A strip steel surface defect detection method based on a multi-task scheduling mechanism

Abstract

A strip steel surface defect detection method based on a multi-task scheduling mechanism is disclosed. The method includes creating a token lock, and initializing a token value L as 1; creating a taskA to respond to an event that image photographing is completed by a camera; creating a defect buffer queue F used for storing historic images; creating an image processing task B(N) which performs thresholding operation on an image N to generate a defect labeled graph; checking the task B(N) with the token lock, if the token value L of the token lock is equal to the N, continuing the task B(N) and changing the token value L of the token lock to be N+1, otherwise, suspending the task B(N); and determining a subsequent processing process of the task B(N) according to whether the defect labeledgraph contains a defect pixel or not, the position of the defect pixel and the state of the defect buffer queue F. The multi-task scheduling mechanism is adopted and parallel computing and task synchronization mechanisms are adopted in a defect detection process, thus ensuring defect detection accuracy and speedability.

Description

technical field [0001] The invention belongs to the field of automatic detection system of steel strip surface quality in the metallurgical industry, especially the field of high-efficiency detection of surface defects, and specifically relates to a method for detecting surface defects of strip steel based on a multi-task scheduling mechanism. Background technique [0002] Strip surface defects are an important factor affecting the surface quality of cold-rolled strip, directly affecting the appearance and performance of the final product. The surface inspection system uses camera sensors to scan the surface of the strip to obtain two-dimensional images of the surface of the strip, and uses machine vision technology to detect and classify surface defects. With the improvement of cold rolling technology, the running speed of strip is getting faster and faster. Faster and faster, this also puts forward higher requirements for the rapid response and real-time processing capabil...

AI Technical Summary

Problems solved by technology

[0007] The technical problem to be solved by the present invention is to provide a method based on multi-task scheduling, aiming at the inadequacy of existing strip steel surface defect detection that cannot detect and report defects in time when the strip speed is fast and surface defects appear intensively. The strip steel surface defect detection method based on the mechanism adopts a multi-task scheduling mechanism, uses parallel computing and task synchronization mechanism in the defect detection process to ensure the accuracy and speed of defect detection, and solves the problem when the strip speed is fast or the defects appear densely. , the accuracy and real-time problems of defect detection

Images