Based On The Three-dimensional Reconstruction Of The Visual Navigation Error Analysis And Post-processing Algorithms Research

This paper made a summary of my academic achievement during the graduate reach in the field of Autonomous Land Vehicle (ALV), which is an intelligent mobile robot that can run autonomously and continuously on road or cross-country. From the specialties of the stereovision applied in vision-based navigation, we discussed a algorithm for error analysis and post-disposal to reconstruction of 3D scene from the disparity map.The prophase algorithms of stereovision were presented at first involved: camera calibration, image rectification and stereo matching. The camera' s interior and exterior parameters are acquired by the calibration. Through image rectification, the corresponding epipolar lines are made parallel the horizontal scan lines of image pairs. Through stereo matching, we gain the dense disparity map from two 2D images of the scene fast and accurate. These algorithms are indispensable preparation to the 3D reconstruction.On the base of existing solution to reconstruct 3D scene from the disparity map .we analysis various error sources and character of binocular stereo system on pinhole camera model for estimation of the depth .As a result various errors are compensated. Thus, more practical and robust reconstruction result was obtained. At the same time, we discussed a algorithm for post-disposal to reconstruction of 3D scene.The experiment results proved that our algorithm can meet the vision-based navigation' s request for real-time and validity.