摘要
该文提出了一种新颖的在基于K元2-立方体网络的SIMD计算机上运行的图像分块模板匹配并行算法。该算法对于N×N的图像、M×M的模板和K×K的处理元阵列(M<K,N=BK,B>1),可通过将图像分成B×B个图像块的分块模板匹配来实现。与已知的各算法相比,该算法具有可处理比处理元阵列尺寸大的图像的优越性。
This paper proposes a novel parallel algorithm for template matching of image division block based on a K-ary 2-cube network SIMD computer. For the template matching on an N by N image, M by M template and K by K processing unit array (M < K, N = BK, B > 1), the algorithm is able to realize template matching by dividing an image into blocks. Compared with known algorithms, the algorithm is able to process an image whose size is larger than that of the processing unit.
出处
《电子与信息学报》
EI
CSCD
北大核心
2001年第9期868-874,共7页
Journal of Electronics & Information Technology
基金
国家部级项目资助
关键词
图像分块
模板匹配
SIMD计算机
并行算法
Image division block, Template matching, SIMD computer, Parallel algorithms
