一種二階段方式校正影像接圖失真的方法

A Two-Step Rectification Method for Image Registration

舒忠(Jung Shu)

指導教授 ： 林啟芳

元智大學/資訊學院/資訊工程學系/碩士(2000年)

摘要

本研究擬提出一種以二維變形為基礎的技術，來矯正影像接圖失真的問題。此項技術包含兩個部份：(1)格狀型的2-D變形，用以矯正接合影像的幾何失真及; (2)亮度差異值的消除，用以矯正接合影像間的亮度失真。在目前的接圖技術中，常常會有不良的接圖現象存在於局部性區域中，而接合影像間的亮度偏差也是需要克服的問題之一。為了解決這類問題，本研究擬提出一矯正方法，利用二維變形技術來克服接合影像間的幾何失真和亮度失真，以提供更真實的影像接合結果。

關鍵字

影像接合 ； 校正 ； 幾何失真 ； 亮度不連續

並列摘要

This study presents a 2-D deformation-based technique for the rectification of image registration. In most current panorama stitching methods, there will invariably exist a serious problem presented in the final registration result. The problem is mainly caused by the mis-registrations and discontinuities in intensity between the registered images. To reduce the problem, a two-step rectification method utilizing the traditional 2-D deformation technique is developed for correcting the distortions of geometric and intensity discontinuities. The experimental results revealed in this study demonstrate the effectiveness of the proposed method

並列關鍵字

Image Registration ； Rectification ； Geometric Distortion ； Intensity Discontinuity

參考文獻

[1]R. Szeliski and H. Y. Shum, “Creating full view panoramic image mosaics and environment maps,” Proceedings SIGGRAPH’97, pp. 251-258, 1997.

[2]R. Szeliski, “Video mosiacs for environments,” IEEE Computer Graphics and Applications, pp. 44-54, 1996.

[3]G. Wolberg, Digital Image Warping, IEEE Computer Society Press, Los Alamitos, California, 1990.

[5]Q. Tian and M. N. Huhns, “Algorithms for subpixel registration,” Computer Vision, Graphics, and Image Processing, Vol. 35, pp. 220-223, 1986.

[6]H. Y. Shum and R. Szeliski, “Constructing and refinement of panoramic mosaics with global and local alignment,” Sixth International Conference on Computer Vision, pp. 953—956, 1998.