Workpiece surface quality detection method and device and appearance machine

Abstract

The invention provides a workpiece surface quality detection method and device and an appearance machine. The detection method comprises the steps of dividing a to-be-detected surface of a workpiece into a plurality of to-be-detected areas, wherein each to-be-detected area corresponds to a preset detection defect threshold value; respectively acquiring a first detection image of each to-be-detected area; and determining the quality of the to-be-detected surface according to the first detection image of each to-be-detected area and the corresponding detection defect threshold. According to thedetection method disclosed by the invention, when the color difference of each position of the to-be-detected surface of the workpiece is relatively large due to the influence of external factors (such as cutting marks), the false alarm rate of the machine can be effectively reduced by adopting a regional surface quality detection mode, the product yield is improved, and the productivity of the machine is improved.

Description

technical field [0001] The invention belongs to the technical field of semiconductor equipment, and in particular relates to a method for detecting the surface quality of a workpiece, a device for detecting the surface quality of a workpiece, and an appearance machine. Background technique [0002] When the appearance machine detects the product, it mainly uses the difference in gray scale value to analyze the image collected by the machine. There is no color change in this image, only the level of grayscale is different, the grayscale value is between 0 and 255, 0 is the darkest black, 255 is the brightest white, we put a certain grayscale value in the whole image The proportion is set as the detection threshold, so as to detect different appearance defects on the product. [0003] Generally, packaged products are relatively thin, and can be formed with one knife when cutting. There is basically no color difference on the side. When testing under the same group of lights, ...

AI Technical Summary

Problems solved by technology

[0004] However, for Flip-Chip Quad Flat No lead (FCQFN) packaging products, the thickness is relatively large, and the cutting is divided into two knives. The side color difference between the first cut and the second cut is obvious. Under the same group of lights, the gray scale value is quite different. If the original method is adopted at this time, it will lead to more false alarms on the machine, affect the product yield, and reduce the machine. Taiwan efficiency

Images