A precise visual inspection system and method for concave-convex defect of flexible IC substrate

Abstract

The invention discloses a system and a method for carrying out precision visual detection on the concave-convex mark defects of a flexible integrated circuit (IC) substrate. The system comprises a laser ranging system, an intelligent concave-convex mark detection module and an intelligent concave-convex mark analysis module. The method comprises the following steps: (1) carrying out calibration on the laser ranging system; (2) moving a laser scanning substrate at high speed by using the laser ranging system, and finally enabling the laser scanning substrate to get back to a point distance data set within a measuring range including the substrate; (3) acquiring complete dense point cloud data of the IC substrate; (4) extracting candidate point cloud areas of the concave-convex mark defects on the IC substrate in a way of partition by using the intelligent concave-convex mark detection module; (5) sending all the point cloud areas to the intelligent concave-convex mark analysis module, and analyzing whether the areas belong to the concave-convex mark defects or not; (6) if so, carrying out statistics, comparing the statistical results with national standard requirements, and returning judging results. The system and the method can better avoid defect misinformation caused by rotation, translation, scaling, and the like, and have better recognition capability for the concave-convex mark defects and other defects.

Description

technical field [0001] The invention relates to a defect detection method of a flexible IC substrate, in particular to a precise visual detection system and method for concave-convex defect of a flexible IC substrate. The invention belongs to the technical field of defect detection. Background technique [0002] At present, the method for detecting defects of flexible IC substrates mainly relies on 2D images, especially 2D images obtained by observation under a microscope. The collected images are processed and analyzed to detect the specific location and type of defects on the substrate. However, relying solely on the analysis of 2D images, the distinction between different types of defects is insufficient, and the correct rate of classification and recognition results is insufficient. The detection and analysis method based on the 3D point cloud is a good solution to the above shortcomings: the 3D point cloud coordinate set established by the laser high-speed moving dist...

AI Technical Summary

Problems solved by technology

However, the analysis of 2D images alone makes the distinction between different types of defects insufficient, and the correct rate of classification and recognition results is insufficient.

Images