Robust facial expression recognition using improved sparse classifier

Recently, sparse classifier (SC) has become a promising classification technique and is increasingly attracting attention in signal processing, computer vision and pattern recognition. In this paper, a new classification algorithm based on a weighted sparse representation model, called improved sparse classifier, is proposed for robust facial expression recognition. The effectiveness and robustness of the proposed method is investigated on clean and occluded facial expression images. The performance of the proposed method on robust facial expression recognition is compared with SC, the nearest neighbour (NN), linear support vector machines (SVM) and the nearest subspace (NS). Experimental results on two benchmarking facial expression databases, i.e., the JAFFE database and the Cohn-Kanade database, demonstrate that the proposed method obtains promising performance and a strong robustness to corruption and occlusion on robust facial expression recognition tasks.