Issue No. 05 - May (2000 vol. 22)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/34.857001
<p><b>Abstract</b>—This paper describes an unsupervised region merging technique based on a novel robust statistical test. The merging decision is derived from the mutual inlier ratio (MIR) of adjacent regions. This ratio is computed using robust regression techniques and a novel method to estimate the robust scale of the Gaussian distribution. A discrimination value to recognize identical Gaussian distributions with the MIR is derived theoretically as a function of the sizes of the compared sets. The presented method to test distributions is compared with the established Kolmogorov-Smirnov test and implemented into a segmentation algorithm for planar range images. The iterative region growing technique is evaluated using an established framework for range image segmentation comparison involving 60 real range images. The evaluation incorporates a comparison with four state-of-the-art algorithms and gives an experimental demonstration of the need for robust methods capable of handling noisy data in real applications.</p>
Segmentation, robust statistics, region merging, range images, clustering, least-median-of-squares, segmentation comparison.
K. Köster and M. Spann, "MIR: An Approach to Robust Clustering-Application to Range Image Segmentation," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 22, no. , pp. 430-444, 2000.