Research on contactless control of elevator based on machine vision
DOI:
https://doi.org/10.54097/hset.v7i.1022Keywords:
Gesture recognition, Machine vision, Non-contact elevator button, YOLOv5, COVID-19Abstract
Aiming at the problem of cross-infection caused by elevator public buttons during the COVID-19 epidemic, a non-contact elevator button control gesture recognition system based on machine vision is designed. In order to improve the detection speed of gesture recognition, combined with the Spatial Pyramid Pooling (SPP) and replaced the Backbone in YOLOv5 with the lightweight model ShuffleNetV2, an improved YOLOv5_shff algorithm was proposed. After testing, in the task of recognizing gestures, the detection speed of the YOLOv5_shff algorithm is 14% higher than the original model, and the detection accuracy is 0.1% higher than the original model. Taking the improved YOLOv5_shff algorithm as the core, a gesture recognition system that can be applied to elevator button control is designed. The experimental data shows that the gesture recognition accuracy for controlling elevator buttons reaches 99.3%, which can meet the requirements of non-contact control of public elevators.
Aiming at the problem of cross-infection caused by elevator public buttons during the COVID-19 epidemic, a non-contact elevator button control gesture recognition system based on machine vision is designed. In order to improve the detection speed of gesture recognition, combined with the Spatial Pyramid Pooling (SPP) and replaced the Backbone in YOLOv5 with the lightweight model ShuffleNetV2, an improved YOLOv5_shff algorithm was proposed. After testing, in the task of recognizing gestures, the detection speed of the YOLOv5_shff algorithm is 14% higher than the original model, and the detection accuracy is 0.1% higher than the original model. Taking the improved YOLOv5_shff algorithm as the core, a gesture recognition system that can be applied to elevator button control is designed. The experimental data shows that the gesture recognition accuracy for controlling elevator buttons reaches 99.3%, which can meet the requirements of non-contact control of public elevators.
Downloads
References
Bose S R, Kumar V S. Efficient inception V2 based deep convolutional neural network for real-time hand action recognition [J]. IET Image Processing, 2020, 14(4): 688-696.
Lu Z, Qin S, Li L, et al. One-shot learning hand gesture recognition based on lightweight 3D convolutional neural networks for portable applications on mobile systems [J]. IEEE Access, 2019, 7: 131732-131748.
Oudah M, Al-Naji A, Chahl J. Hand gesture recognition based on computer vision: a review of techniques[J]. journal of Imaging, 2020, 6(8): 73.
Sun Y, Armengol-Urpi A, Kantareddy S N R, et al. Magichand: Interact with iot devices in augmented reality environment[C]//2019 IEEE Conference on Virtual Reality and 3D User Interfaces (VR). IEEE, 2019: 1738-1743.
Katti J, Kulkarni A, Pachange A, et al. Contactless Elevator Based on Hand Gestures During Covid 19 Like Pandemics[C]//2021 7th International Conference on Advanced Computing and Communication Systems (ICACCS). IEEE, 2021, 1: 672-676.
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, 2019, 67(4): 940-948.
Wong W, Juwono F H, Khoo B T T. Multi-features capacitive hand gesture recognition sensor: a machine learning approach [J]. IEEE Sensors Journal, 2021, 21(6): 8441-8450.
Valueva M V, Nagornov 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, 2020, 177: 232-243.
Mao Q-C, Sun H-M, Liu Y-B, et al. Mini-YOLOv3: real-time object detector for embedded applications [J]. Ieee Access, 2019, 7: 133529-133538.
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, 2020, 178: 105742.
Ma N, Zhang X, Zheng H T, et al. Shufflenet v2: Practical guidelines for efficient cnn architecture design[C]//Proceedings of the European conference on computer vision (ECCV). 2018: 116-131.
Howard A G, Zhu M, Chen B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications [J]. arXiv preprint arXiv:17040.4861, 2017, CVPR 2017.
Fronteddu G, Porcu S, Floris A, et al. A Dynamic Hand Gesture Recognition Dataset for Human-Computer Interfaces [J]. Computer Networks, 2022: 108781.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
