Machine Vision-based Quality Inspection Of Disc Brake Pad Appearance

The safety of driving a car is linked to the quality of the brake pads.Currently,disc brake pad quality is primarily determined through manual sampling inspection.The shortcomings of manual detection have been made clear by the ongoing rise in societal productivity,the quickening of the pace of production,and the transformation of the industrial structure.For instance,manual detection has a high expense but poor detection efficiency,and the phenomenon of missing and false detection is common.As a result,it has been challenging to adapt manual detection to the growth of the car industry.It is more challenging to satisfy the growing demand for brake pad detection automation.Machine vision-based detection technology has advanced significantly with the ongoing development of artificial intelligence and intelligent detection,which is crucial for the look quality detection of disc brake pads.This study examines the appearance quality detection of disc brake pads while taking into consideration the effectiveness and accuracy of the detection process.A machine vision-based appearance quality detection device for disc brake pads is created.The following are the primary study findings:(1)The platform for machine vision inspection.Based on the analysis of the requirements for disc brake pad detection,a set of machine vision-based detection platforms was built,providing a platform for the subsequent realization of disc brake pad size detection and defect detection.These platforms included industrial cameras,lenses,light sources,lighting modes,and software.(2)Machine vision-based size recognition of disc brake pads.Based on an analysis of image processing,the impact of image enhancement,threshold segmentation and edge detection algorithms are examined,and the best image processing method and associated parameters are chosen for disc brake pad size detection.The main components include the selection of filtering methods,selecting Gaussian filtering for image noise reduction;selecting histogram equalization for image enhancement and improving image contrast;and selecting the Canny edge detection operator for edge detection of the captured disc brake pad image.The contour detection method is used to obtain the smallest outer rectangle of the part image,determine the edge contour of the friction material part,calculate the number of pixels of the size to be detected and convert it to the real physical size by pixel equivalence.(3)Convolutional neural network-based disc brake defect detection.The convolutional layer,pooling layer,and completely connected layer functions are examined after a brief introduction to the convolutional neural network.To identify disc brake pad surface flaws,the Alex Net6＿BN model,an upgraded version of the Alex Net convolutional neural network,is proposed.The Alex Net6＿BN model modifies the convolutional layer in the traditional Alex Net network model by adding a convolutional layer and a pooling layer,increasing the first layer’s convolution kernel size to 13×13 to extract more significant sample features,and using batch normalization in place of the original local response normalization behind each layer to quicken the network’s convergence.The enhanced network model’s defect identification for disc brake pads can be learned and trained.(4)Disc brake pad prototype system creation and appearance quality testing.The disc brake pad appearance quality detection prototype system is developed using Python under the Py Qt5 platform,and the reliability and effectiveness of the prototype system is confirmed by the actual disc brake pad detection.The prototype system is based on machine vision-based disc brake pad size detection research and convolutional neural network-based disc brake pad defect detection research.This paper proposes a machine vision-based quality inspection system for the appearance of disc brake pads,including the study of dimensional inspection of disc brake pads,defect detection research and prototype system development,and verifies the effectiveness of the proposed method through actual disc brake pad inspection,providing theoretical guidance and technical method support for improving the efficiency of disc brake pad appearance quality inspection.