Security software using neural networks

This paper describes the structure and the theoretical principles of the security software that we developed. This sofware is an industrial application based on the neural networks theory that we describe hereafter. Its aim is to recognize somebody's face and thus to add one more protecting level to the Windows NT and Windows 95 security access. We implemented the face learning phase by using the projection onto eigenvectors matrix and the backpropagation algorithm. We stored in a database, the identification of the faces which have been learned and we added a security protection when opening the personal session of the operating system Windows NT and we created a new level of protection for Windows 95. We tested our algorithms on images of other types than faces and the results allow to use the software in industrial control.