No abstract

This (raper presents sever experimental 'rsnits of using the techniques of set and function nratheuaat 'al nrorplrology (1'Ilv1) for feature extraction Front real and synthetic range imagery. Acore specifically. we consider the problciu of extracting silhouetted appendages from real imagery wide coarse range resolution and hat of extracting appendages and corners front synthetic imagery wine high range resolution. AbstractThis p;ipor presents several experimental vesnlts of using the techniques of set and function mathernai -al morphology (MM) for feat n ro extract ion from real and synthetic range imagery. More specifically. \ve consider the problem of extract-ing silhouetted appendages from real imagery with coarse range resolution and thai, of extracting appendages and corners from synthetic imagery with high range resolution.

Although little known, mathematical morphology offers great potential in the areas of image enhancement, feature extraction, and object recognition. This work explores this growing field through a survey of established morphological algorithms and the development of new morphological algorithms for range image analysis. With range imagery, mathematical morphology is used for noise removal, 2-D feature extraction, 3-D feature extraction, and 3-D corner extraction.