REAL-TIME FACIAL RECOGNITION FOR ANTI-THEFT WITH ML SVM AND HOG FEATUREs

This study presents a robust and efficient real-time anti-theft system employing facial recognition, supported by machine learning techniques such as Support Vector Machines (SVM) and Histogram of Oriented Gradients (HOG) feature extraction. The proposed system aims to enhance security measures by detecting and preventing theft incidents in real-time scenarios. The facial recognition component leverages advanced machine learning algorithms, particularly SVM, to accurately identify and verify individuals. Additionally, HOG feature extraction is employed to capture and represent the distinctive features of facial images, enhancing the overall recognition accuracy. The integration of these technologies results in a responsive and accurate anti-theft solution. The system's real-time capabilities are crucial for prompt threat identification, allowing timely intervention to prevent theft. Experimental results demonstrate the effectiveness of the proposed approach, showcasing its potential for deployment in various security-sensitive environments. This research contributes to the advancement of anti-theft systems by leveraging state-of-the-art technologies in facial recognition and machine learning for enhanced protection.

Key Words: criminal dataset, Hog feature extraction, Machine Learning Algorithms, Support Vector Machine (SVM), Criminal Classification.