Performance Improving of ANN with Preprocessing Stage in Human Face Expression Recognition System
Sumalakshmi C. H.1, P. Vasuki2

1Sumalakshmi C. H.*, Research Scholar, Sethu Institute of Technology, Kariapatti, Tamilnadu, India.
2Dr. P. Vasuki, Professor, Department of ECE, Sethu Institute of Technology, Kariapatti, Tamilnadu, India.
Manuscript received on January 17, 2020. | Revised Manuscript received on January 29I, 2020. | Manuscript published on February 10, 2020. | PP: 3298-3302 | Volume-9 Issue-4, February 2020. | Retrieval Number: D1592029420/2020©BEIESP | DOI: 10.35940/ijitee.D1592.029420
Open Access | Ethics and Policies | Cite | Mendeley
© 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: In many face recognition systems, the important part is face detection. The task of detecting face is complex due to its variability present across human faces including color, pose, expression, position, and orientation. So, by using various modeling techniques it is convenient to recognize various facial expressions. The system proposed consists of three phases, the facial expression database, pre-processing and classification. To simulate and assess recognition efficiency based on different variables (network composition, learning patterns and pre-processing), we present both the Japanese Female Facial Expression Database (JAFFE) and the Extended Cohn-Kanade Dataset (CK+). Comparative approaches of data preprocessing include face detection, translation, normalization of global contrast and histogram equalization. Significant results were obtained with 85.52 percent accuracy particularly in comparison with some other pre-processing phases and raw data in single pre-processing phases. The result indicates the ANN classifier representation produces a satisfactory result which reaches more accuracy. 
Keywords:  Artificial Neural Network, facial Expression, Database, face Detection, Emotion
Scope of the Article: Pattern Recognition