Face Recognition Using Improved Short-Long-Term Memory Neural Network by the Whale Optimization Algorithm

Recognizing the identity of people through their unique characteristics is one of the searches that has received much attention in recent years, and is used in many legal and security cases.Considering that, facial image-based identification methods have complex processes, it is necessary to provide an algorithm that is useful in facilitating the identification process with high accuracy.One of the problems of facial recognition systems is identifying face.Face recognition in different views and poses is related to the limitation of the poses used in the face recognition algorithm in the system.In this research, the identification of the identity using the face is discussed.In order to identify people, feature extraction is done using principal component analysis algorithm and 9 features with high resolution are selected and after the normalization of the selected features using the improved short-long-term memory neural network with the Whale Optimization Algorithm, identity recognition is performed based on the face and the results have been compared with the convolutional neural network and the k-nearest neighbor classifier.Improved short-long-term memory neural network with the Whale Optimization Algorithm performs face-based identity recognition with 0.25 mean square error and 99% accuracy.