Bolt loss defect detection method, system and device

Abstract

The invention provides a bolt loss defect detection method, system and device, and relates to the technical field of computer vision detection and fastener connection, and the method comprises the steps: marking a bolt and/or a bolt loss area through a fuzzy operation and a preset model, intercepting a mask image, extracting an angular point and a direction angle, and obtaining a bolt loss defect; and determining that the bolt is lost according to the angular point and the direction angle. And performing fuzzification operation on the training set of the preset model to perfect the training set. Through the method, the bolt loss condition of the bolt node can be directly and accurately subjected to machine vision judgment, and the problem of low bolt loss defect detection precision at present is solved; especially in the absence of a bolt loss defect image in a real scene, a bolt loss defect detection model with strong robustness can still be constructed, and the problems of weak robustness and poor practicability of a model for detection marking in the prior art are avoided.

Description

technical field [0001] The invention relates to the technical field of computer vision detection and fastener connection, and in particular to a method, system and device for detecting missing bolt defects. Background technique [0002] Bolted joints are one of the most common connection methods for steel structural members, and the use of hexagonal bolts is also the most common. In the process of installing and maintaining bolts, the defect of individual bolt loss due to negligence will threaten the safety and stability of the entire structure, so the detection of bolt loss is of great significance. [0003] At present, with the development of computer vision technology, especially the continuous deepening of research in the field of deep learning models, the accuracy of computer recognition has been greatly improved, and the application scope of computer vision detection technology has been greatly expanded, especially in the field of fastener connection technology. , The...

AI Technical Summary

Problems solved by technology

[0005] The document "Computer Vision-Based Detection for Delayed Fracture of Bolts in Steel Bridges" (English name: Computer Vision-Based Detection for Delayed Fracture of Bolts in Steel Bridges) records that the target detection model is used to detect the defect area after the delayed fracture of the bolt, and Use a variety of data enhancement methods to expand the limited defect data to improve the accuracy of the model. However, due to the fact that there are too few original missing defect images that can be found, the limited data cannot fully express many unknown defect states, thus affecting the model. Detection accuracy

Images