Abstract

FLORES MENDEZ, Alejandro  and  MENDEZ CUANALO, Ana Antígona. Stable Ear Border Recognition in 2D Images. Comp. y Sist. [online]. 2009, vol.13, n.2, pp.195-208. ISSN 2007-9737.

This paper contains a novel method for the stable border recognition of the main features of the ear for RGB images. For the recognition, we first do the segmentation of the image in order to eliminate the hair and any feature other than skin through the data contained in the HSV transformation. The RGB image is converted to a grayscale one, and from it the information of the pixels not related to skin is discarded. Then a series of techniques, mainly morphological filters, are used to enhance the border detection associated to the main features of the ear. The proposed system achieved good performance even in the presence of noise, different light conditions, skin colors, etc. This proposal has been thought as a first step toward human identification and recognition through ears borders.

Keywords : image processing; border recognition; ear.