Intruder Detection System in an A.T.M.

Telecommunications Engineering

In Real World Applications, tracking Targets in Low Resolution Video is a difficult task, in light of the fact that there is loss of discriminative Detail in the Visual Appearance of Moving Object. The current strategies are generally founded on the Enhancement of LR (Low Resolution) Video by Super Resolution Techniques. Be that as it may, these strategies require High Computational expense. This cost further if we are managing Events Detection. The Project presents an Algorithm which Detects Unusual Events without such kind of Conversion. Conventional Low Resolution Cameras are commonly utilized in ATMs because of their Low Cost. Proposed Algorithm uses Rolling Average Background Subtraction Technique to distinguish Foreground Object from Dynamic Background in a Scene. It further uses Viola-Jones Algorithm, to identify faces. The Proposed Algorithm can perceive the event of Unusual Events, for example, Overcrowding or a fight in the Low Resolution recordings. It is quick enough since it forms Low Resolution Frames and could be useful in Surveillance System for upgrading the Security of ATMs where Conventional Camera of Low Resolution is still used.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.