CONICET | Buscador de Institutos y Recursos Humanos

In recent years Bovine Identification has become a topic of intense worldwide interest in the aftermath of terrorist incidents, outbreaks of bovine spongiform encephalitis (BSE) and, more recently, reports of E. coli contamination in beef. In fact, food could be intentionally contaminated as a terrorist act. The most efficient and effective way of countering all emergencies including food terrorism, is through sensible precautions coupled with strong surveillance. Bovine Identification is a means to implement the mentioned surveillance, and is based on different methods. One of this methods is Biometric Iris Recognition, which is based on the iris? random patterns and other particular attributes that have been shown capable of generating highly unique identification codes. It is essential for any biometric recognition method to identify the regions of interest (ROIs) that are going to be analyzed. In this case, it is necessary to identify the iris and pupil borders. For this purpose, this paper presents the curvelet transform as an alternative to existing segmentation methods. Curvelet transform is an extension of wavelet transform which aims to deal with interesting phenomena occurring along curved edges in 2D images. It is a high-dimensional generalization of the wavelet transform designed to represent images at different scales and different orientations (angles). It is viewed as a multi-scale pyramid with frame elements by location, scale, and orientation parameters with needle-shaped elements at fine scales. Curvelets have time-frequency localization properties of wavelets but also shows a very high degree of directionality and anisotropy, and its singularities can be well approximated with very few coefficients. In particular, the discrete curvelet transform is very efficient in representing curve-like edges, as the edges in the bovine eyes. Experimental results support the use of the curvelet transform for bovine eye segmentation methods.