REAL TIME FACE AUTHENTICATION SYSTEM

Abstract: The paper presents a real-time face recognition system leveraging OpenCV, a powerful open-source computer vision library. The system aims to identify and verify individuals in real-time, utilizing modern advancements in machine learning and image processing. The architecture of the proposed system encompasses three primary stages: face detection, feature extraction, and face recognition. In the face detection stage, we employ the Haar Cascade Classifier for its robustness and efficiency in identifying facial regions within the video frames. Once faces are detected, the system proceeds to the feature extraction phase, where facial landmarks and distinctive features are extracted using Local Binary Patterns Histograms (LBPH). This method ensures high accuracy and resilience to variations in lighting and facial expressions. The face recognition stage utilizes a trained machine learning model to compare the extracted features against a pre-established database of known faces. For this purpose, we employ OpenCV's LBPH Face Recognizer, which is known for its balance between performance and computational efficiency, making it suitable for real-time applications. Extensive testing of the system demonstrates its capability to accurately recognize faces in diverse environments, maintaining a high recognition rate while operating at real-time speeds. The system's flexibility allows for integration into various applications such as security systems, access control, and human-computer interaction. The real-time face recognition system, powered by OpenCV, offers a practical and efficient solution for facial identification tasks, demonstrating significant potential for deployment in a wide range of real-world applications.

Keywords: Face recognition, Open CV, Deep Learning.