T Hanumantha Reddy

Work place: Dept. Computer Science and Engineering, RYMEC College of Engineering and Technology, Bellary, India.

E-mail: thrbly@gmail.com


Research Interests: Data Structures and Algorithms, Image and Sound Processing, Computer Networks, Computer Architecture and Organization, Computer systems and computational processes


Dr. T. Hanumantha Reddy, male, is received B.E. (Electronics and Communication) from Gulbarga University, Gulbarga and Master of Engineering in Computer Engineering from Mysore University, Mysore. He has completed his Ph.D. from J. N. T. U., Hyderabad in the area of Image Processing. Currently he is working as Professor and Head in the Department of Computer science and Engineering, R. Y. M. E. C., Bellary. His subjects of interest are Image and Video Processing, Data Communications and Computer Networks, Computer Architectures and Multimedia computation and Communication. He has total of 26 publications in the area of Image Processing.

Author Articles
Hybrid Approach for Facial Expression Recognition using HJDLBP and LBP Histogram in Video Sequences

By Mahesh U Nagaral T Hanumantha Reddy

DOI: https://doi.org/10.5815/ijigsp.2018.02.01, Pub. Date: 8 Feb. 2018

Any kind of compassionate thoughts can't be expressed through words, but it appears on their facial expression. So, the facial expression reveals the emotions of individuals. The recognition of such emotions can be understood correctly or sometimes ambiguously from the opponent. Hence, there is a scope for automatic facial expression recognition (FER) in the context of image processing. The FER system has three different phases: face detection, feature extraction and expression classifi-cation. In face detection phase, Viola Jones face detector is used to crop the original image such that only the face region is retained by removing the unwanted region. In feature extraction stage, High-order Joint Derivative Lo-cal Binary Pattern (HJDLBP) and Local Binary Pattern (LBP) histogram algorithms are used for extracting fea-tures from the cropped image. In last stage, Support Vec-tor machine (SVM) classifier is used in finding the precise facial expression.CK+ dataset has been used for training and testing, which consist of 442 image samples. We have considered six different universal possible ex-pressions such as, happy, anger, disgust, fear, surprised, and sad for identification. The experimental results indi-cate that the overall accuracy of the proposed system was 74.8%, which is high compare to the results available in literature.

[...] Read more.
Other Articles