The Research Of Color Image Registration And Stitching Technology Based On CSIFT

Image mosaic is a kind of image processing technology, stitching a series of imageswhich are in the same scene and have a number of overlap, together into a new highresolution image with larger size, wider viewing angle and almost no distortion from theoriginal image. Image mosaic is a research hotspot in the area of computer vision, imageprocessing and computer graphics. Image mosaic can be used to create images of largeviewing angle. It is also widely used in the field of virtual reality, medical image processing,remote sensing technology and military.The quality of image mosaic depends on the accuracy of image matching. Currently,image matching based on feature points is an important research area of graphic field.Therefore, the study of images matching and mosaic based on the feature points has animportant research value.Scale Invariant Feature Transform (SIFT) is a kind of stable stereo matching algorithmbased on feature points. The study of this paper mainly focus on the improvement of SIFTalgorithm. Color information of the image is added. Alarger comparison range is also selectedto find the feature point with more distinguished feature.Meanwhile, before calculating the transform matrix, the concept of image entropy isintroduced. By comparing mutual information of image entropy, together with usingProgressive Sample Consensus （PROSAC) algorithm, the quality of the feature point isfurther improved. The accuracy of feature points matching is ensured. This high accuracymatching result will also lead to a reliable outcome after transforming matrix.Finally, in the image mosaic process, IHS transform based improved image fusionalgorithm is used. The weight of brightness information of the original images is added. Thestitching line problem caused by the difference of image brightness is solved to some extent.Some satisfied result of image mosaic is achieved.