Research on the Method of Detecting the Rail Marking of Road-rail Vehicle Based on Image Enhancement

Rail marking is an auxiliary identification line for guiding drop of the road-rail vehicle’s steel wheels. In order to accurately detect the rail marking and improve adaptability, an image-based enhancement method for detecting the rail marking was proposed. After the calibration and image distortion correction, the ROI extraction and image smoothing filtering were performed. Image enhancement algorithm based on combination of Laplace operator and Gamma transform was developed to enhance the edge details and remove light reflective areas in the image. Finally, with the help of Hough transform, the detection of the rail marking was better achieved. Experiments and results show that the method can well solve the problems of light reflection and edge interference in the image, and meet detection accuracy requirements of the rail marking when road-rail vehicle drops its steel wheels.