The Study on Video Stabilization Based on Motion Compensation

Autor: Jie-Ru Chen, 陳玠儒
Rok vydání: 2009
Druh dokumentu: 學位論文 ; thesis
Popis: 97
The image and video process algorithm have been research for many years. In our life, we are shown much application not only for civil use but also for military use the technology. Examples of this are segmentation, tracking, counting, image enhance, image registration and etc…. The factor, such as wild, earthquake and malicious damage may make vibrate to capture device of video and cause the recorded video frame unstable. Frame vibration is not comfortable for human visual perception and video stabilization is also very important for the recorded data that is further utilized for computer vision application. The most image process algorithm are not apply for unstable frame and thus may increase the computation time or cause the error which means miscarriage of justice or system crash. In this paper, we propose a real-time video stabilization, if the first frame is instable frame; we can real-time adjustment of shaking the camera shake caused. The basic idea is to analyze the frame, we set Fixed nine observation blocks in the frame, and then, we use histogram and last block selected the best block is the feature block. Make use the feature block to get the moving vector, and uses the vectors regulate the frame. Experimental results show that the proposed algorithm can effectively reduce the frame vibration in real-time for the general vibration situations of moderate degree and compensate 20 pixels.
Databáze: Networked Digital Library of Theses & Dissertations