IMPLEMENTATION OF IMAGE TEXTURE ANALYSIS USING NEIGHBORHOOD GREY TONE DIFFERENT MATRIX METHOD
Texturehas found in wide application in imageprocessing and it is commonly agreed that texture analysis plays a fundamental role in classifying objects or images. Two approaches have evolved over the years for texture analysis, which are called statistical and structural analysis. In recent studi...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Year Project |
| Language: | en |
| Published: |
Universiti Teknologi PETRONAS
2006
|
| Subjects: | |
| Online Access: | http://utpedia.utp.edu.my/id/eprint/9331/1/2006%20%20-%20Implementation%20Of%20Image%20Texture%20Analysis%20Using%20Neighborhood%20Grey%20Tone%20Different%20Matrix%20Me.pdf http://utpedia.utp.edu.my/id/eprint/9331/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Texturehas found in wide application in imageprocessing and it is commonly agreed
that texture analysis plays a fundamental role in classifying objects or images. Two
approaches have evolved over the years for texture analysis, which are called
statistical and structural analysis. In recent studies, the texture classification and
discrimination usually approached by using statistical analysis. Therefore, this project
is carried out to distinguish the application of statistical analysis especially in the
approach of Neighborhood Grey Tone Different Matrix (NGDTM) in image
classification. This project presents the NGTDMapproach of image texture analysis
by using MATLAB. Thus, three types of texture selected, which consist of thirty
images respectively, are analyzed using the algorithm of first and second order
statistic developed in MATLAB. First order statistic extractsthe statistical parameters
from the image and second order statistic specifically NGTDM emphasizes on the
intensities of neighboring pixels. The first and second order characteristics of each
image are varying to each other, which are lead to the classification of texture. As a
result, each of the input images is successfully classified regarding to its type of
texture. Future work on image classification can be conducted for the use of
application, such as medical image processing and remote sensing community. |
|---|