The Key Technology And Implementation Of Image Mosaic Based On Matching

With the development of digital image processing technology, image stitchingtechnology research has become a more and more important field in computer imageprocessing. Due to the limited conditions of image acquiring device, such as poorresolution, image distortion, the obtained images are usually unable to achieve therequirements of precision or viewing angle range. Therefore, it is necessary to adoptimage stitching technology to splice a group of the images with overlapping area into animage which meet the requirements of the viewing angle. The emergence of imagestitching technology allows the acquisition of high-resolution panoramic images hasbecome a possibility. Thus, image stitching technology to become a research hotspot inrecent years.The image stitching technology is widely used in the field of aviation, medicalimage analysis, and virtual reality. The key of image stitching technology is to find theoverlap region of adjacent images. Therefore the main work of the image mosaictechnology is to match similar regions between two adjacent images. Now，There aretwo commonly used matching algorithm, which is the algorithm based on the area, andthe algorithm based on key point.The main content of this paper is as follows:: First, it introduces the basic processesand key technologies of image stitching, and makes an analysis of key technology andclassic method used in each step of image stitching. Second, it makes a research of keymatching technology in image stitching process, This paper studies Area-basedmatching techniques in image matching, the implementation of the technology ofmatching techniques is given too. Also, with the matched images, it introduces themethod of image fusion. In this paper, we use the gauss weighting transition algorithmto stitch images smoothly. Besides, the Key-point based matching technique ispresented too, which use the canny operator-based image stitching technologies. Andthe result is outlined to prove the effectiveness and robustness of the algorithm. Acomparison between the two methods mentioned above is analyzed and presented.Finally, we make a summarization of the algorithms in this paper, and an analysis of the shortcomings of the existing image stitching technology and the difficultproblems in image mosaic，and points out the direction of further research.