期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2021
卷号：99
期号：4
语种：English
出版社：Journal of Theoretical and Applied
摘要：We have created a Deep Learning network to identify the emotion of a person. It is based on seven facial expressions. (angry, disgust, sadness, happy, nature, fear, and surprise). We used extended Cohn�Kanade (CK+) database basis of (10-fold cross-validation) to identify 6 facial expressions. The Deep Learning Network scored a recognition average of 88.9%. As you can see in the confusion matrix, the expressions happy and surprised achieved the best recognition rates 98.92 and 97.23 successively. We also used, in another experiment, the JAFFE database basis of (LOOCV) and it scored a recognition average of 88%. As you can see in the confusion matrix that the expressions of fear and surprise achieved the best recognition rates 93.33 and 93.33, respectively. We compared the performance of the proposed system to similar studies that followed the same databases with the same sample and the same style. The system we used outscored other systems in the other studies. We also compared in detail the percentage of identification performance for each expression in isolation using the extended Cohn�Kanade (CK+) database. We compared our study to other studies and we found that our system did better.
