| The detection technology based on digital image processing has been widely used and researched at home and abroad for its features of non-contact, online detection, rapid response, high precision and strong reliability, especially in industrial inspection. The most commonly used defects detection method of industrial products is manual inspection, which has a lot of disadvantages. As a result, the realization of online defects detection system of industrial products is important to the productivity improvement and cost reduction. As an important part of this system, the research on defects detection algorithms is of great importance and application value.Based on the current research achievements of image processing, the defect online detection algorithms of spring clamps are researched in this paper. The actual application and important meaning of digital image processing are researched based on machine vision technology. Some analysis and research of HT and some improved algorithms based on HT proposed by domestic and foreign scholars are studied. The circle detection algorithms of HT and RHT are studied and applied in the dimension detection of spring clamp images. Experimental results and data statistics are analysed. Improvements are made from several aspects, including the judgment of the collinear, limitation of distance, the calculation of gradient angle and others. The flow chart of the improved algorithm is given, and the new algorithm is realized on MATLAB. The error analysis is carried out at the end. Based on the above key technologies and digital image processing, the online defect detection algorithm of spring clamps is developed.Experimental results show that the improved algorithm has obvious advantages in speed and accuracy and great application value. Comparing to the traditional algorithms, the desighed algorithm in this paper achieves faster operation, higher precision, and stronger robustness. |
PDF Full Text Request