ADAPTIVE VERGENCE FOR THE STEREO MATCHING OF OBLIQUE IMAGERY

The use of oblique imagery for the extraction of man-made features, su ch as buildings for cartographic applications, introduces several prob lems to the process of stereo matching that are not present when tradi tional near-nadir imagery is used. When oblique imagery is brought int o a collinear epipolar alignment, the ground-plane is usually not perp endicular to the camera axis, a simplifying assumption made with most near-nadir imagery. The obliquity of the ground-plane causes the great est problem for stereo matchers. To accommodate the larger range of di sparity, the stereo process must either increase its search window, or utilize an adaptive vergence mechanism to control the search window o n a pixel-by-pixel basis. Increasing the search window increases both the search time and the chance for aliasing to occur. We show that add ing an adaptive vergence mechanism to an existing stereo system provid es a substantial improvement in its ability to process oblique imagery .