The Research And Development Of Surface Defect Detection System For Capacitor Based On Machine Vision

With the rapid development of information technology and industrialized production,the demand of capacitor is also rapidly increasing. In the process of production, we must detect the surface of the capacitor to ensure the quality of the product. In this paper, we design a real?time and automatic detection system, which is capable of detecting the defects of the capacitor.Firstly, this paper analyses the common defect types of the capacitor surface and its surface defects detection requirements, the overall scheme of the capacitor surface defect detection system is designed, and the structure of the machine vision system and the principle and function of every component are studied. According to the characteristics and requirements of the subject, the hardware of the system is selected. This paper choose the Halcon machine vision software as the development platform of software system, and briefly introduces the advantage of Halcon software, interface of Halcon, application of surface defect detection, etc.Secondly, according to the testing requirements, this paper carries on the algorithm design of the capacitor surface defect detection system. The algorithm design of machine vision system is the core part of the whole system. This paper focuses on the image preprocessing algorithm for the capacitor surface defect detection, including filtering,threshold segmentation, morphology, block analysis,etc. This paper proposes an improved algorithm based on Canny operator, and carry out the corresponding analysis and comparison, and then design a preprocessing algorithm which is suitable for the surface defect detection system for capacitor. This paper also studies the extraction and recognition algorithm of the capacitor surface defect and designs the extraction and recognition of surface defect in different angles according to specific conditions of the capacitor.Finally, this paper uses Halcon10.0 and VC++ programming and analyses the software of the capacitor surface defect detection system. A good interface of the capacitor surface defect is built in VC++and we test the capacitor samples. The experimental results show that this system can meet the requirements of the real?time, rapidly and accurately detecting the capacitor surface defects.