A Hybrid Image Segmentation Approach Based on Mean Shift and Fuzzy C-Means

Image segmentation is an important task in many applications. For large-scale, general image dataset, however, there are the competing requirements, including not making complex prior assumptions about the scene, having fast speed and good segmentation quality. In this paper, a hybrid approach for image segmentation is presented that incorporates two famous methods, i.e. Mean Shift (MS) and Fuzzy C-Means (FCM). The first stage extracts many regions by MS approach, which provides an initial over-segmentation. The second stage groups together these primitive regions into meaningful objects to produce the final segmentation results by FCM. The proposed approach is efficient while provide good segmentation performance. The experimental results demonstrate the effectiveness of the proposed approach.