Abstract:

A parallel image matching algorithm based on YHFTQDSP is proposed, which adopts a datalevel parallel method to implement parallel keypoints extraction and matching, and fully develops multilevel parallelism of multicore processors. The parallel algorithms for SIFT, SURF and PCASIFT are implemented and evaluated, and the experimental results show that the parallel image matching algorithms have good adaptability for various distortion images and the image matching ability is close to the sequential algorithm with the average speedup of 3.2.

Key words: parallel;SIFT;image matching;multicore processor