VIDEO SURVEILLANCE SYSTEM FOR REAL TIME PHYSICAL ATTACK DETECTION ON ATMs in Python

By Pallak Singhal

DEEP LEARNING BASED VIDEO SURVEILLANCE SYSTEM FOR REAL TIME PHYSICAL ATTACK DETECTION ON ATMs

India’s banking sector is growing at a fast pace. India has become one of the most preferred banking destinations in the world. In order to gain further access the main idea of this project is to minimize the occurrence of physical attacks and increase the security at minimum cost. To implement this system, we have developed a deep learning model and trained this model on various video streams of ATMs rooms that involve all possible activities in ATM rooms. At first level, the main objective of this idea is to detect activities that occur inside the ATM rooms only. Later the approach can be extended for other activities or surveillance of other locations. Hence, it aims to solve problems, specifically the ATM rooms and ATM crimes.