Optimized for vehicle wheelset wear detection algorithm based on DM6467

ZHU Zhong-ke，WU Kai-hua，CHEN Jie

（College of Life Information Science & Instrument Engineering，Hangzhou Dianzi University，Hangzhou 310018，China）

Abstract：Aiming at how to improve the speed of wheelset detection，methods for wheelset wear detection，wear detection algorithm and algorithm optimization methods based on DSP were researched and generalized. The optimization methods for vehicle wheelset wear detection algorithm based on TI’s fixed- point DSP DM6467 were proposed. The DSP- image processing system was implemented. Depending on the features of detection algorithm and structural characteristics of the hardware. Some algorithm optimization methods had been used to optimize the wear detection algorithm. Software pipelining was used to achieve parallel processing for DSP. Inline functions

were used to improve the efficiency of the code. Memory was optimized to improve the speed of the system. The results indicate that，the efficiency of optimized algorithm is 3.3 times as fast as before. The optimization of result is obvious.

Key words：rail transport；wear measurement of vehicle wheelset；image processing；DM6467；efficiency of detection algorithm；algorithm optimization