Proposal of terrain mapping under extreme light conditions using direct stereo matching methods

One of the biggest problems in applying stereo vision techniques for field robotics is how to acquire information even in extreme sunlight contrast conditions. In this paper instead of using existing lighting enhancement methods such as Exposure Fusion to increase the texture of 2D image, the stereo matching is directly done using the images grabbed with multiple exposures through two methods. To each pixel of the matching image, its matching vector is directly aggregated or aggregated by weight using the matching vectors of the images grabbed with multiple exposures with the same pixel position. To evaluate the performance of the proposed methods, two different stereo matching algorithms were used: a local, window-based method and semiglobal method. Through the experiments in laboratory and outdoor, it is verified with the proposed methods, more valid 3D points can be obtained and the terrain map could be reconstructed more accurately. Especially when the local window-based method was used, the proposed methods perform much better.