Autonomous Multi-Modal Proactive Safety System for Women’s Security using Edge-AI and Smart Embedded Systems

References

[1] J. Redmon and A. Farhadi, "YOLOv3: An Incremental Improvement," arXiv preprint arXiv:1804.02767, 2018.

[2] C. Lugaresi et al., "MediaPipe: A Framework for Building Perception Pipelines," arXiv:1906.08172, 2019.

[3] A. Vaswani et al., "Attention is All You Need," Advances in Neural Information Processing Systems, pp. 5998-6008, 2017.

[4] W. Han et al., "Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding," ICLR, 2016.

[5] Raspberry Pi Foundation, "Broadcom BCM2712 SoC Data Sheet," v1.1, 2024.

[6] M. Abadi et al., "TensorFlow: A System for Large-Scale Machine Learning," 12th USENIX Symposium on Operating Systems Design and Implementation, 2016.

[7] G. Hinton, O. Vinyals, and J. Dean, "Distilling the Knowledge in a Neural Network," arXiv:1503.02531, 2015.

[8] S. Ren, K. He, R. Girshick, and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Trans. Pattern Anal. Mach. Intell., 2017.

[9] B. Jacob et al., "Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference," IEEE CVPR, 2018.

[10] M. Sandler et al., "MobileNetV2: Inverted Residuals and Linear Bottlenecks," IEEE CVPR, 2018.

[11] J. Lin et al., "Edge AI: On-Device Intelligence in the Era of IoT and 5G," Proceedings of the IEEE, 2023.

[12] V. Sze et al., "Efficient Processing of Deep Neural Networks," Synthesis Lectures on Computer Architecture, 2017.

[13] S. Gupta, "Railway Safety Management Systems: A Global Perspective," Journal of Transportation Tech, 2022.

[14] L. Wang, "Human Activity Recognition using Wearable Sensors and Edge Computing," IEEE Sensors Journal, 2021.

[15] R. Singh et al., "Acoustic Event Detection in High-Noise Environments," Int. Conf. on Signal Processing, 2023.

[16] K. He et al., "Deep Residual Learning for Image Recognition," IEEE CVPR, 2016.

[17] SIMCom Wireless Solutions, "SIM800L Series Hardware Design Guide," 2020.

[18] T. Chen and C. Guestrin, "XGBoost: A Scalable Tree Boosting System," ACM SIGKDD, 2016.

[19] P. Warden and D. Situnayake, "TinyML: Machine Learning with TensorFlow Lite on Arduino and Ultra-Low-Power Microcontrollers," O'Reilly Media, 2019.

[20] Z. Zhao et al., "Object Detection with Deep Learning: A Review," IEEE Trans. on Neural Networks and Learning Systems, 2019.

[21] Y. LeCun, Y. Bengio, and G. Hinton, "Deep Learning," Nature, vol. 521, pp. 436-444, 2015.

[22] A. Krizhevsky et al., "ImageNet Classification with Deep Convolutional Neural Networks," Communications of the ACM, 2017.

[23] F. Chollet, "Xception: Deep Learning with Depthwise Separable Convolutions," IEEE CVPR, 2017.

[24] S. Hochreiter and J. Schmidhuber, "Long Short-Term Memory," Neural Computation, 1997.

[25] D. Kingma and J. Ba, "Adam: A Method for Stochastic Optimization," ICLR, 2015.

[26] J. Long et al., "Fully Convolutional Networks for Semantic Segmentation," IEEE CVPR, 2015.

[27] R. Girshick, "Fast R-CNN," IEEE Int. Conf. on Computer Vision, 2015.

[28] K. Simonyan and A. Zisserman, "Very Deep Convolutional Networks for Large-Scale Image Recognition," arXiv:1409.1556, 2014.

[29] C. Szegedy et al., "Going Deeper with Convolutions," IEEE CVPR, 2015.

[30] A. Howard et al., "Searching for MobileNetV3," IEEE ICCV, 2019.

[31] X. Glorot and Y. Bengio, "Understanding the difficulty of training deep feedforward neural networks," AISTATS, 2010.

[32] N. Srivastava et al., "Dropout: A Simple Way to Prevent Neural Networks from Overfitting," JMLR, 2014.

[33] S. Ioffe and C. Szegedy, "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift," ICML, 2015.

[34] H. Zhang et al., "Mixup: Beyond Empirical Risk Minimization," ICLR, 2018.

[35] Y. Gal and Z. Ghahramani, "Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning," ICML, 2016.

[36] G. Huang et al., "Densely Connected Convolutional Networks," IEEE CVPR, 2017.

[37] M. Tan and Q. Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks," ICML, 2019.

[38] B. Zoph and Q. Le, "Neural Architecture Search with Reinforcement Learning," ICLR, 2017.

[39] . Amodei et al., "Deep Speech 2: End-to-End Speech Recognition in English and Mandarin," ICML, 2016.

[40] N. Smith, "Cyclical Learning Rates for Training Neural Networks," IEEE WACV, 2017.

@article{1715210,
      author = {Dr. Ravindra Duche, Aastha Verma, Rishabh Raj, Shweta Yadav, Divya Doke},
      title = {Autonomous Multi-Modal Proactive Safety System for Women’s Security using Edge-AI and Smart Embedded Systems},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {9},
      pages = {1343-1354},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1715210.pdf},
      abstract = {Ensuring public safety for women in transit environments, particularly within isolated train coaches, has emerged as a critical sociotechnical challenge. While contemporary AI-driven surveillance systems offer high accuracy, their deployment is frequently hindered by the prohibitive costs of specialized edge-computing hardware, such as NVIDIA Jetson modules, which limits large-scale implementation in developing infrastructure. This paper presents the design and development of a low-cost, multimodal safety response system optimized for the Raspberry Pi 5 platform. The technical architecture leverages OpenCV and TensorFlow Lite to perform concurrent blink detection (using Eye Aspect Ratio algorithms), human fall detection, and facial recognition to identify potential distress or medical emergencies. These responses include the immediate activation of high-intensity alarm lights and buzzers, the deployment of physical safety partitions through servo-driven mechanisms, and the release of a deterrent fog sprayer to obstruct an aggressor's vision .Furthermore, the system ensures reliable remote notification by employing a SIM800L GSM module to transmit instantaneous SMS alerts and GPS coordinates to railway authorities This study demonstrates that a highly functional, responsive, and realistic safety solution can be built using affordable edge-AI components, offering a scalable blueprint for enhancing public safety in mass transit systems},
      keywords = {Raspberry Pi, Women Safety, Computer Vision, TensorFlow Lite, IoT, Real-time Detection, Emergency Response.},
      month = {March},
      doi = {https://doi.org/10.64388/IREV9I9-1715210}
  }