The safety helmet detection for ATM's surveillance system via the modified Hough transform

The automatic teller machine (ATM) plays an important role in the modern economical activity. It provides a fast and convenient way to process economical transactions between banks and their customers. Unfortunately, it also provides a convenient way for criminals to get illegal money. For the safety reason, each ATM is with the surveillance system to record customer's face information. However, when criminals use the ATM to withdraw illegal money, they usually hide their faces with something (e.g. safety helmets) to avoid that the surveillance system records their face information. That will make the surveillance system decrease their efficiency. We propose a circle/circular arc detection method based upon the modified Hough transform, and apply it to the detection of safety helmet for the surveillance system of the ATM. Since the safety helmet location will be in the set of the obtained possible circles/circular arcs (if any exists). We use geometric features to verify if any safety helmet exists in the set. The proposed method can be applied to the surveillance systems of ATMs and banks, and it can provide the early warning to save-guards when any "customer" tries to avoid his/her face information from surveillance, such as withdrawing money with the safety helmet. That will make the surveillance system more useful. A real ATM image is used to see the performance of proposed method.