基于改进RBF神经网络的人体姿态局部特征识别算法

doi:10.13229/j.cnki.jdxbgxb.20240464

摘要/Abstract

摘要：

以机器人的人体姿态识别问题为核心，为提高识别精度，提出一种基于改进RBF神经网络的人体姿态局部特征识别算法。利用深度相机得到人体关节点三维方位数据，归一化处理方位数据，组建关节点三维坐标；考虑到不同个体之间的差异，为实现对人体姿态数据的非线性映射和优化，准确识别不同个体姿态，采用newrbe函数构建RBF神经网络，提取人体姿态数据特征矢量，以为识别提供重要依据；为增强RBF神经网络在处理不同个体姿态差异方面的能力，确保识别的准确性和自适应性，使用粒子群优化算法改进神经网络，并通过特定概率对粒子实施遗传操作，实现网络优化得到人体姿态局部特征识别结果。实验结果表明：本文算法相对误差均较小，可维持在0.8以下，识别精度高，且在迭代次数达到20时损失函数已降至最低，收敛速度较快，可为农业机械化领域的人机交互提供扎实基础。

关键词: 改进RBF神经网络, 人体姿态, 局部特征识别, 三维坐标, 粒子群优化

Abstract:

Therefore， with the human pose recognition problem of robots as the core， a local feature recognition algorithm for human pose based on an improved RBF neural network is proposed to improve recognition accuracy. Using a depth camera to obtain three-dimensional orientation data of human joint points， normalizing the orientation data， and constructing three-dimensional coordinates of joint points； Considering the differences between different individuals， in order to achieve nonlinear mapping and optimization of human pose data， accurately identify different individual poses， a Newrbe function is used to construct an RBF neural network， extract feature vectors of human pose data， and provide important basis for recognition； To enhance the ability of RBF neural networks to handle different individual pose differences， ensure recognition accuracy and adaptability， particle swarm optimization algorithm is used to improve the neural network， and genetic operations are performed on particles with specific probabilities to achieve network optimization and obtain local feature recognition results of human pose. The experimental results show that the proposed algorithms have relatively low relative errors， can be maintained below 0.8， high recognition accuracy， and the loss function is minimized when the number of iterations reaches 20. The convergence speed is fast， which can provide a solid foundation for human-machine interaction in the field of agricultural mechanization.

Key words: improve rbf neural network, human posture, local feature recognition, three dimensional coordinates, particle swarm optimization

中图分类号:

TP391

李燕飞,吴加宁. 基于改进RBF神经网络的人体姿态局部特征识别算法[J]. 吉林大学学报(工学版), 2025, 55(5): 1749-1755.

Yan-fei LI,Jia-ning WU. Human pose local feature recognition algorithm based on improved RBF neural network[J]. Journal of Jilin University(Engineering and Technology Edition), 2025, 55(5): 1749-1755.

图/表 7

图1

图2

图3

图4

图5

图6

图7

参考文献 20

[1]	陈怀林, 李文欣, 杜歆桐, 等.基于情境感知的智慧农机管控系统交互设计研究[J].包装工程, 2023, 44(22): 123-130.
	Chen Huai-lin, Li Wen-xin, Du Xin-tong, et al. Interaction design of intelligent agricultural machinery managementand control system based on context-awareness[J]. Packaging Engineering, 2023, 44(22):123-130.
[2]	段铭钰, 袁瑞甫, 杨艺. 基于改进RBF神经网络的采煤机截割煤岩性状智能识别[J]. 河南理工大学学报:自然科学版, 2022,41(1): 43-51.
	Duan Ming-yu, Yuan Rui-fu, Yang Yi. Intelligent recognition of coal and rock properties in shearer cutting process based on improved RBF neural network[J].Journal of Henan Polytechnic University (Natural Science), 2022, 41(1): 43-51.
[3]	曾明如, 熊嘉豪, 祝琴. 基于T-Fusion的TFP3D人体行为识别算法[J].计算机集成制造系统, 2023, 29(12): 4032-4039.
	Zeng Ming-ru, Xiong Jia-hao, Zhu Qin. TFP3D human behavior recognition algorithm based on T-Fusion[J].Computer Integrated Manufacturing Systems,2023, 29(12): 4032-4039.
[4]	余金锁, 卢先领. 基于分割注意力的特征融合CNN-Bi-LSTM人体行为识别算法[J].电子测量与仪器学报, 2022, 36(2): 89-95.
	Yu Jin-suo, Lu Xian-ling. Human action recognition algorithm of feature fusion CNN-Bi-LSTM based on split-attention[J].Journal of Electronic Measurement and Instrumentation, 2022, 36(2): 89-95.
[5]	孙剑明, 韩生权, 沈子成, 等. 基于双卷积链的双目人体姿态距离定位识别[J]. 兵工学报, 2022,43(11): 2846-2854.
	Sun Jian-ming, Han Sheng-quan, Shen Zi-cheng, et al. Binocular human pose and distance identification based on double convolutional chain[J].Acta Armamentarii, 2022, 43(11): 2846-2854.
[6]	Nguyen T T, Pham D T, Vu H, et al. A robust and efficient method for skeleton-based human action recognition and its application for cross-dataset evaluation[J].IET Computer Vision,2022,16(8):709-726.
[7]	Huang Y Z, Zhao H M, Zhao X T, et al.Pattern recognition using self-reference feature extraction for?-OTDR[J].Applied Optics,2022,61(35):10507-10518.
[8]	Zeng Y M, Xiang H J, Zheng N, et al. Flexible triboelectric nanogenerator for human motion tracking and gesture recognition[J]. Nano Energy,2022, 91: 106601.
[9]	李新春, 张玉琛, 阳士宇. 融合全局与局部特征的UWB雷达人体动作识别算法[J].重庆邮电大学学报: 自然科学版, 2023, 35(4): 636-645.
	Li Xin-chun, Zhang Yu-chen, Yang Shi-yu. Human action recognition algorithm incorporating global andlocal features by UWB radar[J].Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition),2023,35(4):636-645.
[10]	Wang C, Zhang F, Zhu X T, et al.Low-resolution human pose estimation[J].Pattern Recognition: The Journal of the Pattern Recognition Society, 2022,126: 108579.
[11]	宋玉琴, 曾贺东, 高师杰, 等.基于改进高分辨表征的人体姿态估计算法[J]. 计算机工程与设计, 2022, 43(4): 1045-1051.
	Song Yu-qin, Zeng He-dong, Gao Shi-jie, et al. Human pose estimation algorithm based on improved high resolution representation[J].Computer Engineering and Design,2022,43(4): 1045-1051.
[12]	Li J C, Han R, Feng W, et al. Contactless interaction recognition and interactor detection in multi-person scenes[J]. Frontiers of Computer Science, 2024, 18(5): 185325.
[13]	邓平, 吴明辉. 基于机器学习的人体运动姿态识别方法[J]. 中国惯性技术学报, 2022,30(1): 37-43.
	Deng Ping, Wu Ming-hui.Human motion attitude recognition method based on machine learning[J]. Journal of Chinese Inertial Technology, 2022, 30(1): 37-43.
[14]	Zhang Y Q, Ding K, Hui J Z, et al.Human-object integrated assembly intention recognition for context-aware human-robot collaborative assembly[J]. Advanced Engineering Informatics, 2022,54: 101792.
[15]	Meng C, He X, Luan T L.Gait recognition based on 3D human body reconstruction and multi-granular feature fusion[J].Journal of Supercomputing, 2023, 79(11): 12106-12125.
[16]	刘锁兰, 周岳靖, 王洪元, 等.基于全局图遍历的ST-GCN人体行为识别算法[J]. 扬州大学学报: 自然科学版, 2022, 25(2): 62-68.
	Liu Suo-lan, Zhou Yue-jing, Wang Hong-yuan, et al. ST-GCN human action recognition algorithm based on global graph traversal[J]. Journal of Yangzhou University (Natural Science Edition), 2022,25(2):62-68.
[17]	陈琳, 王子微, 莫玉良, 等. 改进的自适应复制、交叉和突变遗传算法[J]. 计算机仿真, 2022,39(8): 323-326.
	Chen Lin, Wang Zi-wei, Mo Yu-liang, et al. Im⁃proved genetic algorithms for adaptive replication crossover and mutation[J]. Computer Simulation, 2022, 39 (8): 323-326.
[18]	Fang N, Wei J, Feng T, et al. Feature super-resolution based facial expression recognition for multi-scale low-resolution images[J]. Knowledge-based systems, 2022, 236(25): 1076781.
[19]	王亚东, 秦会斌. 结合不确定性估计的轻量级人体关键点检测算法[J]. 电子技术应用, 2023, 49(10): 40-45.
	Wang Ya-dong, Qin Hui-bin.Lightweight human key point detection algorithm with uncertainty[J].Application of Electronic Technique,2023,49(10):40-45.
[20]	Abul A B, Ram K K, Rahul J. A convolutional neural network and classical moments-based feature fusion model for gesture recognition[J]. Multimedia Systems, 2022, 28(5): 1779-1792.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed