This paper identifies a feature space to address the problem of human face recognition from the database images. The face recognition system is based on Principal Component Analysis. By using PCA the features can be extracted. The multi resolution curvelet transform can be used for the efficient face image retrieval. When compared to wavelet transform the curve let transform has better directional and edge representation. The face images can be decomposed when applying the curvelet transform and the curvelet sub bands can be form. In addition the PCA can be used for dimensionality reduction. Then the PCA can be applying for each curvelet sub bands and create feature set. The mahalanobis distance measure can be used to measure the distance between the query and the database images. The well-known face database indicates the potential of this curvelet based feature extraction and gives good retrieval result. The experimental results show that our approach is significantly better than the conventional methods. © 2006-2015 Asian Research Publishing Network (ARPN).