Research On Compression Of Stereo Image

Stereo image technology is the field of images of one of the important trends. Stereo image technology has very broad application prospects, can be used in scientific research, military, educational, industrial, medical, commerce and many other areas. People can see a real sense with depth by the stereo image, just like see the real things. However, this increase in the price is realistic visual images of data than doubled the increase in single-aisle. So, The stereo image compression coding method is one of the hot areas. Although some of new methods have been raised, no one has been recognized by the experts of this field, became the standard compression method.A stereo image from two or more pieces of images on the composition of. These images are among the great binocular relevance, there are a lot of redundancy. So, these can be used to compress the stereo image. This paper propose a compression of stereo image, has some value for transmission and storage of stereo image.This paper first explain the characteristics of people's visual system and the reason of visual depth. From the research of human vision, extract the topological features of the left image. process the left image with Delaunay triangulation base on the feature points, match the feature points of the right image. And then morph the left image get the rebuild right image. Finally, a new stereo composed by the left image and right.Through experiments, according to data and subjective evaluation show that the method is effective. In the stereo image data compression at the same time, the depth of stereo image is still strong, to watch without discomfort, image features the outline of almost no losses.