Research Of Surface Defect Detection Algorithm For Seat Belt Based On Machine Vision

As an important traffic safety assurance tool,seat belt needs to carry on strict quality inspection in the process of its production, and surface defect detection is one of the key links in the quality inspection. At present, the problem of low efficiency and poor stability of the artificial detection method could not meet the needs of the actual detection. Therefore, this paper explores the seat belt surface defect detection algorithm based on machine vision to meet the actual demand, and the seat belt surface defect detection system based on machine vision was constructed. The main research of this paper is as follows:Firstly, in view of the characteristics of image acquisition and surface defect detection, the overall design of the system was proposed to meet the detection demand of high accuracy and high speed. The hardware components of the measurement system were studied and analyzed, which laid the foundation for the subsequent image processing.Secondly, aiming at the texture characteristics of the seat belt, a morphological segmentation method based on Laws texture measure was proposed, which improved the noise immunity of the image segmentation. In view of the edge defects of the texture image, the defect detection algorithms based on morphological processing was studied, and a good detection result was obtained.Thirdly, aiming at the characteristics of surface defect of seat belt, different detection methods were studied from different angles. In the spatial domain, the Blob feature analysis method, the gray level histogram feature analysis method and the GLCM feature analysis method were studied. For every method, the influence of defects on the characteristic parameters was studied and the experimental results were analyzed. In the frequency domain, a defect detection algorithm based on spectrum features of texture image was proposed, and the selection of the filter and the calculation of the parameters are discussed..Finally, the detection results of several algorithms were compared. The experimental results show that the spectrum analysis method has better effect on the detection of surface defects than other detection methods, which could meet the requirements of real-time and accuracy.Finally, a surface defect detection platform for seat belt was built, and seat belt surface defect detection system based on machine vision was built to accomplish the detection of surface defect of seat belt. The speed and stability of defect detection were analyzed and the test results show that the speed and stability of the system can meet the test requirements.