Finding Common Points From Digitized Video Images

This paper reports a method which was developed for finding common points from digitized video images. The process of finding common points is semiautomatic, guided by the user and is based on the least squares image matching method. To find a pair of control points from two images, the location of the first point has to be pointed out manually from the first image and then the corresponding point is automatically located from the second image. As an application a large image mosaic and a smaller stereo image mosaic were produced using consecutive images.