Parallel Implementation Of Stereo Matching In Drive Assistance System

The field of computer vision, Stereo vision-based drive assistance is the rise of research focus. Binocular stereo vision can directly imitate the human eye and human visual perception of three-dimensional, computer vision is the core topic of study. The ultimate goal of technology is recover three-dimisional enviroment from two-dimision image by computer. In our project we use two cameras to achieve three-dimision information around vehicle. Stereo matching is one of the most active research areas in computer vision. It is also the key and the most difficult problem in stereo vision. The correspondence problem is naturally stated in terms of energy minimization. Once the global energy has been defined, a variety of algorithms Can be used to find a minimum, including dynamic programming, graph cut, belief propagation and so on. Its goal is to generate dense depth mapDynamic programming is a classical global optimization algorithm for stereo matching, Which uses smoothness constraints to optimize correspondences in each scan-line by finding the minimum cost path through a disparity space image. Besides this, because of high precision and more delicate, belief propagation has become the most popular global stereo matching algorithm, however, stereo match algorithm still has some shortcoming, it is difficult to achieve real time, robustness and precision at the same time, which is our project DAS require. So this paper studies paper as follows:This paper describes an efficient parallel implementation of dynamic programming and belief propagation algorithms on a cell processor that speeds up stereo image analysis. Furthermore, we define limitations of the Cell architecture for these applications. For evaluation, we use synthetic and real-world image sequences. Real-world images are typically degraded by various types of noise, changes in lighting, differing exposures, and so on. Sobel edge and residual images can improve the stereo matching results compared to the use of original real-world images; our results show that a cell processor also reduces running time for these processes.