基于OpenCV的摄像头动态手势轨迹识别及其应用

【实例简介】
基于OpenCV的摄像头动态手势轨迹识别及其应用，前沿的研究，最新论文
2012 B_0% random t Xo X,XX 4_1、42、B0 Bhattacharyya glsl. 8 gsl_ran_gaussian pp q 1/3 t=∑∵n 8 HSV certer HSV IISV V 4 4 HMM 3 IIMM A=SO4 B 45 0 A iI i j -+ ∑bk 2 HMM 4+1=A_1*X1-X0+A_2*X o1994-2013ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net Opencv lIMM P., v3) A B P(IA IIMM (a)两点之间的方向夹角(b)轨迹的方间夹角分为8等份 HMM Baum-Welch y O a B,i/∑a,iB t HMM A HMM ∑ s: t H q,=======-s,qu i a 6. o, B, 1 j/PolA a)向右的于势(b)向左的手势(c)向上的手势(d向下的手势 动,O:B1j 1+1J ∑,ij timer mero ∑点:1j/∑ b k Firmer limmer=3 ∑y,1/∑y IMM 丌AB 5 12 Baum-welch 4.3 HMⅥ HMM HMⅥ HMM 8 (a)第5 (b)第6帧 第6幀 第6帧 o1994-2013ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net 132 2012 (a第1 (b)第72 (a)第5帧 (b)第60帧 (c}第62帧 (d)第G3帧 (c)第75 d)第呲 5.4 HMM HMM HMM HMI HMM 丌AB HMM a)第4帧 .0 ()第50 d)第54帧 HMM HMM 0.292806 0.56l901 0.294316 0.560653 0.116791 0.190064 0.042868 0.190336 0.04136 0.321721 0.323580 0.017312 0.116605 0.015503 0.070848 0.06968 0.160129 0.041269 0.160040 0.03974 0.074966 0.073851 0.185957 0.146865 0.146616 0.213680 0.158732 0.095492 0.158626 0.09462 0.206988 0.20746 2 HMM 3 0.00l 20.001 0 0.001 0.00I 0.00l 0.00 0.00 50.00l .O00 1.000 3 60.001 0.001 2 0.00I 0.001 0.00l 0.001 IIMM HMM IIMM 320×240 HMM o1994-2013ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net Opencv 133 jck hand recogni tion 2005,2611):29 Openly 获取直方图 轨迹图像 2010,30(71:59-61. 在单击“轨迹「未选择-图像显示 [5 QIN WEN, PENG QICONG. An improved particle filter algorithm 图像时选择Chmm控制 based on neural network for visual tracking[c / International Con 个按钮选 C控制贪食鱼 ference on Communications, Circuits and Sy Las v. 项 训练HMM IEEE.2007:765-768 LD」 停止 退出 2009 [7 YU JINXIA, LIU WENJING, YANG Y. Impmved particle filter al gorithms based on partial systematic resampling C|//IEEE InTerna tional Conference on Intelligent Computing and Intelligent Systems Las vegas: IEEE, 2010: 483-487 8 FAZLI S, POUR II M, BOUZARI ll. Particle filter based object tracking with sift and color feature[C // Second International Con as Vegas: IEEE, 2009: 89-9 (8:62-64. L10」 LD」 HMM ,2004 [ II YUAN LICHI. An improved HMM speech recognition model[c// International Conference on AudiO. las Vegas: IEEE, 2008: 1311 Ll OpenCV LD」 1315. [12 TAN WENJUN, WU CHENGDONG, ZHAO SHUYING. Recogni [2 CHlO JU, JIN S ll, PILAM X D. Object tracking circuit using parti tion of dynamic hand gesture based on SCllMM model[C]//Chi- cle filter with multiple features [C//SICE-ICASE: International nese Control and Decision Conference. Log Vegas: IEEE. 2009 Joint Conference. Las Vegas: IEEE, 2006: 1431-1436 2430-2434 127 2011,28(2):242-244. C,2010,1(1):92-95 SIFT [8 BEIS J S, LOWE D G. Shape indexing using approximate nearest- neighbour search in high-dimensional spaces [C]// Proceedings of the IEEE 1997 Computer Society Conference on Computer vision and Pattern Recognition. Washington, DC: IEEE Computer Socie- ly,997:1000-1006 SIFT 009,29(5):423-226 2009,30(4):857-861 [O] [2] LOWE D G. Distinctive image features from scale-invariant key- points[ J]. International of Computer Vision, 2004, 60(2):91 [11 SHUM H Y, SZELISKI R. Construction of panoramic image mosaics with global and local alignment[ J International Journal of Comp 3 MIKOLAJCZYK K, SClIMID C. A performance evaluation of local er visian,2000,36(2):101-130 descriptors[ J|. IEEE Transactions on Pattern Analysis and Machine [12 ZOGIILAMI 1. FAUGERLAS O, DERICIIE R. Using geometric cor Intelligence,2005,27(10):1615-1630 ners to build a 2D mosaic from a set of images[C]// Proceedings of 4 LINDEBERG T. Seale -space theory a basic foul for analyzing the International Conference on Computer Vision and Pattern Recog- structures at different seales LJ. Journal of Applied Statistic itiL. Washington, DC: IEEE Computer Society, 1997: 167 1994,21(2):224-270 o1994-2013ChinaAcademicJournalElectronicPublishingHouse.Allrightsreservedhttp://www.cnki.net 【实例截图】
【核心代码】