Research On Image Recognition Technology Of High-speed Train Catenary Suspension System Defect

The contact network support and suspension device play an important role in supporting the contact network.The condition of the device affects the stability of the whole suspension system and affects the performance of the contact line.The contact line may not be in good contact with the pantograph and affects the quality of the flow.At present,the status detection of the catenary support suspension device is very inefficient by artificial detection;the existing intelligent detection is also focused on the pantograph identification,the skateboard wear and tear,and the contact network parameters detection.Due to the bad state of catenary supporting suspension device,there are potential safety problems in catenary.Therefore,it is necessary to study real-time and intelligent detection methods.In this paper,taking the insulator of the suspension device in the catenary as an example,the image processing technology is applied to detect the malfunction of the suspension device based on the image processing.In the image recognition,this paper compares the insulator location of the SURF based SVM classifier,and the insulators location based on the HOG based Adaboost,compares the two methods of error recognition,and finally uses the affine transformation method to extract the insulators.In the state detection of insulators,it is pretreated first,then statistics its longitudinal non zero point,and the application of gray statistical minimum to complete the detection of the state of insulators and foreign objects.