Supporting Issues for Expression Recognition and Classifications

  G.P.Hegde,  M. Seetha,  Nagaratna Hegde

Higher energy subspace creation is a challenging task due to corruption of intermediate pixels. This paper switches on to illustrate the basic issues required for efficient recognition of expressions during the construction of linear and non-linear subspace models. Recognition accuracy of facial expression reclined on region of detected area of face. Feature extraction methods expose required features for formation of subspace model. Among multi spaces of high dimensional data one space may have accurate information without any redundant data and it would generate minimum time classification of expression data. Some of the bench mark data bases have been utilized for testing of various subspace methods such as JAFFE, FD and YALE. The performance issues of subspace methods exhibits good results compare to state of art methods.

Keywords-

Unique Identification Number - IJEDR1704123
Page Number(s) - 747-751
Pubished in - Volume 5 | Issue 4 | November 2017
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  G.P.Hegde,  M. Seetha,  Nagaratna Hegde,   "Supporting Issues for Expression Recognition and Classifications", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.5, Issue 4, pp.747-751, November 2017, Available at :http://www.ijedr.org/papers/IJEDR1704123.pdf