Medical Image Tissue Analysis Using Spatial Texture Attributes and ANN

ISBN-13: 9786139893034 / Angielski / Miękka / 2018 / 188 str.

Suhair H. Al-Kilidar; Loay E. George

Opis

Texture Classification is the process which used in various pattern recognition applications and classification textures that possess a characteristic appearance. The analysis of texture parameters is a useful way of increasing the information obtainable from images. It is an ongoing field of research, whether these images are medical images or natural. In this work, many classification systems was studied and some variants are proposed and applied to distinguish different types of texture using different discriminating characteristics of textures. These features are derived from the Gray Level Co-occurrence Matrices (GLCM), Run Length Matrices (RLM), Contrast Matrices (CM), and Absolut Gradient Matrices (AGM). Six different sets of features were introduced, among them are: MGLCM, CtdGLCM, MRLM, pCM, pAGM, and CtdAGM; depending on the four traditional methods that was used. The considered methods were applied on two Datasets. The first one consists of 13 classes of textures belong to three sets taken from Salzburg Texture Image Database (i.e., bark, marble and woven fabric), and the Dataset 2 consist of 4 classes of textures taken from skin tissues (normal and abnormal).

O autorze

Al-Kilidar, Suhair H.Suhair Al-Kilidar Received B.Sc. in Mathematics Science, M.Sc. in Applied Mathematics and Computer Applications. She's a Head of Programmers in IT Directorate, Ministry of Science and Technology, Iraq. Her current research interest includes; Pattern Recognition, Image Processing, Information Security and Identification, Programs Development.