Automatic Environment Reconstruction from Multiple 3D Acoustic Images A technique for three-dimensional reconstruction of an underwater environment from multiple range images acquired by an acoustic camera will be presented. Due to the narrow field of view and the absence of control of the sensor position, no information is available about the degree of overlapping between the range images; further, speckle noise and low resolution make more difficult the registration process. In this context, we propose a preprocessing method which gives a coarse alignment of range images prior to the application of an Iterative Closest Point (ICP)-based algorithm for the accurate registration of views pairs. The pre-alignment is based on the matching between the three-dimensional skeletons extracted from the images. A comparative analysis is presented where our method is compared with classic ICP, and with a technique based on principal components.
This work is part of an EC project whose final goal is the improvement of the understanding of a human operator driving an underwater remotely operated vehicle (ROV) floating around an offshore structure.