Petroleum Pipeline Interface Recognition and Pose Detection Based on Binocular Stereo Vision

Liquified natural gas (LNG) manipulator arms have been widely used in transportation. However, the automatic docking technology of LNG has not yet realized. The first step is to identify and locate target estimate its pose. This work proposes a petroleum pipeline interface recognition pose judgment method based on binocular stereo vision for arms. proposed three main steps, including detection, 3D information acquisition, plane fitting. First, segmented by using color mask. Then, space Hu moment are obtain pixel coordinates contour center interface. semi-global block matching (SGBM) algorithm depth an image. Finally, fitting point estimation random sample consensus (RANSAC) proposed. performs measurement accuracy verification experiment verify method. experimental results show that distance error more than 1% angle less one degree. meets requirements subsequent docking, which proves feasibility provides data support