Defect detection method for irregular metal machining surface based on depth learning

Abstract

The invention relates to a defect detection method of an irregular shape metal working surface based on depth learning, By preprocessing the collected surface images of metal machining, image enhancement, YOLOv3-based network architecture, Deep learning network suitable for defect detection is built. The samples are labeled manually and the deep learning network is trained with labeled samples toobtain defect detection model. Finally, the defect detection model is used to detect the surface image of metal machining, and the defect detection results are obtained. Compared with the prior art, the network structure in the invention has good adaptability to the detection of small objects and small targets, and unifies the four basic steps of the candidate region generation, the feature extraction, the classification and the position refinement of the target detection into the same depth network framework, thereby improving the running speed and making the detection more accurate.

Description

technical field [0001] The invention relates to a method for detecting defects on a metal processing surface, in particular to a method for detecting defects on an irregularly shaped metal processing surface based on deep learning. Background technique [0002] In the field of metal workpiece production and processing, defect detection is an indispensable process; due to the complexity of the current production workpiece structure, defect detection is still carried out by human resources to a certain extent. The mainstream manual visual inspection method in the existing technology is not only inefficient, but also has a large subjective factor in the inspection standard, which affects the automation process of the entire production line to a certain extent. In addition, with the continuous increase of labor costs, the management of the company's human resources is also a test. In recent years, the automatic defect detection method based on machine vision has attracted the a...

AI Technical Summary

Problems solved by technology

For images, the metal processing surface has a variety of shapes and edges, and the detection effect is not ideal simply by using traditional image processing methods.

However, the robustness of template matching method for defect detection on regular-shaped processed surfaces is not strong, and the requirements for imaging environment are relatively high. Detected defects are not well classified

In addition, for metal processing workpieces, there are many detection stations and the surface is relatively complicated. The existing technology does not meet the requirements and has a high processing time, which is not suitable for the construction of a real-time detection system.

Images