The Topological Detection Algorithm of Object Arrays in Noisy Context Based on Fuzzy Spatial Information Fusion and Prim Algorithm
Most computer vision methods deal with the single object recognition problem. If an object is so small that little features can be used to support recognition procedure, the relationship between multi-objects could be helpful. In many cases, small objects are likely to be arranged by some regular shapes. To recognize these arrays, the paper presents a spatial topology detection algorithm. We call it as S-Prim (Spatial Prim) algorithm which is based on classic Prim algorithm, and integrates the fuzzy spatial information. The algorithm evaluates the spatial distribution regularity among neighboring nodes by back searching the path in the found tree when it is growing, and controls its growing direction according to some fuzzy rules to find out the most likely regular spatial topology. The detected tree can be considered as a spanning tree constrained by topological structures.
Spatial Topology Fuzzy Information Fusion Spanning Tree Pattern Recognition
Tao Wusha
Beijing Institute of System Engineering, Beijing 100101, China
国际会议
三亚
英文
1533-1536
2012-01-06（万方平台首次上网日期，不代表论文的发表时间）