Research On Fault Detection Of Landscape Lamp Based On Camera Image Analysis

With the rapid development of social modernization,landscape lamp are widely used in the process of urbanization.Landscape lamp can not only provide lighting effects for the city at night,but also beautify the night view of the city.The current landscape lamp fault detection technology is mainly manual inspection and remote camera monitoring.It has the shortcomings of not timely detection and consumes a lot of manpower.How to quickly and effectively detect the landscape lamp failure is a problem that needs to be solved in the current engineering.Based on the analysis of the existing image detection technology,this thesis proposes a hybrid mode-based automatic detection method for landscape lamp faults.At the same time,an improved image registration method is proposed to solve the problems of jitter and offset that may occur in the actual application of the landscape lamping camera,so as to ensure the consistency of the collected image and the original template,and greatly reduce the detection rate while ensuring the detection rate.The detection false alarm rate is improved.The main work of the thesis includes:(1)Aiming at the problems of lack of timeliness,high efficiency and excessive manual participation in landscape lamp fault detection,a hybrid mode-based automatic detection method for landscape lamp faults is proposed.The method is to construct a feature template for the landscape lamp,and perform accurate and efficient fault detection on the landscape lamp image collected in real time through template matching.In the process of constructing the feature template,by preprocessing the landscape lamp image,the noise information in the landscape lamp image can be effectively eliminated,and the sharpness of the image can also be improved.Secondly,the OTSU algorithm and Bayesian algorithm are combined to segment the landscape lamp area in the landscape lamp image.Finally,the quadtree algorithm and the minimum bounding rectangle algorithm are combined to extract the feature template of the landscape lamp,which improves the running speed and efficiency of the algorithm.The experimental results show that the hybrid detection method has good reliability and detection effect for common landscape lamping detection.(2)Aiming at the problem of jitter when the camera collects landscape lamp images,a landscape lamp registration method based on an improved image registration algorithm is proposed.This method can effectively extract the characteristic information in the landscape lamp image,and register the landscape lamp image through the characteristic information.In this way,while mapping the position of the landscape lamp image to the image template position,it also effectively improves the accuracy of the landscape lamp fault detection.The final experimental results show that this method eliminates the influence of jitter and greatly reduces the false alarm rate of detection.(3)Designed a fault detection system for landscape lamp.Using opencv and other technologies,and the SSM framework to achieve offline template establishment and online detection functions,integrated with the landscape lamp fault detection algorithm proposed in this article,the system has been successfully deployed in many places,achieving real-time,efficient and accurate landscape lamp fault detection effect.