A geometric approach to the segmentation and reconstruction of acoustic three-dimensional data

This paper presents a technique for the segmentation of three-dimensional (3D) images acquired by an acoustical camera. The proposed algorithm identifies first the most reliable image points likely corresponding to man-made objects, and, second, determines the points belonging to the same surface. Actually, it exploits the geometrical properties embedded in the sparse and noisy 3D information available to group the points which better fit the current quadric surface. This algorithm can be applied for the reconstruction of virtual environments from acoustic data, useful for robotic applications (e.g., vehicle navigation). Results on synthetic and real acoustic images are promising