[1]刘学,戚文静.基于外观和深度信息的视觉跟踪算法研究[J].山东建筑大学学报,2016,(02):177-182.
 Liu Xue,Qi Wenjing.Visual tracking algorithm based on appearance feature and depth information[J].Journal of Shandong jianzhu university,2016,(02):177-182.
点击复制

基于外观和深度信息的视觉跟踪算法研究()
分享到:

《山东建筑大学学报》[ISSN:1673-7644/CN:37-1449/TU]

卷:
期数:
2016年02期
页码:
177-182
栏目:
工程实践
出版日期:
2016-04-15

文章信息/Info

Title:
Visual tracking algorithm based on appearance feature and depth information
作者:
刘学1 戚文静23
(1.山东电子职业技术学院 计算机科学系, 山东 济南 250200;2.泰华智慧产业集团股份有限公司博士后工作站,山东 济南 250101;3.山东建筑大学 计算机学院,山东 济南 250101)
Author(s):
Liu Xue1 Qi Wenjing 23
(1.Department of Computer Science, Shandong College of Electronic Technology, Jinan 250200, China; 2. Shandong Taihua Telecommunication Co.,,Ltd, Jinan 250101,China; 3. School of Computer Science, Shandong Jianzhu University, Jinan 250101, China)
关键词:
视觉导航目标跟踪外观特征深度信息
Keywords:
vision-based navigation object tracking appearance feature depth information
分类号:
TP391.4
文献标志码:
A
摘要:
能够实现视觉导航的自主移动机器人具有很好的应用前景,而场景变化、目标运动、障碍、遮档等问题是自主机器人视觉导航过程经常遇到的挑战,结合外观特征和深度信息的目标检测和跟踪算法是提高自主机器人对目标及环境变化的适应能力的重要途径。文章结合人类在跟踪和定位目标时既利用颜色、亮度、形状、纹理等外观特征,又利用物体间距离、深度信息的特点,提出了结合外观特征和深度信息的目标跟踪算法并通过实验验证了该算法对视角、运动、遮挡等因素所引起变化的适应能力,且利用定量的方法对算法的性能进行了评价。
Abstract:
Challenges that robot faces in vision-based navigation include scene change, appearance change, obstacle, occlusion etc. Imitating human vision perception, an object detection and tracking algorithm that combines appearance feature and depth information is proposed. First, RGB image and depth information are captured by the Kinect camera that works as the vision system of robot. Then, an appearance model is created with features extracted from RGB image. A motion model is created on plan-view map produced from depth information and camera parameters, and the estimation of object position and scale is performed on the motion model. Finally, appearance features are combined with position and scale information to track the target. Experimental result show the robustness of our object detection and tracking method to appearance changes arose from view, motion and occlusion factors. It also shows that the object detection efficiency and object tracking accuracy are improved greatly compared with the method that only employ the appearance features.
更新日期/Last Update: 2016-05-16