Segmentation of Underwater Images by Edge Detection on Posteriori Probability.

In this paper we address the problem of the recognition of a sea-pipe, by recovering the relative pose between the underwater vehicle and the pipe, both from a single and a sequence of images. A color digital camera, mounted on the vehicle, allows to collect images which are then processed by a suitable software to perform a real-time detection of the pipe. Firstly, images are segmented in order to identify the pipe of the sea-line by performing edge detection and applying the Hough transform on the posterior probability images computed by a classifier. Each image pixel is grouped according to color and texture features. Once the sea-line has been located, the relative pose and position of the underwater vehicle (UV) is computed.