Based On Image Processing Algorithm The Crane Boom Bending Monitoring Technology Research

With the development of China’s economy and modern industry, the large cranes are widely used in various field, the quantity demand is increasing, in the fields of industrial production and infrastructure, mining enterprises, bridges, ports etc. At the same time, people pay great attention to the safety of crane. Crane boom is an important component of bearing, and the weak point of structure. It could produce the side bending phenomenon. But it is difficult to know the value of crane boom’s stress and horizontal offset. And it seriously affects the normal operation of the crane. We are unable to know the working condition of crane boom in real time. Therefore, we cannot be evaluated in the boom structure strength.In order to monitor the bending degree of crane boom, a measurement method to monitoring horizontal offset of crane boom is proposed, which based on computer vision technology and combined with the optical, mechanical design, image processing, computer technology, through the value of crane boom horizontal offset to reflect the bending degree. This measurement method has the advantages of simple operation, high reliability, highly accuracy. And the system improves the crane operation safety, so as to reduce the occurrence of accidents. The main contents and results of the paper are as follows:First of all, this paper is based on the actual planning project of science and technology as research background. Aiming at the security problem of large crane boom, it provides a scheme for the boom side bending degree online monitoring system, which based on infrared laser tracking and recognition. According to the plan, we put forward some technical difficult points.Second, study on the cause and hazard analysis of the boom side bending phenomenon. Design the calculation method of crane boom horizontal offset. In the image processing technology, this paper mainly describes some commonly used image processing methods, such as the mean filter, median filter, adaptive threshold image segmentation, watershed region segmentation and so on. In the algorithm of target recognition and tracking, mainly describes the related content of template matching tracking algorithm. Research and analysis, obtained the advantages and disadvantages, combined with the requirements of the actual project, determine to use which one of the algorithms of target recognition and tracking in the paper.Again, according to the requirements of the system working environment, combined with the above algorithm, we design a set of large crane boom side bend of online monitoring system. The working includes image acquisition, image processing, target recognition and tracking, data computation, and communication etc. After repeated testing and debugging, it can realize the real-time detection of horizontal offset of crane boom. Transmit real-time data to the display interface.Finally, through the scene test results, the actual measurement data is accurate, whit in the range of error allowed, the system is easy to operate, high reliability, real-time, can greatly improve the safety factor of the crane, and has good prospects for practical application. The paper summarizes some deficiencies in the system, and puts forward corresponding methods and ways to improve the system.