Printed circuit board(PCB)in the manufacturing process will inevitably produce a variety of defects.In order to improve production efficiency and product quality,detection and classification should be done for the common defects in PCB manufacturing.The main purpose of this paper is to study how to construct a deep learning model and adopt image processing technology to carry out comprehensive and efficient defect detection on PCB images.Using a large amount of training data,the model can learn the characteristics of various defects,including but not limited to short circuit,open circuit,poor welding,etc.The classification of PCB defects is studied,and its methods are illustrated with examples,derivation and demonstration.On the basis of the ingenious construction of deep learning models and the optimization application of classification algorithms complementing each other,it provides feasible solutions are provided for improving production efficiency and product quality,promoting the development of PCB manufacturing industry in the direction of intelligence.
NETL
NSTL
万方数据
