Recognizing human emotions from facial images with the help of Alexnet neural network and genetic algorithm and decision tree classification

Facial emotion recognition is a technology that analyzes facial expressions from still images and videos to reveal information about a person’s emotional state. The feeling is one of the essential characteristics of every person. These characteristics will cause different conditions in the body and mind to occur in this structure. In our nature, the main challenge in facial emotion recognition is to achieve fast and reliable performance. Consequently, in this structure, using an optimization algorithm to select the most important features will speed up and reduce the cost of training. For this purpose, we use the genetic algorithm, which still shows promising results in many fields, to obtain the more essential features extracted from the neural network. We use AlexNet to recognize human emotions better. The obtained accuracy results show that our proposed method has shown a significant improvement in facial emotion recognition compared to the two compared methods. In the proposed method, we reached 91.61% accuracy, which compared to the two compared methods, improved the diagnosis by 20%, which shows the superiority of the proposed method compared to the reviewed articles.