Face recognition techniques attempt to identify faces according to the patterns of mouth, lip, eyes and so on. However, the effectiveness of existing approaches degrades in presence of uncontrolled conditions such as variations of background light and image sizes. To deal with this problem, we propose a novel approach based on Classified Vector Quantization (CVQ) technique. The new approach divides images into some blocks and each block is classified into several patterns. Then, the Vector Quantization (VQ) technique is applied on the vectors of each pattern. In order to evaluate our approach, we have conducted a family of experiments on some standard image databases, MIT, YALE, and AR. The results demonstrate that the new approach is steadily capable of identifying faces in different situations.