This paper discusses the development of an efficient and automated system for the recognition of facial expressions, which is essentially an application augmented with many multimedia computing systems. The proposed scheme works in three stages. In the first stage, ripplet transform type II (ripplet-II) is employed to extract the features from facial images because of its efficiency in representing edges and textures. In the next stage, a principal component analysis (PCA)+linear discriminant analysis (LDA) approach is utilized to obtain a more compact and discriminative feature set. In the final stage, classification is performed using a least squares variant of support vector machine (LS-SVM) with radial basis function (RBF) kernel. The proposed system is validated on two benchmark datasets namely the Extended Cohn-Kanade (CK +) and Japanese female facial expression (JAFFE). The experimental results demonstrate that the propose system yields superior performance as compared to other state-of-the-art schemes. © 2017, Springer Science+Business Media, LLC, part of Springer Nature.