WLD DESCRIPTOR WITH KNN CLASSIFIER FOR GENDER RECOGNITION FROM FACE IMAGES
DESIGN DETAILS
Gender classification is an important task which in turn can enhance the performance of a wide range of applications including identity authentication, human-computer interaction, access control, and surveillance, involving frontal facial images. A large majority of gender classification approaches are based on extracting features from face images and then giving these features to a binary classifier. The feature extraction phase has been carried out by using either (a) appearance-based methods or (b) geometric methods. In appearance-based methods, the whole image is considered rather than the local features corresponding to different parts of the face. While, in geometric based methods, the geometric features like distance between eyes, face length and width, etc., are considered. For classification purposes, mostly neural networks, nearest neighbor method, linear discreminant analysis, and other binary classifiers are used.
This Matlab design is based on gender classification rate using the textural properties of the faces. Texture descriptor WLD (Weber Local Descriptor) is used for gender recognition. The basic WLD descriptor is a histogram where differential excitation values are integrated according to their gradient orientations. The differential excitation values are concatenated irrespective of their spatial location and so WLD behaves like a holistic descriptor. By extending it to enhance its discriminatory power by embedding the local spatial information and call it Spatial WLD (SWLD) and will refer to the basic WLD as holistic WLD (HWLD). This design is mainly contributing in: (i) Exploiting WLD as local feature extractor (ii) finding the best combination of parameters and (iii) finding optimal gender recognition rate with two different distance measures: City-block distance (L1) and Chi-Square distance (CS) using Matlab 2015b simulation.
REFERENCES
Reference Paper-1: Gender Recognition from Face Images with Local WLD Descriptor
Author’s Name: Ihsan Ullah, Muhammad Hussain, Ghulam Muhammad, Hatim Aboalsamh, George Bebis and Anwar M. Mirza.
Source: IEEE
Year: 2012
Request source code for academic purpose, fill REQUEST FORM below,
http://www.verilogcourseteam.com/requ...
You may contact +91 7904568456 by WhatsApp Chat, for paid services.
We are available on Telegram and Signal.
Visit Website: http://www.verilogcourseteam.com/
Visit Our Social Media
Like our Facebook Page: / verilogcourseteam
Subscribe: / verilogcourseteamelectricalprojects
Subscribe: / verilogcourseteammatlabproject
Subscribe: / verilogcourseteam