Extracting Contours of Hydrothermal Chimney Undersea Images Based on Machine Vision

Contour information is regarded as important characteristic in computer vision. It is difficult to extract contour information from the object of underwater images such as hydrothermal chimney due to its complicated structure. In this paper we discuss of the problem of the contour extraction of the hydrothermal chimney from underwater images. The variety of image quantity of information was denoted dynamically by multi-scale entropy (MSE). and the multi-scale entropy difference (MSED) can present the break part of the image gray information and recognize the boundary between object and background effectively. The non-structured object contours underwater can be extracted with the advantages of the maximal multi-scale entropy difference (MMSED). From the experimental results, the approach is effective.