Arcing Detection And Recognition Of Pantograph And Catenary Based On Image Processing

With the rapid development of economy,the cross-regional flow of social labor factors is becoming larger and larger,so the demand for travel tools is increasing accordingly.At present,high-speed railway can satisfy the demand of speed,comfort and safety,and it has become the first choice for people to travel.The railway construction in our country is developing towards heavy-load and high-speed.With the increase of speed,the requirement of safe and stable operation is higher and higher.The reliability and continuity of power supply is also called the quality of power supply.The quality of power supply is affected by the contact between pantograph and catenary.Arcing of Pantograph and catenary is an important performance of off-line pantograph and catenary.Therefore,the detection of Arcing of Pantograph and catenary is directly related to the safe and stable operation of the train,this is also an important reason why people pay more and more attention to arc detection in pantograph-catenary system.This thesis studies the current arc detection technology of pantograph and catenary at home and abroad,and focuses on the detection structure and principle of the dynamic detection device(3C)for vehicle-borne catenary,it is found that the 3C detection device has the disadvantages of high miss detection rate except for the advantages of low miss detection rate.At the same time,the arcing characteristics of pantograph-catenary system are studied,and the characteristics of high temperature,strong light and high radiation in the arcing process of pantograph-catenary system are analyzed.Therefore,this thesis puts forward that when the infrared image and the visible image are reflected by the high temperature bloom characteristic of the burning arc image,the arc spot of the pantograph and catenary will appear in the visible image corresponding to the infrared image with the temperature exceeding the limit in the high temperature region,so the pixel-based fusion algorithm is used to fuse the infrared image and the visible light image,and the number of connected areas of the fusion image is used to realize the automatic checking of arc in pantograph-catenary system and reduce the false detection rate.In order to get the connected region of the fused image,firstly,the traditional Otsu is improved by the way of weight,which makes up for the disadvantage of the traditional Otsu in the segmentation of the histogram with unimodal distribution or near unimodal distribution,segmentation of high temperature region in infrared image.Secondly,the visible light image is binarized by the method of automatic binarization,and the spot area of the burning arc image is recognized.Finally,the infrared segmented image and visible binary image are fused by the fusion method based on the Minimum Pixel.Then,the concept of connected area is used to realize the automatic rechecking of the burning arc.From the simulation results,it can be seen that the proposed method can be used to detect the arc-firing of the pantograph ingeneral working conditions and under the influence of tunnel inlet and outlet and strong light.