Face recognition has been widely studied due to its importance in smart cities applications. However, the case when both training and test images are corrupted is not well solved. To address such a problem, this paper proposes a locality constrained low rank representation and dictionary learning (LCLRRDL) algorithm for robust face recognition. In particular, we present three contributions in the proposed formulation. First, a low-rank representation is introduced to handle the possible contamination of the training as well as test data. Second, a locality constraint is incorporated to acknowledge the intrinsic manifold structure of training data. With the locality constraint term, our scheme induces similar samples to have similar representations. Third, a compact dictionary is learned to handle the problem of corrupted data. The experimental results on two public databases demonstrate the effectiveness of the proposed approach. Matlab code of our proposed LCLRRDL can be downloaded from https://github.com/yinhefeng/LCLRRDL.