Thresholding of a sonar image from a small underwater object using the maximum entropy of the one-dimensional bound histogram

The concept of the one-dimensional bound histogram was given. It is the one-dimensional histogram bound by some prior knowledge, and it can make some image processing methods be simple and feasible. Furthermore, a thresholding method based on the maximum entropy of the one-dimensional bound histogram is proposed. In the method, Pun entropy function is established by means of the bound histogram instead of the usual one. In applying the method to thresholding a sonar image of a small underwater object, the bound set is constructed according to the restriction in the gray-level values of both pixels and their neighborhood averages, and the one-dimensional bound histogram corresponding to that bound set is established according to the concept of the one-dimensional bound histogram. The experimental results show that the proposed method can succeed in thresholding a sonar image of a small underwater object. The proposed method is applicable to the image in which there is prior knowledge.