All Issue

2023 Vol.19, Issue 4 Preview Page

Original Article

An Untrained Person's Posture Estimation Scheme by Exploiting a Single 24GHz FMCW Radar and 2D CNN

단일 24GHz FMCW 레이더 및 2D CNN을 이용하여 학습되지 않은 요구조자의 자세 추정 기법

장경석1ㆍ주준호2ㆍ손초3ㆍ김영억4*

Kyongseok Jang1, Junhao Zhou2, Chao Sun3, Youngok Kim4*

1Master Course, Department of Electronic Engineering, Kwangwoon University, Seoul, Republic of Korea
2Ph.D Course, Department of Electronic Engineering, Kwangwoon University, Seoul, Republic of Korea
3Ph.D, Department of Electronic Engineering, Kwangwoon University, Seoul, Republic of Korea
4Professor, Department of Electronic Engineering, Kwangwoon University, Seoul, Republic of Korea
31 December 2023. pp. 897-907
PDF Citation
Abstract
Purpose: In this study, We aim to estimate a untrained person's three postures using a 2D CNN model which is trained with minimal FFT data collected by a 24GHz FMCW radar. Method: In an indoor space, we collected FFT data for three distinct postures (standing, sitting, and lying) from three different individuals. To apply this data to a 2D CNN model, we first converted the collected data into 2D images. These images were then trained using the 2D CNN model to recognize the distinct features of each posture. Following the training, we evaluated the model's accuracy in differentiating the posture features across various individuals. Result: According to the experimental results, the average accuracy of the proposed scheme for the three postures was shown to be a 89.99% and it outperforms the conventional 1D CNN and the SVM schemes. Conclusion: In this study, we aim to estimate any person's three postures using a 2D CNN model and a 24GHz FMCW radar for disastrous situations in indoor. it is shown that the different posture of any persons can be accurately estimated even though his or her data is not used for training the AI model.
Keywords
Disaster
Emergency
FMCW Radar
Deep Learning
FFT
Posture Estimation
연구목적: 본 연구에서는 단일 24GHz FMCW레이더를 사용하여 수집된 적은 양의 학습데이터로 학습된 AI 모델을 사용하여 학습되지 않은 사람의 3가지 자세를 구분하고자 한다. 연구방법: 실내에서 학습 대상자들의 3가지 자세(서기, 앉기, 눕기)에 대한 FFT데이터를 수집하여 2D 이미지로 변환시킨 후 제안하는 2D CNN 모델로 학습시켜 학습에 사용되지 않은 새로운 대상자들의 자세를 잘 구분할 수 있는지 실험을 통해 정확도를 분석하였다. 연구결과: 제안하는 기법을 통해 3가지 자세의 평균 정확도가 89.99%임을 보였고, 기존의 1D CNN이나 SVM 보다 성능이 향상되었다. 결론: 실내에서 재난이 발생하는 경우 단일 FMCW 레이더와 AI 기법을 통해 요구조자의 자세를 추정하고자 하였으며, 학습되지 않은 대상자의 자세도 높은 정확도로 추정이 가능함을 실험을 통해 확인하였다.
키워드
재난
응급
FMCW 레이더
딥러닝
FFT
자세 추정
References
  1. Abdu, F.J., Zhang Y., Deng Z. (2022). "Activity classification based on feature fusion of FMCW radar human motion micro-doppler signatures." IEEE Sensors Journal, Vol. 22, No. 9, pp. 8648-8662. 10.1109/JSEN.2022.3156762
  2. Chin, L.C.K., Eu, K.S., Tay, T.T., Teoh, C.Y., Yap, K.M. (2019). "A posture recognition model dedicated for differentiating between proper and improper sitting posture with kinect sensor." 2019 IEEE International Symposium on Haptic, Audio and Visual Environments and Games (HAVE), Subang Jaya, Malaysia, pp. 1-5. 10.1109/HAVE.2019.8920964
  3. Cui, M., Fnag, J., Zhao, Y. (2020). "Emotion recognition of human body's posture in open environment." 2020 Chinese Control and Decision Conference (CCDC), Hefei, China, pp. 3294-3299. 10.1109/CCDC49329.2020.9164551
  4. Khatun, M.A., Yousuf, M.A., Ahmed, S., Uddin, M.Z., Alyami, S.A., Al-Ashhab, S., Akhdar, H.F., Khan, A., Azad, A., Moni, M.A. (2022). "Deep CNN-LSTM with self-attention model for human activity recognition using wearable sensor." IEEE Journal of Translational Engineering in Health and Medicine, Vol. 10, pp. 1-16. 10.1109/JTEHM.2022.3177710 35795873 PMC9252338
  5. Lee, J., Park, K.E., Kim, Y. (2021). "A study on indoor positioning based on pedestrian dead reckoning using inertial measurement unit." Journal of the Society of Disaster Information, Vol. 17, No. 3, pp. 521-534.
  6. Luo, F., Poslad S., Bodanese, E. (2019). "Kitchen activity detection for healthcare using a low-power radar-enabled sensor network." 2019 IEEE International Conference on Communications (ICC), Shanghai, China, pp. 1-7. 10.1109/ICC.2019.8761484
  7. Nalci, D., Akgul, Y.S. (2022). "Human action recognition with raw millimeter wave radar data." 2022 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA), Ankara, Turkey, pp. 1-5. 10.1109/HORA55278.2022.9800009
  8. Neili, S., Gazzah, S., El Yacoubi, M.A., Ben Amara, N.E. (2017). "Human posture recognition approach based on ConvNets and SVM classifier." 2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), Fez, Morocco, pp. 1-6. 10.1109/ATSIP.2017.8075518
  9. Tong, L., Ma, H., Lin, Q., He, J., Peng, L. (2022). "A novel deep learning Bi-GRU-I model for real-time human activity recognition using inertial sensors." IEEE Sensors Journal, Vol. 22, No. 6, pp. 6164-6174. 10.1109/JSEN.2022.3148431
  10. Yang, S., Kim, Y. (2023). "Single 24-GHz FMCW radar-based indoor device-free human localization and posture sensing with CNN." IEEE Sensors Journal, Vol. 23, No. 3, pp. 3059-3068. 10.1109/JSEN.2022.3227025
  11. Yun, Y., Sohn, J.W., Kim, Y. (2023). "A weak signal detection algorithm in clutter environment for indoor location estimation based on IR-UWB radar." Journal of the Society of Disaster Information, Vol. 19, No. 1, pp. 10-17. 10.15683/kosdi.2023.3.31.010
Information
  • Publisher :The Korean Society of Disaster Information
  • Publisher(Ko) :한국재난정보학회
  • Journal Title :Journal of the Society of Disaster Information
  • Journal Title(Ko) :한국재난정보학회논문집
  • Volume : 19
  • No :4
  • Pages :897-907
  • DOI :https://doi.org/10.15683/kosdi.2023.12.31.897
Journal Informaiton Journal of the Society of Disaster Information Journal of the Society of Disaster Information
  • NRF
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close