图像灰度匹配在多核环境下的优化

随着计算机技术的不断发展,图像匹配已经成为图片处理的一个重要部分。在图像匹配中,图像的灰度匹配虽然具有匹配精度高的优点但却需要大量的计算时间,且计算时间随着使用模版的增大而大幅度增长。文章在多核的环境下使用内存优化与处理器亲和力优化方法来解决计算时间长的问题。实验结果表明,使用并行技术并结合内存优化与处理器优化方法可大幅度减少计算时间、提高缓存的命中率、避免乒乓效应的产生使并行程序的加速比与并行效率有所提高。

Image Grey Match Base on Cache optimization and CPU affinity in Multi Core Environment

With the rapid development of science and technology,image matching technology has become very important in the modern image processing.In image matching filed,image gray matching is a method with high accuracy but cost a lot of time on calculation.And it will cost more time when template is too large.Point to this problem,parallel computing technology based on Cache optimization and CPU affinity optimization in multi core environment are used in this paper to reduce the computation time.Test show that Cache optimization and CPU affinity optimization can reduce the matching time in image gray matching through improving the Cache hit rate.