Defect Detection of Electronic Connector Plastic Shell Based on Deep Learning

For the traditional manual inspection of the quality of plastic case of electronic connector, there are problems such as low detection efficiency, low accuracy, easy visual fatigue, excessive dependence on personal subjective feelings, and easy to cause false detection. In order to achieve the defect detection accuracy, this paper uses CMOS area array scanning camera to construct the defect detection data set of electronic connector plastic shell, and proposes a deep convolution neural network detection algorithm to detect the defect of electronic connector plastic shell. Comprehensively compare the detection accuracy under different convolution neural network structures, and optimize the network parameters. Finally, through experimental verification, the optimized network model can achieve 100% accuracy and good reliability in defect detection.