FICS golden finger defect detection system and detection method based on bp neural network

Abstract

The invention discloses a BP neural network-based FICS golden finger defect detection system and a detection method. The system includes an image acquisition subsystem, a defect detection module and a data analysis module. The method includes: 1) the image acquisition subsystem controls the microscope to position and take photos of the flexible substrate, and the target position is the area where the gold finger is located; 2) through the "input image protection mechanism", it is ensured that the input image of the defect detection method is a gold finger pattern; 3 ) Extract the gold finger area through morphological processing; 4) Analyze each gold finger separately to analyze whether there is a defective area in it; 5) Calculate nine types of feature values ​​for the pictures that are determined to contain defects; The category corresponding to the item is returned as the final determined cheating defect category. The invention and the system can better avoid false alarms of defects in gold finger imaging detection under different light intensities, and have a higher accuracy rate for simultaneous detection of multiple types of gold finger defects.

Description

technical field [0001] The invention belongs to the technical field of defect detection, and relates to a BP neural network-based FICS golden finger defect detection system and method. Background technique [0002] The research on FICS without film is mainly focused on two types, one is the finished flexible substrate that does not require film, also called bare board, and the other is the part that does not need to be filmed on the finished flexible substrate with film, that is, for soldering components. Goldfinger. The gold finger is the key part with the highest circuit density, which plays a decisive role in the connection reliability between FICS and IC and peripheral devices. With the increase of FICS usage rate and the improvement of quality requirements, people pay more and more attention to the appearance quality inspection of FICS. At present, there are very few detection systems or methods for defects such as indentation, scratches, and exposed copper on gold fi...

AI Technical Summary

Problems solved by technology

At present, there are very few detection systems or methods for defects such as indentation, scratches, and exposed copper on gold fingers. Even the detection systems for other types of defects are usually based on single-threaded multi-tasking processing, and the work efficiency is not high.

Images