Face detection using fuzzy granulation and classifier fusion in color images

Abstract EN

Face Detection is the process of determining the face location, size and number.

It can be considered as a classification problem in the sense that a given image region can be classified as face or non-face classes.

In this paper, we propose a method based on skin color segmentation and classification with Fuzzy Information Granulation (FIG) for robust and fast face detection in color images.

The proposed FIG-classifier constructs fuzzy granules based on pixels of image train data and classifies image regions using these fuzzy granules.

We have used a classifier fusion method to select the best classifier.

For each sub-window on the train data, the FIG classifiers are generated and the classifier which has the highest detection rate is selected as final classifier.

Face detection task is performed based on classification of normalized skin color segments using the final classifier.

Experimental results show effectiveness of the proposed method in comparison with the previous methods.