دراسة مقارنة لطريقة العزوم الثابتة Hu في درجة وضوح منخفضة لتمييز الصور الرقميـة تحت تأثير حالات مختلفـة لموضع الصورة
Abstract
Image recognition is the best subject that has been studied in last decades and there are many methods for digital image recognition the more generally used is invariant moment Hu , since this method work on digital image recognition for many different cases such as changing scale of the image size , rotate image , change or translation of image position and image noise , so this method work on recognition of seven invariant properties for this differences on the shape of original image , but the recognition for this multiple differences for digital image needs image which having a high obviousness degree more than 150 dpi so that more obviousness degree for digital image high method of recognition more accurate and effective , but obtaining of high obviousness digital image would not be available to us for many reasons there for we cannot know the effect digital image with low obviousness on the method of invariant moment Hu and for many different cases for above mentioned difference.
So that this research focus on the study of method of recognition according to invariant moment Hu approach with low obviousness degree less than 150 dpi for different digital image cases as regard changing image scale size , image rotation ,changing or translation image position and image noise so that we can know the accuracy of recognition method Hu under those different cases which change the shape of original image also in this research a suggestion have been done for improving Hu method to invariant moment to get more accuracy and efficiency concerning recognition of digital image with more invariance for different digital image cases as regard changing image scale size , image rotation ,changing or translation image position and image noise , so that we try to know about accuracy of improving Hu recognition method and comparing its results with results of original Hu method .
