Abstract:
Stereo vision has traditionally been and continues to be one of the most extensively investigated topics in computer vision. Since stereo can provide depth information, it has potential uses in many visual domains such as autonomous navigation, 3D reconstruction, object recognition and surveillance systems. At present, few high-performance implementations of stereo vision algorithms exist. The key challenge in realizing a reliable embedded real-time stereo vision system is keeping the balance of execution time and the quality of the matching results. In this paper we have designed a real-time stereo vision model based on adaptive chromosome aggregation approach. When performing cost aggregation, the support from an adjacent pixel is valid only if such pixel has same variation. The way to choose proper support is a key factor of the correlation technique. For this purpose, an adaptive support weight (AW) algorithm is proposed to carry out aggregation on the appropriate support. This adaptive support weight approach begins from an edge-preserving image smoothing method called bilateral filtering. It unites gray levels or colors based on both their geometric closeness and their photometric similarity and prefers near values to remote values in both domain and range. The final disparity selection in our proposed method is performed with the help of the genetic algorithm which is an optimization technique that helps to select the best disparity value for further processing and results proved that our method is very efficient with 80% of bad pixel error reduction compared with other state of art algorithms and it attains 49% of PSNR .
