Pipeline weld defect detection method and related device

Abstract

The invention relates to the field of image processing, and provides a pipeline weld defect detection method and a related device, and the method comprises the steps: obtaining an auxiliary picture set and a detection picture set; inputting the auxiliary picture set and the detection picture set into a feature extraction network to obtain a first feature map corresponding to the auxiliary picture set and a second feature map corresponding to the detection picture set; processing the first feature map and the second feature map by using a regional suggestion network to obtain a defect part predicted in the detection picture set; determining the similarity between the defect part marked in the auxiliary picture set and the defect part predicted in the detection picture set; and outputting a detection result according to the similarity. According to the technical scheme of the embodiment of the invention, the pipeline weld defect detection efficiency and detection effect can be improved.

Description

technical field [0001] The present application relates to the field of image processing, and in particular to a method for detecting defects of pipeline weld seams and related devices. Background technique [0002] The radiograph of the pipeline weld is an important document for the quality inspection of the pipeline weld. The radiograph of the pipeline weld is obtained through the ray detection method. Through the collection of the effective data of the radiograph, the operation status of the pipeline can be known in time, and leakage and fracture can be avoided. ACCIDENT. [0003] At present, when using the ray detection method to detect pipeline weld defects, the usual method is: use the conventional target detection method based on deep convolutional network for the ray pictures of pipeline welds to detect defects. Due to the use of conventional target detection methods based on deep convolutional networks, a large amount of sample data is required in the stage of train...

AI Technical Summary

Problems solved by technology

Due to the use of conventional target detection methods based on deep convolutional networks, a large amount of sample data is required in the stage of training the model, and the training process is cumbersome.

Moreover, for a specific pipeline type, there are few samples available for training the model, so the trained model has a poor detection effect for a specific pipeline type, and the model generalization performance is poor

Images