An active contour model based on Jeffreys divergence and clustering technology for image segmentation

Classic active contour models (ACMs) generally implement Euclidean distance to measure the gap between true image and fitted one, which may cause issues such as edge leakage and falling into false boundary. In addition, some existing ACMs are sensitive to noise and different initial contours. To resolve these problems, this study raises an ACM based on Jeffreys divergence (KJD) and clustering technique for image segmentation. Firstly, the K-medoids clustering algorithm is deployed to cluster the foreground and background pixels into two sets, which forms original data -driven term and is further embedded into the theory of Jeffreys divergence to formulate a KJD energy. Next, a regularization function regularizes the ranges of optimized data -driven term and level set function respectively, which essentially produces a more stable and robust evolution environment and improves segmentation precision. In contrast with fitting function -based and recently developed ACMs on three types of images, KJD model not only raises segmentation precision, but also reduces computation expense. Experimental outcomes also verify that this model can resist different noise within limits and adapts to various initial contours.