A self-organizing neural network for hierarchical range image segmentation

Range image segmentation is the process of partitioning a range image represented by a two-dimensional pixel array into geometric primitives so that all the image pixels are grouped into clusters with a common geometric representation or property that could be used by higher-level cognitive processes. A self-organizing neural network for range image segmentation is proposed and described. The multi-layer Kohonen´s self-organizing feature map (MLKSFM) which is an extension of the traditional single-layer Kohonen´s self-organizing feature map (KSFM) is seen to alleviate the shortcomings of the latter in the context of range image segmentation. The problem of range image segmentation is formulated as one of vector quantization and is mapped onto MLKSFM. The MLKSFM is currently implemented on the Connection Machine CM-2 which is a fine-grained single instruction multiple data computer. Experimental results using both synthetic and real range images are presented