We consider the task of automatic detection of low-contrast objects against the background of the sea surface in this paper. The analysis of video data containing information about the sea surface in order to detect objects is relevant for the development of security and monitoring systems, navigation and meteorological systems, etc. We study the features of the subband matrices mathematical apparatus used for solving the problem of analyzing the sea surface images frequency properties. We proposed the decision function in the detection problem which is based on the analysis of subband frequency properties within the cosine transform of the individual components of the sea surface image sequence (video frames) presented in the form of vectors containing pixel values having the same coordinates in the image set. The method of slow-moving objects subband detection on the sea surface is described. The computational experiments results of the slow-moving objects detection on the sea surface are presented, they demonstrate the efficiency of the developed method.
Volume 12 | 05-Special Issue
Pages: 1314-1325
DOI: 10.5373/JARDCS/V12SP5/20201891