Facial Expression Recognition using Fusion of LBP and HoG Features
Mamta Santosh1, Avinash Sharma2
1Mamta Santosh, Computer Science &Engg, MMDU/MMCE, Ambala, India.
2Dr. Avinash Sharma, Computer Science &Engg, MMDU/MMCE, Ambala, India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 296-300 | Volume-8 Issue-8, June 2019 | Retrieval Number: H6336068819/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Human Machine Interaction has gained significant attention in the previous years. The communication through facial expression plays a vital role in the interaction since the expressions convey more information than spoken words and it allows to recognize the emotions without wearing sensors and monitoring devices. So it is important to classify the emotions correctly. In this paper, we have proposed an effective framework for Emotion Recognition using Facial Expressions which recognizes 96.2% of emotions accurately. Voila Jones Algorithm is used for face detection followed by splitting the face in parts which are dominant for expressions and applying Gauss Filtering to these parts. Combination of HOG and LBP features is used to extract the features. PCA is used for dimensionality reduction and these features are classified using Canberra Distance Classifier. The framework performs well on JAFFE Dataset and gives promising results.
Keyword: Emotion Recognition, Facial Expression Recognition, LBP, HOG, Canberra Distance Classifier
Scope of the Article: Image Processing and Pattern Recognition