Level set evolution with locally linear classification for image segmentation

This paper presents a novel local region-based level set model for image segmentation. In each local region, we define a locally weighted least squares energy to fit a linear classifier. With level set representation, these local energy functions are then integrated over the whole image domain to develop a global segmentation model. The objective function in this model is thereafter minimized via level set evolution. In this process, the parameters related to the locally linear classifier are iteratively estimated. By introducing the locally linear functions to separate background and foreground in local regions, our model not only achieves accurate segmentation results, but also is robust to initialization. Extensive experiments are reported to demonstrate that our method holds higher segmentation accuracy and more initialization robustness, compared with the classical region-based and local region-based methods.

► Develop a novel level set model for image segmentation. ► Define a locally weighted least squares energy to fit a linear classifier. ► Introduce a discriminative model to level set framework. ► Demonstrate its superior in terms of accuracy and robustness.