Design,development and evaluation of an online grading system for peeled pistachios equipped with machine vision technology and support vector machine

作     者：Hosein Nouri-Ahmadabadi Mahmoud Omid Seyed Saeid Mohtasebi Mahmoud Soltani Firouz 

作者机构：Department of Agricultural Machinery EngineeringFaculty of Agricultural Engineering and TechnologyUniversity of TehranKarajIran 

出 版 物：《Information Processing in Agriculture》 (农业信息处理（英文）)

年 卷 期：2017年第4卷第4期

页      面：333-341页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：support provided by the Research Department of University of Tehran Iran under Grant No.1305051.6.34 is duly acknowledged 

主　　题：Pistachio kernel Sorting Machine vision Sensitivity analysis Support vector machine 

摘      要：In this study,an intelligent system based on combined machine vision(MV)and Support Vector Machine(SVM)was developed for sorting of peeled pistachio kernels and *** system was composed of conveyor belt,lighting box,camera,processing unit and sorting unit.A color CCD camera was used to capture *** images were digitalized by a capture card and transferred to a personal computer for further ***,images were converted from RGB color space to HSV color *** segmentation of the acquired images,H-component in the HSV color space and Otsu thresholding method were applied.A feature vector containing 30 color features was extracted from the captured images.A feature selection method based on sensitivity analysis was carried out to select superior *** selected features were presented to SVM *** SVM models having a different kernel function were developed and *** SVM model having cubic polynomial kernel function and 38 support vectors achieved the best accuracy(99.17%)and then was selected to use in online decision-making unit of the *** launching the online system,it was found that limiting factors of the system capacity were related to the hardware parts of the system(conveyor belt and pneumatic valves used in the sorting unit).The limiting factors led to a distance of 8 mm between the *** overall accuracy and capacity of the sorter were obtained 94.33% and 22.74 kg/h,respectively.