Face Recognition using Smaller Database Machine Learning Approach

Face detection, registration, and recognition have become a fascinating field for researchers. The motivation behind the enormous interest in the topic is the need to improve the accuracy of many real-time applications. In this research, we presented an enhanced approach to improve human face recognition using a deep learning and features extraction based on the correlation between the training images. A key contribution of this work is the generation of a new set called the “datastorage” from the original training data set, which is used to train the system. We generated the “datastorage” using the correlation between the training images without using a common technique of image density. The correlated “datastorage” provides a high distinction layer between the training images, which helps the system to converge faster and achieve better accuracy. Data and features reduction is essential in the face recognition process, and researchers have recently focused on the modern neural network. We applied 4 features and then combined them to obtain the “datastorage”, which we fed into the system. We achieved higher face recognition accuracy with less computational time compared with the current approach by using reduced image features. We test the proposed framework on a dataset with 140 to 150 screenshots of the object personnel. We achieved tremendous accuracy. In addition, we presented an enhanced framework to improve the face registration (collection) using deep learning model. We used deep architectures to train our method.