In recent years, an explosion in research on pattern recognition systems using neuralrnnetwork methods has been observed. Face Recognition (FR) is a specialized patternrnrecognition task for several applications such as security: access to restricted areas,rnbanking: identity verification and recognition of wanted people at airports.rnThis thesis will explain what is involved in FR task and outline a complete FacernRecognition System (FRS) based on Artificial Neural Network (ANN). In this work, twornFRS are developed. The first model uses Principal Component Analysis (PCA) forrnfeature extraction from the face images and ANN for the classification purpose. In thernsecond model, combination of Gabor Filter (GF) and PCA are used for feature extractionrnand ANN for the classification.rnIn the first approach, the face images are projected into subspace called eigenspace,rnconsisting of the eigenvectors from the covariance matrix of the face images. Thernprojection of an image into eigenspace will transform the image into a representation of arnlower dimension which aims to hold the most important features of the face. Thesernfeature vectors are classified into training, validation and testing set. The training andrnvalidation set are used during the training of ANN. The testing set is used to evaluate thernrecognition performance of the model.rnIn the second approach, Gabor feature vectors are derived from a set of downsampledrnGabor wavelet representations of face images, then the dimensionality of the vectors isrnreduced by means of Principal Component Analysis (PCA), and finally ANN is used forrnclassification. The Gabor filtered face images exhibit strong characteristics of spatialrnlocality, scale, and orientation selectivity. These images can, thus, produce salient localrnfeatures that are most suitable for FR.rnExperimentation is carried out on FRS by using Olivetti Research Laboratory (ORL)rndatasets, the images of which vary in illumination, expression, pose and scale. The resultrnshows the feasibility of the methodology followed in this thesis work. Model 1 achieves arnrecognition rate of 76.6% whereas model 2 achieves 88.3% of correct classification andrnperformed very efficiently when subjected to new unseen images with a false rejectionrnrate of 0% during testing. The high recognition rate of model 2 shows the efficiency ofrnGF in feature extraction.rnKey wordsâ€”Face recognition, biometrics, artificial neural network, Gabor filter andrnprincipal component analysis.