Weld defect multi-scale feature extraction module based on lightweight cavity convolution

Abstract

The invention discloses a weld defect multi-scale feature extraction module and method based on lightweight cavity convolution, and belongs to the related technical field of weld defect detection. The module comprises a plurality of branches and a first superposition layer which are connected in parallel, each branch comprises the lightweight cavity convolution and a space pooling pyramid which are connected, the lightweight cavity convolution is used for extracting weld defect features of different receptive fields from the weld defect feature map to generate a first feature map, and the spatial pooling pyramid is used for enhancing the expression ability of the first feature map for the weld defect features of different scales through pooling layers of different sizes and then generating a second feature map; the expansion rates of the lightweight cavity convolution of different branches are different; and the first superposition layer is used for superposing the second feature maps output by all the branches to generate a final feature map. According to the method, lightweight improvement is carried out on the model, the robustness of the neural network model for large size change of the weld defect is enhanced, and the neural network model can be conveniently and quickly embedded into various models.

Description

technical field [0001] The invention belongs to the technical field related to weld defect detection, and more specifically relates to a multi-scale feature extraction module and method for weld defects based on lightweight void convolution. Background technique [0002] There are many types of weld defects, and the shape and size are also very different. It is difficult to learn feature information that can deal with different defect types at the same time when performing feature extraction in the convolutional neural network model. In the case of large differences in target size, the information extracted by the neural network model needs to have different receptive fields and multi-scale feature sources. In addition, in industrial applications, the complexity of the model needs to be considered, and the model should be improved as much as possible to improve production efficiency. [0003] The patent (Application No. 202011206236.2) discloses a method for detecting small...

AI Technical Summary

Problems solved by technology

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a multi-scale feature extraction module and method for weld defects based on lightweight hollow convolution, thereby solving how to make the weld defect information extracted by the module have different Receptive fields and multi-scale feature sources, while making lightweight improvements to the technical issues of the module

Images