A novel active contour model driven by J-divergence entropy for SAR river image segmentation

It is of great difficulty to utilize the existing active contour models (ACMs) to achieve accurate segmentation of synthetic aperture radar (SAR) river images. To address this problem, a novel ACM driven by J-divergence entropy is proposed. The external energy constraint term of the proposed model is defined by the J-divergence entropy, which differs from those of many existing ACMs defined by the Euclidean distance. Moreover, the median absolute deviations of pixel grayscale values inside and outside the curve are utilized as energy weights, which can adaptively adjust proportions of region energies inside and outside the curve, leading to the improvement in segmentation efficiency. Experiments are performed on a large number of SAR river images, and the results demonstrate that, compared with the existing ACMs, the proposed model shows clear advantages in terms of both segmentation performance and segmentation efficiency.