Date of Award

Spring 2003

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Electrical & Computer Engineering

Program/Concentration

Electrical Engineering

Committee Director

Vijayan K. Asari

Committee Member

Stephen A. Zahorian

Committee Member

Ravindra P. Joshi

Call Number for Print

Special Collections LD4331.E55 H37 2003

Abstract

Automatic detection of faces from video sequences is an important task in security applications. The number, location, size and orientation of human faces in a video frame are unpredictable and can vary from frame to frame. A face detection algorithm for color images in the presence of varying lighting conditions and complexity in background relying upon color and statistical analysis is presented in this thesis. The new method detects skin regions over the entire image and then classifies the skin regions as faces and non-faces. Segmentation of skin regions is performed by a novel color space merging procedure named Integrated Fault Tolerant Skin Segmentation (IFTSS) wherein, the preliminary skin regions threshold from a two dimensional histogram representation of Hue-Saturation plane and skin regions extracted from a modified chrominance map are integrated. It is observed that IFTSS is capable of extracting skin regions with varying skin colors from images captured at natural environments. The classification of the segmented regions as faces and non-faces is based on statistical measures. A set of vectors extracted by principal component analysis and another set of vectors obtained from a priori training procedures are used to calculate the statistical measures used for face classification. Tolerance to adverse lighting conditions is achieved by a robust lighting compensation scheme applied to the input image. The new face detection method is found invariant to pose, lighting and size of the face regions.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/mam4-8j36

Download

Share

COinS