Delamination is a principal mode of failure in laminated composites that adversely affect structural integrity and may lead catastrophic failure of the structures. In this work, structural vibration response is processed through wavelet packet transform and linear discriminant analysis to identify appropriate discriminative features for the presence of delamination in the structure. A supervised machine learning technique is trained and tested with the discriminative features. Numerical results showed that the proposed approach could successfully identify the presence of delamination from structural vibration response of smart composite laminates.