Hybrid Tow Feature Extraction Descriptor For Shape Pattern Recognition

Features extracted from the image. In this work a novel method is proposed to extract an effective and sufficient features from the images. Applying normalizing technique to the images. Finally we normalize the extracted features and limited their values in the range of “zero” and “one” and “one”. This work has contributed in extracting an excellent features that fully describe the shape of the images in any orientation using a hybrid method of Gabor texture features and grey-level co-occurrence matrices.