Panoramic image patching algorithm based on global optimization

被引:0
作者
Gong, Yan [1 ]
Xie, Hong [1 ]
Xie, Wu [1 ]
机构
[1] Information and Communication Engineering, Harbin Engineering University, Harbin
来源
Journal of Information and Computational Science | 2015年 / 12卷 / 14期
关键词
Global optimization; Panoramic image stitching; Time and space domain patch;
D O I
10.12733/jics20150019
中图分类号
学科分类号
摘要
Considering the process of panoramic image stitching, there are problems such as loss of image information, redundant objects, scene repeated, etc. In this paper, we propose the space-time domain patching algorithm based on global optimization of panoramic image stitching. It uses the SFM algorithm to get the correspondence position relation between the target area and the reference area. And the visual consistency of the image patches is improved by constructing the MRF global optimization model. By increasing the reference sample search, to improve the robustness of the algorithm; by rationally defining the label set and setting up the energy consumption function, the time and space domain patch are organically unified for the same process. Through the experiment, the proposed algorithm can modify the existing panoramic stitching images, optimize and realize the natural transition, achieve good visual effect and strong robustness. Copyright © 2015 Binary Information Press.
引用
收藏
页码:5523 / 5530
页数:7
相关论文
共 9 条
  • [1] Li Z., Hu W., A novel method for exemplar-based image inpainting, Journal of Information and Computational Science, 9, 3, pp. 761-769, (2012)
  • [2] Komodakis N., Tziritas G., Image completion using efficient belief propagation via priority scheduling and dynamic pruning, IEEE Transactions on Image Processing, 16, 11, pp. 2649-2661, (2007)
  • [3] Zhang G., Qin X., Hua W., Wong T., Et al., Robust metric reconstruction from challenging video sequences, Proc. Computer Vision and Pattern Recognition, pp. 1-8, (2007)
  • [4] Yang Y., Zhu Y., Peng Q., Image completion using structural priority belief propagation, Proc. of the 2009 ACM Multimedia Conference, pp. 717-720, (2009)
  • [5] Li S., Markov Random Field Modeling in Image Analysis, (2009)
  • [6] Gallup D., Frahm J., Mordohai P., Et al., Real-time plane-sweeping stereo with multiple sweeping directions, Proc. Computer Vision and Pattern Recognition, pp. 1-8, (2007)
  • [7] Agarwala A., Dontcheva M., Agrawala M., Et al., Interactive digital photomontage, ACM Transaction Graphics, 23, 3, pp. 294-302, (2004)
  • [8] Amirshahi H., Kondo S., Ito K., Et al., An image completion algorithm using occusion-free images from internet photo sharing sites, IEICE Trans. Fundamentals, E91-A, 10, pp. 2918-2927, (2008)
  • [9] Wu Y., A novel underwater image mosaicing approach based on improved SURF, Journal of Information and Computational Science, 12, 4, pp. 1381-1389, (2015)