Research Of Structure Light Measurement Method In Axle Of EMU

Axle is a major part of the fuselage carrying loads, lack of precision can lead to uneven distribution of axle loads and thus lead to the axles serious wear and tear, if discovery not timely may lead to fracture of the axle, the train derailed. In addition, the locomotive driving and braking also through the wheel axle for work, not enough to affect the accuracy of locomotive axle braking and driving. Therefore, the axle is directly related to the accuracy of the safe of EMU.Based on the measurement of the efficiency of EMU axle needs and precision requirements, proposed EMU axle binocular vision structured light measurement method. The system mainly consists of two same cameras, multi-line structured light laser, coding to identify targets, computer. Two cameras take the projected axle pictures, after matching the feature points of the same name on the structure line, then restore the three-dimensional coordinates information of the feature points, multiple the feature points on structure line,you can restore the three-dimensional surface of the axle.Binocular vision measurement system needs calibrate the parameters of the cameras, the absolute outer exterior orientation parameters, and structural parameters; in addition, in order to reduce mismatch, improve efficiency and accuracy, highly accurate extraction of structured light stripe is needed; To complete the structure matching feature points on the light bar, solve the fundamental matrix is needed.Based on the target or target ruler identified by sub-pixel algorithm, extracted precision target pixel coordinate, since the positional relationship between the relative encoded targets and different points relationship between the encoded identification is known, restore the camera's definitely exterior orientation parameters; correspondence between feature points on both the images to strike the structural parameters and the basis matrix equation.Detect structured light by edge detection algorithm to extract the precise of the light strip edges, by curve fitting methods to improve the edge location accuracy, thereby further improving the accuracy of the measurement system.Then analyzes the rotational movement of the axle and a binocular cameras movementaxial scan to determine the absolute outer bearing binocular camera self-calibration method,so that binocular camera data measured under one of the world coordinate system can be more accurate unified.