Empowering Women’s Safety: Deep Learning Approaches to Combat Violence

DOI

10.2991/978-94-6463-700-7_22How to use a DOI?

Keywords

Data Augmentation; Annotation; Gender Classification; Alert system; SOS signals; Yolov8; Object Detection

Abstract

The “Empowering Women’s Safety” project helps in growing the safety concerns for women by sensing real-time threats and their prevention. The system recognizes dangerous situations, such as a lone woman at night or a woman surrounded by men, by continuously monitoring public places through person detection, gender classification, and gesture recognition. It also generates alerts that can be acted on at the correct time to prevent incidents from getting worse. The major challenge addressed is the lack of correct safety measures. Although the traditional methods are indeed reactive, in that they react to crimes after they have occurred, the early warning nature of the system limits the opportunity for violence against women. These innovative analytics, done by using object detection model YOLOv8 and deep learning techniques augmentation and annotation, can be used to highlight gender distribution at particular sites and identify hotspots where there is a high possibility of threats towards women and send an alert to the control systems to safeguard women. This will benefit women themselves, law enforcement, and city planners with respect to public safety. The project works with the trend of violence against women in cities and advances technology that can detect all forms of threats before they occur. Maintaining constant observation and monitoring of places leads to proactive solutions. As a result, safer places are created for women, and crimes are eradicated before happening. The system’s effectiveness was validated using advanced metrics, demonstrating impressive performance with a mean Average Precision (mAP@50) of 0.864 across all classes. The model showcased particularly high precision for individual categories, such as 0.873 for lone women and 0.94 for women surrounded by men, indicating its robustness in diverse scenarios. These results underscore the model’s superior ability to accurately detect threats in real-time, surpassing traditional approaches and setting a new standard for proactive safety measures.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - B. Ch. S. N. L. S. Sai Baba
AU  - Jagana Deepak Kumar
AU  - J. Vinay Siva Subhash Kotha
AU  - Javvadi Karthik
AU  - Jamula Subhashini
PY  - 2025
DA  - 2025/04/19
TI  - Empowering Women’s Safety: Deep Learning Approaches to Combat Violence
BT  - Proceedings of the International Conference on Advancements in Computing Technologies and Artificial Intelligence (COMPUTATIA-2025)
PB  - Atlantis Press
SP  - 271
EP  - 287
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-700-7_22
DO  - 10.2991/978-94-6463-700-7_22
ID  - SaiBaba2025
ER  -