Animal Face Detection Technique Using DCT

Abstract

The problem of facilitating machine to learn & detect patterns of animals and categorizing them in the rapidly expanding image databases has become increasingly challenging for image retrieval systems. The problem to be solved is detection of animal faces in an image. In this regard, we propose transform domain techniques to identify the low frequency components present in an image signal to generate highly discriminative geometrical features preserving color and texture information which are analogous to human vision perceptual models. A human can do this easily, but a computer needs precise instructions and constraints. Animal detection based researchers are useful for many real life applications. Animal detection methods are helpful on the research related to locomotive behavioral of targeted animal and also to prevent dangerous animal intrusion in residential area. There are a few branches of research related to animal detection. Therefore, in this paper, Face Detection is done by Discrete Cosine Transform (DCT) approach.