Integración ConvNeXt-YOLO mediante CVV para detectar caídas en robot social

  1. Sánchez-Girón, Celia 1
  2. García Gómez, Miguel 1
  3. Duque Domingo, Jaime 1
  4. Gómez García-Bermejo, Jaime 1
  5. Zalama Casanova, Eduardo 1
  1. 1 Universidad de Valladolid
    info

    Universidad de Valladolid

    Valladolid, España

    ROR https://ror.org/01fvbaw18

Journal:
Jornadas de Automática
  1. Cruz Martín, Ana María (coord.)
  2. Arévalo Espejo, V. (coord.)
  3. Fernández Lozano, Juan Jesús (coord.)

ISSN: 3045-4093

Year of publication: 2024

Issue: 45

Type: Article

DOI: 10.17979/JA-CEA.2024.45.10788 DIALNET GOOGLE SCHOLAR lock_openOpen access editor

Abstract

More and more older adults are choosing to live at home, which has created a critical need to ensure safe environmentsfor this population. 50 % of people over the age of 80 experience at least one fall per year. This study seeks to detect falls by implementing a vision system, providing a rapid response in case of emergency, so that the fallen person has assistance if he/she suffers an accident. Here we propose a methodology based on the use of deep learning models, specifically using the Cross Validation Voting(CVV) technique, to improve generalization and accuracy in detecting falls from images. The proposed model achieved an accuracy of 92.95 % and a loss of 0.1885 for the test set. The fall detection system has been integrated into theTemi social robot, which will be introduced in the users’ home to continuously monitor their well-being and provide immediate assistance in case a fall is detected.

Bibliographic References

  • Antonello, M., Carraro, M., Pierobon, M., Menegatti, E., 2017. Fast and robust detection of fallen people from a mobile robot. In: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, pp. 4159–4166. DOI: https://doi.org/10.1109/IROS.2017.8206276
  • Anwary, A. R., Rahman, M. A., Muzahid, A. J. M., Ul Ashraf, A. W., Patwary, M., Hussain, A., 2022. Deep learning enabled fall detection exploiting gait analysis. In: 2022 44th Annual International Conference of the IEEE Engineering in Medicine Biology Society (EMBC). pp. 4683–4686. DOI: 10.1109/EMBC48229.2022.9871964 DOI: https://doi.org/10.1109/EMBC48229.2022.9871964
  • Domingo, J. D., Aparicio, R. M., Rodrigo, L. M. G., 2022. Cross validation voting for improving CNN classification in grocery products. IEEE Access 10, 20913–20925. DOI: https://doi.org/10.1109/ACCESS.2022.3152224
  • EIAROB, 2022. Ecosistema de inteligencia ambiental para el apoyo a los cuidados de larga duración en el hogar mediante uso de robots sociales. https://www.itap.uva.es/en/eiarob/.
  • INE, 2023.
  • Jocher, G., Chaurasia, A., Qiu, J., Jan. 2023. Ultralytics YOLO. URL: https://github.com/ultralytics/ultralytics
  • Kwolek, B., Kepski, M., 2014. Human fall detection on embedded platform using depth maps and wireless accelerometer. Computer Methods and Programs in Biomedicine 117 (3), 489–501. DOI: https://doi.org/10.1016/j.cmpb.2014.09.005
  • Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., Xie, S., 2022. A ConvNet for the 2020s. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11976–11986. DOI: https://doi.org/10.1109/CVPR52688.2022.01167
  • Maldonado-Bascón, S., Iglesias-Iglesias, C., Martín-Martín, P., Lafuente-Arroyo, S., 2019. Fallen people detection capabilities using assistive robot. Electronics 8 (9), 915. DOI: https://doi.org/10.3390/electronics8090915
  • Martínez-Villasenor, L., Ponce, H., Brieva, J., Moya-Albor, E., Núñez-Martínez, J., Penafort-Asturiano, C., 2019. Up-fall detection dataset: A multimodal approach. Sensors 19 (9), 1988. DOI: https://doi.org/10.3390/s19091988
  • Merino-Fidalgo, S., Zalama, E., Gómez-García-Bermejo, J., Duque-Domingo, J., Gómez, R., Viñas, P., García, D., Uruena, H., May 2023. Sistema de monitorización no intrusiva para vivienda de personas mayores. Jornadas Nacionales de Robótica y Bioingeniería 2023: Libro de actas, 115–121. No publicado. DOI: 10.20868/UPM.BOOK.74896 DOI: https://doi.org/10.20868/UPM.book.74896
  • Quinayas Burgos, C. A., Quintero Benavidez, D. F., Ruíz Omen, E., Narváez Semanate, J. L., 2020. Sistema de detección de caídas en personas utilizando vídeo vigilancia. Ingeniare. Revista chilena de ingeniería 28 (4), 684–693. DOI: https://doi.org/10.4067/S0718-33052020000400684
  • Rao, S., Pramod, N., Paturu, C. K., 2008. People detection in image and video data. In: Proceedings of the 1st ACM Workshop on Vision Networks for Behavior Analysis. pp. 85–92. DOI: https://doi.org/10.1145/1461893.1461909
  • SAM, Apr. 2024. Pose detection dataset. https://universe.roboflow.com/sam-vcqdz/pose-detection-gwapj, visited on 2024-05-16. URL: https://universe.roboflow.com/sam-vcqdz/pose-detection-gwapj
  • Schaffer, C., 1993. Selecting a classification method by cross-validation. Machine Learning 13, 135–143. DOI: https://doi.org/10.1007/BF00993106
  • Sánchez-Girón, C., García-Gómez, M., Duque-Domingo, J., Gómez-García-Bermejo, J., Zalama, E., May 2024. Detección de caídas con un robot social aplicando visión artificial. Jornadas Nacionales de Robótica y Bioingeniería 2024: Libro de actas. No publicado.
  • Temi, R., 2024. URL: https://www.robotemi.com/product/temi/
  • Workspace, Jan. 2024. Test2 dataset. https://universe.roboflow.com/workspace-3pvv7/test2-fqv2c, visited on 2024-05-16. URL: https://universe.roboflow.com/workspace-3pvv7/test2-fqv2c
Data source: Dialnet