
Smart CCTV Security System Using Facial Recognition Through Deep Learning
Software as a Service
NoCategory
AI & Machine LearningClients
Chung Yuan Christian University
Techstack
PythonOpenCVdlibRaspberry PiPyQt4
Purpose
Provide entrance surveillance that does more than record, detecting faces in real time, recognizing specific persons of interest, and logging exactly who was seen, where, and when.
Description
A 720p camera on a Raspberry Pi 3B watches an entrance while a PyQt GUI shows the live feed. Haar cascade detection finds faces, and a Histogram of Oriented Gradients plus linear SVM pipeline built on dlib and OpenCV recognizes them. When a watched person appears, the system captures the face, records a 9-10 second clip, and appends a log entry with the person's identity, timestamp, and location.




