Although discriminative locality alignment (DLA), which is based on the idea of part optimization and whole alignment, has better performance than classical methods in feature extraction, DLA is too overly sensitive to the values of the parameters and falls short of exploiting the full supervision information. We propose a novel supervised feature extraction method, named enhanced discriminative locality alignment (EDLA), for robust feature extraction. EDLA is not sensitive on the choice of the parameters, and both the local structure and class label information are taken into consideration in EDLA algorithm. Moreover, a kernel version of EDLA, named kernel EDLA, is developed through applying the kernel trick to EDLA to increase its performance on nonlinear feature extraction. Experiments on the face databases demonstrate the effectiveness of our methods.