Deep Convolutional - Optimized Kernel Extreme Learning Machine Based Classifier for Face Recognition

Face recognition task is an active area of research in recent years in the field of computer vision and biometric. Feature extraction and classification are the most significant steps for accurate face recognition system. Conventionally, eigenface approach or frequency domain features were used for feature extraction, but they are not invariant to outdoor conditions like, lighting, pose, expression and occlusion. In the present work, multiple convolutional and pooling layers of Deep Learning Networks (DLN) will extract efficiently the high level features of the face database. These features are given to the Kernel Extreme Learning Machine (KELM) classifier whose parameters are optimized using Particle Swarm Optimization (PSO). The proposed Deep Convolutional-Optimized Kernel Extreme Learning Machine (DC-OKELM) algorithm leads to better performance results and fast learning speed compared to classification using deep neural networks. The performance of DC-OKELM is evaluated on four standards face database such as AT&T, CMU-PIE, Yale Faces and UMIST. Experimental results are compared with other state of the art classifiers in terms of error rate and network training time which shows the effectiveness of the proposed DC-OKELM classifier. © 2020 Elsevier Ltd