Design Of Air Tightness Detection System Of Pressure Vessels Based On Machine Vision

With the development of science and technology and the improvement of industrial production, the air tightness detection of containers has been paid more attention by the manufacturer. Greater demands were being placed on the leakage detection, in order to avoid the explosion, the fires and the accidents caused by the leakage. No longer can manufacturers be content with traditional detection method and high price of instrumentation. Therefore, it is necessary to develop a device which has virtue of strong function, low price and high automation to satisfy the industrial demand. For the sake of automatically detection of pressure vessels in Jiaxing and demand for leak detection, a moving bubbles detection method and a blurred image restoration algorithm is proposed in this thesis. The air tightness detection system of pressure vessels based on machine vision has been designed, which can raise the productivity, save labor and improve product quality.This work was partially supported by the Science and Technology Department of Zhejiang Province and Graduate Innovation Project of Zhejiang Sci-Tech University. The main research contents and achievements are as follows:(1) First of all the current situation and the developing trend of air tightness detection are reviewed. And then describe the advantages of the machine vision. The air tightness detection scheme based on machine vision is designed after introduction of the process of the air tightness detection and analysis of the drawbacks of the common detection methods. According to the functional requirements of the system and the specifications, the overall structure of the air tightness detection system is presented.(2) The article designs the mechanical system of the air tightness detection, which includes the supporting frame, the horizontal transmission device, the vertical transmission device and the fixed device. And then the parameters of the cylinder and the motor are calculated and analyzed. Besides, the schematics of the electrical control cabinet and the PLC hardware are designed. Finally, the software design of the electrical control system is accomplished, and the control system has the function of automatic movement.(3) The point spread function of uniform linear motion blur is introduced to deal with the degradation of detection images. And an image semi-blind restoration method with edge regularization is proposed, the effectiveness of the method is verified by the simulation results. Besides, a moving target detection algorithm is studied for the moving bubbles in leak detection. An improved frame difference method on the basis of the Kirsch edge detection and the background difference is presented, which is designed for dealing with the exists of the overlapped parts and the incomplete object contours. The target regions detected by this method are more completed, and then the recognition and detection of moving bubbles based on machine vision is realized.(4) A prototype system is built in this paper. After that the functional and performance testing is conducted in the laboratory. The results show that the automatic detection function is implemented. So, the system is characterized by high efficiency, low cost and high degree of automation. It can effectively reduce the mistake rate and miss rate of the air tightness detection. As a result, it has very wide popularization and application prospects in the companies of pressure vessel.