Symmetric color ratio in spiral architecture

Abstract

Color ratio gradient (CRG) is a robust method used for color image retrieval and object recognition. It has been proven to be illumination-independent and geometry-insensitive when tested on scenery images. However, the color ratio gradient produces unsatisfying matching results when dealing with an object which appears rotated by a certain relative angle in the model and target images. In this paper, we adopt the idea of color ratio gradient and develop a new method called Symmetric Color Ratio (SCR) based on a hexagonal image structure, the Spiral Architecture (SA). We focus on license plate images and our aim is to achieve a higher matching rate between the SCR histogram of the images within same class in order to separate different classes of images. Our experimental results demonstrate that the proposed SCR is robust to changes over view angles. © Springer-Verlag Berlin Heidelberg 2006.