This paper aims to develop a platform that allows face features to be extracted faster using multiple algorithms for looking up people in a large database. We will be presenting an enhanced technique for human face recognition where we will be using an image-based approach (process of using two-dimensional images to create three-dimensional models) towards artificial intelligence by extracting features from face images by using Principle Component Analysis, Local Directional Pattern and SVM Machine Learning. Up until now, studies focusing on face recognition rely on the fusion of PCA (Principle Component Analysis) and LBP (Local Binary Pattern) for feature extraction, PCA and LBP were used for global feature extraction of the whole image and the features of the mouth area separately. Results show that this method was susceptible to random noise and resulted in a performance rate of 89.64% [1]. Also, recent studies have shown the fusion of PCA (Principle Component Analysis) and LDP (Local Directional Pattern) for feature extraction [2]. First, PCA is adopted to extract global features of facial images, then LDP operator is used to extract local texture features of eyes and mouth area and these areas are calculated by comparing the relative edge response value of a pixel in different directions. This fusion resulted in a performance rate of 91.61%. The results of PCA and LDP method show that it is more effective than adopting the fusion of PCA and LBP. It's more robust to noise and improves the rate of facial recognition. However, both methods still suffer from changes in illumination, pose changes, random noise, and aging. In this paper, we propose using a set of trained images to make the facial recognition process faster and provide more accurate results.