Non-uniform texture small defect detection method based on improved Faster R-CNN model

Abstract

The invention discloses a non-uniform texture small defect detection method based on an improved Faster R-CNN model, and the method comprises the steps: obtaining an image of a to-be-detected object with a to-be-detected defect, and carrying out the feature extraction of the image of the to-be-detected object through the improved Faster R-CNN model; the improved Faster R-CNN model is specificallycharacterized in that features are extracted by embedding a feature pyramid multi-scale fusion structure of a residual complementary attention gate module in the last three stages of a VGG16 network structure. The improved Faster R-CNN model integrates feature map information of different scales and receptive fields of each stage from top to bottom in a feature pyramid manner; the judging capacityfor different types of defects is enhanced, particularly the judging capacity for wrinkles of different degrees is remarkably improved, the overall recall rate is remarkably increased, and the industrial precision requirement is met. The residual complementary attention gate module can guide multi-scale feature fusion to suppress complex background information, and then context information is captured globally to locate small defects more accurately.

Description

technical field [0001] The invention relates to technical fields such as computer vision, deep learning, and target detection, and in particular to a method for detecting small non-uniform texture defects based on an improved Faster R-CNN model. Background technique [0002] As one of the important materials in modern industry, lithium batteries are widely used in new energy vehicles, construction, industrial manufacturing and other fields. Before lithium batteries are put into use, the quality of lithium batteries needs to be tested. There are many kinds of defects in lithium batteries, mainly including edge folds, pole sheet scratches, exposed foil, particles, perforations, dark spots, foreign objects, as well as surface dents, stains, bulges, coding deformation, etc., and the defects are random in shape and size. In addition, the surface of lithium batteries has a complex background with non-uniform textures, which bring great challenges to the quality inspection of lithi...

AI Technical Summary

Problems solved by technology

There are many kinds of defects in lithium batteries, mainly including edge folds, pole sheet scratches, exposed foil, particles, perforations, dark spots, foreign objects, as well as surface dents, stains, bulges, coding deformation, etc., and the defects are random in shape and size. In addition, the surface of lithium batteries has a complex background with non-uniform textures, which bring great challenges to the quality inspection of lithium batteries

[0003] The existing lithium battery quality inspection method is to sample the product through manual comparison with the naked eye and hand touch feeling. Artificial has the problem of fatigue, and it is easy to make mistakes in the face of long-term detection; artificial judgment standards are subjective, and different time periods Similar samples will have different judgments; due to the small size of some defects in lithium batteries, many details cannot be accurately identified by human eyes; the efficiency of manual inspection is very low, and real-time inspection is not possible

[0004] The literature "Research on Surface Defect Detection of Lithium Batteries" proposes a lithium battery surface detection system, which can realize the automatic detection of defects in the appearance and size of mobile phone lithium batteries, foreign matter defects in peak seals, and poor coding; Therefore, it is difficult to find a set of algorithms that can characterize all defect features, especially some defects look very small and are seriously affected by complex backgrounds, and the effect of defect recognition will be significantly reduced

[0005] He Kaiming and others proposed a two-stage target detection framework Faster R-CNN (Ren S, He K, GirshickR, et al.Faster R-CNN: Towards Real-Time Object Detection with Region ProposalNetworks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149.), but it is mainly aimed at public data sets, and the target pixels are large (for example, the target pixel in the VOC2007 data set is 500*366), which is difficult to apply to industrial defect detection

Large images and small defects are a major feature of industrial defect detection. For example, the defects on a lithium battery image with a resolution of 1 million are only dozens of pixels or even a dozen pixels, which can easily cause inaccurate positioning of small defects. , cannot meet the requirements of industrial detection accuracy, so the existing Faster R-CNN cannot be directly applied to the surface defect detection of lithium batteries

Images