摘要
Kinect相机不仅能够获取彩色图像,还可以得到相应的深度信息。本文探讨了一种通过Kinect相机来实现里程计的方法,该方法由Kinect相机获取周围环境的连续帧信息,提取并匹配连续帧间的SIFT特征点,获取图像帧的描述子,通过特征点的位置变化计算机器人的位姿、里程编码等,从而实现视觉里程计的功能。
Kinect camera can not only get the color image,but also get the depth information.In the paper,a method of implementing o dometer through Kinect camera is discussed.The continuous frame information of the surrounding environment is obtained by the Kinect camera,and the SIFT feature points are extracted and matched,and the image frame descriptors are obtained.Then the robot pose corn puting and the mileage encoding are achieved through the position change of feature points.So that the function of visual odometer is achieved.
出处
《单片机与嵌入式系统应用》
2017年第9期64-66,共3页
Microcontrollers & Embedded Systems
关键词
定位
视觉里程计
导航
深度
ROS
localization
vision odometer
navigation
depth
ROS
