Application of Machine Vision Technology in Grading Detection Cases
This paper expounds a classification detection algorithm based on machine vision.Firstly,the collected transmitted light images of leather eggs are segmented to obtain the de-bottomed image;then,the features of the de-bottomed image are extracted and selected,and the features are downscaled by linear discriminant analysis,so as to obtain the key features that can characterize the different classes of leather eggs;finally,the downscaled features are inputted into a classification model,and the performance of the three machine vision algorithms,namely,logistic regression,support vector machine,and random forest,is observed,and it is found that the performance of the random forest algorithm is good.The performance of the three machine vision algorithms,logistic regression,support vector machine and random forest,was found to be the best,with an accuracy rate of 95.65％,which can realize the effective classification of leather eggs images.
NETL
NSTL
万方数据
