Implementation of surveillance system through face recognition using HOG algorithm

Abstract

Recent years have seen a sharp rise in crime, raising concerns about neighborhood safety among the general public. A number of security technologies, such as digital door locks, closed-circuit televisions (cctvs), and alarm systems, have been developed and implemented. Even though most buildings have cctv installed, manually monitoring the footage takes a lot of manpower. The hog (histogram of oriented gradients) algorithm is widely used for object detection in computer vision. In particular, it shows good performance in face recognition and motion detection. In addition to identifying the user, the proposed system allows for mobile engagement, allowing the user to see the video, set off the alarm, and even receive notifications when an unknown person approaches the residence. The system proposed in the paper seeks to improve the performance of the current surveillance system by applying additional functions. The system was implemented to recognize people through real-time monitoring and send warnings if they are not registered users. Additionally, the system can be used to quickly report to the police.