efficient image stitching in the presence of dynamic objects and structure misalignment高效的图像缝合在动态对象的存在和结构失调.pdfVIP

Journal of Signal and Information Processing, 2011, 2, 205-210 205 doi:10.4236/jsip.2011.23028 Published Online August 2011 (http://www.SciRP.org/journal/jsip) Efficient Image Stitching in the Presence of Dynamic Objects and Structure Misalignment 1 2 1 1 Chao Tao , Hanqiu Sun , Changcai Yang , Jinwen Tian 1The Institute for Pattern Recognition and Artificial Intelligence, Huazhong University of Science and Technology, Wuhan, China; 2The Department of Computer Science and Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong (China). Email: kingtaochao@126.com Received March 9th, 2011; revised April 20th, 2011; accepted April 29th, 2011. ABSTRACT This paper presents a new method for simultaneously eliminating visual artifacts caused by moving objects and struc- ture misalignment in image stitching. Given that the input images are roughly aligned, our approach is implemented in two stages. In the first stage, we discover motions between input images, and then extract their corresponding regions through a multi-seed based region growing algorithm. In the second stage, with prior information provided by the ex- tracted regions, we perform a graph cut optimization in gradient-domain to determine which pixels to use from each image to achieve seamless stitching. Our method is simple to implement and effective. The experimental results illus- trate that the proposed approach can produce comparable or superior results in comparison with state-of-the-art me- thods. Keywords : Image Stitching , Motion Estimate, Region Growing, Graph Cut 1. Introduction it does not establish the correspondence for regions re- lating to the same object. Thus, amb