Synthesizing the Whole View of a Large-Scale Object

Abstract

This paper presents the effective method for the whole view synthesizing of the large-scale object. This synthesizing method is developed based on the image registration approach. The foreground object is extracted from each image by color image processing. The information of the feature points of an object is obtained from a foreground image and establishing the correspondence among views. The translation and rotation parameters between two successive images are estimated by the computed camera calibration parameters. The second image is transformed by using these parameters and combine to first image. The integrated image is obtained by combining the overlapping parts of these images. A whole view of the large scale object is reconstructed image by applying the above process successively. Preliminary investigations with sampled images of a real scene have been done to confirm the effectiveness of the proposed method.

DOI: http://dx.doi.org/10.11591/ij-ict.v1i1.597