An image segmentation method for detecting objects in images with textural background

In this paper, we propose a novel method for detecting objects in images which have a textural background. We combine two filter banks to smooth the textural regions and preserve edges in the image then, we use Chan and Vese active contours without edges model. In fact our model is a modification of the above said active contours model and we make it suitable for detecting objects in the images which have textures. The image is passed through a symmetric bank of Gabor filters and filtered images that possess a significant component of the original image are then subjected to a bilateral filter. The output filtered images which are smooth will use as inputs of Multi channel C-V active contours model. Results of the proposed method are presented for different kind of images with different textures in background to demonstrate the robustness of this method.