一種矯正影像接圖失真的三維變形技術

A 3-D Deformation Based Technique for Rectifiying Registration Errors

林學億(Hsueh-Yi Lin)

指導教授 ： 林啟芳

元智大學/工程學院/電機與資訊工程研究所/碩士(1998年)

摘要

本研究擬提出一種以三維變形為基礎的新技術來矯正影像接圖失真。此項技術包含兩個部份：(i) 二維變形技術，用以矯正接合影像的幾何失真及; (ii) 可變形的亮度參數曲面(第三維)，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案利用三維變形技術來克服接合影像間的亮度失真和幾何失真，以提供更真實的接合環場影像。由於本研究係透過特徵點來矯正影像的接痕現象，因此特徵點的擷取方法在本研究中也一併被提出。本研究的目的著重於接圖失真的彌補，以產生接近於完美無縫的環場影像。

關鍵字

三維變形 ； 影像接圖 ； 二維變形 ； 幾何失真 ； 亮度失真 ； 接圖失真

並列摘要

In this study, we develop a novel 3-D deformation-based technique that can be utilized to rectify the distortion caused by image registration. The technique is mainly comprised of the following two processes: (1) the process of rectifying edge discontinuities, and (2) the process of rectifying intensity discrepancies. In current panorama stitching techniques, there will invariably be geometrical discontinuities and intensity discrepancies, both are caused by registration errors. To reduce those problems, a rectification method utilizing 3-D deformation technique is proposed, which can greatly improve the quality of the registering result. A coarse-to-fine stitching scheme is also presented to generate seamless panoramas, thereby enabling the proposed rectification method to be integrated into the current stitching techniques. Experimental results will reveal the advantages of the proposed method.

並列關鍵字

3-D Deformation ； Image Registration ； 2-D Warping ； Geometrical Discontinuities ； Intensity Discrepancies ； Registration Errors

參考文獻

[2] T. Beiter and S. Neely, “Feature-based image metamorphosis,” Proceedings SIGGRAPH’92, pp. 35-42, 1992.

[4] R. Jain, R. Kasturi, and B. G. Schunck, Machine Vision, McGraw-Hill, Inc., New York, U.S.A., 1995.

[6] H. C. Huang, S. H. Nain, and Y. P. Hung, “Disparity-based view morphing─a new technique for image-based rendering,” Proceedings of VRST’98, pp. 9-16, Taipei, Taiwan, November, 1998.

[7] 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

[10] R. Y. Tsai, “A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses,” IEEE Journal of Robotics and Automation, pp. 323-344, August 1987.