Edge accurate dense disparity estimation is of great importance to applications, such as augmented reality where the geometry relationships among objects in a scene need to be presented precisely. Binocular stereo is a promising approach to get 3D depth information of a real scene from 2D images, but faces the issue of difficult to achieve high accurate disparity edges with reasonable computation complexity. A depth edge preserving dense stereo matching method is presented in this paper aiming to alleviate this problem. By taking a sparse-to-dense route for disparity estimation, depth edges corresponding to object boundaries are distinguished from the texture edges based on sparse disparities which can be obtained efficiently. With a designed disparity filling strategy, these extracted edges are used to refine the dense disparities and align the depth discontinuity edges with corresponding object boundaries. Disparities obtained by this work can faithfully conform to the scene geometry recorded in the input images only with a relative small increase about 13% in computational complexity. The effectiveness of the proposed method is verified through experiments and contrastive analysis.
In this paper a new pattern or feature abstraction algorithm was developed simulating our human eyes scan on situation leapingly and focus its attention only to limited space, in order to make a physiognomic analysis of the remote sensing images of the earth's surface, in the end to acquire a description like that in where the foothill, the forest, the desert or the alluvial pie slice was, and so on. At first a size-changeable and edge-fuzzy window was designed to get many samples of the earth's surface through sliding around the image, all these samples served for the learning of a Support Vector Machine model, which was designed to make pattern's classifications. This process was repeated in different area, with different sampling size, to different pattern and lasting different times. Once some distinct local patterns were found and mastered, a self-organizing of comparability assembling will happen based on the similarity of some types of local patterns to form a holistic description or understanding of the remote sensing image. Our aim was to compartmentalize the image by physiognomic features. At the end of this paper the results of classification experiment and application of this method to some actual visible light images were presented. This method was suitable to extend to other pattern recognition problems with texture property.