Open AccessMonocular vision based pose measurement of 3D object from circle and line features
Author(s) -
Weixin Fu,
Baofu Li
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/677/5/052093
Subject(s) - artificial intelligence , computer vision , monocular vision , feature (linguistics) , projection (relational algebra) , constraint (computer aided design) , computer science , object (grammar) , focal length , line (geometry) , monocular , pose , 3d pose estimation , mathematics , algorithm , geometry , optics , physics , philosophy , linguistics , lens (geology)
An approach for estimating object pose from establishing equations between the features and their projection is proposed. In the case that the focal length of camera is unknown, this paper establishes the mathematical model between the circular feature of 3D Object and its projection through a series of basic changes, and adds common linear feature to provide spatial geometric constraint for pose measurement, then provides the nonlinear equations with seven unknown parameters. The equations are transformed into objective function, then the algorithm of the One-Dimensional search is applied to obtain the optimal value of the focal length. The other parameters can be calculated by the focal length. Experimental results verify the validity and accuracy of the method.