This document summarizes an automated criminal identification system using face detection and recognition. The proposed system uses a Python implementation with OpenCV to detect faces in video frames from a surveillance camera. It then compares the detected faces to images stored in three databases: a citizen database, local watchlist database of known criminals, and an international watchlist database. If a match is found in the watchlists, details of the criminal and detection time are sent to police. The system uses a Haar cascade classifier for face detection and Local Binary Patterns Histograms for face recognition. It analyzes video frames in real-time to identify potential criminals and notify authorities.