Downloads: 129 | Views: 396
Research Paper | Electronics & Communication Engineering | India | Volume 7 Issue 12, December 2018 | Popularity: 6.2 / 10
Modified PSO Based Facial Emotion Recognition System
Sarath Lal T S, Anoop K Johnson
Abstract: Facial emotion recognition (FER) is an important topic in the fields of computer vision and artificial intelligence owing to its significant academic and commercial potential. In this paper, the system employs a facial emotion recognition system which accurately identifies the nine facial emotions such as happy, sad, akward, depressed, surprise, disgust, anger, terror and courage. The facial emotion recognition system mainly consist of three steps namely, feature extraction, feature optimization and emotion recognition. In feature extraction, a horizontal and vertical neighbourhood pixel comparison LBP is used for generating the LBP operation. A micro genetic algorithm embedded with the particle swarm optimization is implemented to generate the discriminative facial contents in the feature optimization process. Sparse classifier is used for the classification of the facial expressions in the emotion recognition process, which is a highly accurate one. The modified PSO is highly efficient in generating the facial expressions as compared to the conventional PSO.
Keywords: Feature selection, Sparse classifier, Particle swarm optimization
Edition: Volume 7 Issue 12, December 2018
Pages: 1083 - 1090
Please Disable the Pop-Up Blocker of Web Browser
Verification Code will appear in 2 Seconds ... Wait