摘要
非局部均值滤波算法已经在多种图像去噪场合得到了验证,并且显示出了优异的性能。在核磁共振图像去噪中,该算法存在计算量大的问题。针对该问题,将多线程技术引入到算法实现中,提高了非局部均值滤波算法的计算效率。仿真数据和真实数据的实验结果表明:在保证图像去噪效果不变的情况下,四核处理器主机内,多线程算法较之于单线程算法的运行速度提高了2.53倍。
Non-local means（ NLM） algorithm has been successfully applied in various image denoising scenarios and can produce satisfactory denoising results. However,it is not readily applicable to magnetic resonance image（ MRI） denoising due to the large computational cost of the algorithm. To speed up NLM,a multi-threaded NLM is designed and implemented. Experimental results on both simulated and real data show that our multi-threaded implementation can run 2. 53 times faster than single-threaded implementation without loss of denoising efficacy on a processor with four cores.
出处
《科学技术与工程》
北大核心
2014年第15期185-188,共4页
Science Technology and Engineering
