Image fractal dimension value
19 views (last 30 days)
Show older comments
I have interest in fractal dimension (FD) as I was reading this paper https://translational-medicine.biomedcentral.com/track/pdf/10.1186/1479-5876-8-140.pdf
I run the code below on the right figure (Koch snowflake) https://translational-medicine.biomedcentral.com/articles/10.1186/1479-5876-8-140/figures/1 but I get different results, any idea why? Any recommendation of introductory book covering this issue with matlab?
clc;
clear all;
close all;
fileName='E:/fd126.png'
I = (imread(fileName));
I = imfill(I,'holes');
figure
imshow(I);
BW = imbinarize(I);
BoxCountfracDim(BW) %using Hausdorff (Box-Counting) Fractal Dimension with multi-resolution calculation version 1.0.0.0 by Tan Nguyen
hausDim(BW) %using Hausdorff (Box-Counting) Fractal Dimension version 1.2.0.0 by Alceu Costa
2 Comments
John D'Errico
on 15 Feb 2022
What result do you get? How far is it off? Do you accept that this computation could some some degree of error in it? How can we know what you think is the wrong number?
Image Analyst
on 16 Feb 2022
If you want us to run the code you should attach BoxCountfracDim and hausDim. Otherwise I'm not doing anything.
Accepted Answer
yanqi liu
on 16 Feb 2022
yes,sir,may be should use image as same,such as
clc; clear all; close all;
fileName='12967_2010_Article_2115_Fig1_HTML.webp.jpg';
[I,map] = imread(fileName,'jpg');
if ~isempty(map)
I = ind2rgb(I,map);
end
figure
imshow(I);
BW = im2bw(I,0.9);
bws = imfill(~BW,'holes');
bw1 = bwperim(bwareafilt(bws,1));
[r,c] = find(bw1);
bw1 = bw1(min(r):max(r),min(c):max(c));
bw2 = bwperim(bwareafilt(bws,1,'smallest'));
[r,c] = find(bw2);
bw2 = bw2(min(r):max(r),min(c):max(c));
figure;
subplot(1,2,1); imshow(bw1);
subplot(1,2,2); imshow(bw2);
BoxCountfracDim(bw1) %using Hausdorff (Box-Counting) Fractal Dimension with multi-resolution calculation version 1.0.0.0 by Tan Nguyen
hausDim(bw1) %using Hausdorff (Box-Counting) Fractal Dimension version 1.2.0.0 by Alceu Costa
BoxCountfracDim(bw2) %using Hausdorff (Box-Counting) Fractal Dimension with multi-resolution calculation version 1.0.0.0 by Tan Nguyen
hausDim(bw2) %using Hausdorff (Box-Counting) Fractal Dimension version 1.2.0.0 by Alceu Costa
ans =
1.0495
ans =
1.0495
ans =
1.2097
ans =
1.2097
>>
now,we can see
first, 1.0495
second, 1.2097
as similar with 1 and 1.26
More Answers (0)
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!