Cloth defect detection model and method based on improved YOLOv4-tiny

Abstract

The invention provides a cloth defect detection model and a cloth defect detection method based on improved YOLOv4-tin. According to the detection model, a dense connection convolution block CSPDenseBlock is added to a residual block in a YOLOv4-tiny backbone network, and an spp module is added to the backbone network finally, so that a new feature extraction network is formed; the new feature extraction network outputs two feature maps with different scales; and the two feature maps with different scales are processed by respective corresponding convolution blocks and then enter respective YOLO layers to predict a target. According to the model and method, the backbone network is adjusted by adding the dense connection convolution block into the original residual block, so that the neuron receptive field of the detection model is expanded, and the extraction of shallow information is facilitated; a deep backbone network framework is constructed by using dense convolution blocks, so that a plurality of targets which are difficult to distinguish can be identified, and the accuracy of model detection is improved; the whole model can be used for optimizing a cloth defect detection task in a complex scene, and the detection precision is better than that of a traditional model.

Description

technical field [0001] The invention relates to the technical field of defect detection, in particular to a cloth defect detection model and method based on improved YOLOv4-tiny. Background technique [0002] The detection of cloth defects in actual factories is usually carried out by artificial vision, which is helpful to correct defects in time, but due to the fatigue of human beings, this will lead to human errors, and it is usually difficult for human eyes to detect to small defects. For long-term applications in industry, this method is inefficient and less accurate. [0003] Compared with the end-to-end training and detection method in deep learning, the traditional defect detection technology often requires manual definition of features and feature extraction methods, which means that effective features need to be designed to Different defects in fabrics are characterized. Although challenging, many researchers have made great efforts to solve these problems. Trad...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problem that the traditional cloth defect detection algorithm cannot be applied to the existence of various cloth defect types, the present invention provides a cloth defect detection model and method based on improved YOLOv4-tiny

Images