计算机视觉
结构光
人工智能
校准
特征(语言学)
交叉口(航空)
直线(几何图形)
视野
平面的
计算机科学
平面(几何)
点(几何)
光场
机器视觉
摄像机切除
光学
数学
物理
几何学
计算机图形学(图像)
工程类
语言学
统计
哲学
航空航天工程
摘要
A calibration method for line structured light vision sensor with large Field of View(FOV) by using 1D target with unconstrained combination is proposed.Several 1D targets,each of which has at least 3 feature-points aligned on a feature-line,are combined freely according to the measuring range of the sensor.All primitive parameters of the line structured light vision sensor are determined with observation of the target shown at a few(at least two) different viewpoints.The intrinsic parameters are solved using the invariability of angle between every two one-dimensional targets in different views.The coordinates of feature-points in the camera coordinate frame are computed in accordance with the property of the vanishing point and the known distance between the feature points.The coordinates of the control points which are the intersection points of the light plane and the feature lines are obtained based on the cross-ratio invariability.The light plane is fitted with at least three nonlinear control points.The experimental results show that the proposed approach can achieve the same calibration accuracy as the planar-target-calibration approach.Compared with existing methods,the proposed method is fast with high accuracy,which is suitable for the field calibration of line structured light vision sensor with large FOV.
科研通智能强力驱动
Strongly Powered by AbleSci AI