Method for creating digital radiographic inspection model for weld defects

Abstract

The invention provides a method for creating a digital radiographic inspection model for weld defects. A weld joint in an image is extracted by using a digital image processing technology; the numberof samples of the weld joint image is expanded by applying an image enhancement technology; the built deep learning model and the designed loss function are used for training the model, defect rules can be learned from existing defect images, all to-be-detected images can be processed in batches more flexibly by using the deep neural network, the method is more flexible, the performance is more stable, and the high recall rate and certain accuracy of the model are guaranteed; and the model is packaged and stored, so that the model is convenient to call and use. Manual workload is reduced, andworking efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of image detection model creation methods, in particular to a digital ray detection model creation method for weld defects. Background technique [0002] In recent years, artificial intelligence has brought new developments to the field of computer vision, especially in image classification and object detection. Commonly used image classification models such as Alexnet, VGG, Resnet, etc. can handle extremely complex image data and are often used for classification of public datasets. The performance and accuracy of these models have been continuously improved. However, the data in actual production is not as easy to distinguish as common public data sets. For example, in the field of weld inspection, there is a problem of difficult film evaluation. There are certain defects in the welded joints, such as cracks, pores, slag inclusions, lack of penetration, lack of fusion, etc. The existence of these defects ...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiments of the present invention is to provide a method for creating a digital ray detection model of weld defects, so as to solve the problem that the neural network model used in the public data set in the prior art is not effective for such images with inconspicuous features, and Problems that cannot meet the requirements of defect detection for high recall rate

Images