Compressive Sensing Based Expression Recognition
Suparna Parua Biswas, Jaya Sil, Santi P. Maity · 2019 Second International Conference on Advanced Computational and Communication Paradigms (ICACCP) · 2019
In this paper a novel expression recognition method is presented by integrating curvelet transform, compressive sensing and Principal Component Analysis. Here curvelet transform has been used to represent the face images with prominant edges, curvatures, boundaries and to offer sparse representation to apply compressive measurements on detailed subband. The compressive sensing offers the compressive measurements of face images leading to a significant reduction in feature space. To extract the feature vector, Principal Component Analysis is then applied on the reconstructed detailed subband. Experimental results shows improved performance applied on Cohn-Kanade database compare to other methods. The proposed approach provides maximum of 99.73% recognition rate on Cohn Kanade database.