Abstract：The DSConvBiGRU network model which is suitable for embedded systems and combines depthwise separable convolutional neural networks and bidirectional gated recurrent units for the classification of dynamic gesture sequences is proposed. A dynamic gesture recognition solution which utilizes a low-resolution thermopile array sensor is designed and implemented. An experimental dataset comprising various dynamic gestures has been constructed and publicated on open website. The deployment of the pre-trained network model on the Raspberry Pi edge device has been accomplished. The system preprocesses 20 consecutive temperature matrices exported by the sensor through interval mapping, background subtraction, Lanczos interpolation, and Otsu thresholding to obtain a single dynamic gesture sequence. Subsequently, the pre-trained DSConvBiGRU network is employed for the classification. Experimental results demonstrate that the network model achieves an accuracy of 99.291% on test dataset. The time comsunption of preprocess and inference on the edge device is 5.513ms and 8.231ms respectively. The system meets the design requirements for low-power consumption, high precision, and real-time performance.
TIAN Q H, YANG H M, LIANG Q L, et al. Overview on vision-based dynamic gesture recognition［J］. Journal of Zhejiang Sci-Tech University(Natural Sciences), 2020, 43(4):557-569.
CHENG S H, YANG Z H, WANG C. Research on Gesture Recognition Algorithm under Multi-channel Fusion and Design of Ship Virtual Interaction Platform［J］. Acta Metrologica Sinica, 2022, 43(7): 856-862.
LIU X H, DENG B S, PEI Y, et al. Research on Wearable Gesture Interaction System and Recognition Algorithm［J］. Journal of Chinese Computer Systems, 2020, 41(11):2241-2248.
QU L L, WU H X, YANG T H, et al. Hand Gesture Recognition Method Using FMCW Radar Based on Dual Channel CNN［J］. Chinese Journal of Electron Devices, 2022, 45(6):1469-1475.
［11］
TATENO S, ZHU Y W, MENG F X. Hand Gesture Recognition System for In-car Device Control Based on Infrared Array Sensor［C］//58th Annual Conference of the Society of Instrument and Control Engineers of Japan. Hiroshima, Japan, 2019:701-706.
［13］
余丹. 基于热像仪的手势识别研究［D］. 南京:东南大学,2019.
［15］
KEN T. Filters for common resampling tasks［J］. Graphics gems, 1990: 147-165.
CHOLLET F. Xception:Deep learning with depthwise separable convolutions［C］//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA, 2017: 1251-1258.
TRAN D, BOURDEV L, FERGUS R, et al. Learning spatiotemporal features with 3d convolutional networks［C］//IEEE International Conference on Computer Vision(ICCV). Santiago, Chile,2015: 4489-4497.
［23］
HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetV3［C］// IEEE/CVF International Conference on Computer Vision (ICCV). Seoul, Korea, 2019: 1314-1324.
VANDERSTEEGEN M, REUSEN W, VAN BEECK K, et al. Low-latency hand gesture recognition with a low resolution thermal imager［C］//IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Seattle, USA, 2020:98-99.
［22］
MUTHUKUMAR K A, BOUAZIZI M, OHTSUKI T. A novel hybrid deep learning model for activity detection using wide-angle low-resolution infrared array sensor［J］. IEEE Access, 2021, 9: 82563-82576.
［4］
JACOB A, KOSHY M, NISHA K K, et al. Real Time Static and Dynamic Hand Gestures Cognizance for Human Computer Interaction［C］//International Conference on Advances in Computing and Communications (ICACC). Kochi, Kakkanad, India, 2021:1-6.
LI S R. Infrared Image Rnhancement Technology Based on Improved Histogram Equalization［J］. Industrial Control Computer, 2022, 35(12):52-53, 56.
［14］
LIANG Q S, YU L, ZHAI X P, et al. Activity Recognition Based on Thermopile Imaging Array Sensor［C］//IEEE International Conference on Electro/Information Technology(EIT). Rochester, USA, 2018: 770-773.
ZHANG L F, WANG Z, WU S C. Flow Pattern Recognition of Gas-Liquid Two-phase Flow Based on Convolutional Neural Network and Gated Recurrent Unit［J］. Acta Metrologica Sinica, 2022, 43(10):1306-1312.
［21］
FAN X, ZHANG H, LEUNG C, et al. Robust unobtrusive fall detection using infrared array sensors［C］//IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI).Daegu, Korea, 2017: 194-199.
WANG Z C, YANG X W, HUANG J Y, et al. Fault Diagnosis of Turnout Switch Machine Based on ITD-SDP Image Features and DSCNN［J］. Journal of the China Railway Society, 2023, 45(5): 65-71.
2024, Vol. 45 
