The Fault Diagnosis Of Crossbar And Tighten Fasteners Based On Digital Image Processing

With the continuous development of the freight trains, high speed, heavy duty,high density open line has become today’s major trends, the old man column pickingup way no longer to adapt to the development. So, the late1990s TFDS system.After decades of development, the TFDS already development became an importantpart of one of the vehicles safe operation of the detection system, and its produce isthe premise of rail freight to achieve modernization. Cross-rod and tighten fastenersis an important part of the freight car bogie and the capacity to reduce the event offailure will allow freight car bogie freight security under serious threat.The main contents of this article is to design two sets of the TFDS imagerecognition system for the cross-rod bending and detained the bolt loose faultidentification algorithm, the idea of digital image processing and pattern recognitioncombine to design two sets of precision, high efficiency automatic recognitionalgorithm for the the TFDS system to provide strong technical support.How to design a strong robustness, high precision automatic identificationalgorithm is the difficulty of the subject and focus on algorithm design andimplementation process, the accuracy rate is the main objective, fast and as areference, the full use of digital image processing.The technology of image rotation,image enhancement, image segmentation, the gray-scale mapping, edge detection,image extraction and splicing, Hough transform classic,and summarize theoptimization algorithm in the design process of fault diagnosis algorithm.Brief introduction to algorithm design and implementation process designed toimitate TFDS image recognition platform test form, and then listed in the the TFDSsystem for automatic identification algorithm interface, according to the way designpackage recognition algorithm general process and produce a DLL.Finally, introduce the two sets of algorithms in the the TFDS system test results,test data to calculate the rate of false positives and false negative rate of thealgorithm in full compliance with the the TFDS system algorithm proposedrequirements, analysis of false positives picture false positive reasons that can bychanging the algorithm to solve the problem and suggest improvements. Brieflyaddressed the several optimization algorithms in the program design process.