE, a parental portal that provides a referral point for all School/College related activities. Activities assigned to students or notifications from the School/College to the parents are sent through this portal. EDUMERGE App provides Institutions a communication &Learning platform among their students, teachers, parents and management. Fig.1 Architecture of Educational App Benefits of Mobile Apps in Education In this tech-savvy era, students are more determined towards using a mobile phone for every purpose, and in this state, the educational apps can be the perfect way to attract the students and persuade them to study. The students can get access to any information from anywhere with the educational apps. Therefore, the mobile apps are the most interactive and constructive way to attract the students towards studies and enhance their productivity. Accelerated development: The process a learner comprehends, and processes information will vary depending on their cognitive learning styles. The app can facilitate the learning style according to curriculum requirements and it will accelerate the learning process and as a result prove to be beneficial to the user. Cognitive improvement: An app with quality content can challenge a learner’s problem solving skills through the use of virtual rewards, incentives etc. that will motivate them to persist and continually develop themselves in the process. Educational standards changing: The future of education belongs in technology, specifically mobile technology and apps. So, the best educational apps will surface and be readily available to all learners. Research has shown that the integration into the curriculum can and will raise educational standards. If every learner using these apps sees improvement, the landscape of education could be enhanced for the better in schools and in society. 198

Remote Learning: The use of educational apps facilitates remote learning, and the learning is just a click away! Informal Learning: The entertaining graphics and attractive illustrations are way better than regular study patterns. Comfortable atmosphere of learning apps helps children towards enhanced learning outcomes. In brief, we can see that the quality of an educational app has huge benefits and can do a great deal for a learner. Through the use of tailored curriculum requirement and relevance in terms of the subject it is obvious that such an app could accumulate success amongst personal development and accomplishment. 24/7 Availability School and special classes have time limits for study. A child can study in a particular timing slot of school and tuitions; they have to clear their doubts in that time.Now education app for students can help to clear doubts anytime and anywhere as it doesn’t have any time limits and need of teachers. Systematic Learning Activated Education apps enable both the things together, smart and systematic learning. Entire content on an education app is arranged in such a way that increases the student’s desire to learn more things and that too in a systematic way.A high-end education app enable students to follow a smooth and logical flow without making much efforts. Advantages from Contributors Viewpoint The contributors will be able to showcase their courses offered, monitor the progress of the students, and facilitate a reward based fun learning. It creates a platform for the efficient content management. Advantages from the customer/learner point of view The learner/customer can view the various courses, take up tests, monitor the performance. The App facilitates a remote learning for the learner. OBJECTIVES  To understand the requirements of remote learning.  To create mobile application to facilitate remote learning.  To design a creative user interface.  To create dashboard with the courses, contents and performance tracker. REQUIREMENT ANALYSIS A study on the features to be included in the development of the app is summarized as follows.  Dashboard For the Summary view of Performance, Attendance, Online Tests, Fee &Notices etc. Tile-based layout for each functional area.  Notifications To Get instant mobile alerts on your mobile/Tablet. Date-wise view of all the notifications received from the Institution.  Send Notifications Send instant notifications to all Parents & staff or students Groups. Send unlimited alerts on events, notices, academic details, dues etc. Feature enabled for Institution& Administration Team.  Learning Center To access content shared by the teacher. Content can be pdf, images, tasks, Tests and blogs.  Students access Learning Content shared by the teacher. Access of tasks, pdfs, images, videos & Audios, Tests & Surveys to Blogs. Students work based on the defined learning path.  Online Test - To facilitate the students, take the Test & review their test results with answer keys. 199

IMPLEMENTATION Creative user interface of Application: Login/ Signup It’s better to offer various options to sign up. For example, an account can be created by email, phone number, or via social networks like Facebook, Twitter, LinkedIn and others. To make the app more user- friendly, you need to verify the account and add the “Forgot password” option. User profile Through these dashboards, users can check which courses they participate in and adjust to their schedules. Here, they also can add their account details like subject, age, education, etc. Learning materials First, learning materials must be proven scientifically. it can vary from different types:  Video/ audio materials  Interactive exercises to test knowledge.  Online courses Push notifications Proved to be useful for user engagement. Notifications can be for an opportunity to get a discount, new interesting features, any course updates, etc. Social networks integration Today it’s impossible to imagine a mobile app without the implementation of such a feature. In fact, it has several tasks, namely – to allow the user to share their achievement with friends. Search functionality In addition, the faster the user gets the desired result, the better the chances they keep your educational app. Students can find online courses they need or come back to completed soon. For teachers, they are able to adjust to the programs they teach. SCREENSHOTS Fig.2 Sign in/Register Page Fig.3 Dashboard 200

Fig.4 Profile View Fig.5 Attendance Tracker Fig.6 Feeds and Notifications Fig.7 Help Desk CONCLUSION With this app, we are primarily focussing on Learner satisfaction with respect to remote learning. The learner can learn the course contents which is just a click away without actually going to a study centre. Consultancy Project Period: 2017-2018 Ref: 17-18-CP/CSE/002 Project Name: Online Promotional Mobile Apps Funding Agency: K2 LEARNING Development Leads: Mr. Puneet S Palagi, Ms. Yogitha Ms. Nidhishree, Ms. Harshitha INTRODUCTION The widespread access to the internet has started a new era of digitization in every field. The fast and easy accessibility of the internet has made it possible to reach out from the most developed region to the remotest regions in the world with the same information. Due to this, it’s now easy to access social sites or entertainment sites or news sites. There is a unified platform for every field which can be accessed by anyone in the world. Education has also seen a lot of changes in the past few decades with the start of digitization. Education apps have been developed to encourage students of all ages to learn and explore more in the field and generate more interest to make learning fun. The eLearning industry has grown exponentially in the last two decades and it continues to grow with better accessibility and improved education apps that cater to the requirements of students of all age groups. Advantages of Educational mobile apps There are uncountable advantages & importance of educational mobile apps for students and school management. Education apps improve education system, it will make easier for students to learn things and 201

remember it for a long time. Education app is easier to understand and use as compared to books and lectures that students attend in the classrooms. Various Benefits of Education Apps are given below:  Students prefer apps. Today’s generation like to spend their time on mobile. If we connect their boring activities into interactive activities, then they will love study.  Leisure Hours Utilization eLearning on a smartphone is the best way to utilize children’s time effectively. Children can utilize their leisure hours to learn new things using a mobile application for education.  Reducing the Gaps Advancement in technology has helped to remove various glitches that are present in the education system. Different apps and websites can be developed to fill the interaction gap between students & teachers and parents & teachers. Students and parents will be informed about any event or change in schedule or announcement.  Quality Interaction As various surveys conducted, education apps are proved to be helping students to enhance their interest in studies. They are assisting the students to become more concerned about their education.  24/7 Availability School and special classes have time limits for study. A child can study in a particular timing slot of school and tuitions; they have to clear their doubts in that time.Now education app for students can help to clear doubts anytime and anywhere as it doesn’t have any time limits and need of teachers.  Systematic Learning Activated Education apps enable both the things together, smart and systematic learning. Entire content on an education app is arranged in such a way that increases the student’s desire to learn more things and that too in a systematic way.A high-end education app enable students to follow a smooth and logical flow without making much efforts.  Sustainability Mobile learning apps are more sustainable as compared to traditional learning methods that consist of various papers, pencils, and pens. Moreover, these learning apps allow students to obtain reference notes quickly by just downloading them. Also, completing a lesson on mobile app is effective and simple. A hidden advantage is that this will result in less number of trees cut down every year.  Portability and Mobility Mobile phones are portable. Students and parents can carry mobile phone anywhere and use education apps in any setting in which their child feels comfortable. Using education apps, learning is not just restrained to classrooms, or homes. Students can use them anywhere they like. OBJECTIVES  To understand the requirements of remote learning.  To create mobile application to facilitate remote learning.  To design a creative user interface.  To create dashboard with the courses, contents and performance tracker. 202

Survey Results on the Study on the Features of Education App Syllabus It is important that we chalk out the syllabus of the course that you intend to offer in the learning app. An education app is also used in smart classrooms to study full course with special teaching skills. There are many schools that are using education app for improve the experience of students and make study enjoyable. Notifications and course activities The app should include a good notification feature that will help to inform the user with all the required updates. It is a great idea to break the learning monotony with various course related activities. The users can be notified about the latest events related to the course taken in an activity forum. UI UX of the app Educational app should be more engaging and simpler to use. Quality of the content The content of the page should not be very extensive that it gets strenuous to the eyes of the user. Since it is a mobile app the view able area is relatively small hence with too much content you may lose the interest of the user. It is good to use animation and graphics whenever possible to maintain interest throughout the course. Using instructional videos is a great way to teach and keep a human approach to the whole learning process. Live Tutorials and Interactive Sessions There should be an option to allow your students to join live lectures and teaching sessions of their tutors. These live interactive sessions allow a student to clear any doubts or get answers to any query which they may have on a specific topic. Virtual classrooms with live tutorials and chat sessions with the tutors take the learning experience to a whole new level. IMPLEMENTATION On the client side we have a Login portal, once done; it displays the home screen. The history of the visit is maintained on the server side. It has mainly two modules. Client Module In this module, we have an optional login module. If not registered, a signup module is designed. The login is not mandatory, just as an option to improve the customer choices. Login Module/Signup Module Display K2 Learning Server Module: (Admin Module) The Server module has two sub modules -one for updating customer particulars to offer K2 learning services and a record visit module to improvise the user experience. Update student Particulars Record Visits made (History) UI DESIGN 203

Fig.1 Home Page Fig.2 Sign in/Register Page. Fig.3 Login Page Fig.4 The Side Pane View Fig.5 Course content Details CONCLUSION With this app, we are primarily focussing on Learner satisfaction with respect to remote learning. The learner can learn the course contents which is just a click away without actually going to a study centre. 204

Consultancy Project Period: 2018-2019 Ref: 18-19-CP/CSE/001 Project Name: Online Evaluation Tools Funding Agency: SARAH ENTERPRISES Development Leads: Dr. Nilima Kulkarni, Ms. Asha Rani Borah, Mr. Rohit Mullay, Ms. Shriya B INTRODUCTION People who love reading books have different genres of books or literature books that they have stopped reading and these books are laying on their bookshelves unused. These books can be re-sold, exchanged and donated and the proceeds can be used to buy another book the person currently needs. This application will serve anybody who wants to buy or get second hand books or wants to empty their bookshelves and make financial gain from it. These books can just simply be uploaded to the application and another person can have access to buy the book by just using the application. Fig.1 Architecture Diagram Books are essential to every process of education, information, creativity and development. There two groups of bookaholics first group of people cannot afford to buy as many books as they want and second group who has huge number of books that there is no place left on their shelves due to which they sometimes throw these books which can be very precious to others. The gap between these two groups is fulfilled in this platform, where anyone can buy, donate and resell books. The power of technology upon education has been immense over the past few decades. Great education for students is no more a dream with the availability of the various mobile applications available at the finger tips. Choosing the Right Application can revolutionize the way a student looks at the process of learning. For Education through mobile apps, there’s a diverse range of software that falls under the umbrella of education. Mobile technology is an amazing revolutionary productivity tool of all time. One can use smart phones and tablets to teach new skills and learn new things. The quality of an educational app is a pivotal attribute in the ever progressing world of teaching and education, and it can add a great deal of value to the learner’s educational experience. The major aspects with respect to an app in education are the quality content with the mobile app features. While developing mobile application for K2 learning one of the challenges was to ensure that this app will stand out among all other hundreds of thousands of apps which are available to any learner on their preferred platform. 205

OBJECTIVES  To create mobile application which allows its users to resell, exchange and donate their books.  To design a creative user interface. ANALYSIS AND DESIGN Algorithm / Pseudo code: As soon as the app is launched the user gets log in screen. If it is a new user then user has to click sign up button to register. Activities in Sarah Enterprises:  Activity 1 – Welcome page which consists of two options I.e. Login or Register.  Activity 2 – The new user can register by filling the details like name, email and password to proceed further. Already registered users can log indirectly.  Activity3–AftersuccessfulloginHomescreenisdisplayed.Homescreenconsists of list of books that is available for reselling, donation or exchange along with search option to search for any particular book.  Activity 4 – Profile activity displays the account details of user. User can check its activity on the app, edit account details, update location and share this app with friends.  Activity 5 – My Books activity displays the books that are put on sale under resell, exchange and donate category by the user and shows the status if book is sold or still waiting in the list.  Activity 6 – Wish list activity shows the books that are wish listed by the user and can buy or exchange in future.  Activity 7 – My Orders activity displays all the books that are ordered by the user.  Activity 8 – Add Book screen is where user can enter the details of the book like book name, author name, publisher, page number and small description about the book. After user adds this book it will be stored in database and it will be visible to all users of this application. IMPLEMENTATION Sign Up for Users: User details like name email, mobile number and password are taken and location is marked by the user on Google Maps, which is stored like an address in firebase database in the form of key value pair. Email authentication is used for authenticating each user. User Log in : User log in using the email and password set by that user. After verifying the credentials, the user is logged in successfully and enters the Home page of the app where all books available for resell, donate and exchange are displayed. SNAPSHOTS Fig.1 Forgot Password Fig.2 Welcome Page 206

Fig.3 Add New Book for post Fig.4 Upload Multiple Images of Book CONCLUSION The goal of this project is to create a user-friendly application that will help the readers as well as learners to get the second-hand books and novels easily at reasonable prices or free sometimes. Books that are read once are left on shelves or thrown in dustbin, with this application any book knowledge will not go in dustbin and will be passed further to others who are really in need of the books. It has been achieved as it applies the android application development knowledge and java programming language and Firebase Database, to build the theoretical and practical aspects of the learning process of Mobile App Development and its concepts. Consultancy Project Period: 2018-2019 Ref: 18-19-CP/CSE/002 Project Name: Online Promotional Mobile Apps Funding Agency: K2 LEARNING Development Leads: Dr. Vishwanath.Y, Ms. Yogitha Ms. Nidhishree, Ms. Harshitha INTRODUCTION The power of technology upon education has been immense over the past few decades. Great education for students is no more a dream with the availability of the various mobile applications available at the fingertips. Choosing the Right Application can revolutionize the way a student looks at the process of learning. For Education through mobile apps, there’s a diverse range of software that falls under the umbrella of education. Mobile technology is an amazing revolutionary productivity tool of all time. One can use smart phones and tablets to teach new skills and learn new things. The quality of an educational app is a pivotal attribute in the ever-progressing world of teaching and education, and it can add a great deal of value to the learner’s educational experience. The major aspects with respect to an app in education are the quality content with the mobile app features. While developing mobile application for K2 learning one of the challenges was to ensure that this app will stand out among all other hundreds of thousands of apps which are available to any learner. 207

Benefits of Mobile Apps in Education In this tech-savvy era, students are more determined towards using a mobile phone for every purpose, and in this state, the educational apps can be the perfect way to attract the students and persuade them to study. The students can get access to any information from anywhere with the educational apps. Therefore, the mobile apps are the most interactive and constructive way to attract the student’s towards studies and enhances their productivity. Accelerated development: The process a learner comprehends, and processes information will vary depending on their cognitive learning styles. The app can facilitate the learning style according to curriculum requirements and it will accelerate the learning process and as a result prove to be beneficial to the user. Cognitive improvement: An app with quality content can challenge a learner’s problem-solving skills through the use of virtual rewards, incentives etc. that will motivate them to persist and continually develop themselves in the process. Educational standards changing: The future of education belongs in technology, specifically mobile technology and apps. So, the best educational apps will surface and be readily available to all learners. Research has shown that the integration into the curriculum can and will raise educational standards. If every learner using these apps sees improvement, the landscape of education could be enhanced for the better in schools and in society. Remote Learning: The use of educational apps facilitates remote learning, and the learning is just a click away! Informal Learning: The entertaining graphics and attractive illustrations are way better than regular study patterns. Comfortable atmosphere of learning apps helps children towards enhanced learning outcomes. In brief, we can see that the quality of an educational app has huge benefits and can do a great deal for a learner. Through the use of tailored curriculum requirement and relevance in terms of the subject it is obvious that such an app could accumulate success amongst personal development and accomplishment. Advantages from Contributors Viewpoint The contributors will be able to showcase their courses offered, monitor the progress of the students, and facilitate a reward based fun learning. It creates a platform for the efficient content management. Advantages from the customer/learner point of view The learner/customer can view the various courses, take up tests, and monitor the performance. The App facilitates a remote learning for the learner. OBJECTIVES  To create mobile application to facilitate remote learning  To design a creative user interface.  To create dashboard with the courses, contents and performance tracker ANALYSIS AND DESIGN Algorithm / Pseudo code Step 1: Start If New user Register Else Login If (Login Successful) Go to 2 Else 208

Print(“Login Failed ! Try again”) Step 2: Home page Select any class to continue If(Selected class==”CPT”) Go to 3 Else if(Selected class==”IPCC”) Go to 4 Else if(Selected class==”Basic Mathematics”) Go to 5 Else if(Selected class==”Business Studies”) Go to 6 Else if(Selected class==”Economics”) Go to 7 Else if(Selected class==”Statistics”) Go to 8 Step 3: CPT dashboard ->Select any subject to submit response Go to 10. Step 4: IPCC dashboard ->Select any subject to submit response. Go to 10. Step 5: Basic Mathematics dashboard ->Select any subject to submit response. Go to 10. Step 6: Business Studies dashboard ->Select any subject to submit response. Go to 10. Step 7: Economics dashboard ->Select any subject to submit response. Go to 10. Step 8: Statistics dashboard ->Select any subject to submit response. Go to 10. Step 9: Selected subject page If (Toggle on) Send responses. Else Don’t send any responses. Step 10: End SNAPSHOTS 209

Fig.1 Home Page Fig.2 Sign in/Register Page. Fig.3 Login Page Fig.4 Dashboard Fig.5 The Side Pane View Fig.6 The Course Display Module CONCLUSION With this app, we are primarily focusing on Learner satisfaction with respect to remote learning. The learner can learn the course contents which is just a click away without going to a study centre. 210

Consultancy Project Period: 2019-2020 Ref: 19-20-CP/CSE/001 Product Name: Mobile App Development Funding Agency: ITCA, CSPD, Serbia, UNISEC India, TSC P Ltd Development Lead: Dr. Thirukumaran R, Mr. Santhosh Kumar B, Mr. Sanketh Huddar ABSTRACT Due to the growing need for global coverage of reliable antenna networks for communication with satellites we need to establish a complete platform of a satellite ground-station network. The scope of the project is to create a full stack of open technologies based on open standards, and the construction of a full ground station as a showcase of the stack. INTRODUCTION Sticking to modular architecture based on network communications we ensure remote access and interchangeable design meeting all possible needs. We designed and created a global management interface (SatNOGS Network) to facilitate multiple ground station operations remotely. An observer is able to take advantage of the full network of SatNOGS ground stations around the world. Due to the growing need for global coverage of reliable antenna networks for communication with satellites we need to establish a complete platform of a satellite ground-station network. The scope of the project is to create a full stack of open technologies based on open standards, and the construction of a full ground station as a showcase of the stack.  Reliable global coverage (real-time communication, regardless of the position of the satellite);  Support and strengthening of SATNOGS network;  Raising radio amateurism to a higher level;  Gaining importance of the operators/owners/participating countries;  Global benefit;  A global player;  Cooperation (unity) and assistance (help);  Motivation of young people for this area, etc. Our Ground Station is a modular and scalable stack for implementation. Fully proprietary,it provides interoperability with existing or future subsystems. Estimated Cost of Full Stack Ground Station Setup with Software and Mobile App. IMPLEMENTATION TSCSatNav: Amateur Radio and Weather Satellite Tracker and Passes Predictor for Android. The Libre Space Foundation team is also behind the epic SatNOGS project that provides an extremely easy to use API and DB with a huge amount of information about satellites, their telemetry and transmitters, which the app uses under the hood. For TLE data calculation and passes prediction TSC SatNav uses the mavenized version of predict4java library, making this library efficient and easy to use! The app is built using Dagger2, Retrofit2, Kotlin and Kotlin coroutines, Architecture Components and Jetpack Navigation.  Calculating satellite passes for up to one week (168 hours)  Calculating passes for the current or manually entered location 211

 Showing the list of currently active and upcoming satellite passes  Showing the active pass progress, polar trajectory and transceivers info  Showing the satellite positional data, footprint and ground track on a map Offline first: passes predictions are made offline. Weekly updates of TLEs and transceivers DB are recommended. RESULTS Snapshots of TSC SatNav (Mobile App): Fig 1. Mobile Snapshot Fig2. Settings Fig3.Position Fig4. Mobile App Snapshot 212

Consultancy Project Period: 2019-2020 Ref: 19-20-CP/CSE/002 Product Name: Mobile App Development and Customer Satisfaction Survey Development Lead: Dr. Vinodha .K, Ms. Pramila Rani.K, Mr. Edwin Benny, Mr. Anand R Patil INTRODUCTION As the world is moving forward with technology, the distance between buyer and seller has become smaller and, in turn, making commerce better for everyone. Say hello to [Fitness], a digital shopping application and your own fitness buddy available at the fingertips. This application makes it easier for users to order their fitness accessories and kits easily and track their orders. Also, this application rewards you based on how much you have done efforts to keep your body fit. With rewards you will be able to get discounts on the accessories and kits in the shop which is a win-win situation. It allows users to set fitness goals, track activity. Characteristics such as a user-friendly interface, automatic tracking, and security are main characteristics of the app. combining the popularity of mobile devices with the on-going search for fitness is the purpose of this application. Benefits of Mobile Apps in Fitness and shopping Fitness apps are applications designed to keep you fit and healthy. The aim of the app is to make you lifestyle healthier. Shopping apps are designed to make it easier for users to purchase their required accessories and kits easily at their fingertips. Users can buy their accessories as their need from the app. Their order will be delivered by the delivering companies to their doorsteps. Set realistic fitness goals: Setting up an unreachable goal at the beginning of our workout regime can be the catalyst for starting the fitness regime but is hardly sustainable. Apps help to set up realistic goals within our desired time frame. They take us to the next level of exercise only after we achieve the previous target. This helps in sustaining our exercise regime which is the hardest challenge for beginners. Easily keep Track of your Progress: Of course, you have set the goals. However, you must keep an eye on the day-to-day results so as to check if it is working for you or not. The app helps you keep track of your progress. It lets you know whether you are going in the right direction or not. If there is progress, continue with your plans, otherwise, it is time to change them. Better efficiency: Mobile applications are more flexible and user friendly. A good app with right concept and functionality brings more clients and interests people to start their fitness journey. Detailed Analytics: Data is easy to gather and track in the application. Mobile functionality allows you to monitor user’s interaction and gives you useful information about them such as responsiveness to content and features, feedback, session length, audience composition. Cognitive improvement: Setting goals helps user to improve them physically and mentally. Also, rewards keep them interested in the workouts. 213

Advantages from Contributors Viewpoint The contributors can add their products provide their products. It creates a platform for the efficient content management. Advantages from the customer/user’s point of view The customer/user can view their orders; track their steps through the app. OBJECTIVES  To create mobile application for fitness and for making order on sports accessories.  To design a user-friendly interface.  To create dashboard to track steps and make it easier for user to buy their fitness accessories. IMPLEMENTATION First, User needs to sign up with proper credentials. After signing up successfully, user need to login. After login with correct credentials, they would be taken to dashboard where they will be able to track their steps and see their orders. In Shop section, user can choose the category they need and order their product easily with their fingertips. In Settings section, user can change their profile details. RESULTS Fig.1 Starting Page Fig.2 Register Page Fig.3 Login Fig.4 Dashboard 214

Fig.5 Side Pane View Fig.5 Category Page Fig.6 Place Order Fig.7 Account Settings CONCLUSION The main aim of this app is to keep users fit with the help of their mobile phone. Also, to provide a digital shop for users this will be available at their fingertips. 215

Consultancy Project Period: 2019-2020 Ref: 19-20-CP/CSE/003 Product Name: Development of Software and Mobile App & Survey Funding Agency: Sodexo Development Lead: Dr. Rachana Ms. Alpha Vijayan, Ms. Moni Krithika, Ms. Divya ABSTRACT: Development of Software for Sodexo is a dedicated feedback or data collection tool that brings the actionable information through online surveys or forms. This platform aim to become customer-centric and feedback-driven to drive success and achieve growth. This survey tools helps the company to accomplish that. This web application provides an easy-to-use platform that helps even a novice to create surveys with little effort. Mobile Application Module conduct surveys anywhere, even off-site and on the go! Survey responses are saved on your device until they can be synced into a database via an internet connection. INTRODUCTION The online surveys are a simple and easy way to get in contact with a targeted audience and collect feedback. There are several reasons businesses send out online surveys, including:  Tracing product feedback  Conducting market research  Obtaining feedback on customer service  Measuring customer and employee satisfaction IMPLEMENTATION The system has two types of users which is the admin and subscribers/users. The admin is the one who manages all the data of this system also the user role who is permitted to create a new user. The subscribers/users are those system users that answer the survey. The system has a simple form builder feature to create a survey questionnaire. The survey form builder has only three options to choose from for the type of inputs on how the users/subscribers answer the questions. The system generates a printable report for the result of each survey set. Modules: All Users: 1. Login Page a. The page where system users will submit their credentials to access the data and functionalities of the system. 2. Dashboard Page a. The page where the system users will be redirected by default after logging into the Online Survey System. 3. Manage Account Modal a. The popup modal where the system users update their system credentials such as their email and password. 216

Admin Side: 1. New User Page a. The page where can system admin create a new user. 2. Manage Users a. This feature includes View, Edit, and Delete for the user list. 3. New Survey Page a. The page where can system admin create a survey set. 4. Manage Survey a. This feature includes View, Edit, and Delete for the survey list. 5. View Survey Page a. The page where the admin can manage the questionnaire of the selected survey set. 6. Survey Report a. The page where the admin can view and print the result of the survey. Users Side:  Survey List  The page where the active survey sets are listed.  Answer Sheet Page  The page that shows the question of the selected survey set. This where the user answer or take the survey. Mobile App: Collect Data Store in Database RESULTS All the necessary tests have been conducted and the software has cleared all the tests and the software is ready for direct deployment. Fig.1 New User Registration Fig.2 New Survey Creation 217

Fig.3 Insert New Questions Fig.4 Survey Report Fig.5 Mobile Application Survey CONCLUSION Using a survey is an interactive way to introduce a new feature to your users and easily crystallize for them both the way to use it and the value. Consultancy Project Period: 2019-2020 Ref: 19-20-CP/CSE/004 Project Name: Web Portal Development Funding Agency: Veerabhdareshwara Enterprises Development Leads: Dr.Sridevi S, Ms.Rajitha, Mr.Rahul,Mr.Vinay INTRODUCTION In the current COVID-19 situation, offline stores have not been able to get/ fulfill orders and this caused a drop in the store’s sales. The need of the hour is that the store is available to customers without violating the COVID norms. A solution to this is an online store, this allows customers to place orders with the store without having to step outside their house. This will also help the store increase sales and attract more customers since they do not need to pay a visit physically to get things done and is also available 24/7. OBJECTIVES The main objective of this web application is to bring up a virtual store, a store where customers can shop just like a physical store and place orders for ready-made/custom-made products. These benefits can be availed right from one’s home. The website can also be used by the back-end team for analysis and getting an overview on the store’s performance over the years. IMPLEMENTATION Python Web Application: This web application is coded in python with flask as its back-end framework. The web-app works with 2 major modules: - For logged-in users: navbar consists of catalog, cart, account and logout tabs, every user who logs in has a record in the database that is retrieved and matched across entered credentials and a space on the store website that help provide customized products, offers and promotions. They can then place orders on the site. - For not-logged-in users: navbar consists of home, login, register, contact us, about us tabs, user doesn’t have a personalized space in the website and can just get an overview on the store, to purchase a product, he/she needs to register on the site. Flowchart: 218

Fig.1 Flow Chart Dashboard The Dashboard.html page is accessible only by the Admin post login with secret credentials, the admin receives a similar login page as the customers. If the user ID matches ‘admin’ and password match to that with the database record, dashboard is accessible where the admin can perform actions on behalf of the store, he/she can fulfil orders, generate invoices and trigger emails right from the dashboard, and also view statistical data for the store over time. The shopping cart page is available only after login or registration and it contains the user details that are to be retrieved when the user places an order. RESULTS The Website lays its foundation on the index.html page and branches across various WebPages and items within the website. This website also has a user page which access after logging in or registration this user page is redirected to the login page if there is no user ID present in the browser's cache. Fig.2 Home Page Fig.3 Gallery 219

Fig.4 Index Page Fig.5 Sign up/Register Fig.6 Login page Fig.7 Dashboard CONCLUSION This web application uses python 3.9 with Flask as its front-end framework; it is based on the login system where users have an account on the portal that they create the first time they visit the site and then login when they visit the site after that. A logged-in user gets personalised feed on the catalog page, and can access pages like ‘cart.html’, ‘my-account.html’ etc which is not available for Users that are not logged in. The dashboard shows an overview on the annual sales, daily agenda etc. which the store caretaker/backend team fulfils. Consultancy Project Period: 2019-2020 Ref: 19-20-PD/CSE/005 Product Name: Evaluation of Projects: Feasibility Studies Development Lead: Dr. Clara Kanmani ,Ms. Rajitha Nair, Ms. Nisha Ms. Thrupthi INTRODUCTION Catering is the business of providing food service at a remote site or a site such as a hotel, hospital, pub, aircraft, cruise ship, park, filming site or studio, entertainment site, or event venue. Catering has become very important and a necessity in any event. Swadishta Catering offers pure veg catering service to remote places. It aims at offering the best quality and taste food to events. This Catering Service provides various varieties of foods with different tastes. 220

OBJECTIVES * Offering remote food service to different variety of events to different areas. * Offering a quality and pure vegetarian food service. IMPLEMENTATION FEATURES * The app is developed using Android Studio. * Java is the programming language used. *In the home screen: -Linear layout is used to fit all the views. -Text views are used to display texts. -Image view is used to represent the image. -A button view is used to go to menu page. -Intent is used to navigate from one class to other *In menu page: -Linear layout is used. -Text views and button views are used. -List views will be used for list of food items. -Buttons are used to lead us to next page Results: Fig.1 Home Page Fig.2 Menu Fig.3 Finalize Order Fig.4 Bill Payment 221

Consultancy Project Period: 2019-2020 Ref: 19-20-CP/CSE/006 Project Name: Web Portal Development for Online Survey Funding Agency: Swathi Glass & Ply Boards Development Lead: Dr.Jaya.R Mr.Sivabalan, Mr.Suhas, Mr.Ayush ABSTRACT Online Survey web portal for Swathi glass and ply boards is a dedicated feedback or data collection tool that brings the actionable information through online surveys or forms. Every businesses in the world aim to become customer-centric and feedback-driven to drive success and achieve growth. This survey tools helps the company to accomplish that. This web application provides an easy-to-use platform that helps even a novice to create surveys with little effort. INTRODUCTION The online surveys are a simple and easy way to get in contact with a targeted audience and collect feedback. There are several reasons businesses send out online surveys, including: ● Tracing product feedback ● Conducting market research ● Obtaining feedback on customer service ● Measuring customer and employee satisfaction IMPLEMENTATION The system has two types of users which is the admin and subscribers/users. The admin is the one who manages all the data of this system also the user role who is permitted to create a new user. The subscribers/users are those system users that answer the survey. The system has a simple form builder feature to create a survey questionnaire. The survey form builder has only three options to choose from for the type of inputs on how the users/subscribers answer the questions. The system generates a printable report for the result of each survey set. Modules: All Users: ● Login Page o The page where system users will submit their credentials to access the data and functionalities of the system. ● Dashboard Page o The page where the system users will be redirected by default after logging into the Online Survey System. ● Manage Account Modal o The popup modal where the system users update their system credentials such as their email and password. Admin Side: ● New User Page o The page where can system admin create a new user. ● Manage Users o This feature includes View, Edit, and Delete for the user list. ● New Survey Page o The page where can system admin create a survey set. ● Manage Survey 222

o This feature includes View, Edit, and Delete for the survey list. ● View Survey Page o The page where the admin can manage the questionnaire of the selected survey set. ● Survey Report o The page where the admin can view and print the result of the survey. Users Side: ● Survey List o The page where the active survey sets are listed. ● Answer Sheet Page o The page that shows the question of the selected survey set. This where the user answer or take the survey. RESULTS ● All the necessary tests have been conducted and the software has cleared all the tests and the software is ready for direct deployment. Fig.1 New User Registration Fig.2 New Survey Creation Fig.3 Insert New Questions Fig.4 Survey Report 223

Product Development Period: 2017-2018 Ref: 17-18-PD/CSE/001 Product Name: Virtual Motion Detector to Detect Crime in Real-Time And Alert Authorized Persons Using Binary Comparison Development Lead: Ms. Jaya.R, Ms. Sheba Pari ABSTRACT The Virtual Motion Detector System aims at providing an understanding vision to the user’s computer. This proposed software would revolutionize the technology and it will replace the existing technology by providing strong means of security which was lacked in the previous security system by implementation of the concepts like Computer Vision, Object recognition, video tracking and advanced image processing with object detection systems. The proposed system is highly user friendly and portable it can convert any normal pre- existing security cam or the web cam into Artificial intelligence camera which would add the quality of recognizing objects in them which would set a bench mark in security systems and in the fields of computer vision and artificial intelligence. The software has two different zones the red and the green zones. When the cameras are set to the green zone it behaves like a normal CCTV camera and records footage., When its set to the red mode the normal CCTV cameras turn into an artificial intelligence camera which would activate the artificial intelligent brain for the CCTV cameras to detect any intrusions. The proposed software also lets the user to set both the green and red zones simultaneously. INTRODUCTION Computer foresight isan associative handle which deals amidst how CPUs could be dupe promote extraordinary figuring out deriving out of abacus impressions or televisions. From the attitude of planning, it seeks to brutalize tasks that one the character beheld process can do. Computer for sight tasks consist of methods for acquiring, processing, analyzing and figuring out microcomputer drawings, and pedigree of high-dimensional input of your world of nature so as to cultivate scientific or typical science, e.g., within the styles of decisions. Understanding during this background mode the shift of imaged impressions (the goods of your retina) within descriptions of one's everyone that may impart amidst diverse excogitation and evoke correct case. This impression figuring out may be seen because the disentangling of emblematic info coming out of perception input the use of models constructed including the aid of algebra, physics, data, and research premise. Image processing is a method to convert an image into digital form and perform some operations on it, in order to get an enhanced image or to extract some useful information from it. The purpose of image processing is divided into 5 groups. They are:  Visualization - Observe the objects that are not visible.  Image sharpening and restoration - To create a better image.  Image retrieval - Seek for the image of interest.  Measurement of pattern – Measures various objects in an image. Object detection is a PC development related to PC vision and picture taking care of those courses of action with recognizing cases of semantic objects of a particular class, (for instance, individuals, structures, or automobiles) in mechanized pictures and accounts. 224

Machine research could be the subfield of robotics that offers clones the power to be informed externally soul deliberately edited. Evolved of the learn about of system credit and computational schooling understanding easy savvy, mechanical device schooling explores the find out about and development of finding which could contribute and perform predictions on goods– such conclusion overcome following strictly static prioritize instructions by making picture driven predictions or decisions, through building a model from sample inputs. Machine research is employed in a range of computing tasks where designing and prioritizing explicit break through his infeasible; example applications include spam filtering, detection of network intruders or malevolent cabal engaged against an info disregard, optical character acknowledgment (OCR), see’s and mac understanding. Computer Vision Software has enjoyed excellent R and D Expenditures and method of applications that fact entrepreneurs are conclusion for this one automation leave again fuelled which expel accelerate. There are breakneck advances and uses of that program in just about each dominant Industry:  Transportation  Communication  Banking  Education  Prisons  Courts  Construction  Disaster Relief  Space This software mainly aims for the betterment of the usability and even the focus on improvising the technology few of the main advantages are as follows:  The sole purpose of a motion detector and its biggest benefit is providing further security against burglars. A burglar cannot easily sabotage a wireless motion detector alarm by cutting the wires connecting the device to the alarms in wireless motion detector alarm has no wires that could be cut. Fig.1 Project Sequence 225

Fig.2 Algorithm Architecture IMPLEMENTATION AND RESULTS The security and the protection are the significant worry in any task created. The security which has been executed is:  Windows login: In PC security, signing in is the procedure by which an individual accesses a PC by distinguishing and confirming themselves. The client certifications are normally some type of \"username\" and a coordinating \"watchword\", and these accreditations themselves are now and then alluded to as a login. By and by, present day secure frameworks additionally frequently require a moment factor for additional security. At the point when get to is never again required, the client can log out. Signing in is typically used to enter a particular page, which trespassers can't see. Once the client is signed in, the login token might be utilized to track what moves the client has made while associated with the site. Logging out might be performed unequivocally by the client taking a few activities.  Bitlocker: Bit Locker (codenamed Cornerst one and once in the past known as Secure Startup )is a full plate encryption highlight included with select versions of Windows Vista and later. It is intended to secure information by giving encryption to whole volumes. Of course it utilizes the AES encryption calculation in figure piece fastening (CBC)orXTSmodewitha128-pieceor256-piece key, and furthermore the Elephant diffuser (on Windows Vista and Windows 7) for extra circle encryption security not gave by AES. CBC isn't utilized over the entire plate, just for every individual circle segment. Fig.3 Welcome Screen 226

Fig.4 Motion Detector 1 Fig.5 Motion Detector2 Fig. 6 Motion Detector3 227

Fig.7 Motion Detector 3 Optimized Fig.8 Motion Detector 4 CONCLUSION Machine learning approaches connected in methodical audits of complex research fields, for example, quality change may aid the title and theoretical consideration screening process. Machine learning approaches are quite compelling considering consistently expanding seek yields and openness of the current proof is a specific test of the examination field quality change. The favorable position we are having is, a picture is made of pixels. So much of the time we know the area of next point, it will be associated with our present pixel. Beginning with circles, take a picture, changeover it to dim scale, and recognize edges. The created programming is joined of the systems like picture handling channel usage, question recognition and fundamentals parts of machine realizing which all together will make an in-fact sound application which will likewise be a decent check in the better of the innovation. 228

Product Development Period: 2017-2018 Ref: 17-18-PD/CSE/002 Product Name: Predicting Air Pollution Trend from Historical Data collected from cities Development Lead: Ms. Sheeba P, Ms. Asha Rani Borah ABSTRACT Communities across the world are battling Climate change and Pollution. Leaders of the world host summits and sign pacts to mitigate the possibility of a dreadful and grave future caused by Climate change and Pollution. The question we need to ask is: Is this enough, or is it our responsibility as well, to do our bit and protect our environment, our world and our heritage? This Project is a small step in that direction. Here, we aim to tackle a serious type of Pollution: Air pollution. Air pollution is the introduction of particulates, biological molecules, or other harmful materials into the Earth’s atmosphere, causing diseases to humans, damage to other living organisms such as food crops, or damage to the natural or man-made environment. Particles like (O3), fine particulate matter (PM2.5), carbon monoxide (CO), nitrogen dioxide (NO2), sulphur dioxide (SO2) are hazardous and cause environmental concerns like acid rain and loss of biodiversity, and cause an enormous amount of harm and adverse health conditions to humans and other life forms as well. This project helps in identifying the most polluted places of the cities by employing Machine Learning. We can get real time information about the visibility index of the air and the general quality of the air. This will enable us to reduce the number of road accidents by using the government air quality index (AQI) datasets. Air quality has a tremendous impact on health as well, and this project will give real time information on statistics on that front as well. INTRODUCTION The main purpose of the project is to predict the air quality of the city that helps to find the concentration of the air pollutants. By the help of this project we will be able to find the current environmental situation related to the air pollution. We can improve the quality of the air by being aware of the situation of the air pollution. By knowing the particle matter (PM 2.5) we will be able to reducing the pollution creating substances in the urban areas. After studying the Meteorological condition of the air, we can try to improve the condition of the air. We select the meteorological data that will help to predict the affect air pollution including the wind speed, temperature, pressure etc. By the help of this project, we can ably find out the place where people should transplant the industries and check the urbanization based on that people can transplant the industry. This is possible through Machine Learning. The parameters used for analyzing the performance are prediction pattern accuracy and root mean squarer error (RMSE). A modular approach has been used to tackle the various problems: Greenhouse gas estimation Transportation is one of the main sources of emission of greenhouse gases. The two approaches are used to estimate the emission of greenhouse effect are bottom up and top down model and these models are proposed by IPCC. In top down approach the total CO2 level is calculated from total fuel consumption in the urban areas. And these carbons are transmitted into transportation sectors. Whereas in bottom up each economic level the fuel level is calculated then only the carbon emission of the fuel is calculated. 229

Matern function based extended fractional Kalman filtering This technique was used to predict the air pollution emission. Unpredictably, data involves the measurement error or the instrument precision related error. Here, the input data are only observed, and filtering techniques suffer from few obstacles. FCM-HMM Clustering and TS Fuzzy Inference Algorithm The multi model framing system is based on FCM-HMM clustering technique and the TS Fuzzy algorithm model is used to predict the air pollution index. Here, the data used were meteorological and climatic factors. But as we see HMM have different unstructured parameters and dependencies between the hidden state cannot be expressed Fig.1 System Flow Diagram Fig.2 System Architecture Fig.3 Process Design Architecture IMPLEMENTATION AND RESULTS Fetching Pollution Data: Fetching some components like SO2, NO2, PM, Ozone, Air Quality etc. Parameters used are SO2, NO2, Respirable Suspended Particulate Matter PM10 (Particulate Matter), Particulate Matter /PM 2.5. Each parameter will be used for analysis of air pollution in the city. Pre-processing The data we get from different sources may contain inconsistent data, missing values and repeated data. Also, redundant data must be removed or eliminated. Some datasets may have some outlier or extreme values which have to be removed to get good prediction accuracy. Classification and clustering algorithms and other data mining methods will work well only if all this pre-processing is done on the data. 230

Building the classification model First, we have to divide the data set into training and testing set. The predicting model is first trained with the training dataset. Later it will be tested with the testing set. Otherwise k-fold cross validation can also be used. After testing the model, the accuracy of the model is estimated by using different parameters. Existing data along with their air quality parameter will be provided in training set. The output of the training set will be air quality index. Data will be provided in time series in the form of SO2, NO2, RSPM, PM, air quality. Finally, once the model is competent, deployed and modelled to do the predictions and the pursuits and ambitions because of the inconsistency in historic knowledge on financial institution accountant for this reason participate in an analysis of the given dataset and describe how to repair it routinely. Recognition for future data points using Time Series Analysis ANN model is used for recognition for future data points. Based on the model, new values will be provided to get the future AIR quality index and you can predict next month next year wise air quality prediction using this MLP ANN model. The following Machine Learning Algorithms are implemented during the building of the product.  Linear Regression: A Supervised Learning Algorithm used in Machine Learning which perform a regression task by which a machine can predict a real number sets as outputs.  Extra tree classifier: It is a type of ensemble Learning technique which aggregates the results of multiple de- correlated decision trees collected in a “forest” to output its classification results.  Random forest classification: This is for ensemble methods used for tree classification and used *h (x, θk), k=1,2,3….+ where the θk value are not dependent and distributed in the random vector and the value of x is the input, h (x, θk) is finally generated classifier.  Lasso regression: This is a type of linear regression that uses shrinkage. Shrinkage is where data values are shrunk towards a central point, like the mean. The lasso procedure encourages simple and sparse  Elastic net regression combines the power of ridge and lasso regression into one algorithm. What this means is that with elastic net the algorithm can remove weak variables altogether as with lasso or to reduce them to close to zero as with ridge. All of these algorithms are examples of regularized regression. Models. Fig.4 Neural Network Regression Fig.5 Boosted Decision Tree 231

Fig.6 Time Series Fig.7 Decision Tree Predicted Fig.8 Lasso Prediction Fig.9 XGBoost Prediction Fig.10 Output Based on Month wise 232

Fig.11 Elastic Net Regression CONCLUSION With this model we can forecast the AQI and alert the respected region of the country also it a progressive learning model it is capable of tracing back to the particular location needed attention provided the time series data of every possible region needed attention. Air quality checking and investigation stage, and incorporates the normal every day fine particulate issue (PM2.5), inhalable particulate issue (PM10), ozone (O3), CO, SO2, NO2 fixation and air quality record(AQI). The analytical procedure from information cleaning and processing of incomplete records, detailed evaluation and in the end, model constructing, and evaluation are carried out. This application can help India meteorological division in predicting the way forward for air nice and its reputation and will depend on that they are able to take motion. Product Development Period: 2017-2018 Ref: 17-18-PD/CSE/003 Product Name: Reversible Data Hiding in Encrypted Images using Interpolation-based Distributed Space Reservation. Development Lead: Ms. Alpha Vijayan, Mr. Vijay Kumar ABSTRACT Reversible data hiding (RDH) in encrypted images has attained more attention recently in research community. Reversible data hiding is a type of data hiding techniques whereby the host image can be recovered exactly. Being lossless makes this technique suitable for medical and military applications. Privacy protection of additional data as well as cover media makes it attractive for applications in medical imaging, cloud storage, forensics etc. In this paper, a new method for reversible data hiding in encrypted images (RDH-EI), is proposed. Our method adopts the approach of reserving sufficient space for the additional data before encrypting the cover image. The proposed method is simple and intuitive. Experimentally results show that our method outperforms the state-of-the-art methods for reversible data hiding in encrypted images. 233

INTRODUCTION In this generation of secured data, stenography is a field gaining utmost popularity. The data should be hidden in almost all cases and data should be made private so that there is no intrusion in our privacy. This project aims at hiding the data in the images which is also called as stenography.The focus of this project is to hide the data in the images by the method of interpolation of the last bits so that the encrypted image is not so different from the original image but encrypted enough so as to protect the privacy of the sender to the receiver. First the image in which the data is hidden is encrypted by interpolating the last bits of the image and then the data is again encrypted in the image in which we use the implementation of the XOR gate and also use the AES encryption algorithm to encrypt the image and then add the data into the encrypted image which can be only decrypted if the receiver has the encryption key for both the data and the image. Data privacy helps in securing the privacy of the user and the communication protocols such that it aims in protecting the sensitive information which can be very vulnerable in many cases. It aims at reducing the human intervention for encryption as much as possible and replacing it with computer technology, for encrypting and sending the data. This in turn makes the information more secure and then also helps in developing the algorithms which are much faster, and also improves in the data privacy of more sensitive materials. Although the idea of stenography has been around earlier, actual cipher techniques for the image encryption are getting developed in recent days and will continue to evolve further. The system is capable of collecting the input from the user using the jsp code for image browsing in the system and then the image being encrypted has a feature to add the data inside the encrypted image so that the same image can be sent to verified users only. This makes sure that the data sent to the receiver is a verified user, and also makes sure that the data cannot be misused as the data is only sent to the verified user. We have embedded a SMTP protocol so that we can send the mail to the verified users, where we can attach the encrypted images with data. Here, we are using a reversible data hiding technique using image interpolation method to hide the data in the image and also decrypt the image and get the data back using the encryption key.In this project we have used eclipse for the coding part which involves servlets and java. The project is running under TOMCAT server and uses MySQL for data storing. Fig.1 System Architecture IMPLEMENTATION AND RESULTS There are modules for login for admin as well as the user. Login Details for the application 234

User ID and Password for Admin Login: User Id : admin Password : admin User ID and Password for Member Login: User Id : Password : *There is no user id and password for end users. The user has to first register into the system. Fig.2 Login pages for both admin and User Fig.3 Admin Home Page and User login page Fig.4 Show Profile page for displaying profiles. Fig.5 User List 235

Fig.6 User Home Page Fig.7 Encryption Process Fig.8 Send mail Fig.9 Decryption Process CONCLUSION In this Project, a new RDH-EI protocol for three parties is proposed. Main improvement in thie project is extending the traditional recovery to the progressive based recovery. The progressive recovery based RDH- EI provides a better prediction way for estimating the LSB-layers of the original image using three rounds, which outperforms state-of-the-art RDH-EI methods. Since RDH-EI is equivalent to a rate-distortion problem, capability of the method should be evaluated by both the distortion and the embedding rate. Moreover, this method is simple and easy to implement compared to its predecessors. This is because we are eliminating the need for restructuring the image unlike the other methods in literature. For most of the images the embedding capacity for additional data is improved beyond 10000 bits. At the same time, if we compare the PSNR of stego image by keeping the data embedded constant, there are notable improvements in PSNR for all the images over the stat of-the art methods. 236

Product Development Period: 2018-2019 Ref: 18-19-PD/CSE/001 Product Name: Detection of Fraud Behavior in Water Consumption Development Lead: Ms. Alpha Vijayan, Mr. Siddarth Konar ABSTRACT Fraudulent behavior in drinking water consumption is a significant problem facing water supplying companies and agencies. This behavior results in a massive loss of income and forms the highest percentage of non-technical loss. Finding efficient measurements for detecting fraudulent activities has been an active research area in recent years. Intelligent data mining techniques can help water supplying companies to detect these fraudulent activities to reduce such losses. This research explores the use of two classification techniques (SVM and KNN) to detect suspicious fraud water customers. The main motivation of this research is to assist Yarmouk Water Company (YWC) in Irbid city of Jordan to overcome its profit loss. The SVM based approach uses customer load profile attributes to expose abnormal behavior that is known to be correlated with non-technical loss activities. The data has been collected from the historical data of the company billing system. The accuracy of the generated model hit a rate of over 74% which is better than the current manual prediction procedures taken by the YWC. To deploy the model, a decision tool has been built using the generated model. The system will help the company to predict suspicious water customers to be inspected on site. INTRODUCTION Water is an essential element for the uses of households, industry, and agriculture. Jordan, as several other countries in the world, suffers from water scarcity, which poses a threat that would affect all sectors that depend on the availability of water for the sustainability of activities for their development and prosperity. The mentioned Irregularities are known as non-technical losses (NTLs). NTLs originating from electricity theft and other customer malfeasances are a problem in the electricity supply industry.NTL is a problem in water supply industry too because of the similarity between water and electricity distribution systems in depending on meter technology and load profiling concept. NTLs include the following activities 1) Losses due to faulty meters and equipment. 2) Tampering with meters so that meters record low rates of consumption. 3) Stealing by bypassing the meter or otherwise making illegal connections. 4) Arranging false readings by bribing meter readers. 5) Arranging billing irregularities with the help of internal employees by means of such subterfuges as making out lower bills, adjusting the decimal point position on the bills, or just ignoring unpaid bills. Fraud is a serious problem face information system that implemented in various domains. Credit card transactions as a financial system branch had a total loss of 800 million dollars of fraud in U.S.A. and 750 million dollars in U.K. in the year 2004 [1]. In the area of health care according to transparency international [2], the total expenditure exceeds the amount of 3 trillion euro worldwide. That size in the health care industry induces several actors in the field to make a profit by using illegal means, forbidden financial operation committing health care fraud. This water crisis situation has been aggravated by the rapid population growth and mismanagement. Efforts of the water suppliers to improve water and sanitation services are faced by managerial, technical and financial determinants and the limited amount of renewable freshwater resources. A. Real Application on Nontechnical Losses Detection 237

The main objective of data mining techniques is the evaluation of data sets to discover relationships in information. These relationships may identify anomalous patterns or patterns of frauds. Fraud detection is a very important problem in telecommunication, financial and utility companies. B. Artificial Neural Networks and Support Vector Machines for Water Demand Time Series Forecasting Water plays a pivotal role in many physical processes, and most importantly in sustaining human life, animal life and plant life. Water supply entities therefore have the responsibility to supply clean and safe water at the rate required by the consumer. Water plays a pivotal role in many physical processes, and most importantly in sustaining human life, animal life and plant life. Water supply entities therefore have the responsibility to supply clean and safe water at the rate required by the consumer. The modelling of water resource variables is a very active field of study and definitely there still is a lot of work to be done. In the initial stages, modelling of water resource variables was done using the traditional statistical models. The modelling of water resource variables is a very active field of study and definitely there still is a lot of work to be done. In the initial stages, modelling of water resource variables was done using the traditional statistical models. C. Machine Learning Algorithm for Efficient Power Theft Detection Using Smart Meter Data Electricity Theft is one of the major problems of electric utilities. The dishonest electric power users produce financial loss to the utility companies. Machine learning algorithm is used for this purpose the trustworthiness of customer is verified and is selected for theft program. This analysis is carried out by tweaking the actual smart meter data to create fraudulent data. D. An Approach To Detection Of Tampering In Water Meters Meter tampering is nothing but fraudulent manipulation which explains a service that is not billed by a utility company. It is a lack of consumption for the utility company and a main problem because they represent an important loss of income. The algorithms were generated and program after data mining process from the database of the company. They detect three types of consumption patterns. The Module descriptions of the methodology are as follows: A. Customer Data The customers those who are willing to get water through agencies are registered with system. The only ways for user to consume water by customers are through this registration. Customer request for admin to get water and to generate bills. B. Verify Feedback Bills are generated after checking the limit by on field executives after check the limit. The quantity that they consumed must be equal to noted details by admin. The fraud details can be check through this process. The bills were uploaded after this and find the fraudulent among the customers. C. Action Against Fraudulent The fraud customers who illegally consumes more water than they used or may be requires can be found by admin and bills also verified by them. Fraud details are set to block by the user and let them not provide any more water to them again and the details handover to cops to punish them with legally. D. Graph Analysis The graphs are handy to understand the data and based on this analysis admin can find the fraud customers. The business gradually improves as per their understand of where exactly problem arises and to find the place improve and lack. This will give the clear picture about the current and past picture from the dataset. 238

Fig.1 System Architecture The customers those who are willing to get water through agencies are registered with system. Upload the water consumption details through this registration. On field executives checking the limit according to branch wise and collect the feedback from the customers through this process. Admin can prepare the data to analyses the positive and negative feedbacks. Based on their feedback Fraud details are set to block by the user and let them not provide any more water to them again. Admin can find the fraud customers where exactly problem arises. This will give the clear picture about the current & past picture from the dataset. E. Algorithm In this work the experiments are performed on two important and well-known classification algorithms K- Nearest Neighbor (KNN) &SVM are applied to the water customer’s dataset which is taken from the Executives. There accuracy is obtained by evaluating the datasets. Each algorithm has been run over the training dataset and their performance in terms of accuracy is evaluated along with the prediction done in the testing dataset. It is one of the world’s most popular and most used open source data mining solutions. It has a comfortable user interface, where in a process view analyses are configured. It uses a modular concept, where respective operators are used in the analysis process. These operators have input and output ports through which the operators can communicate with the other operators in order to receive input data or pass the data and generated models over to the following operator. In this way, the entire analysis process creates a data flow. K-Nearest Neighbor makes predictions based on the outcome of the K neighbors closest to that point. Therefore, to make predictions with KNN, we need to define a metric for measuring the distance between the query point and cases from the examples sample. Fig.2 KNN Algorithm Fig.3 SVM Algorithm 239

IMPLEMENTATION AND RESULTS Fig.4 User Information Page Fig.5 User-Branch Information Page Fig.6 Positive Analysis Graph Fig.5 Positive Analysis Graph CONCLUSION In this Project, we have used the data mining classification techniques for the purpose of detecting customers with fraud behavior in water consumption. We used SVM and KNN classifiers to build classification models for detecting suspicious fraud customers. The models were built using the customers’ historical metered consumption data; the Cross Industry Standard Process for Data Mining (CRISP-DM). This phase took a considerable effort and time to pre-process and format the data to fit the SVM and KNN data mining classifiers. The conducted experiments showed that a good performance of Support Vector Machines (SVM) and K-Nearest Neighbours (KNN) had been achieved with overall accuracy around 70% for both. In Future accuracy of the same can be improved with the help of improved techniques. With the use of the proposed model, the water utilities can increase cost recovery by reducing administrative Non-Technical Losses (NTL’s) and increasing the productivity of inspection staff by onsite inspections of suspicious fraud customers. 240

Product Development Period: 2018-2019 Ref: 18-19-PD/CSE/002 Product Name: Handwriting recognition system using machine learning Development Lead: Ms. Sheba Pari, Mr. Keerthan Rao ABSTRACT Handwriting recognition has been an active and challenging area of research. Handwriting recognition system plays a very important role in today’s world. This is extremely popular and computationally expensive work. At present time it is very difficult to find correct meaning of handwritten documents. There are many areas where we need to recognize the words, alphabets and digit. There are many application postal addresses, bank cheque where we need to recognize handwriting. This project is concentrated in finding the handwritten content in the document and converting it into digital form. Handwriting recognition system can be used to solve many complex problems and can make human’s work easy. There are many applications where we need handwriting recognition system like bank cheque, postal addresses, and form documents. In all the techniques main stage is feature extraction. This project may be used in all the other fields and can be helpful in converting handwritten text into typed format. INTRODUCTION The main objective of this proposed project is to develop software that read the handwritten content, recognize it, converts it into the text and displays in typed format. This involves in gathering the handwritten data from the user in the form of image and pre-processing the gathered image by binarizing it , by removing all the stains in the image and then the system recognizes the image by using various machine learning techniques and converts it into text and displays in the typed format. Although we have large number of technologies for typing and writing people still use pen and paper to store the information which is gathered the form of handwritten format. This format is very difficult to store and access and it’s also difficult to share the information with others ,this project helps in converting the handwritten data into the text format so that it can be stored in any physical device or in the cloud in small amount of time and can be accessed whenever is needed and the information can be easily shared with others, it will also can be used in various fields such as verification of bank details, scanning the hardcopy of the notes and sending the softcopy of it to the students etc. So the aim is to develop software that read the handwritten content, recognize it, converts it into the text and displays in typed format. Fig.1 System Architecture 241

IMPLEMENTATION AND RESULTS This project uses the concept of neural networks. The input grey scale image will be magnified to a size of 128 *32, and each pixel in the image will be considered as one neuron and will be given with unique grey scale value, each pixel of these values will be fed into further layers of neural network and weights will be assigned to each neuron which then will be sent to the next layer. It consists of five convolution neural network layers and two CTC layer, Layers with 256 units spread data the grouping through and guide the arrangement to a framework of size 32x80. Every grid component speaks to a score for one of the 80 characters at one of the 32 time-steps. CTC layer mainly calculates the lost value of the matrix and it decodes the matrix to the final text with the help of beam. In the pre-processing step, the image is checked for the damage. If the image is damaged then the black image is used as default and then the target image is created with help of copying sample image on removing the noise present in the sample image the new image formed will be magnified to the new size of respective width and height and will be passed on to the next phase. Testing is further used to avoid errors. 1. Unit testing: Checks whether the program inputs produce legitimate yields. 2. Integration testing: Intended to test incorporated programming segments to decide whether they really keep running as one program. 3. Validation testing: Validation is a Quality confirmation procedure performed on designing models Fig.2 Handwriting Recognition CONCLUSION Handwriting recognition system’s aim is in converting handwritten data into digital text format with the help of various machine learning concepts such as deep learning, neural networks etc. and converts it into digitalized text format. Apart from converting the handwritten input to digital format, this project will also finds its applications in many fields such as financial and banking sectors to identify the amount written in check leaves. It can also be used to identify the postal address of the customer. This project can be used in converting huge volume of handwritten data into digital format in a small period of time efficiently. This can also be used to recognize the prescriptions given by the doctors if the handwriting is not recognizable. 242

Product Development Period: 2018-2019 Ref: 18-19-PD/CSE/003 Product Name: Skin Disease Classification Using Deep Learning Development Lead: Ms. Kamatchi Priya L, Mr. Adithya Nambiar ABSTRACT Skin diseases are very common in people’s daily life. Each year, millions of people are affected by all kinds of skin disorders. Diagnosis of skin diseases sometimes requires a high-level of expertise due to the variety of their visual aspects. As human judgments are often subjective and hardly reproducible, to achieve a more objective and reliable diagnosis, a computer aided diagnostic system should be considered. In this project, we investigate the feasibility of constructing a universal skin disease diagnosis system using Deep learning. The key part of architecture is a Convolution Neural Network that is trained on a skin disease image database. The dataset is obtained from skin disease database available openly HAM10000 dataset. Seven classes of diseases are predicted. It uses soft max layer of Deep learning for disease prediction. Our project can achieve as high as 90% accuracy. The accuracy can be further improved if more training images are used. INTRODUCTION Direct digital imaging for medical diagnosis has become a popular method with modern computation power and machine learning methods. Today, various deep learning models have been created and applied in the field of medical diagnosis due to their ability to recognize patterns in digital images. Deep learning are currently the best-performing technique for image classification. As such, deep learning have led to breakthroughs in many medical image analysis tasks such as classification and detection of illnesses. For example, Deep learning models have been used to classify malignant and benign lesions from dermoscopy images of human skin. Skin diseases are one of the most commonly seen infections among people. Due to the disfigurement and associated hardships, skin disorders cause lots of trouble to the sufferers. Speaking of skin cancer, the facts and figures become more serious. In United States, skin cancer is the most common form of cancer. According to a 2012 statistics study, over 5.4 million cases of nonmelanoma skin cancer, including basal cell carcinoma and squamous cell carcinoma, are treated among more than 3.3 million people in America. In each year, the number of new cases of skin cancer is more than the number of the new incidence of cancers of the breast, prostate, lung and colon in combined. Research also shows that in the course of a lifetime one-fifth of Americans will develop a skin cancer. Skin diseases intend to be prevalent due to climatic as well as the living situation of the vast majority of people. Skin disease doesn’t just affect the skin. It can have a huge impact on a person’s day-to-day life, crush self-confidence, restrict their movement, and lead to depression and even ruin relationships. So it is needed to take skin disease seriously. In the field of medical science there is a great demand for computer-aided tools to facilitate many tasks. Many things that were done manually using traditional equipment have been replaced with automated systems. Modern medical science is looking for solution which could assist the doctors with any aspect of work using the new technology. One of the common approaches used in these areas are digital Image processing and Data mining. Our proposed system enables user to recognize skin diseases and provide user advises or treatments in a shorter time period. We build our skin disease dataset from skin disease dataset HAM10000 dataset which is one of the largest photo dermatology source that is available publicly. At first, we Pretrain Deep Learning with VGG16/VGG19 models then extract features with Deep Learning for the whole train set each feature vector is of 4096. It Classifies each skin into one of n classes: Specific skin disease sets and trains fully connected layers with skin images, 7 classes of diseases are predicted. It uses SoftMax layer of Deep Learning for prediction. 243

Fig.1 System Architecture IMPLEMENTATION AND RESULTS DATASET Training of neural networks for automated diagnosis of pigmented skin lesions is hampered by the small size and lack of diversity of available datasets of dermatoscopic images. We use the HAM10000 (“Human Against Machine with 10000 training images”) dataset. We collected dermatoscopic images from different populations acquired and stored by different modalities. Given this diversity we had to apply different acquisition and cleaning methods and developed semi- automatic workflows utilizing specifically trained neural networks. The final dataset consists of 10015 dermatoscopic images which are released as a training set for academic machine learning purposes and are publicly available through the ISIC archive. This benchmark dataset can be used for machine learning and for comparisons with human experts. Cases include a representative collection of all important diagnostic categories in the realm of pigmented lesions. More than 50% of lesions have been confirmed by pathology, while the ground truth for the rest of the cases was either follow-up, expert consensus, or confirmation by in-vivo confocal microscopy. FEATURE EXTRACTION Feature extraction is one of the most important phases of image processing which requires extensive domain knowledge to help in classification phase. Deep learning, a specific type of deep learning algorithm, overcome the problem in traditional machine learning algorithms which required manual feature extraction before the classification process. Deep Learning not only perform classification, but they can also learn to extract features directly from raw images. Deep Learning provides the flexibility of extracting intrinsic and discriminating features from images which are the most suitable for classification. Deep Learning consist of four types of layers are convolution layers, pooling/subsampling layers, non- linear layers, and fully connected layers. The convolution layer of Deep Learning uses multiple learned filters to obtain multiple filter maps detecting low-level filters, and then the pooling layer combines them into higher-level features. The Deep Learning used in the proposed model by using a pertained model and adapt it for the proposed system. There are several pre-trained networks that have gained popularity. CONVOLUTIONAL LAYERS Plays the role of feature extractor but they are not hand designed. Through the training process the convolution filter kernel weights are decided on. The first convolution layer extracts low-level features like edges, lines, and corners. Higher-level convolution layers extract higher- level features. POOLING LAYERS Makes the feature representations smaller and more manageable. It makes the features robust against noise and distortion. Its function is to reduce to spatial size of the representation to reduce the amount of parameters and computation in the network. Pooling layer operates on each feature maps independently with the most common approach used in pooling is max pooling. 244

CLASSIFICATION USING RANDOM FOREST Once features are extracted, a classifier can be trained to classify a test sample as a member of one of the known classes. In this work the images have been classified using Random Forest. Random Forest is a flexible, easy to use machine learning algorithm that produces, even without hyper-parameter tuning, a great result most of the time. It is also one of the most used algorithms, because it’s simplicity and the fact that it can be used for both classification and regression tasks. Random Forest is a supervised learning algorithm. Like you can already see from it’s name, it creates a forest and makes it somehow random. The “forest” it builds, is an ensemble of Decision Trees, most of the time trained with the “bagging” method. The general idea of the bagging method is that a combination of learning models increases the overall result. To say it in simple words: Random forest builds multiple decision trees and merges them together to get a more accurate and stable prediction. One big advantage of random forest is, that it can be used for both classification and regression problems, which form the majority of current machine learning systems. Fig.2 Training and Validation Accuracy Training accuracy shows how the model learned out of the training data given testing accuracy shows how much the model is able to cope up with unknown data. Variation of them over epoch shows how its happening over each epoch and improving g the learning. Fig.3 Training and Validation Loss CONCLUSION Deep learning differs to other forms of Artificial Neural Network in that instead of focusing on the entirety of the problem domain, knowledge about the specific type of input is exploited. This in turn allows for much simpler network architecture to be set up. Future work may include increasing the dataset size and trying this technique on greater number of images. Different machine learning algorithm will be investigated in order to improve the accuracy. 245