Chen, Y.-T.Y.-T.ChenLin, T.-P.T.-P.LinFENG-LI LIAN2020-06-112020-06-112017https://www.scopus.com/inward/record.uri?eid=2-s2.0-85040589252&doi=10.1109%2fIIAI-AAI.2017.105&partnerID=40&md5=dedaf4c4050d476465a2b0a5c534321eEnvironment reconstruction from a monocular camera has been a popular research topic. This technique can be applied to automatic navigation, environment exploration and automatic obstacle avoidance. This paper proposes stereo matching and aims to match the low gradient area around the high gradient area between two images correctly by using the epipolar geometry. The experimental results demonstrate that the proposed regularization algorithm can eliminate most of the noises and reconstruct a more clearly point cloud. © 2017 IEEE.Camera pose estimation; Monocular simultaneous localization and mapping; Nonlinear optimization; Regularization algorithm; Semi-dense map reconstructionCameras; Nonlinear programming; Optimization; Camera pose estimation; Map reconstruction; Non-linear optimization; Regularization algorithms; Simultaneous localization and mapping; Stereo image processingconference paper10.1109/IIAI-AAI.2017.1052-s2.0-85040589252