基于多特征融合与支持向量机的手势识别

›› 2015, Vol. 28 ›› Issue (5): 127-.

基于多特征融合与支持向量机的手势识别

吴健健,陈玮

(上海理工大学光电信息与计算机工程学院,上海 200093)

出版日期:2015-05-15 发布日期:2015-05-19
作者简介:吴健健(1991—),男,硕士研究生。研究方向:嵌入式系统,模式识别,图像处理。Email:wujianjian111@126.com。陈玮(1964—),女,副教授。研究方向:计算机控制技术,模式识别,嵌入式系统。
基金资助:
沪江基金资助项目(C14002)

Hand Gesture Recognition Based on Multifeature Fusion and Support Vector Machines

WU Jianjian,CHEN Wei

(School of OpticalElectrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)

Online:2015-05-15 Published:2015-05-19

摘要/Abstract

摘要：

针对手势识别中人的手部特征描述易受到环境因素影响,手势识别率低等问题,并考虑到单个特征的局限性,提出了一种基于Hu矩和HOG特征融合的支持向量机手势识别新方法。该方法首先对处理后的手势图像提取局部的HOG特征,然后针对手势的轮廓提取全局Hu矩特征,再将两种特征融合成混合特征,并通过主成分分析法对混合特征进行降维形成最终分类特征,并将新特征输入到支持向量机中进行识别。实验表明,该方法具有较好的鲁棒性和较高的识别率。

关键词: 手势识别, Hu矩, 梯度直方图, 主成分分析法, 支持向量机

Abstract:

Hand gesture features are easily affected by environmental factors and the gesture recognition rate is therefore low.And single features are much limited.This paper presents a novel hand gesture recognition algorithm based on multifeature fusion and support vector machines.The fusion features include HOG features and Hu moments.First,the gesture image is set to extract the local HOG features.Then Hu moments are extracted from the gesture contour.After that,the dimensions of the fusion features of HOG and Hu moments are reduced by using Principal Component Analysis (PCA).Finally,the fusion features are put into support vector machines for identification.The experimental results show that the method proposed has better robustness and a higher recognition rate.

Key words: gesture recognition;Hu moments;histograms of oriented gradients;principal component analysis;support vector machines

中图分类号:

TP391.41

吴健健,陈玮. 基于多特征融合与支持向量机的手势识别[J]. , 2015, 28(5): 127-.

WU Jianjian,CHEN Wei. Hand Gesture Recognition Based on Multifeature Fusion and Support Vector Machines[J]. , 2015, 28(5): 127-.

参考文献

［1］Akmeliawati R,Jimoh M,Salami E,et al.Vision-based hand posture detection and recognition for sign language study［C］.New York:Proceedings of the fourth International Conference on Macaronis,IEEE,2011:1-6.

［2］Chen Wenher,Ching Tang Heish,Tsun Teliu.A real time hand gesture recognition system based on DFT and SVM［J］.Applied Mechanics and Materials,2013(2):284-287.

［3］Zhang Qiuyu,Wang Daodong,Zhang Moyi,et al.Hand gesture recognition based on bag of features and support vector machine［J］.Journal of Computer Applications,2012,32(12):3392-3396.

［4］Amin M A,Yan H.Sign language finger alphabet recognition from Gabor-PCA representation of hand gestures［C］.New York:Proceedings of the Sixth International Conference on Machine Learning and Cybernetics,IEEE,2007:2218-2223.

［5］Liu Yun,Yin Yanmin,Zhang Shujun.Hand gesture recognition based on hu moments in interaction of virtual reality［C］.New York:Proceedings of the fourth International Conference on Intelligent Human-Machine Systems and Cybernetics,IEEE,2012:145-148.

［6］Feng Kaiping,Yuan Fang.Static hand gesture recognition based on HOG characters and support vectormachines［C］.New York:Proceedings of the second International Symposium on Instrumentation and Measurement,Sensor Network and Automation,IEEE,2013:936-938.

［7］汪成亮,周佳,黄晟.基于高斯混合模型与PCA-HOG的快速运动人体检测［J］.计算机应用研究,2012,29(6):2157-2160.

［8］向征,谭恒良.改进的HOG和Gabor,LBP性能比较［J］.计算机辅助设计与图形学学报,2012,24(6):787-792.

［9］Hu M K.Visual pattern recognition by moment invariants［J］.IRE Transactions on Information Theory,1962,8(2):179-187.

［10］兰添才.基于多特征融合和支持向量机的媒矸石自动分选研究［J］.科技信息,2008,24(20):7-8.

［11］尹飞,冯大政.基于PCA算法的人脸识别［J］.计算机技术与发展,2008,18(10):31-33.

［12］曾志强.支持向量分类机的训练与简化算法研究［D］.浙江:浙江大学,2007.

［13］Laganiere R.Open CV2 computer vision application programming cookbook［M］.Birmingham:Packt Publishi Ltd,2011.

[1]	伍伟嘉,杨俭,袁天辰,邵志慧. 基于支持向量机的轨道结构病害识别研究[J]. 电子科技, 2022, 35(2): 27-33.
[2]	邵志慧,杨俭,袁天辰,伍伟嘉. 基于排列熵和支持向量机的轨枕病害诊断[J]. 电子科技, 2022, 35(2): 52-58.
[3]	冯先丁,魏镜弢,吴张永,钱杰,浦友尚. 基于PCA-PSO-SVM的球磨机负荷预测研究[J]. 电子科技, 2022, 35(1): 29-34.
[4]	张朋飞,魏存伟,刘先康,刘安然,杨欧,林永霖. 窄带舰船目标识别方法研究[J]. 电子科技, 2021, 34(6): 50-55.
[5]	顾伟,李菲菲,陈虬. 基于多特征融合的行人检测方法[J]. 电子科技, 2021, 34(5): 29-34.
[6]	刘姝,邵杰,张颐婷,张善章. 基于双谱和谱特征的心电信号分类方法[J]. 电子科技, 2021, 34(5): 42-46.
[7]	李兆基,王海鹏,阮伟华,黄书豪. 辅助虚拟现实数据手套及电刺激力触觉增强反馈电路系统设计[J]. 电子科技, 2021, 34(3): 28-36.
[8]	彭荣杰,彭亚雄,陆安江. 基于改进PCA+SVM的人脸识别系统[J]. 电子科技, 2021, 34(12): 56-61.
[9]	巨志勇,翟春宇,张文馨. 基于SVM与区域生长的彩色商品标签图像分割方法[J]. 电子科技, 2021, 34(10): 69-74.
[10]	金鑫,冯毅,尤雪汐,王佳欣. 基于机器学习的信息安全设备调配保障技术研究[J]. 电子科技, 2020, 33(8): 80-86.
[11]	万东燕,许桢英,杨卿,吴梦琪. 基于PCA与SVM的焊缝缺陷信号分类方法[J]. 电子科技, 2020, 33(5): 9-14.
[12]	陈磊,李菲菲,陈虬. 基于迁移学习与权重支持向量机的图像多标签标注算法[J]. 电子科技, 2020, 33(3): 12-16.
[13]	杨海柱,江昭阳,李梦龙,康乐. 基于改进人工鱼群-蛙跳算法优化LSSVM参数短期负荷预测[J]. 电子科技, 2020, 33(12): 67-74.
[14]	王贺贺,李菲菲,陈虬. 基于稠密轨迹特征的动态手势识别研究[J]. 电子科技, 2020, 33(12): 17-21.
[15]	缪冉,李菲菲,陈虬. 基于卷积神经网络与多尺度空间编码的场景识别方法[J]. 电子科技, 2020, 33(12): 54-58.