|
|
|
(3.128.78.30)
|
|
Users online: 12316
|
|
|
|
|
|
|
Ijournet
|
|
|
|
|
|
|
Dual Local Binary Pattern based Gender Detection from Facial Image Chitra*, Dr. Balakrishnan** *Assistant Professor, SASTRA University, India **Director, Indra Ganesan College of Engineering, India Online published on 9 November, 2016. Abstract Gender recognition using facial images plays the extreme part in biometric technology. Gender identification from detected face features, helps in knowing the person's behaviour and possible actions being produced. In addition, different facial expressional classes give a scope for identifying the appearance and age estimation. Local binary pattern improves the images with neutral expressions which are used for recognition and provide better performance. But face based matcher is complex in producing better performance for inconsistent codes and improve robustness while analyzing the face images. In this study, a novel method called Dual Local Binary Pattern Classifier (DLBPC) from facial image is presented. The DLBPC method uses Spectral Color Clustered Face Detection, Dual Local Binary Pattern Classifier and Wrinkle Textured Seed Point Age Estimation with preliminary input face image. The Spectral Color Clustered Face Detection is used to extract facial portion from human face. Next, the Dual Local Binary Pattern is applied that uses both the sign and weight to ascertain the classification. Finally, the Wrinkle Textured Seed Point performs age estimation for the input image. The performance of the DLBPC has been investigated through extensive experiments. Comparisons with some of the existing method have shown a marked reduction in gender detection time used per image to produce results more efficiently and without loss of accuracy for gender classification. Top Keywords Gender recognition, Dual Local Binary Pattern Classifier, Spectral Color, Wrinkle Textured Seed Point. Top | |
|
|
|
|
║ Site map
║
Privacy Policy ║ Copyright ║ Terms & Conditions ║
|
|
|
749,877,757 visitor(s) since 30th May, 2005.
|
|
All rights reserved. Site designed and maintained by DIVA ENTERPRISES PVT. LTD..
|
|
Note: Please use Internet Explorer (6.0 or above). Some functionalities may not work in other browsers.
|