Illumination Insensitive Reconstraction and Pattern Recognition Using Spectral Manipulation and K-Factor Spatial Transforming

Abstract

Image recognition under various changing illumination conditions is an important problem being frequently addressed. The paper presents a new approach based upon combination between spectral manipulation called the HSV and spatial transformation called the K-factor that is applied over the HSV components. Such manipulation allows composing image which is both insensitive to illumination and contains the significant spatial details of the original pattern. A useful application of this algorithm can be applied to pattern recognition problems under variable illumination. Numerical simulations as well as experimental results demonstrate the capability of the proposed algorithm to obtain reduced sensitivity to illumination variations and to increase probability of detection while maintaining the same level of false alarm rate.