Paper Title
Dipper Throated Optimization for Feature Selection in Facial Expression Recognition

Abstract
Facial Expression Recognition (FER) is now the most popular and well-known part of cognitive and affective computing, owing to its wide range of applications. FER has been the subject of a number of investigations, as well as a number of review publications. In this paper, we propose the application of the recently published dipper throated optimization for selecting the significant features in facial images. In addition, we proposed the application of this optimizer in optimizing the parameters of a neural network for boosting the classification accuracy. The features extracted are based on the VGG16 deep network. This deep network is pre-trained on a huge dataset and based on transfer learning, it is used to extract the main features of facial expressions. Several experiments were conducted to assess the performance of the proposed approach. These experiments include the comparison with other feature selection methods. The comparison also includes other optimization methods that can be used to optimize the parameters of the employed neural network. The recorded results confirmed that the proposed approach outperforms the other approaches based several evaluation metrics. Keywords - Dipper throated optimization, Neural network, Facial expression, Feature selection, Parameter’s optimization.