Among the many issues facing the world, the issue of traffic accidents, many security facilities have developed research on fingerprints, palmistry biometrics and other biometrics. In modern technologies, facial and eye detection algorithms have been developed and identified. This paper includes comparisons of face and eye tracking methods, which are extremely useful for identifying individuals in a variety of situations. Facial biometrics are more stable due to their distinct features, which gives them an advantage over other biometrics such as palmistry and fingerprint. This work dealt with 74 studies of face and eye detection and a comparison of algorithms that were used in previous researches, where the Viola-jones algorithm was identified as the best facial algorithm in terms of accuracy up to 98% for different types of face databases because this algorithm depends on geometry of face and also able to detect all faces in the same image, Furthermore, it has been determined that SVM is the best eye detection algorithm due to the accuracy of its characteristics and the speed of execution in real time so, the highest accuracy achieved was 98% accurate .

We study the supervised learning of a two-dimensional patches detector. We compute the generalization error of the network with the annealed approximation; a first-order phase transition appears at all temperatures on the generalization error curve as a function of the size of the training set. We show that the system is subject to dynamical effects: the earlier story of presentation of the patterns to the system can be relevant, and a hysteresis appears on the generalization error curve. We show Monte Carlo simulations which agree with the theoretical predictions.