Apple Appearance Grading Detection Based on Machine Vision and Improved Support Vector Machine
In order to solve the problems of low efficiency and low accuracy in traditional apple appearance quality classifica-tion,an apple appearance classification method based on machine vision and improved support vector machine is proposed.The ap-ple color classification image database is established by video image method.The apple is separated from the background by mathe-matical morphology method,and the apple color,size,defect,texture and other appearance feature vectors are extracted as the in-put feature vectors of support vector machine(SVM).Aiming at the problems of noise influence and data imbalance,an improved adaptive penalty support vector machine classifier(DCSVM)is used to detect apple appearance quality classification.The results show that the classifier has a good classification effect(accuracy of 93.56%),which is better than the traditional support vector ma-chine classification model(FSVM),and verifies the effectiveness of the classification model applied to apple appearance quality classification.
国家科技期刊平台
NSTL
万方数据
