Online monitoring method and system for glue climbing height of production line chip

Abstract

The invention discloses a production line chip glue climbing height on-line monitoring method. The method comprises the following steps: collecting glue climbing images in four directions at the same moment; carrying out image preprocessing on the images of the image data set, and carrying out manual annotation on the preprocessed rubber climbing images; inputting the preprocessed carved and climbed glue image as a training set into a generative adversarial network to obtain the generative adversarial network; inputting a to-be-detected image into the optimized generative adversarial network to obtain a predicted silica gel frame; extracting a silica gel boundary from a region in the silica gel region prediction frame obtained by prediction of the generative adversarial network through a Canny operator; calculating the distance between any two pixel points on the boundary of the silica gel region extracted by the Canny operator, and obtaining the glue climbing height of the chip; and comparing the obtained rubber climbing height with a rubber climbing height limit value summarized by experience when the chip is laminated, and outputting a signal to adjust the mechanical arm in time. Compared with the prior art, the method has the advantages of being more accurate, not prone to fatigue, labor-saving, loss-reducing and cost-reducing.

Description

technical field [0001] The invention relates to the field of chip production measurement, in particular to an on-line monitoring method and system for the height of a chip on a production line. Background technique [0002] The climbing height of the chip during lamination is one of the important evaluation indicators for the quality of the chip. According to the experience of front-line production workers, the glue climbing height of the chip can be determined by the maximum distance between the climbing glue outline and the edge of the chip base. Controlling the appropriate glue climbing height is of great significance to the quality of the chip. The traditional method of observing the height of the glue with the naked eye or ordinary optical equipment is time-consuming and labor-intensive, and is now out of date. Image processing based on artificial intelligence has the advantages of high precision and low latency, and can be applied to the real-time monitoring of the ch...

AI Technical Summary

Problems solved by technology

The traditional method of observing the climbing height with the naked eye or ordinary optical equipment is time-consuming and labor-intensive, which is outdated at present

Images