Metal drop weight tear DWTT fracture image automatic evaluation method

Abstract

The invention provides a metal drop weight tear DWTT fracture image automatic evaluation method, and can solve the problems in the conventional method that a series of tests are required for testing ductile-brittle transition temperature, the process cost is high and automatic image discrimination cannot be realized. Accurate segmentation of the boundary of a ductile area and a brittle area of a fracture image and accurate type identification are realized by using an image segmentation technology based on a minimum spanning tree and a machine learning technology based on a support vector machine SVM; accurate boundary segmentation is performed by using the image segmentation technology based on the minimum spanning tree so that subareas are formed through segmentation and the type of the subareas cannot be distinguished; the feature data of the subareas are extracted by using a gray scale co-occurrence matrix, Fourier analysis and laws texture energy; and the type of the subareas is identified by using sampling through the feature data and the machine learning technology based on the support vector machine SVM. The new input subareas of the unknown type can be rapidly, effectively and accurately classified into the ductile area or the brittle area; and operation and use are easy, the effect is substantial and practicality is high.

Description

[0001] design field [0002] The invention belongs to the technical field of metal fracture detection and analysis, and relates to an image automatic assessment method of metal fracture fracture, in particular to an automatic assessment method for metal fracture DWTT fracture image. Background technique [0003] The detection of high-strength and high-toughness indicators is the bottleneck of my country's high-end steel production and development of the international market; steel toughness analysis is very important. Historically, there have been countless accidents and disasters caused by ignoring toughness properties. High toughness means that cracks are not easy to expand or crack It needs to absorb more energy, so around the performance testing of metal materials, a variety of material testing standards, instruments and products have been formed, forming a market with a scale of tens of billions, among which high-strength and high-toughness testing is the current hot spot; ...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the problem of testing the ductile-brittle transition temperature in the traditional method that requires a series of tests, the process consumes a lot of money, and cannot realize automatic image discrimination, the present invention proposes an automatic evaluation method for fracture images, and its specific technical scheme is as follows:

Images