Glass fragment segmentation method based on edge enhancement and gradient-guided secondary segmentation

Abstract

The invention provides a glass fragment segmentation method based on edge enhancement and gradient guide secondary segmentation, which is used for fragment image segmentation in toughened glass test. The method comprises the following steps: step 1, adhering an adhesive tape to a single-sided area of toughened glass, and then crushing the toughened glass by using a knocker for testing to form a crushed glass aggregate fixed by the adhesive tape; step 2, placing the fragmented glass aggregate on a transparent platform capable of emitting light, and acquiring a scanning image of the fragmented glass aggregate by using a line scanning camera positioned above the platform; step 3, identifying the glass fragment edge in the scanning image by using an edge enhancement method and a gradient guide method, and obtaining the fragment segmentation state after the glass is crushed; according to the method, the image of the glass area is captured through the line scanning camera, and the problems that in practical application, edge discontinuity is caused by image noise and uneven illumination, image edges shake due to other factors, and a unified threshold value is difficult to segment can be solved.

Description

technical field [0001] The invention relates to the technical fields of image generation and image retrieval, in particular to a glass shard segmentation method based on edge enhancement and gradient-guided secondary segmentation. Background technique [0002] Since May 1, 2003, my country has implemented a mandatory inspection system for tempered glass, requiring inspection and testing of glass in buildings, cars and other scenes. Tempered glass segmentation technology is a necessary technology for performance testing of tempered glass. Through communication with the manufacturer, when producing tempered glass on the production line, at least break the first piece of glass, the middle glass, and the last piece of glass produced, and count the densest and sparsest areas of the glass. The factory has a limited budget for ordinary quality inspection equipment. Using deep learning algorithms to segment the entire image requires a huge amount of GPU memory, which will increase ...

AI Technical Summary

Problems solved by technology

[0006] The present invention proposes a glass shard segmentation method based on edge enhancement and gradient-guided secondary segmentation. The image of the glass area is captured by a line-scan camera, which can solve the problem of edge discontinuity caused by image noise, uneven illumination, and other factors in practical applications. The edge of the image shakes, and it is difficult to use a unified threshold to separate the problem

Images