Sea-sky line detection method combined with local texture feature filtering
In view of the problems of low contrast between sea and sky,image blur caused by interference conditions such as sea objects,sea fog,waves,and clouds,and the obscuring of sea-sky-line in the image detected under video surveillance,a method for detecting sea-sky-line in the gradient domain based on the gray level co-occurrence matrix contrast filtering of the eccentric neighborhood is proposed.The method design uses two directional eccentric box filters to obtain the contrast gradient value map of the image,and uses a sliding window to search for the gradient maximum candidate position in the block image.According to the design strategy,a probability statistical model is constructed to extract the fitted straight line as the detection result.Simulation experiments show that the proposed method can improve the accuracy of detection in complex scenes with multiple objects and other interference under visual images,with a 7.69%improvement compared to similar methods.
万方数据
维普
