Battery defect detection method based on lightweight neural network

Abstract

The invention relates to a battery defect detection method based on a lightweight neural network. The method comprises the following steps: constructing a defect detection model based on the lightweight neural network, wherein the defect detection model takes ShuffleNetV2 as a backbone network, a refined cross-stage local mechanism is fused into a ShuffleNetV2 network, a refined cross-stage local ShuffleNetV2 network is obtained, and a parallel grouping attention module is further fused into the refined cross-stage local ShuffleNetV2 network; and refining an output feature map of the cross-stage local ShuffleNetV2 network, inputting the output feature map into a recommended region in a region recommendation network after the output feature map passes through a fusion module guided by low-layer features, and classifying and regressing the recommended region to obtain a defect category and a defect position. A parallel grouping attention module realizes shallow-layer and deep-layer feature fusion, and a fusion module guided by the low-layer features expands the receptive field; and the method solves the problem that the lightweight network detection effect is not ideal.

Description

technical field [0001] The invention relates to the technical field of battery defect detection, in particular to a battery defect detection method based on a lightweight neural network. Background technique [0002] Batteries are one of the important factors restricting the development of new energy technologies. The quality of batteries directly affects the performance and service life of new energy equipment. Surface defect detection is an important part of the battery manufacturing process. Existing manual detection methods are greatly affected by factors such as workers' subjectivity and prone to fatigue, and it is difficult to meet the needs of large-scale production. Therefore, defect detection algorithms based on machine vision are applied. The defect features extracted by machine vision detection algorithms mainly rely on manually defined rules, and only shallow features are extracted, which limits its wide application in industrial inspection. And neural network-b...

AI Technical Summary

Problems solved by technology

However, expanding the architecture of the neural network usually brings more calculations and requires higher hardware resources, resulting in a higher cost of the system that can deploy the deep neural network, which greatly limits the application of the object detection model based on the deep neural network. Applications in practical engineering, so it is necessary to develop a lightweight network

Due to the multi-scale characteristics and strong background interference of defects on the battery, such as random distribution of background texture shapes, surface defects overlapping with the background in strength and shape, etc., the lightweight network is not effective in detecting complex background and small target defects. Ideal, so developing a lightweight network to detect battery defects is difficult to implement

Images