This paper proposes a sparsity enforcing based face detection with three kinds of classifiers. The feature extraction is done on the basis of rectangular feature from the integral image of the face. Sparsity enforcing method is applied for the dimension reduction on rectangular features from the generated Integral image. The performance of the classifier such as Support Vector Machines, Bayesian and AdaBoost are analyzed with the proper testing procedure. The analysis is also made, how the classifier performs according to the training sample size. Experimental results on three widely used face databases are presented to demonstrate the efficiency of the proposed approach.