AUTHOR=Zhang Yawei , Gu Jin , Rao Tao , Lai Hanrong , Zhang Bin , Zhang Jianfei , Yin Yanxin TITLE=A Shape Reconstruction and Measurement Method for Spherical Hedges Using Binocular Vision JOURNAL=Frontiers in Plant Science VOLUME=Volume 13 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.849821 DOI=10.3389/fpls.2022.849821 ISSN=1664-462X ABSTRACT=The center coordinate of a spherical hedges and its radius are the basic phenotypic parameters for automatic pruning. A binocular vision-based shape reconstruction and dimension measurement system for front-end vision information acquisition is built in this paper. A parallel binocular vision system is used as detector and image processing functions are called to obtain 2D image extraction of target spherical hedges, and then stereo rectification algorithm is conducted to keep two cameras to be perfectly parallel. Moreover, an improved semi-global block matching(SGBM) algorithm is studied and used to obtain disparity map. According to the disparity map and parallel structure of binocular vision system, the 3D point-cloud of target are obtained. On this basis, the center coordinate of a spherical hedges and its radius can be measured. Laboratory and outdoor tests on shape reconstruction and dimension measurement are conducted. In detection range of 2000-2600mm, laboratory test demonstrates that the average absolute error and average relative error of standard spherical hedges radius are 1.58 mm and 0.53%, respectively; the average location deviation of center coordinate of spherical hedges is 15.92 mm in range of 2000-2600 mm. Outdoor test shows that the average absolute error and average relative error of spherical hedges radius by the proposed system are 4.02 mm and 0.44%, respectively; the average location deviation of center coordinate of spherical hedges is 18.29 mm. This study provides important technical support for phenotypic parameter detection in the study of automatic trimming.