Circuit board defect detection system based on deep learning and detection method thereof

Abstract

A circuit board defect detection system based on deep learning comprises an image acquisition unit, the communication output end of the image acquisition unit is connected with the communication receiving end of an image processing unit, and the communication output end of the image processing unit is connected with the communication receiving ends of a motion control unit and a display module, the motion control unit controls the circuit board to transmit and detect, and the display module displays the received circuit board defect information and the defect position contour; the system obtains a target detection model by making a circuit board defect data set and carrying out Faster R-CNN training on defect data, then collects a circuit board image, carries out feature extraction after preprocessing the circuit board image, carries out comparison identification according to the extracted features and the target detection model, obtains a defective circuit board image and position contour information, and displays the defective circuit board image and the position contour information on a display screen. The method adapts to detection requirements of different circuit boards, and is high in recognition rate, practical, efficient and low in cost.

Description

technical field [0001] The invention relates to a circuit board defect detection system, in particular to a circuit board defect detection system and a detection method based on deep learning. Background technique [0002] At present, the country is vigorously developing integrated circuits, artificial intelligence and other fields. In the future, the printed circuit board industry will definitely develop rapidly, and circuit board defect detection is the top priority. The quality of the circuit board directly affects the performance of the product. Circuit board defect detection can enable enterprises to avoid huge loss of life and property. However, the existing detection methods are mostly carried out manually, which is prone to missed detection and false detection. Research on an efficient and fast circuit board defect detection system is an urgent need of the society. Contents of the invention [0003] In order to overcome the deficiencies of the above-mentioned prior...

AI Technical Summary

Problems solved by technology

[0002] At present, the country is vigorously developing integrated circuits, artificial intelligence and other fields. In the future, the printed circuit board industry will definitely develop rapidly, and circuit board defect detection is the top priority. The quality of the circuit board directly affects the performance of the product. Circuit board defect detection can enable enterprises to avoid huge loss of life and property. However, the existing detection methods are mostly carried out manually, which is prone to missed detection and false detection. It is an urgent need for the society to study an efficient and fast circuit board defect detection system.

Images