Research And Implementation On Medical Image Mosaic Based On ARM9

In recent 20 years, embedded system has been the hot topic in research area, and becomes another technologic way after the net communication technology. For the advantages of more compact, lower power consumption, more stabile and oriented to application, embedded system has been widely used in many fields. However, the usage of embedded-system in medical treatment and public hygiene is still rare. In fact, considering the special properties of embedded-system, the potential contribute can be expected. The purpose of this study is what deigns a embedded-system of Medical image mosaic based on ARM9 chip. this system can integrate local organic images into a fully part image with high resolution.At the beginning of this dissertation introduces embedded-system concept and reviews the background and current situation of research on image mosaic at home and abroad ,also elaborated the purpose and significance of this study. Then, this dissertation makes a guide of ARM structure and details two important driver design and implementation: the design of BootLoader on S3C2410 chip and USB Camera driver on Windows CE platform. About BootLoader, this dissertation discusses BootLoader stage 1. About USB Camera driver, this dissertation takes OV511 as a example to guide a implementation of USB Camera driver .Based on these, this dissertation mostly focus on image alignment and stitching algorithm. we uses SIFT(Scale invariant feature transform) to extract interests points from images, then matchs points and purifies matching points as the condition to solve transform parameters used The Nearest Neighbors and RANSAC algorithm. And transform Model uses the classical 8 parameter projection model. After image stitching, we take mutli-resolution Spline to eliminate the edge effect. Finally, the summary collects some problems about this study, these should be improved in behind works.