Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/4886
標題: KINECT於即時指尖偵測之研究
Real-time Fingertips Detection Using KINECT
作者: 黃華晨
Huang, Hua-Chen
關鍵字: KINECT
KINECT
指尖偵測
深度影像
深度資訊
fingertip detection
depth image
depth information
出版社: 通訊工程研究所
引用: [1] 維基百科,“KINECT”, http://zh.wikipedia.org/wiki/Kinect . [2] msdn KINECT for windows 開發,http://msdn.microsoft.com/zh-tw/hh367958.aspx . [3] Herrera, C., and Juho Kannala. "Joint depth and color camera calibration with distortion correction." Pattern Analysis and Machine Intelligence, IEEE Transactions on 34.10 (2012): 2058-2064. [4] Soo-Chang Pei, Yu-Ying Wang, Fong-Jou Hsieh. "Auditory depth images by speech navigation for visually impaired users using KINECT depth camera." CVGIP. 2012. [5] Oikonomidis, Iason, Nikolaos Kyriazis, and Antonis A. Argyros. "Efficient model-based 3D tracking of hand articulations using Kinect." BMVC. 2011. [6] Yu Tu , Chih-Lin Zeng , Che-Hua Yeh , Sheng-Yen Huang, Ting-Xin Cheng and Ming Ouhyoung. "Real-time head pose estimation using depth map for AVATAR control." Jia-Yi, Taiwan, 2011. [7] Takimoto, Hironori, et al. "Robust fingertip tracking for constructing an intelligent room." RO-MAN, 2012 IEEE. IEEE, 2012. [8] Lee, Lae-Kyoung, Su-Yong An, and Se-Young Oh. "Robust fingertip extraction with improved skin color segmentation for finger gesture recognition in Human-robot interaction." Evolutionary Computation (CEC), 2012 IEEE Congress on. IEEE, 2012. [9] Panwar, Meenakshi, and Pawan Singh Mehra. "Hand gesture recognition for human computer interaction." Image Information Processing (ICIIP), 2011 International Conference on. IEEE, 2011. [10] Panwar, Meenakshi. "Hand gesture recognition based on shape parameters." Computing, Communication and Applications (ICCCA), 2012 International Conference on. IEEE, 2012. [11] Chaudhary, Ankit, Jagdish L. Raheja, and Shekhar Raheja. "A Vision based Geometrical Method to find Fingers Positions in Real Time Hand Gesture Recognition." Journal of Software 7.4 (2012): 861-869. [12] Dawod, Ahmad Yahya, Junaidi Abdullah, and Md Jahangir Alam. "Fingertips detection from color image with complex background." The 3rd International Conference on Machine Vision (ICMV 2010). 2010. [13] Chaudhary, Ankit, et al. "A Vision-Based Method to Find Fingertips in a Closed Hand." JIPS 8.3 (2012): 399-408. [14] Panwar, Meenakshi. "Hand gesture based interface for aiding visually impaired." Recent Advances in Computing and Software Systems (RACSS), 2012 International Conference on. IEEE, 2012. [15] Nguyen, Dung Duc, Thien Cong Pham, and Jae Wook Jeon. "Fingertip detection with morphology and geometric calculation." Intelligent Robots and Systems, 2009. IROS 2009. IEEE/RSJ International Conference on. IEEE, 2009. [16] Li, Yi. "Hand gesture recognition using Kinect." Software Engineering and Service Science (ICSESS), 2012 IEEE 3rd International Conference on. IEEE, 2012. [17] He, Guan-Feng, et al. "Real-time gesture recognition using 3D depth camera." Software Engineering and Service Science (ICSESS), 2011 IEEE 2nd International Conference on. IEEE, 2011. [18] Raheja, Jagdish L., Ankit Chaudhary, and Kunal Singal. "Tracking of Fingertips and Centers of Palm Using KINECT." Computational Intelligence, Modelling and Simulation (CIMSiM), 2011 Third International Conference on. IEEE, 2011. [19] OpenNI官方網站, http://www.openni.org/ . [20] KINECT calibration, "http://nicolas.burrus.name/index.php/Research/KinectCalibration". [21] Suzuki, Satoshi. "Topological structural analysis of digitized binary images by border following." Computer Vision, Graphics, and Image Processing 30.1 (1985): 32-46. [22] 維基百科,“高斯模糊”,https://zh.wikipedia.org/wiki/高斯模糊 .
摘要: 指尖偵測(fingertips detection)在手勢辨識以及多點觸碰(multi-touch)等相關研究中扮演非常重要的角色,本文利用KINECT所擷取的深度影像以及深度資訊提出一個完整的指尖偵測方法。 本文所提出的指尖偵測方法第一步是擷取KINECT的深度影像以及深度資訊,並且將深度影像的對比強化;第二步透過形態學修補深度影像;第三步使用邊界跟隨找出手勢外部輪廓的順序,再利用手指細長的特徵找出候選點,分群後找出每群中距離掌心最遠的即為手勢外部指尖;第四步利用索貝爾運算子(Sobel operator)找出手勢內輪廓,將一個設計過的單位圓遮罩與手勢內輪廓做高斯模糊(Gaussian blur)後進行摺積(convolution),並設定門檻值篩選出候選點,標記分群後找出每群中像素值最大的點即可篩選出手勢內部指尖。 最後由實驗證明本文所提出的方法不僅能偵測到所有指尖,同時不會受到手勢旋轉角度、手勢大小的影響。由於本文的方法並未使用色彩資訊因此同樣不會受到光線以及膚色的影響。此方法日後可以用於提高手勢辨識的準確率以及應用在人機互動(human computer interaction)相關主題上。
Fingertips detection plays an important role in hand gesture recognition, multi-touch and other related research. This paper presented a fingertips detection algorithm by using KINECT. First of all, we captured depth images and information using KINECT, and pre-processed gesture images with morphological operation. We searched the order of hand contour by contour tracing, and found candidate points by geometrical features of fingers. Then we clustered the candidate points and looked for the open fingertip points which had the furthest distance between the palm centers in each clusters. Next, we blurred the designed unit circle mask and the edges inside the gesture by Gaussian operator. We used the mask convolved the edges inside the gesture, and found fingertips candidate points. Finally, clustering the candidate points and looking for the closed fingertip points which had the biggest pixel value in each clusters. The experimental results proved the proposed algorithm was rotation invariance and scale invariance. Moreover, the algorithm did not use color information, so it will not be influenced by light condition and skin-color. In the future, this algorithm can be used to improve the accuracy of hand gesture recognition and applied to related topics in human computer interaction.
URI: http://hdl.handle.net/11455/4886
其他識別: U0005-0508201316121000
文章連結: http://www.airitilibrary.com/Publication/alDetailedMesh1?DocID=U0005-0508201316121000
Appears in Collections:通訊工程研究所

文件中的檔案:

取得全文請前往華藝線上圖書館



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.