Earticle

초록

영어

Panoramic image registration is the core content of virtual three-dimensional system based on images. According to the shortage of fisheye images registration using the traditional image registration method, this paper put forward strategy of panoramic image registration, which was divided into three steps: project the fisheye images on the sphere, extract and match the feature points, transform the coordinates of panoramic images. The focus was improving the SIFT algorithm and RANSAC algorithm for image feature exaction and matching, whose target was to meet the demand of the fish-eye image registration. Finally, experiments were designed to verify the effectiveness of supposed algorithm and fisheye image registration, and the future direction of panoramic images registration was also pointed out.

목차

Abstract
 1. Introduction
 2. Spherical Projection of Fisheye Image
 3. Image Feature Extraction and Matching
  3.1. Feature point extraction based on SIFT
  3.2. Feature point matching based on RANSAC
 4. Coordinate Transformation of Panoramic Image
 5. Experiment Results
 6. Conclusion
 References

키워드

저자

• Xiaohui Wang [ Postdoctoral Mobile Research Station of Management Science and Engineering, North China Electric Power University, Beijing 102206 P. R. China ]
• Kehe Wu [ School of Control and Computer Engineering, North China Electric Power University, Beijing 102206 P. R. China ]
• Shengzhuang Wang [ School of Control and Computer Engineering, North China Electric Power University, Beijing 102206 P. R. China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Signal Processing, Image Processing and Pattern Recognition

• 간기

  격월간

• pISSN

  2005-4254

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605