Parallel Algorithms / Architecture Synthesis, AIZU International Symposium on (1995)
Aizu-Wakamatsu, Fukushima, Japan
Mar. 15, 1995 to Mar. 17, 1995
M. Hamdi , Dept. of Comput. Sci., Hong Kong Univ., Hong Kong
The paper proposes a new architecture, which has the potential to support low-level image processing as well as intermediate and high-level vision analysis efficiently. The integrated architecture consists of a mesh of processors enhanced with an efficient recursive network. Low-level image processing is performed on the mesh processor, while intermediate and high-level vision analysis is performed on the recursive network. The interaction between the two levels is supported by parallel data and control paths. To illustrate the power of such a two-level system, we present efficient parallel algorithms for a variety of problems from low-level image processing to high-level vision. Representative applications include image component labeling convexity problems, and histogram-type operations. Through time complexity analysis, we show that the integrated architecture meets the requirements of many image understanding tasks.
image processing equipment; computer vision; parallel algorithms; computational complexity; computer architecture; hypercube networks; image processing; modified mesh-connected computers; image processing applications; architecture; low-level image processing; intermediate-level vision analysis; high-level vision analysis; integrated architecture; processor mesh; recursive network; parallel data paths; control paths; parallel algorithms; image component labeling convexity problems; histogram-type operations; time complexity analysis; image understanding
M. Hamdi, "Modified mesh-connected computers for image processing applications," Parallel Algorithms / Architecture Synthesis, AIZU International Symposium on(PAS), Aizu-Wakamatsu, Fukushima, Japan, 1995, pp. 139.