Real Time Personal Data Protection Using Image Processing

The use of personal mobile devices is increasing at a rapid rate due to digitalization of the world. These devices are portable and can be used at various locations governing access to these devices is a challenging task. To solve this problem, an application has been developed that helps in implementing the authorization mechanism on a device using image processing and machine learning techniques. Face recognition is one of the best methods to govern access on a device. Authorization mechanism has implemented using Local Binary Pattern Histograms (LBPH) along with Haar-cascade for face detection. The algorithm has trained with the users' database that has face data of all authorized users. The algorithm generates a configuration model as output which is used to recognize the current user of a device. When unauthorized access or spoof attack is detected the application will automatically implement the alerting mechanism.