ROBUST ADAPTIVE SEGMENTATION OF RANGE IMAGES

We propose a novel image segmentation technique using the robust, adap tive least Mh order squares (ALKS) estimator which minimizes the Mh or der statistics of the squared of residuals. The optimal value of k is determined from the data, and the procedure detects the homogeneous su rface patch representing the relative majority of the pixels. The ALKS shows a better tolerance to structured outliers than other recently p roposed similar techniques: Minimize the Probability of Randomness (MI NPRAN) and Residual Consensus (RESC). The performance of the new, full y autonomous, range image segmentation algorithm is compared to severa l other methods.