GENDER CLASSIFICATION FROM FACIAL IMAGES USING LBP

Adithya Krishna Kumar; Abhishek Malik; Akshay Kumar Bhardwaj; Sachin Wakurdekar

GENDER CLASSIFICATION FROM FACIAL IMAGES USING LBP

Authors: Adithya Krishna Kumar, Abhishek Malik, Akshay Kumar Bhardwaj , Sachin Wakurdekar

Unique Paper ID: 148060
Volume: 5
Issue: 12
PageNo: 206-209

Keywords: Accuracy Gender Classification LBP

Abstract:
Gender is an important demographic attribute of people. This project provides an approach to human gender recognition in computer vision using a facial recognition technique called Local Binary Patterns Histogram (LBPH). In this technique, the face area is divided into small regions from which local binary pattern (LBP) histograms are extracted and concatenated into a single vector efficiently representing a facial image. A working model has been created using a combination of Python and OpenCV, along with a multistage Haar Cascade Classifier in order to identify the major datapoints needed. The model, as well as a review of contemporary approaches exploiting information from facial images is presented. We highlight the advantages of using LBPH over Eigenfaces an Fisher Faces, besides the various challenges faced and survey the representative methods of these approaches. Based on the results, good performance has been achieved for datasets captured under controlled environments, but there is still much work that can be done to improve the robustness of gender recognition under real-life environments where various environmental conditions such as lighting, distance to subject, background etc. cannot be controlled.

Download article

email to a friend

Cite This Article

ISSN: 2349-6002
Volume: 5
Issue: 12
PageNo: 206-209

GENDER CLASSIFICATION FROM FACIAL IMAGES USING LBP

Available:https://ijirt.org/Article?manuscript=148060

Impact Factor
8.01 (Year 2024)

UGC Approved
Journal no 47859

Join Our IPN

IJIRT Partner Network

Submit your research paper and those of your network (friends, colleagues, or peers) through your IPN account, and receive 800 INR for each paper that gets published.

Join Now