.png){kind=icon}
.png){kind=icon}
Return
An efficient facial emotion recognition using convolutional neural network with local sorting binary pattern and whale optimization algorithm
.png){kind=icon}
PRE
AI
DOI：10.1007/s41060-024-00601-1.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Facial emotion recognition is one of the fields of machine learning and pattern recognition. Facial expression recognition is used in a variety of applications. For robust automatic facial emotion recognition, feature extraction from input image data is challenging. To address this issue, we propose an emotion recognition system based on a new feature extraction method, whale optimization algorithm, and convolutional neural network. In the feature extraction phase, a new efficient human face descriptor is expressed using a local sorting binary pattern and a convolutional neural network. Also, the hyperparameters of the convolutional neural network are optimized using the whale optimization algorithm. Then, the convolutional neural network is applied for classification. The performance of the proposed method is evaluated using three well-known face databases CK+ (extended Cohn-Kanade) with facial expressions (happiness, sadness, fear, anger, disgust, contempt, and surprise), JAFFE (Japanese female facial expression) with (happiness, sadness, anger, fear, neutral, disgust, and surprise), and MMI (MMI facial expression database) with facial emotions (happiness, sadness, anger, fear, disgust, and surprise). The accuracies with CK+, JAFFE, and MMI are 100%, 99.93%, and 99.83%, respectively. Experimental results demonstrate that the proposed model can provide better performance compared to alternative methods.
Keywords:
Feature extraction
Convolutional neural network
Local sorting binary pattern
Facial emotion recognition
Deep learning
Whale optimization algorithm
Journal
I
IF:
2.8
Papers:
1.0K
Citations:
1.3K