Apparatus for optimizing wide-range image collection based on machine vision and detection method thereof

Abstract

The invention relates to an apparatus for optimizing wide-range image collection based on machine vision. The apparatus comprises an image acquisition unit, an image preprocessing unit, an image processing unit and an image output unit. The image acquisition unit includes a light source generation module and an image acquisition module. During the defect detection process, the image acquisition module receives reflected light from a visual light source and performs photoelectric conversion to output original image data; and the mage preprocessing unit receives an original image sent pixel grayvalue sent by the image acquisition module, carries out equalization on the original image sent pixel gray value, and then inputs the processed original image sent pixel gray value into the image processing unit. Therefore, on the basis of the image equalization and light source intensity correction methods, the image gray error after processing is reduced by several magnitude orders by being compared with the original image, so that the overall image uniformity is improved effectively. Moreover, the apparatus being suitable for the image with the poor light source uniformity is not limited to the column pixels and can be extended to make a light source intensity map being associated with the image gray, so that the flaw detection accuracy is improved.

Description

technical field [0001] The invention belongs to the field of integrated circuit manufacturing, relates to machine vision technology, and more specifically relates to an optimization device for collecting large-scale images by machine vision and a detection method thereof. Background technique [0002] At present, the design of semiconductor devices is rapidly developing in the direction of high density and high integration, which puts forward higher and higher requirements for new technology, new technology and new equipment of semiconductor integrated circuits. The popularity of machine vision applications is mainly reflected in the semiconductor and electronics industries, of which about 40%-50% are concentrated in the semiconductor industry. In particular, it plays an important role in the manufacturing process of silicon wafers as a flaw detection in the integrated circuit production line process. [0003] Defect detection is an extremely important step in the quality i...

AI Technical Summary

Problems solved by technology

like figure 2 As shown, the gray scale error range of the collected original image 4 is 80 gray levels, and its gray scale changes in an arc shape, and the error range is 80 gray levels. If image processing is performed on this basis, it is easy to generate local fixed errors

Images