A method for identifying fracture information based on monitoring video of material stretching process

Abstract

A method for identifying fracture information based on monitoring video during a material stretching process belongs to the technical field of material fracture analysis and image recognition. The present invention first uses SIFT feature points for time matching, calculates the number of feature points that can be matched by adjacent frame images, and According to the number of matching feature points, the time point of fracture occurrence is determined, and then the SIFT feature points are used for spatial matching, and the area with the least number of matching feature points is counted, so as to determine the area where the fracture occurs. Afterwards, a mixed Gaussian model is used to model the small area taken out from the fracture area, and the internal stress situation and overall change trend are analyzed. Compared with the traditional digital image correlation method, the technical effect obtained by the present invention can obtain more detailed internal changes of the material through the analysis of the parameters of the established model under the condition of ensuring accuracy. development prospects.

Description

technical field [0001] The invention belongs to the technical field of material fracture analysis and image recognition, and particularly relates to a method for recognizing fracture information according to a monitoring video of a material stretching process. Background technique [0002] Digital Image Correlation (DIC) technology, also known as digital speckle correlation method, was first proposed by Yamaguchi and Peters in the 1980s. Deformation information for the region of interest. The basic principle of this method is to mesh the region of interest in the image before deformation, treat each sub-region as a rigid motion, and then for each sub-region, through a certain search method, according to the predefined correlation function Carry out correlation calculation, in the deformed image, find the area with the maximum mutual pivot coefficient with the sub-area, that is, the position of the sub-area after deformation, and then obtain the displacement of the sub-area....

AI Technical Summary

Problems solved by technology

[0003] At present, digital image correlation technology is mainly used for fracture identification of various materials, but there are still some deficiencies in its basic principles: 1) It cannot effectively estimate the shape function error; 2) For complex deformations, traditional algorithms cannot take into account both speed and 3) The deep mechanism of the interpolation deviation is unknown, and there is a lack of experimental measurement methods; 4) The lack of a complete speckle quality evaluation standard; 5) The lack of a theoretical model for speckle optimization; 6) The traditional two-dimensional DIC compensation method has a negative effect on image noise It is very sensitive, and these shortcomings make the digital image correlation technology unable to adapt to various situations. The conclusions obtained cannot always guarantee its accuracy, and are greatly affected by external factors. The stress situation and other internal information inside the material, and only the macroscopic image of the material can be processed, these limitations make a new method urgently needed to solve these current problems

Images