A novel SRC based method for face recognition with low quality images

Sparse representation-based classification (SRC) shows a good performance for face recognition in recent years, but SRC can not be suitable for low quality data with disguise or noise, which are often presented in the practical applications. To solve the problem, in this paper, we propose a novel SRC based method for face recognition with low quality images named sparse low-rank component based representation (SLCR). In SLCR, we utilize the low-rank component from training dataset to construct dictionary. The dictionary composed of low-rank component and non-low-rank component is able to describe the face feature better, especially for low quality training samples. Our recognition rule is based on the minimum class-wise reconstruction residual which leads to a substantial improvement on the proposed SLCR's performance. Extensive experiments on benchmark face databases demonstrate that the proposed method consistently outperforms the other sparse representation based approaches for disguised and corrupted face recognition.