Positive and Negative Recognition of Watch Shell Blank Using Backpropagation Algorithm Based on Matlab

In this paper, a machine vision recognition method for identifying the two sides of a watch shell used for mechanical arm loading and unloading is present, aiming at many problems such as the production efficiency, failure rate and labor injury caused by manual loading and unloading in the process of watch shell processing. Because the traditional method of image feature extraction is difficult to recognize two sides, the backpropagation algorithm of neural network is used to identify the positive and negative of the watch shell. The algorithm is realized by programming in Matlab, used on the collected picture of the watch shell, and good experimental results have been obtained.