Prepreg, also known as PP film, is the core material in the production of multilayer boards. It is mainly composed of resin and reinforcing materials, of which reinforcing materials are divided into various types such as glass fiber cloth, paper base, composite materials, etc. When making multi-layer printed boards, we usually choose prepregs with glass fiber cloth as the reinforcing material, that is, adhesive sheets.
As the demand for prepregs is increasing, the production process of prepregs is constantly improving and improving in order to meet the market demand. The traditional prepreg production process uses the hot pressing method, which requires a lot of heat and pressure, which not only consumes energy, but also has low production efficiency. In order to solve this problem, some enterprises have begun to adopt new production processes, such as cold pressing method, which is a new type of prepreg production process, which uses low temperature and low pressure for production.
With the continuous improvement of multilayer board manufacturing technology, the quality and performance requirements of prepregs are also getting higher and higher. In order to meet the market demand, improve the production efficiency of prepregs and reduce costs, more and more enterprises have begun to use machine vision technology to detect surface defects of prepregs.
Machine vision technology is based on image processing and computer vision technology, by obtaining a surface image of a prepreg, analyzing and processing the image, so as to detect whether there are defects on the surface. Compared with traditional inspection methods, machine vision technology has higher inspection efficiency and accuracy, and can greatly reduce inspection costs.
Convolutional neural network (CNN) is a very effective machine vision algorithm for surface defect detection in pregs. CNN is a deep learning algorithm that extracts image features by learning a large amount of image data, so as to achieve image classification, recognition, and detection. CNNs can quickly process large amounts of image data, accurately detect surface defects, and can automatically identify different types of defects. We believe that with the continuous development of machine vision technology, the surface defect detection of prepreg machine vision based on convolutional neural network will be more widely used in prepreg production.