This paper proposed a system for the recognition of the facial expression by using cross-correlation of optical flow and mathematical models from the facial points. That defined these facial points of interest in the first frame of an input face sequence image, which utilized manually marker. The facial points were automatically tracked by using a cross-correlation based on optical flow, and extracted feature vectors. The mathematical model extracted features from feature vectors. We used an ELMAN neural network for classifying expressions. The performances of the proposed facial expressions recognition were computed using Cohn-Kanade facial expressions database. The proposed approach achieved a high recognition rate.