Abstract:
In This paper presents a efficient texture based recognition on multi scale local binary pattern (LBP) texture features .It�s a simple and fast for implementation, To extract useful representative features, �uniform� LBP was proposed and its effectiveness has been evaluated. However, all �non-uniform� patterns are grouped into one pattern, so a lot of useful information is lost. In this study, propose to build a modified multiscale LBP histogram for a palmprint image. The palmprint image is divided into several regions from which the LBP feature distributions are extracted and concatenated into an enhanced feature vector to be used as a image descriptor. The useful information of �non-uniform� patterns at large scale is dug out from its counterpart of small scale; the performance of the proposed method is that it can be fully utilize LBP features. Other applications and several extensions are also discussed.
Keywords:Facial image representation, local binary pattern, multiscale, component based face recognition, texture features, face misalignment