Provendo um modelo automático de detecção de quedas baseado em rede adversária generativa para assistência de idosos

• Allan Costa N. dos Santos UFF
• Flávio Luiz Seixas UFF
• Natalia Castro Fernandes UFF

DOI: https://doi.org/10.5753/sbcas.2022.222471

Abstract

Falls are a serious public health problem and people over 65 are among the most vulnerable to serious injury from a fall. This article proposes and evaluates a model of neural network architecture, called CNN Video Stream Combination (CVSC), to improve the monitoring and safety of the elderly. It is proposed to use a generative neural network to calculate the anomaly score. A model capable of working simultaneously with RGB and infrared cameras is proposed, as falls of older people often occur in low-light environments. The CVSC using an RGB camera presented 97.00% of sensitivity and with an infrared camera it obtained 94.00% of sensitivity.

References

Belasco, A. G. S. and Okuno, M. F. P. (2019). Reality and challenges of ageing.

Chen, Z., Wang, Y., and Yang, W. (2021). Video based fall detection using human poses.

Dai, J., Li, Y., He, K., and Sun, J. (2016). R-fcn: Object detection via region-based fully convolutional networks.

Khan, S. S., Nogas, J., and Mihailidis, A. (2020). Spatio-temporal adversarial learning for detecting unseen falls. Pattern Analysis and Applications, 24(1):381-391.

Lin, T.-Y., Maire, M., Belongie, S., Bourdev, L., Girshick, R., Hays, J., Perona, P., Ramanan, D., Zitnick, C. L., and Dollár, P. (2015). Microsoft coco: Common objects in context.

Liu, W., Luo, W., Lian, D., and Gao, S. (2018). Future frame prediction for anomaly detection - a new baseline. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 6536-6545.

Mehta, V., Dhall, A., Pal, S., and Khan, S. S. (2021). Motion and region aware adversarial learning for fall detection with thermal imaging. In 2020 25th International Conference on Pattern Recognition (ICPR), pages 6321-6328.

Mshali, H., Lemlouma, T., and Magoni, D. (2018). Adaptive monitoring system for e-health smart homes. Pervasive and Mobile Computing, 43:1 - 19.

Nogas, J., Khan, S., and Mihailidis, A. (EasyChair, 2019). Fall detection from thermal camera using convolutional lstm autoencoder. EasyChair Preprint no. 824.

OpenCV (2022). OpenCV modules. https://docs.opencv.org/4.x/. Acessado em: 31 de Janeiro de 2022.

Pannurat, N., Thiemjarus, S., and Nantajeewarawat, E. (2014). Automatic fall monitoring: A review. Sensors, 14(7):12900-12936.

Quigley, P. A., Campbell, R. R., Bulat, T., Olney, R. L., Buerhaus, P., and Needleman, J. (2012). Incidence and cost of serious fall-related injuries in nursing homes. Clinical nursing research, 21(1):10-23.

Ramachandran, A. and Karuppiah, A. (2020). A survey on recent advances in wearable fall detection systems. BioMed Research International, 2020.

Shahroudy, A., Liu, J., Ng, T.-T., and Wang, G. (2016). Ntu rgb+d: A large scale dataset for 3d human activity analysis. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 1010-1019.

Shojaei-Hashemi, A., Nasiopoulos, P., Little, J. J., and Pourazad, M. T. (2018). Video-based human fall detection in smart homes using deep learning. In 2018 IEEE International Symposium on Circuits and Systems (ISCAS), pages 1-5.

Xu, Q., Huang, G., Yu, M., and Guo, Y. (2020). Fall prediction based on key points of human bones. Physica A: Statistical Mechanics and its Applications, 540(C).