Discriminant splitting for feature extraction applicable to view-independent object recognition

This paper presents a novel object recognition technique that is based on discriminant image splitting for feature extraction. Spatially homogeneous and discriminant regions for each object class are generated. The classical image splitting technique is used in order to determine those regions. Thus, each object class is characterized by a unique region pattern which consist of homogeneous and discriminant 2-D regions. The mean intensities of these regions are used as features. The proposed method has been tested in view independent object recognition at the COIL-100 database. The obtained results demonstrate that the method can achieve a very satisfactory recognition rate.