Mohammad Elahi - Department of Computer Science, Islamic Azad University, Science and Research branch, Tehran, Iran.
Ehsan Fotoohabadi - Department of Computer Science, Islamic Azad University, Sepidan, Iran.

Face recognition is a common problem in artificial intelligence. Several smartphones are unlocking phones with face recognition capabilities to protect private details and use Facebook to instantly recognize when Facebook users appear in images. So far, several methods have been proposed for face recognition, but it is not easy in real world conditions. A basic technique for distinguishing individuals depends on various conditions, including illumination and variation in body posture. This paper presents a method based on recurrent neural network (RNN) algorithms for face recognition. The genetic algorithm is used to improve the neural network weights' accuracy. The proposed algorithm is then compared with other known face recognition algorithms, including Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). The implementation results of the proposed method on the Face Detection Data Set and Benchmark (FDDB) dataset indicate that the proposed method succeeded in improving the accuracy of detecting and recognizing faces in images by ۱.۴% compared to the convolutional neural network.

Genetic algorithm, Feature extraction, Face detection, Face recognition, Machine learning, Deep learning