Women's Safety Through Real-Time Gesture Recognition Using Mediapipe and Machine Learning Algorithms

Women's safety has to get better as technology shapes human communication, particularly with alternative signals. Protection at night attends to women's particular threats and vulnerabilities. Public safety concerns include emergency management, harassment, and sexual assault. MediaPipe and machine learning approaches form our sophisticated safety solution. We use gradient boosting, Gaussian Naive Bayes, and a decision tree to teach the model. Tests then assess the optimal performance of the model with regard to F1 score, precision, recall. These new technologies are meant to empower women with security, awareness, and tools for environmental change. MediaPipe clustered hand tracking and key point identification help to improve sign language recognition. Open CV records webcam video for real-time model comparison. It promises to be sign language-based in predicting. To increase system dependability, we included a database matching tool alerting and send pictures should any indication match recommended motions. This function fast alerts authorities to fix safety concerns. At last, the real-time response and gesture detection of the system will safeguard women.