To search, Click below search items.


All Published Papers Search Service


Rotation Estimation Between Two Images Using Feature Analysis


Mohammed I. Abd-Almajied, Loay E. George, Kamal M. Abood


Vol. 19  No. 1  pp. 22-27


Image matching, Analysis, and registration required two images of the same aligned scene. Scaling, rotation, and translation are the most artifacts which a companied with them, so the image transformation estimation is the most effective step in image matching. A proposed algorithm has been done to estimate the rotation between images of the same scene. Features or segments play an important part in evaluating the appropriate degree of rotation and do not need for control points or image regression for finding transformation. These features have different aspect in different images and influence on the image matching. A computer simulation was applied on images for angles of (1o-90o). Different types of images may affects the goodness of rotation estimation. Satellite images of planet differs from others in their contrast and resolutions. Two different images have been tested for the rotation process.


reference point, feature, center feature.