No abstract

With the upgrading of mobile phone equipment, automatic detection of mobile phone film defects has been paid more and more attention in industrial production quality. Mobile phone film defect detection is a huge workload and challenging problem. Traditional methods can also detect some industrial identification defects, but these methods can only detect defects under specific conditions, such as obvious defect outline, strong contrast, low noise conditions. The defect detection method of mobile phone film proposed in this paper is to locate the target area with input images obtained from the industrial environment, remove the background, and then classify them into their designated classes through convolutional neural network. Experimental results show that this method can meet the robustness and accuracy of mobile phone film defect detection.