[1]韦知辛,方勇纯.融合人类认知模式的自主发育网络及其在手势识别的应用[J].智能系统学报,2023,18(1):144-152.[doi:10.11992/tis.202212002]
WEI Zhixin,FANG Yongchun.Autonomous developmental network incorporating human cognitive modes and its application in gesture recognition[J].CAAI Transactions on Intelligent Systems,2023,18(1):144-152.[doi:10.11992/tis.202212002]
点击复制
WEI Zhixin,FANG Yongchun.Autonomous developmental network incorporating human cognitive modes and its application in gesture recognition[J].CAAI Transactions on Intelligent Systems,2023,18(1):144-152.[doi:10.11992/tis.202212002]
融合人类认知模式的自主发育网络及其在手势识别的应用
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
18
期数:
2023年第1期
页码:
144-152
栏目:
人工智能院长论坛
出版日期:
2023-01-05
- Title:
- Autonomous developmental network incorporating human cognitive modes and its application in gesture recognition
- Keywords:
- autonomous developmental network; gesture recognition; human cognitive modes; top-k competition mechanism; neuron cognitive ability feedback; synaptic readjustment; incremental hierarchical discriminant regression; lobe component analysis
- 分类号:
- TP183
- 摘要:
- 自主发育算法在智能机器人等领域有着广阔的应用前景，考虑到现有方法的局限性，本文提出了一种融合人类认知模式的自主发育神经网络，并将该方法应用于手势识别任务中。通过动态改变神经元预响应值计算过程中的人为指导部分来模拟人类的学习方式，提出了基于动态 k值的top-k 竞争机制，并模拟人脑对知识的接收和记忆功能实现优胜神经元突触权重的更新，最后基于神经元认知能力反馈进行突触重调整。对比实验结果表明，与原有方法相比，经该文改进后的自主发育网络在手势识别任务中具有更好的学习效果和更高的识别率。
- Abstract:
- The autonomous development algorithm has broad application prospect in the intelligent robots and other fields. Considering the limitation of existing methods, this paper proposes an autonomous development neural network integrating human cognitive mode, and applies the method to gesture recognition tasks. Specifically, the proposed method dynamically changes the human-guided part for the computation of neuron pre-response values to simulate the way humans learn. In addition, we propose a top-k competition mechanism based on the dynamic k value. Besides, in the method, human brain’s ability of receiving and memorizing knowledge is simulated to update the synaptic weights of the winning neurons, and neuron cognitive ability is utilized as the feedback for synaptic readjustment. Comparative experimental results show that, compared with the original method, the improved autonomic development network in this paper has better learning effect and higher recognition rate in gesture recognition task.
- 参考文献/References:
-
[1] FENG Zhao, WU Jinlong, NI Taile. Research and application of multifeature gesture recognition in human-computer interaction based on virtual reality technology[J]. Wireless communications and mobile computing, 2021, 10: 1–5.
[2] LI Jiayao, RAY S, RAJANNA V, et al. Evaluating the performance of machine learning algorithms in gaze gesture recognition systems[J]. IEEE access, 2021, 10: 1020–1035.
[3] HUANG Dengyuan, HU W C, CHANG S H. Gabor filter-based hand-pose angle estimation for hand gesture recognition under varying illumination[J]. Expert systems with applications, 2011, 38(5): 6031–6042.
[4] TARVEKAR M P. Hand gesture recognition system for touch-less car interface using multiclass support vector machine[C]//2018 Second International Conference on Intelligent Computing and Control Systems. Madurai: IEEE, 2019: 1929?1932.
[5] RAHIM M A, MIAH A S M, SAYEED A, et al. Hand gesture recognition based on optimal segmentation in human-computer interaction[C]//2020 3rd IEEE International Conference on Knowledge Innovation and Invention. Kaohsiung: IEEE, 2021: 163?166.
[6] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the acm, 2017, 60(6): 84–90.
[7] GUO Xing, XU Wu, TANG Wen quan, et al. Research on optimization of static gesture recognition based on convolution neural network[C]//2019 4th International Conference on Mechanical, Control and Computer Engineering. Hohhot: IEEE, 2020: 398?3982.
[8] SIMONYAN K, ZISSERMAN A. Two-stream convolutional networks for action recognition in videos[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems?Volume 1. Montreal: ACM, 2014: 568?576.
[9] TRAN D, BOURDEV L, FERGUS R, et al. Learning spatiotemporal features with 3D convolutional networks[C]//2015 IEEE International Conference on Computer Vision. Santiago: IEEE, 2016: 4489?4497.
[10] MOLCHANOV P, GUPTA S, KIM K, et al. Multi-sensor system for driver’s hand-gesture recognition[C]//2015 11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition. Ljubljana: IEEE, 2015: 1?8.
[11] OYEDOTUN O K, KHASHMAN A. Deep learning in vision-based static hand gesture recognition[J]. Neural computing and applications, 2017, 28(12): 3941–3951.
[12] ELBADAWY M, ELONS A S, SHEDEED H A, et al. Arabic sign language recognition with 3D convolutional neural networks[C]//2017 Eighth International Conference on Intelligent Computing and Information Systems. Cairo: IEEE, 2018: 66?71.
[13] CARREIRA J, ZISSERMAN A. Quo vadis, action recognition? A new model and the kinetics dataset[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 4724?4733.
[14] FANG Wei, DING Yewen, ZHANG Feihong, et al. Gesture recognition based on CNN and DCGAN for calculation and text output[J]. IEEE access, 2019, 7: 28230–28237.
[15] KOLLER O, NEY H, BOWDEN R. Deep hand: how to train a CNN on 1 million hand images when your data is continuous and weakly labelled[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 3793?3802.
[16] JAHANDAD, SAM S M, KAMARDIN K, et al. Offline signature verification using deep learning convolutional neural network (CNN) architectures GoogLeNet inception-v1 and inception-v3[J]. Procedia computer science, 2019, 161: 475–483.
[17] WENG Juyang, LUCIW M. Dually optimal neuronal layers: lobe component analysis[J]. IEEE transactions on autonomous mental development, 2009, 1(1): 68–85.
[18] WENG Juyang, MCCLELLAND J, PENTLAND A, et al. Autonomous mental development by robots and animals[J]. Science, 2001, 291(5504): 599–600.
[19] WENG Juyang, HWANG W S. Incremental hierarchical discriminant regression[J]. IEEE transactions on neural networks, 2007, 18(2): 397–415.
[20] 黄敏, 路飞, 李晓磊, 等. 基于IHDR算法和BP神经网络复合框架的机器人服务自主认知和发育系统[J]. 机器人, 2019, 41(5): 609–619
HUANG Min, LU Fei, LI Xiaolei, et al. Autonomous cognition and development system of robot service based on a composite framework combining IHDR algorithm with BP neural network[J]. Robot, 2019, 41(5): 609–619
[21] LU Fei, HUANG Min, LI Xiaolei, et al. Learning and development of home service robots’ service cognition based on a learning mechanism[J]. Applied sciences, 2020, 10(2): 464.
[22] 钱夔, 宋爱国, 章华涛, 等. 基于自主发育神经网络的机器人室内场景识别[J]. 机器人, 2013, 35(6): 703–708,743
QIAN Kui, SONG Aiguo, ZHANG Huatao, et al. Robot indoor scenes recognition based on autonomous developmental neural network[J]. Robot, 2013, 35(6): 703–708,743
[23] 王东署, 谭达佩, 韦晓琴. 基于发育网络的人脸朝向识别研究[J]. 郑州大学学报(工学版), 2017, 38(5): 23–27
WANG Dongshu, TAN Dapei, WEI Xiaoqin. Study of face orientation recognition based on development network[J]. Journal of Zhengzhou university (engineering science edition), 2017, 38(5): 23–27
[24] CASTRO-GARCIA J L, WENG Juyang. Emergent multilingual language acquisition using developmental networks[C]//2019 International Joint Conference on Neural Networks. Budapest: IEEE, 2019: 1?8.
[25] ZHENG Zejia, WU Xiang, WENG Juyang. Emergent neural Turing machine and its visual navigation[J]. Neural networks, 2019, 110: 116–130.
[26] 余慧瑾, 方勇纯. 基于改进型自主发育网络的机器人场景识别方法[J]. 自动化学报, 2021, 47(7): 1530–1538
YU Huijin, FANG Yongchun. A robot scene recognition method based on improved autonomous developmental network[J]. Acta automatica sinica, 2021, 47(7): 1530–1538
[27] 余慧瑾, 方勇纯, 韦知辛. 基于多传感融合的自主发育网络场景识别方法[J]. 机器人, 2021, 43(6): 706–714
YU Huijin, FANG Yongchun, WEI Zhixin. A scene recognition method of autonomous developmental network based on multi-sensor fusion[J]. Robot, 2021, 43(6): 706–714
[28] 崔丽, 曾凤章. 基于“学习曲线”效应的长期质量成本模型[J]. 商业研究, 2004(24): 46–48
CUI Li, ZENG Fengzhang. A model of long-term quality cost based on learning curve effect[J]. Commercial research, 2004(24): 46–48
[29] MEMO A, MINTO L, ZANUTTIGH P. Exploiting silhouette descriptors and synthetic data for hand gesture recognition[C]//Smart Tools and Applications in Graphics - Italian Chapter Conference 2015. Verona: The Eurographics Association, 2015.
[30] MEMO A, ZANUTTIGH P. Head-mounted gesture controlled interface for human-computer interaction[J]. Multimedia tools and applications, 2018, 77(1): 27–53.
[31] BAKHEET S, AL-HAMADI A. Hand gesture recognition using optimized local Gabor features[J]. Journal of computational and theoretical nanoscience, 2017, 14(3): 1380–1389.
[32] BAKHEET S, AL-HAMADI A. Robust hand gesture recognition using multiple shape-oriented visual cues[J]. EURASIP journal on image and video processing, 2021, 2021(1): 26.
[33] TASMERE D, AHMED B, DAS S R. Real time hand gesture recognition in depth image using CNN[J]. International journal of computer applications, 2021, 174(16): 28–32.
[34] SAHANA T, BASU S, NASIPURI M, et al. MRCS: multi-radii circular signature based feature descriptor for hand gesture recognition[J]. Multimedia tools and applications, 2022, 81(6): 8539–8560.
- 相似文献/References:
-
[1]余思泉,曹江涛,李平,等.基于空间金字塔特征包的手势识别算法[J].智能系统学报,2015,10(3):429.[doi:10.3969/j.issn.1673-4785.201405054]
YU Siquan,CAO Jiangtao,LI Ping,et al.Hand gesture recognition based on the spatial pyramid bag of features[J].CAAI Transactions on Intelligent Systems,2015,10():429.[doi:10.3969/j.issn.1673-4785.201405054]
[2]贾鹤鸣,朱传旭,张森,等.对偶树复小波与空域信息的手势识别分类研究[J].智能系统学报,2018,13(4):619.[doi:10.11992/tis.201708003]
JIA Heming,ZHU Chuanxu,ZHANG Sen,et al.Research on gesture recognition and classification of dual-tree complex wavelet and spatial information[J].CAAI Transactions on Intelligent Systems,2018,13():619.[doi:10.11992/tis.201708003]
[3]董旭德,许源平,舒红平,等.基于质心分水岭算法的静态手势分割算法模型[J].智能系统学报,2019,14(2):346.[doi:10.11992/tis.201804028]
DONG Xude,XU Yuanping,SHU Hongping,et al.Static gesture segmentation algorithm model based on centroid watershed algorithm[J].CAAI Transactions on Intelligent Systems,2019,14():346.[doi:10.11992/tis.201804028]
备注/Memo
收稿日期:2022-12-02。
基金项目:国家自然科学基金面上项目(61873132);先进计算与关键软件海河实验室项目(22HHXCJC00003).
作者简介:韦知辛,硕士研究生,主要研究方向为机器视觉及发育神经网络;方勇纯,教授,博士生导师,南开大学人工智能学院院长,主要研究方向为机器人视觉控制、欠驱动吊运系统控制、仿生机器人运动控制和微纳米操作。主持国家重点研发计划项目、国家自然科学基金重点项目、"十二五"国家技术支撑计划课题、国家基金仪器专项等国家级项目10多项。获吴文俊人工智能自然科学奖一等奖、天津市专利奖金奖、天津市自然科学一等奖、高等教育教学成果一等奖等多项奖励,发表学术论文100余篇
通讯作者:方勇纯.E-mail:fangyc@nankai.edu.cn
更新日期/Last Update:
1900-01-01