Target for Flipbook PDF

The Definitive Guide to Digital Sorting for Food Processors

CopyrightCopyright 2016 Key Technology, Inc. All rights reserved.Information in this document is subject to change without notice. The software described inthis document is furnished under a license agreement or nondisclosure agreement. The soft-ware may be used or copied only in accordance with the terms of those agreements. No part ofthis publication may be reproduced, stored in a retrieval system, or transmitted in any form orby any means electronic or mechanical, including photocopying and recording for any purposeother than the purchaser's personal use without the written permission of Key Technology, Inc.Key Technology, Inc.150 Avery StreetWalla Walla, WA 99362www.key.netTrademark AcknowledgmentsUNITED STATES OF AMERICA TRADEMARKSIntel® is a registered trademark of Intel Corporation, Santa Clara, CaliforniaCore™ is a trademark of Intel Corporation, Santa Clara, CaliforniaVisual Basic® is a registered trademark of Microsoft Corporation, Redmond, WashingtonJava® is a registered trademark of Oracle Corporation, Redwood City, CaliforniaLinux® is a registered trademark of William R. Della Croce, Jr., Boston, Massachusetts

First edition published in 1988: \"The Food Processor's Guide to Optical DefectDetection\" by Bill Shaw and Neil Burck. Technical Contributors:Danielle WilsonTim JusticeBob RhodesRod FazzariRichard PhillipsDave FoxBrian HowtonJohn KadingerEditor:Sandra McBride

Table of Contents 1 1Digital Sorting Overview 2 Introduction 3 Value for the Processor Principles of Operation

Chapter 1 Digital Sorting OverviewD i gi tal S orti ng O vervi ewIntroductionBecause of the rising cost of labor, the dimin-ishing availability of laborers and more demand-ing consumers who expect high quality products,digital sorters have replaced manual sorting inmost food processing applications. From receiv-ing raw product to inspecting finished productand points in between, digital sorters are placeddirectly into production lines where they achievenon-destructive inspection at full production volumes. Capable of recognizingeach object’s color, size, shape, structural properties and chemical com-position, digital sorters detect and remove defective products and foreignmaterial (FM) from the main product stream. The technology can also be usedto separate product by different grades. Payback is realized through improvedproduct quality, increased yields and reduced operational costs. 1

2 The Definitive Guide to Digital Sorting for Food ProcessorsValue for the Processor Digital sorting systems mimic manual sorting in that cameras and/or laser scanners see defects and FM much like human eyes and intelligent software and algorithms make accept/reject decisions much like a human brain. However, digital sorting achieves many significant improve- ments compared to manual inspection.While laborers are subjective and inconsistent, digital sorters do not sufferfrom fatigue; they are objective and consistent, making thousands ofdecisions every second for an indefinite period of time. This consistencyenables food processors to use digital sorters to produce the exact same qual-ity of product day after day and at every location around the world, despite fluc-tuations in the quality of incoming product.Sophisticated sensor technology enables digital sorters to greatly exceed thecapabilities of humans. Color cameras can accurately distinguish millions ofdifferent colors in the visible light spectrum. Cameras and laser scanners canidentify infrared (IR) and ultraviolet (UV) light, which human eyes can’t see.Laser scanners can recognize surface structural properties, and hyperspectralsensing technology can differentiate chemical composition, which are bothbeyond human detection.Digital sorters with intelligent software and algorithms deliver additional valueby making new and more complex sorting decisions using hundreds of datapoints. Sort-to-Grade is one example. Like humans, sorters make accept/re-ject decisions by comparing each defect to predetermined criteria. Whilehumans make single, one-off decisions regardless of final in-the-bag qualityresults, smart sorters can make relative decisions and grade by count, accept-ing some lower grade product with minor defects in order to maximize yieldwhile controlling the quality of the output to a defined grade.

Digital Sorting Overview 3When product specifications change, simple adjust-ments on a digital sorter’s user interface (UI) instant-aneously change the sort parameters, from the most-subtle differences to entirely new categories ofproduct defects and FM. This operational flexibilityenables food processors to perfectly match a widevariety of final product specifications.With throughputs of more than 25 metric tons per hour on high-capacity sort-ers, this automated technology can do the work of dozens of manual laborerswhile achieving superior results. As such, digital sorters generate immediatecost reductions and protect a food processor from growing labor shortages.Principles of OperationIn its simplest form, a digital sorting system contains several basic elements(Figure 1).Figure 1: 1-infeed conveyor, 2-presentation conveyor, 3, 5-camera and lasersensors, 4-reject mechanism, 6-reject collection conveyor, 7-accept collection con-veyor. 

4 The Definitive Guide to Digital Sorting for Food ProcessorsThe user interface (Figure 2) enables the food processor to define exactly whatthe digital sorter will eject from the product stream. Criteria can include thecolor, size, shape, structural properties and chemical composition of products,defects and FM as well as the location of a defect on the product.Figure 2: User Interface Example The cameras and/or laser scanners, or eyes of the system, are designed tooperate within the wavelengths that achieve the greatest contrast betweengood and bad objects for a given application. The most basic sorters operateusing a single color band while the majority of sorters use multiple colorbands, and hyperspectral sensors collect data from hundreds of wavelengths.The controller, or brain of the system, combines powerful data processing hard-ware with intelligent software and algorithms to compare the objects that thesensors see to the user-defined thresholds to make accept/reject decisions innanoseconds. Data processing hardware and software also manage the userinterface and other machine functions.The sorter’s mechanical platform presents product to the inspection zone andejection system and helps separate defects and FM from good product.

Digital Sorting Overview 5Figure 3: Typical Configuration of a Chute-fed Sorter A total sorting system might include mechanical grading systems and vibratoryshakers that transport product to the digital sorter and convey ‘good’ and ‘bad’streams to the next process after sorting.Beyond the primary purpose of removing defects and FM or separatingproduct of different grades, digital sorters can be leveraged to collect, analyzeand share data across the food processor’s enterprise to better manage rawmaterials and optimize processes to maximize yield, ensure quality andenhance productivity.