Welding visual detection method and device based on convolutional neural network

Abstract

A welding visual detection method based on a convolutional neural network includes the following steps that firstly, at the training stage, a training sample is input into the convolutional neural network, and the connection weight and offset value of the convolutional neural network are obtained; secondly, at the testing state, a welding picture is read in and preprocessed with the digital picture processing technology, and a region of interest is extracted and then subjected to picture size normalization processing to serve as input of the convolutional neural network. The invention further provides a welding visual detection device based on the convolutional neural network. The welding visual detection device based on the convolutional neural network comprises a crawling mechanism, a power transmission mechanism, visual detection equipment and a weld defect detection and analysis system. By means of the welding visual detection method and device based on the convolutional neural network, the automaton and intelligence level is improved, the detection precision is effectively improved, and the detection speed is effectively increased.

Description

technical field [0001] The invention belongs to the application of crawler, omnidirectional visual sensor, wireless communication, deep learning and computer vision technology in weld defect recognition and detection, and relates to a welding visual detection method and device. Background technique [0002] In the manufacturing industry, welding is the basis of many manufacturing methods, and its quality determines the reliability of the product. Welding quality inspection is an important means to ensure welding quality. At present, the link quality inspection methods mainly include: destructive inspection and non-destructive inspection. Compared with non-destructive testing, destructive testing has higher reliability, but the workpiece must be destroyed. As the most widely used non-destructive inspection method, visual inspection method can realize the quality analysis of weld surface topography, surface geometry and discontinuities or defects existing on the surface. [...

AI Technical Summary

Problems solved by technology

The main problem of this invention is that the judgment of welding effect is to compare the picture information of solder joints during and after welding with the standard formula data of solder joints during and after welding through the visual system, and the accuracy cannot be guaranteed. requirements

However, explicit feature extraction is not easy and is not always reliable in some applied problems

Images