[1]黄金鑫,赵 勇.一种改进的未知环境无人机三维地图实时创建方法[J].机械与电子,2015,(01):76-80.
　HUANG Jinxin,ZHAO Yong.An Improved Real-time 3D Reconstruction Method of Unknown Environment Based on UAV[J].Machinery & Electronics,2015,(01):76-80.
点击复制

一种改进的未知环境无人机三维地图实时创建方法
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
期数:
2015年01期
页码:
76-80
栏目:
智能工程
出版日期:
2015-01-20

文章信息/Info

Title:
An Improved Real-time 3D Reconstruction Method of Unknown Environment Based on UAV
文章编号:
1001-2257(2015)01-0076-05
作者:
黄金鑫; 赵 勇
(西北工业大学航空学院,陕西 西安 710072)
Author(s):
HUANG Jinxin;  ZHAO Yong
(School of Aeronautics,Northwestern Polytechnical University, Xi'an 710072,China)
关键词:
RGB-D传感器;  摄像机标定;  点云;  地图构建
Keywords:
RGB-D sensor;  camera calibration;  point-cloud;  mapping
分类号:
TP391.4
文献标志码:
A
摘要:
提出一种针对RGB-D传感器获取未知环境三维地图的方法。该方法将RGB-D传感器获得的数据信息实时地转化为三维点云地图。在扫描环境之前,引入径向畸变和切向畸变模型,完成对RGB-D传感器彩色摄像头和深度摄像头的标定,获得摄像机参数; 同时,对RGB-D传感器红外模式和深度模式下的图进行偏差矫正,提高三维点云计算的精度。实时地图构建时,利用投影和反投影将获得的彩色图片和深度图片转换成局部三维地图; 在RGB-D传感器移动过程中,实时提取RGB图片上的特征进行位姿估计和关键帧处理,从而获得整个地图。实验表明,
Abstract:
This paper presents a mapping method of RGB-D sensor in unknown three-dimension environment. This method converts the data obtained by RGB-D sensor to three-dimension point-cloud map. Before the scanning, we introduce the radial distortion model and the tangential distortion model calibrating the RGB camera and Depth camera of the sensor to get the camera parameters; at the same time, combining the color image and the depth image to get the local point-cloud map by projection and back-projection; at the same time, we make a deviation correction of the images obtained under the infrared mode and depth mode of the RGB-D sensor, thus the accuracy of the three-dimension point-cloud can be improved. When mapping in real-time, we convert the RGB images and the depth images to the local three-dimension map by projection and back-projection; during the movement of the RGB-D sensor, the position estimation and key frame processing are executed using the features on color images to get the whole map. This method is applicable to cluttered unknown environment.

参考文献/References:

[1] Endres F, Hess J, Engelhard N, et al. An evaluation of the RGB-D SLAM system[C]// 2012 IEEE International Conference on Robotics and Automation(ICRA), 2012:1691-1696.
[2] Besl P J, McKay N D.A method for registration of 3-D shapes[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992, 14(2):239-256.
[3] Huang A S, Bachrach A, Henry P, et al.Visual odometry and mapping for autonomous flight using an RGB-D camera[C]//International Symposium on Robotics Research(ISRR),2011:1-16.
[4] Neumann D, Lugauer F, Bauer S, et al. Real-time RGB-D mapping and 3-D modeling on the GPU using the random ball cover data structure[C]// 2011 IEEE International Conference on Computer Vision Workshops(ICCV Workshops),2011:1161-1167.
[5] Zhang Zhengyou. A flexible new technique for camera calibration[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[6] Smisek J, Jancosek M, Pajdla T. 3D with Kinect[M]//Consumer Depth Cameras for Computer Vision. London:Springer, 2013:3-25.
[7] Shi J, Tomasi C.Good features to track[C]//1994 IEEE Computer Society Conference on Computer Vision and Pattern Recognition,1994:593-600.

备注/Memo

备注/Memo:
收稿日期:2014-09-18
作者简介:黄金鑫(1992-),女,湖北随州人,硕士研究生,研究方向为计算机图形图像处理、模式识别; 赵 勇(1992-),男,湖南衡阳人,硕士研究生,研究方 向为机器人自主导航、图形图像处理。
更新日期/Last Update: 2015-01-25