Face recognition is one of the important applications of identifying and verifying an individual based on the facial image of the person himself, since there is difficulty in detecting data when using various patterns of facial image capture, especially when there is a large overlap between the patterns, it takes time to determine person’s membership to recognize the individual, so many statistical methods have been proposed in recent years, and in this research, the Hidden Markov Model (HMM) will be used, which was one of the most popular models in statistics that are widely used in sequence prediction and natural language processing tasks, and particularly useful in face recognition, so this model will be compared with the Support Vector Machine (SVM), which is a statistical model for face recognition that has been used for classification and regression tasks, due to its wonderful advantages and the ability to quickly solve. When dealing with high-dimensional data, the problem of face recognition becomes more complex, so we proposed in our research the Multi-Dimensional Scaling method with the Support Vector Machine (MDS-SVM) to manage it in two steps: first, reduce the dimensions using the MDS method, and then use SVM as a classifier for the recognition task. Different methods will be compared using two different sets of data and different samples (number of images), the first group represented the ORL database, which was made up of 40 individuals, while the second group represented the real data, which was made up of 100 individuals. The results showed that the MDS-SVM achieved the lowest mean square error (MSE) for both sets of data, followed by SVM and HMM and all three methods reached the highest accuracy rate of 100% for facial image recognition on real data.