Human Face Detection In Color Images

Document Type

Article

Publication Date

1-1-2004

Abstract

In this paper we have used a simple and efficient color-based approach to segment human skin pixels from background, using a 2D histogram-based approach as a preprocess stage for human face detection. For skin segmentation, a total of 446,007 skin samples from the training set is manually cropped from the RGB color images, to calculate three lookup tables based on the relationship between each pair of the triple components (R, G, B). Derivation of skin classifier rules from the lookup tables are based on how often each attribute value (interval) occurs, and their associated certainty values. For face detection, we assume the face-appearance as blob-like, and that the face has an approximately elliptical shape. Accordingly, an ellipse-fitting algorithm is appropriate, which is based on statistical moments, and those blobs that have an elliptical shape are retained as face candidates.

Keywords

Skin segmentation, histogram-based approach, lookup table, skin classifier, ellipse fitting, face detection, feature-based approach.

Divisions

ai

Publication Title

Advances in Complex Systems

Volume

7

Issue

3

This document is currently not available here.

Share

COinS