Center Plate Bolts on the train after the photos were analyzed to observe the morphological characteristics, a gray map based on computer vision recognition algorithms. According to the physical characteristics of the train bogie image positioning, Gray mapping used in conjunction with the gradient mapping and transformation using adaptive threshold filter out irrelevant features, and image segmentation, Hough transform to extract the main features of application Lines, and finally under the Center Plate Bolts train characteristics of the geometry of fault interpretation. Algorithm design greatly enhanced the detection speed, accuracy and reliability.