Research on Accurate Gesture Recognition Algorithm in Complex Environment Based on Machine Vision

doi:10.16182/j.issn1004731x.joss.21-FZ0668

Abstract

Abstract: To address the issue of cross infection caused by elevator public buttons during COVID-19, a software algorithm based on machine vision for non-contact control of public buttons by gesture recognition is designed. In order to improve the accuracy of gesture recognition, an improved YOLOv4 algorithm is proposed. A Ghost module is designed based on attention mechanism, and the ResBlock module in YOLOv4 is improved to Ghost module. The experimental results show that, in the task of gesture recognition, the detection speed is improved by 14% and the detection accuracy is improved by 0.1% compared with the original model. The improved YOLOv4 algorithm is applied to the non-contact elevator buttons control system based on vision. The experimental results show that the detection accuracy reaches 98%, which meets the requirements of non-contact control for public elevators.

Key words: Gesture recognition, YOLOv4, Attentional mechanism, Ghost module

CLC Number:

TP391

Xu Sheng, Feng Wenyu, Liu Zhicheng, Tu Xintao, Fei Minrui, Zhang Kun. Research on Accurate Gesture Recognition Algorithm in Complex Environment Based on Machine Vision[J]. Journal of System Simulation, 2021, 33(10): 2460-2469.

References

[1] Liu S Q, Zhang J C, Zhu R.A Wearable Human Motion Tracking Device Using Micro Flow Sensor Incorporating a Micro Accelerometer[J]. IEEE Transactions on Biomedical Engineering (S0018-9294), 2019, 67(4): 940-948.
[2] Wong W K, Juwono F H, Khoo B T T. Multi-features Capacitive Hand Gesture Recognition Sensor: a Machine Learning Approach[J]. IEEE Sensors Journal (S1530-437X), 2021, 21(6): 8441-8450.
[3] Valueva M V, Nagornov N N, Lyakhov P A, et al.Application of the Residue Number System to Reduce Hardware Costs of the Convolutional Neural Network Implementation[J]. Mathematics and Computers in Simulation (S0378-4754), 2020, 177: 232-243.
[4] Cao Z, Hidalgo G, Simon T, et al.OpenPose: Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2019, 43(1): 172-186.
[5] Mao Q C, Sun H M, Liu Y B, et al.Mini-YOLOv3: Real-time Object Detector for Embedded Applications[J]. IEEE Access (S2169-3536), 2019, 7: 133529-133538.
[6] Wu D, Lv S, Jiang M, et al.Using Channel Pruning-based YOLO v4 Deep Learning Algorithm for the Real-time and Accurate Detection of Apple Flowers in Natural Environments[J]. Computers and Electronics in Agriculture (S0168-1699), 2020, 178: 105742.
[7] 朱洪堃, 殷佳炜, 冯文宇, 等. 一种轻量化实时人体姿势检测模型研究与应用[J]. 系统仿真学报, 2020, 32(11): 2155-2165.
Zhu Hongkun, Yin Jiawei, Feng Wenyu, et al.Research and Application of a Lightweight Real-time Human Pose Detection Model[J]. Journal of System Simulation, 2020, 32(11): 2155-2165.
[8] 冯文宇, 朱洪堃, 殷佳炜, 等. 无人CT智能姿态识别算法研究[J]. 仪器仪表学报, 2020, 41(8): 188-195.
Feng Wenyu, Zhu Hongkun, Yin Jiawei, et al.Research on Intelligent Attitude Recognition Algorithm of Unmanned CT[J]. Chinese Journal of Scientific Instrument, 2020, 41(8): 188-195.
[9] Fu L, Feng Y, Wu J, et al.Fast and Accurate Detection of Kiwifruit in Orchard Using Improved YOLOv3-tiny Model[J]. Precision Agriculture (S1385-2256), 2021, 22(3): 754-776.
[10] Bao C, Xie T, Feng W, et al.A Power-Efficient Optimizing Framework FPGA Accelerator Based on Winograd for YOLO[J]. IEEE Access (S2169-3536), 2020, 8: 94307-94317.
[11] Li Y, Li S, Du H, et al.YOLO-ACN: Focusing on Small Target and Occluded Object Detection[J]. IEEE Access (S2169-3536), 2020, 8: 227288-227303.
[12] Zhang X, Zhou X, Lin M, et al.Shufflenet: An extremely Efficient Convolutional Neural Network for Mobile Devices[C]// IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City: CVPR, 2018: 6848-6856.
[13] Han K, Wang Y, Tian Q, et al.Ghostnet: More Features from Cheap Operations[C]// IEEE/CVF Conference on Computer Vision and Pattern Recognition. ELECTR NETWORK: CVPR, 2020: 1580-1589.
[14] Hu J, Shen L, Albanie S, et al.Squeeze-and-Excitation Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2020, 42(8): 2011-2023.
[15] Fan H, Mei X, Prokhorov D, et al.Multi-level Contextual RNNS with Attention Model for Scene Labeling[J]. IEEE Transactions on Intelligent Transportation Systems (S1524-9050), 2018, 19(11): 3475-3485.
[16] 石敏, 姚瀚钦, 李淳芃, 等. 基于深度Alignment网络的足部测量[J]. 系统仿真学报, 2020, 32(7): 1267-1278.
Shi Min, Yao Hanqin, Li Chunpeng, et al.Foot Measurement Based on Deep Alignment Network[J]. Journal of System Simulation, 2020, 32(7): 1267-1278.
[17] Duan P, Wang T, Cui M, et al.Multi-person Pose Estimation Based on a Deep Convolutional Neural Network[J]. Journal of Visual Communication and Image Representation (S1047-3203), 2019, 62: 245-252.
[18] Shankar V, Roelofs R, Mania H, et al.Evaluating Machine Accuracy on Imagenet[C]// International Conference on Machine Learning. Salt Lake City: PMLR, 2020: 8634-8644.
[19] Suh S, Lee H, Lukowicz P, et al.CEGAN: Classification Enhancement Generative Adversarial Networks for Unraveling Data Imbalance Problems[J]. Neural Networks (S0893-6080), 2021, 133: 69-86.
[20] Sandler M, Howard A, Zhu M, et al.Mobilenetv2: Inverted Residuals and Linear Bottlenecks[C]// IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City: CVPR, 2018: 4510-4520.
[21] Tan M, Le Q.Efficientnet: Rethinking Model Scaling for Convolutional Neural Networks[C]// International Conference on Machine Learning. Long Beach: LCML, 2019: 6105-6114.
[22] Huang G, Liu S, Van D M L, et al. Condensenet: An Efficient Densenet Using Learned Group Convolutions[C]// IEEE Conference on Computer Vision and Pattern Recognition. Houston: Medical Imaging Conference-Image-Guided Procedures, 2018: 2752-2761.