Kinect彩色图像在光标移动控制中的应用
|
摘要
传统光标控制方式不能较好满足残疾、野外工作等人员的工作需求,而用Kinect识别人体不同的动作类型来实现光标控制时,由于控制方式过于单一,存在动作类型过多且相互干扰的情况。为此,提出将Kinect提取的彩色图像和骨骼数据相结合的方法实现光标的移动控制。从彩色图像上提取可变的像素颜色,以此减少动作类型数量,同时避免动作之间的相互干扰。给出光标移动控制中涉及到的坐标映射相关理论的计算公式。实验结果表明,将感应器彩色图像应用于光标移动控制中是可行的,与单一控制方式相比,具有更好的移动控制性能,通过调节公式中的相关参数更加灵活地实现鼠标控制。
Traditional cursor control ways cannot better meet the job requirements of disabled persons,field work personnel,etc. When Kinect is used to identify different types of human actions to achieve cursor control,there are too many types of actions and interference between each other due to the single control mode. So this paper proposes the method of combining color image and skeleton data extracted by Kinect to implement the cursor movement control. The extracted pixel colors are variable,which can reduce the number of action types while avoiding the interference between actions. The calculation formulas of coordinate mapping theory related to cursor movement control are given. By adjusting the related parameters in the formula,the mouse can be controlled more flexibly. Experimental results showthat it is feasible to use Kinect color image in the cursor movement control. Compared with the single control mode,the proposed method has better mobile control performance.
Traditional cursor control ways cannot better meet the job requirements of disabled persons,field work personnel,etc. When Kinect is used to identify different types of human actions to achieve cursor control,there are too many types of actions and interference between each other due to the single control mode. So this paper proposes the method of combining color image and skeleton data extracted by Kinect to implement the cursor movement control. The extracted pixel colors are variable,which can reduce the number of action types while avoiding the interference between actions. The calculation formulas of coordinate mapping theory related to cursor movement control are given. By adjusting the related parameters in the formula,the mouse can be controlled more flexibly. Experimental results showthat it is feasible to use Kinect color image in the cursor movement control. Compared with the single control mode,the proposed method has better mobile control performance.
引文
[1]余振中,邓为凑.基于Kinect的移动机器人实时局部路径规划[J].计算机工程,2013,39(4):1-2.
[2]李小龙.基于Kinect的手势识别在虚拟人体解剖教学系统的设计与实现[D].北京:北京工业大学,2014:1-3.
[3]SZILVIA S,ZICHAR M,ATTILA F.Gesture-based Computer Mouse Using Kinect Sensor[C]//Proceeding of the 5th IEEE International Conference on Cognitive Infocommunications.Washington D.C.,USA:IEEE Press,2014:419-424.
[4]马风力.基于Kinect自然人机交互系统的设计与实现[D].杭州:浙江大学,2016:10-11.
[5]REN Zhou,YUAN Junsong,ZHANG Zhengyou.Robust Hand Gesture Recognition Based on Finger-earth Mover’s Distance with a Commodity Depth Camera[C]//Proceedings of the 19th ACM International Conference on Multimedia.New York,USA:ACM Press,2011:1093-1096.
[6]KLOMPMAKER F,NEBE K,FAST A.d Sensing NI:A Framework for Advanced Tangible Interaction Using a Depth Camera[C]//Proceedings of the 6th International Conference on Tangible,Embedded and Embodied Interaction.New York,USA:ACM Press,2012:217-224.
[7]YEO Hui-shyong,LEE Byung-gook,LIM Hyotaek.Hand Tracking and Gesture Recognition System for Human-computer Interaction Using Low-cost Hardware[J].Multimedia Tools and Applications,2015,74(8):2687-2715.
[8]QIN Shuxin,ZHU Xiaoyang,YANG Yiping,et al.Realtime Hand Gesture Recognition from Depth Images Using Convex Shape Decomposition M ethod[J].Journal of Signal Processing Systems,2014,74(1):47-58.
[9]MICHAEL A,YANG Yi,LI Lian.Joint Filtering:Enhancing Gesture and M ouse M ovement in M icrosoft Kinect Application[C]//Proceedings of the 12th International Conference on Fuzzy Systems and Know ledge Discovery.Washington D.C.,USA:IEEE Press,2015:2528-2532.
[10]邵浩,侯文立.Kinect用于其他应用时坐标系转换问题研究[J].现代计算机,2014(4):42-45.
[11]陈建军.基于Kinect手势识别的网页软件控制设计[D].太原:太原理工大学,2015.
[12]李恒.基于Kinect骨骼跟踪功能的骨骼识别系统研究[D].西安:西安电子科技大学,2013:19-21.
[13]WANG Yali,SONG Guangming,QIAO Guifang,et al.Wheeled Robot Control Based on Gesture Recognition Using the Kinect Sensor[C]//Proceedings of IEEE International Conference on Robotics and Biomimetics.Washington D.C.,USA:IEEE Press,2013:378-383.
[14]宋诗超,禹素萍,许武军.基于Kinect的三维人体扫描、重建及测量技术的研究[J].天津工业大学学报,2014,31(5):4-5.
[15]DUTTA T.Evaluation of the Kinect TM Sensor for 3-D Kinematic M easurement in the Workplace[J].Applied Ergonomics,2012,43(4):645-649.
[16]李红波,李双生.基于Kinect骨骼数据的人体动作姿势识别方法方法[J].计算机工程与设计,2016(4):1-3.
[17]Microsoft.Kinect for Windows 2.0主要性能改进[EB/OL].(2014-09-16).http://www.cnkinect.com/article-6772-1.html.
[2]李小龙.基于Kinect的手势识别在虚拟人体解剖教学系统的设计与实现[D].北京:北京工业大学,2014:1-3.
[3]SZILVIA S,ZICHAR M,ATTILA F.Gesture-based Computer Mouse Using Kinect Sensor[C]//Proceeding of the 5th IEEE International Conference on Cognitive Infocommunications.Washington D.C.,USA:IEEE Press,2014:419-424.
[4]马风力.基于Kinect自然人机交互系统的设计与实现[D].杭州:浙江大学,2016:10-11.
[5]REN Zhou,YUAN Junsong,ZHANG Zhengyou.Robust Hand Gesture Recognition Based on Finger-earth Mover’s Distance with a Commodity Depth Camera[C]//Proceedings of the 19th ACM International Conference on Multimedia.New York,USA:ACM Press,2011:1093-1096.
[6]KLOMPMAKER F,NEBE K,FAST A.d Sensing NI:A Framework for Advanced Tangible Interaction Using a Depth Camera[C]//Proceedings of the 6th International Conference on Tangible,Embedded and Embodied Interaction.New York,USA:ACM Press,2012:217-224.
[7]YEO Hui-shyong,LEE Byung-gook,LIM Hyotaek.Hand Tracking and Gesture Recognition System for Human-computer Interaction Using Low-cost Hardware[J].Multimedia Tools and Applications,2015,74(8):2687-2715.
[8]QIN Shuxin,ZHU Xiaoyang,YANG Yiping,et al.Realtime Hand Gesture Recognition from Depth Images Using Convex Shape Decomposition M ethod[J].Journal of Signal Processing Systems,2014,74(1):47-58.
[9]MICHAEL A,YANG Yi,LI Lian.Joint Filtering:Enhancing Gesture and M ouse M ovement in M icrosoft Kinect Application[C]//Proceedings of the 12th International Conference on Fuzzy Systems and Know ledge Discovery.Washington D.C.,USA:IEEE Press,2015:2528-2532.
[10]邵浩,侯文立.Kinect用于其他应用时坐标系转换问题研究[J].现代计算机,2014(4):42-45.
[11]陈建军.基于Kinect手势识别的网页软件控制设计[D].太原:太原理工大学,2015.
[12]李恒.基于Kinect骨骼跟踪功能的骨骼识别系统研究[D].西安:西安电子科技大学,2013:19-21.
[13]WANG Yali,SONG Guangming,QIAO Guifang,et al.Wheeled Robot Control Based on Gesture Recognition Using the Kinect Sensor[C]//Proceedings of IEEE International Conference on Robotics and Biomimetics.Washington D.C.,USA:IEEE Press,2013:378-383.
[14]宋诗超,禹素萍,许武军.基于Kinect的三维人体扫描、重建及测量技术的研究[J].天津工业大学学报,2014,31(5):4-5.
[15]DUTTA T.Evaluation of the Kinect TM Sensor for 3-D Kinematic M easurement in the Workplace[J].Applied Ergonomics,2012,43(4):645-649.
[16]李红波,李双生.基于Kinect骨骼数据的人体动作姿势识别方法方法[J].计算机工程与设计,2016(4):1-3.
[17]Microsoft.Kinect for Windows 2.0主要性能改进[EB/OL].(2014-09-16).http://www.cnkinect.com/article-6772-1.html.