DocumentCode
3545860
Title
A novel approach for panic-face extraction based on mutation
Author
Hossain, Md Aynal ; Samanta, Debabrata ; Sanyal, Goutam
Author_Institution
Dept. of Comput. Applic., Calcutta Inst. of Technol., Kolkata, India
fYear
2012
fDate
23-25 Aug. 2012
Firstpage
473
Lastpage
477
Abstract
The facial expression of mankind is one of the vital evidence of the surrounded circumstances that a person is facing or overwhelmed to be faced and Panic is a spontaneous occurrence that emerges through any person´s face depending on environmental state of affairs or circumferences. The Mutation can have an imperative role in the way of image recognition and its features extraction. It may carefully be measured in the structure of matrices so as to plot a space similar to Euclidean-space. The use of covariance features has better and faster results as compared to most of the methodology usually are in use on the way of object/image identification. In this paper we present a different technique for the recognition of facial expression of any person´s Panic moment by Mutation. We tried to consider Mutation based on span-space of covariance matrices to replicate the belongingness of errors that could occur less, probably, during the phase of repetition. The pixels span-space-points and Mutation coefficients are engaged to outward appearance of the covariance matrices. The results set of the Experiment expose the reward of this anticipated methodology.
Keywords
covariance matrices; face recognition; feature extraction; Euclidean-space; covariance features; covariance matrices; environmental state; facial expression recognition; image recognition; matrices structure; mutation coefficients; mutation-based panic-face extraction; object identification; person panic moment; pixels span-space-points; repetition phase; Face recognition; Humans; Image recognition; Reliability; Solids; Mutation; span-space-points covariance; weighted coefficients vector;
fLanguage
English
Publisher
ieee
Conference_Titel
Advanced Communication Control and Computing Technologies (ICACCCT), 2012 IEEE International Conference on
Conference_Location
Ramanathapuram
Print_ISBN
978-1-4673-2045-0
Type
conf
DOI
10.1109/ICACCCT.2012.6320825
Filename
6320825
Link To Document