AOI (Automated Optical Inspection)-based panel detecting judging method and device

Abstract

The invention discloses an AOI (Automated Optical Inspection)-based panel detecting judging method and an AOI-based panel detecting judging device. The method comprises the following processes: performing primary defect detection on a panel to determine different types of main defects; performing surface dust filtering on all detected defects; respectively performing disaggregated classification on various types of the main defects in sequence to form various sub-defects according to a defect classification standard; re-judging various sub-defects by adopting a re-judging mechanism, and comparing a re-judging result with a defect level to obtain disaggregated classification of defect levels. According to the AOI-based panel detecting judging method and the AOI-based panel detecting judgingdevice, after the defects are detected primarily, the defects are judged secondarily through the re-judging mechanism, so that accurate and detailed level judgement output can be performed on the defects of the panel; manpower cost is greatly reduced; detection rate is improved.

Description

technical field [0001] The invention belongs to the technical field of panel automatic defect detection, and in particular relates to an AOI-based panel detection and judgment method and device. Background technique [0002] With the development of the LCD panel inspection industry, manufacturers have higher and higher requirements for panel inspection, and have different degraded specifications for different types of defects. In order to correctly degrade the output of the panel, the AOI equipment needs to accurately classify the detected defects, and degrade the classified defects according to the customer's requirements. [0003] For example, for dark spot defects, a single dark spot (0.5-1.5) is judged as an OK level, two consecutive dark spots (1.5-2.5) is judged as an L level, three consecutive dark spots (2.5-3.5) is judged as a K level, and multiple dark spots (2.5-3.5) are judged as a K level. Dark (>3.5) is judged as C level. However, how to distinguish double...

AI Technical Summary

Problems solved by technology

[0007] 1. For large-size LCD panels, due to the large area, the LCD panel and the camera cannot be accurately parallel

Therefore, there are also differences in the focusing effect of different positions, resulting in defects of the same area showing different sizes on the camera.

[0008] 2. There is a problem of strength and weakness of the defect. The bright spot area of ​​the L0 screen may be small, but the luminous intensity is very large, and the halo is very serious, resulting in a large imaging area on the camera

At the same time, the bright spots of some L0 images may have a large area, but the luminous intensity is very small, resultin

Images