Robot Abrasive Automatic Switching Method Based on Machine Vision
Aiming at the problem of low switching accuracy of existing methods,a robot abrasive automatic switching method based on machine vision is proposed.First,use a CMOS camera to capture abrasive images and preprocess the images.Then,the grey correlation degree is used to reflect the characteristic information of abrasives,calculate the correlation degree of feature parameter points,and establish an abrasive recognition optimization model to extract abrasive features by calculating the relative membership degree of similarity between the identified and standard abrasives.Finally,the BP neural network is used to train the model,input new abrasive images or data into the model,and automatically output classification results to control the robot to complete the automatic switching of abrasives.The results showed that there was no abnormal recognition phenomenon among different types using the proposed method,and the switching accuracy of all 10 groups was 100%,which met the expectations.
NETL
NSTL
万方数据
