In this study, we develop an unsupervised coarse-to-fine video analysis framework and prototype system to extract a salient object in a video sequence. This framework starts from tracking grid-sampled points along temporal frames, typically using KLT tracking method. The tracking points could be divided into several groups due to their inconsistent movements. At the same time, the SLIC algorithm is extended into 3D space to generate supervoxels. Coarse segmentation is achieved by combining the categorized tracking points and supervoxels of the corresponding frame in the video sequence. Finally, a graph-based fine segmentation algorithm is used to extract the moving object in the scene. Experimental results reveal that this method outperforms the previous approaches in terms of accuracy and robustness.
Selecting regions of interest (ROI) of the medical images is an important task in medical image processing. Manual selection of ROIs serves as the main method for single images and it has a high accuracy. However, it will become infeasible to manually segment ROIs on a large number of images. Observing this problem, this paper proposes a fast and accurate segmentation method to obtain ROIs on a batch of medical images. Firstly, we segment the standard brain image St which has not been injected with tracer. Secondly, we use a B-Spline elastic registration method to get the inverse-registration parameters. Thirdly, we get the template image Te with the registration parameters. Finally, we search the target region by template matching. Experimental results show that the proposed method performs well on medical image segmentation.