基于Kinect改进的增量PCA扭锁在线识别

doi:10.13229/j.cnki.jdxbgxb201603032

吉林大学学报(工学版) ›› 2016, Vol. 46 ›› Issue (3): 890-896.doi: 10.13229/j.cnki.jdxbgxb201603032

基于Kinect改进的增量PCA扭锁在线识别

马爽^{1, 2}, 周长久², 张连东², 洪伟¹, 田彦涛¹

1.吉林大学通信工程学院,长春 130022;
2.新加坡理工学院先进智能机器人与智能控制中心,新加坡139651

收稿日期:2014-11-26 出版日期:2016-06-20 发布日期:2016-06-20
作者简介:马爽(1985),女,博士研究生.研究方向:模式识别,机器人视觉,认知学习,神经网络,智能信息处理.E-mail:mashuang11@mails.jlu.edu.cn
基金资助:
新加坡创新基金项目(11-27801-36-R140); 新加坡教育部创新基金项目(MOE2013-TIF-1-G-057, MOE2013-TIF-2-G-040); 国家自然科学基金项目(51275065).

Twist-lock online recognition based on improved incremental PCA by Kinect

MA Shuang^{1, 2}, ZHOU Chang-jiu², ZHANG Lian-dong², HONG Wei¹, TIAN Yan-tao¹

1.College of Communication Engineering, Jilin University, Changchun 130022, China;
2.Advanced Robotics and Intelligent Control Centre, Singapore Polytechnic, Singapore 139651,Singapore

Received:2014-11-26 Online:2016-06-20 Published:2016-06-20

摘要/Abstract

摘要： 以海港集装箱扭锁自动化装卸过程中扭锁的认知识别为研究背景,采用Kinect传感器感知环境及物体信息,提出基于改进的增量主成分分析(Principal component analysis,PCA)方法构建实时认知识别系统。在系统在线学习阶段,基于新样本与已有特征重建样本之间的差异程度来感知、监测新类别输入,控制特征向量增量式更新;基于类内距离比较,优化特征向量组合,并自适应地更新类内距离阈值;将高维视觉信号转化为低维的机器人内部表达形式,从而在线实时地学习、更新、累积特征知识,同时完成模式识别任务。实验结果表明:该方法在有效提高视觉系统实时性、自适应性、稳定性及识别准确率的同时,控制了特征维度,从而减少了数据处理量及存储空间。

关键词: 计算机应用, 在线学习, 增量主成分分析, 自适应特征更新, 感知识别

Abstract: Research of the cognitive recognition of twist-lock automation handling system is conducted. In this research, Kinect is employed to collect environment and objects information, and an improved incremental Principal Component Analysis (PCA) is proposed to build real-time cognitive recognition system. In online learning phase, the new class is monitored and feature vectors are updated incrementally based on the difference between the new input and the reconstruction one using current eigenvectors; the feature vectors are optimized and the inner-class distance threshold is updated adaptively based on comparison of inner-class distance. Thereby, the proposed algorithm can convert high-dimension information to low-dimension machine expression, learn, update and accumulate feature knowledge online, and complete pattern recognition task at the same time. Experiment results show that the proposed algorithm can improve the adaptability, robustness, recognition rate and real-time performance of a visual system, Moreover, calculation and storage space can be reduced by controlling the feature space dimension.

Key words: computer application, online learning, incremental principal component analysis(PCA), adaptive feature update, cognitive recognition

中图分类号:

TP391.4

马爽, 周长久, 张连东, 洪伟, 田彦涛. 基于Kinect改进的增量PCA扭锁在线识别[J]. 吉林大学学报(工学版), 2016, 46(3): 890-896.

MA Shuang, ZHOU Chang-jiu, ZHANG Lian-dong, HONG Wei, TIAN Yan-tao. Twist-lock online recognition based on improved incremental PCA by Kinect[J]. 吉林大学学报(工学版), 2016, 46(3): 890-896.

参考文献

[1] Chellappa M. The weakest link[C]∥International Maritime-Port Technology and Development Conference,Singapore,2011:278-282.
[2] 宋怀波,史建强. 应用PCA理论进行多人脸姿态估计的方法[J]. 吉林大学学报:工学版,2013,43(增刊1):43-46.
Song Huai-bo, Shi Jian-qiang. Pose estimation of varied human face based on PCA method[J]. Journal of Jilin University (Engineering and Technology Edition), 2013, 43(Sup.1):43-46.
[3] Yuan Y, Pang Y W, Pan J,et al. Scene segmentation based on IPCA for visual surveillance[J]. Neurocomputing,2009,72(10-12):2450-2454.
[4] 杨仁杰,刘蓉,杨延荣,等. 用二维相关近红外谱和多维主成分分析判别掺杂牛奶[J]. 光学精密工程,2014,22(9):2352-2358.
Yang Ren-jie, Liu Rong,Yang Yan-rong, et al. Classification of adulterated milk by two-dimensional correlation near-infrared spectroscopy and multi-way principal component analysis[J]. Optics and Precision Engineering, 2014,22(9):2352-2358.
[5] 刘志强, 尹建芹, 张玲, 等. 基于Kinect数据主成分分析的人体动作识别[J]. 光学精密工程,2015,23(10):702-711.
Liu Zhi-qiang, Yin Jian-qin, Zhang Ling, et al. Human action recognition based on Kinect data principal component analysis[J]. Optics and Precision Engineering,2015,23(10):702-711.
[6] Zhao H T, Yuen P C, Kwork J. A novel incremental principal component analysis and its application for face recognition[J]. IEEE Transactions on Systems, Man and Cybernetics, Part B(Cybernetics),2006,36(4):873-886.
[7] 李大健,齐敏,郝重阳. 模式识别导论[M]. 北京: 清华大学出版社, 2009.
[8] Oja E, Karhunen J. On stochastic approximation of the eigenvectors and eigenvalues of the expectation of a random matrix[J]. Journal of Mathematical Analysis and Applications,1985,106(1):69-84.
[9] Sanger T D. Optimal unsupervised learning in a single-layer linear feedforward neural network[J]. Neural Networks,1989,2(6):459-473.
[10] Weng J Y, Zhang Y L, Hwang W S. Candid covariance-free incremental principal component analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(8):1034-1040.
[11] Hall P, Marshall D, Martin R. Adding and subtracting eigenspaces with eigenvalue decomposition and singular value decomposition[J]. Image and Vision Computing,2002,20(13-14):1009-1016.
[12] 黄诚,沈昱明,刘华平,等. 基于增量PCA的目标跟踪算法[J]. 江南大学学报:自然科学版,2013,12(6):647-652.
Huang Cheng,Shen Yu-ming,Liu hua-ping,et al. Target tracking algorithm based on the incremental PCA[J]. Journal of Jiangnan University(Natural Science Edition),2013,12(6):647-652.
[13] Skocaj D, Leonardis A. Weight and robust incremental Method for subspace learning[C]∥9th IEEE International Conference on Computer Vision,Nice, France,2003:1494-1501.
[14] Neto H V, Nehmzow U. Incremental PCA: an alternative approach for novelty detection[C]∥Towards Autonomous Robotic System, London,2005:227-233.
[15] Qu X Y, Yao M H. Adaptive subspace incremental PCA based online learning for object classification and recognition[C]∥4th International Congress on Image and signal Processing, Shanghai,China,2011:1494-1498.
[16] Ma S, Zhou C J, Zhang L D,et al. 3D irregular object recognition for twist-lock handling system[C]∥26th Chinese Control and Decision Conference,Changsha,China,2014:2729-2734.

相关文章 15

[1]	刘富,宗宇轩,康冰,张益萌,林彩霞,赵宏伟. 基于优化纹理特征的手背静脉识别系统[J]. 吉林大学学报(工学版), 2018, 48(6): 1844-1850.
[2]	王利民,刘洋,孙铭会,李美慧. 基于Markov blanket的无约束型K阶贝叶斯集成分类模型[J]. 吉林大学学报(工学版), 2018, 48(6): 1851-1858.
[3]	金顺福,王宝帅,郝闪闪,贾晓光,霍占强. 基于备用虚拟机同步休眠的云数据中心节能策略及性能[J]. 吉林大学学报(工学版), 2018, 48(6): 1859-1866.
[4]	赵东,孙明玉,朱金龙,于繁华,刘光洁,陈慧灵. 结合粒子群和单纯形的改进飞蛾优化算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1867-1872.
[5]	刘恩泽,吴文福. 基于机器视觉的农作物表面多特征决策融合病变判断算法[J]. 吉林大学学报(工学版), 2018, 48(6): 1873-1878.
[6]	欧阳丹彤, 范琪. 子句级别语境感知的开放信息抽取方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1563-1570.
[7]	刘富, 兰旭腾, 侯涛, 康冰, 刘云, 林彩霞. 基于优化k-mer频率的宏基因组聚类方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1593-1599.
[8]	桂春, 黄旺星. 基于改进的标签传播算法的网络聚类方法[J]. 吉林大学学报(工学版), 2018, 48(5): 1600-1605.
[9]	刘元宁, 刘帅, 朱晓冬, 陈一浩, 郑少阁, 沈椿壮. 基于高斯拉普拉斯算子与自适应优化伽柏滤波的虹膜识别[J]. 吉林大学学报(工学版), 2018, 48(5): 1606-1613.
[10]	车翔玖, 王利, 郭晓新. 基于多尺度特征融合的边界检测算法[J]. 吉林大学学报(工学版), 2018, 48(5): 1621-1628.
[11]	赵宏伟, 刘宇琦, 董立岩, 王玉, 刘陪. 智能交通混合动态路径优化算法[J]. 吉林大学学报(工学版), 2018, 48(4): 1214-1223.
[12]	黄辉, 冯西安, 魏燕, 许驰, 陈慧灵. 基于增强核极限学习机的专业选择智能系统[J]. 吉林大学学报(工学版), 2018, 48(4): 1224-1230.
[13]	傅文博, 张杰, 陈永乐. 物联网环境下抵抗路由欺骗攻击的网络拓扑发现算法[J]. 吉林大学学报(工学版), 2018, 48(4): 1231-1236.
[14]	曹洁, 苏哲, 李晓旭. 基于Corr-LDA模型的图像标注方法[J]. 吉林大学学报(工学版), 2018, 48(4): 1237-1243.
[15]	侯永宏, 王利伟, 邢家明. 基于HTTP的动态自适应流媒体传输算法[J]. 吉林大学学报(工学版), 2018, 48(4): 1244-1253.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于Kinect改进的增量PCA扭锁在线识别

Twist-lock online recognition based on improved incremental PCA by Kinect

RICH HTML

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 10