The Design And Implementation Of A Stereo GUI System Based On Adaptive Windows

Stereo matching is one of the most active research areas in computer vision. The main problem in stereo matching by computing correlation or sum of squared differences (SSD) lies in selecting an appropriate window. The window size must be large enough to include enough intensity variation for reliable matching, but small enough to avoid the effects of projective distortion. For this reason, a window size must be selected adaptively depending on local variations of intensity and disparity.This paper designed a GUI (Graphic User Interface) system to implement our stereo algorithm based on adaptive window and several other traditional window-based algorithms. We designed a method to select an appropriate window by evaluating the local variation of the intensity and the disparity. This method estimated the disparity of every pixel by considering boundary, occlusion. However, by restriction of computation complexity, this paper only selects the size of the window.Furthermore, an evaluation model has been designed to compare the performance of our algorithm and traditional fixed window stereo algorithms. Qualitative and quantitative results obtained for image pairs show that the proposed algorithm outperforms most state-of-the-art local matching algorithms without efficiency loss as global algorithms.Our system is designed to assess the different components and design decisions made in stereo algorithms. Thus, we build a matlab implementation by using process-orient design technology for its modality and extendability. The modality guarantees the stand-along test for each block of algorithm. And the extendability let us could absorb other algorithms (such as global algorithms) into this system. Also, in the future, we could develop other applications in this area such as multiview image synthesis and coding by the extendability of our system.