The worry of the security matter in the world is increased; in general this lead to interest of the biometrics. Face recognition technique is a type of the biometrics has attracted importance in several security applications. In this thesis, a security of computers network system depends upon online face recognition technique is considered. The proposed work is consisting of two sections. The first section represents the face recognition technique and the second section represents the computers network system. Face recognition technique consists of four main phases which are image acquisition phase, preprocessing phase, features extraction phase, and classification phase. Acquisition means capture the face images of the users via webcam and preprocessing is used to enhance the images. Features extraction is used to get features from face images, Principle component analysis is used to extract the best features that represent the users face image. Artificial feed forward neural network with back propagation algorithm is used in training and testing of the users face images in recognition stage. Computers network system is consisting of a number of clients connected to web server on internet. In addition of computers network system being secured by face recognition technique, the connection between the clients and web server is secured via Hypertext Transfer Protocol Secure (HTTPS) to secure exchange of data. Face recognition is examined for three different cases. In the first case, six persons are connected to the server and face recognition rate of 97.222 % is achieved. In the second case twelve persons connected to the server and face recognition rate of 94.444% is achieved. While in the third case eighteen persons are connected to the server and face recognition rate of 91.667 % is achieved.