Research Of Railway Fastener State Recognition Based On LabVIEW

With the rapid growth of the national economy of China,the railway played an increasingly important role;people also pay more and more attention to the development of railway;the railway safety issues attract more attention;the maintenance of railway track components, becomes one of the key problem. Fastener is one of the important components of track, is used to connect the rail and sleeper,make the rail fixed on the sleeper, Whether the fastener state is good to maintain the safety of railway transportation, lack of fasteners, loosening, fracture and other phenomena will cause a great threat to the safety of train traffic, so the research on the inspection of the fastener state has become a hot spot. At present, our country for the status of the fastener is still based on manual inspection, it is the use of patrol workers experience observation and judgment of fastener and track, this method has a lot of disadvantages, on the one hand, it consumes a lot of manpower and time, resulting in slow speed, low efficiency and so on, on the other hand, the influence of subjective factors is great, the false positive rate and the false positive rate are very high, and the reliability is low.Therefore, the realization of the automatic detection technology of the fastener state will put forward a new change for the rapid detection of rail fastener failure.In recent years, the rapid development of digital image processing technology, the development and research of track component inspection system based on computer vision has become the focus of research at home and abroad, the system has the advantages of high speed and high precision. In this paper, based on the status of the status of the fastener state detection, a series of algorithms for precise location, feature extraction and state identification of fasteners are studied:Detailed understanding and analysis of several existing methods of fastener positioning, and in view of the deficiencies and the characteristics of the fastener structure, use the "cross across method" to locate the fastener area, first,use a projection operation to the image preprocessing of two valued image,confirm the approximate area of the fastener,and then use the canny operator to detect the edge of the fastener, statistical edge highlights to determine the derailment of sleeper sleeper and rail boundaries, accurate positioning of fasteners. It is proved that the algorithm has good stability and robustness.According to the research on the recognition algorithm of the fastener, the existing methods of state identification of several fasteners are compared and analyzed, the method of principal component analysis(PCA algorithm) has been successfully applied in many fields, reduct the dimension of the precision positioning of the fastener image, extract the feature of the fastener image, this method can greatly reduce the computational complexity of the identification of fasteners. Then set up the training set and test set, using minimum distance classifier to recognize the state of the fastener. Compared to other algorithms, PCA algorithm has the advantages of high recognition accuracy, fast calculation speed, simple calculation process and so on.This topic uses the LabVIEW host computer programming, the above algorithm is tested through the building of the module, verified the validity of the algorithm, realize the positioning and identification of fasteners, and has achieved good results. The experimental results show that the localization algorithm adopted in this paper has good stability, the recognition of the fastener has some robustness, these algorithms can meet the needs of practical application.