Abstract

The development of an effective and efficient face recognition system has always been a challenging task for researchers. In a face recognition system, feature selection is one of the most vital processes to achieve maximum accuracy by removing irrelevant and superfluous data. Many optimization techniques, such as particle swarm optimization, genetic algorithm, ant colony optimization, etc., have been implemented in face recognition systems mainly based on two feature extraction methods: discrete cosine transform and principal component analysis. In this research, a nature-inspired well-known algorithm, namely cuckoo search, has been implemented for face recognition. Further, a hybrid method consisting of DCT and PCA is applied to extract the various features by which recognition can be made with a high rate of accuracy. To validate the proposed methodology, the results are also compared with the existing methodologies, such as PSO, differential evolution, and GA.