Development of an Enhanced Human Face Recognition Model

Abstract Face recognition plays an important role in various areas of life and has attracted the interest of many scholars in the domain of computer vision, pattern recognition, and artificial intelligence. Most of the existing techniques adopted in the recognition of humans through facial images were found to be efficient but are computationally expensive due to the high dimension of facial images, which in turn increases processing speed and memory consumption. In this research, an enhanced model is proposed. Consequently, 6 facial images from 60 individuals were locally acquired using a Canon digital camera of default size 1200 × 1200. Thereafter, 240 images were used for training, while 120 images were used for testing. The enhanced model reduced the high dimension of images with improved feature extraction accuracy. Results showed significant improvement when compared to the existing ones.