How do I detect and find the discolored area as a percentage of selected area

Question

Dinuka Ratnasinghe 2021-2-22

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/752359-how-do-i-detect-and-find-the-discolored-area-as-a-percentage-of-selected-area

评论： KALYAN ACHARJYA 2021-2-24

please could you help me to find the discolored area (dark gray) as a percentage. (Compared to the full area of the sample - inner circle)

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

KALYAN ACHARJYA 2021-2-22

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/752359-how-do-i-detect-and-find-the-discolored-area-as-a-percentage-of-selected-area#answer_629754

编辑：KALYAN ACHARJYA 2021-2-22

在 MATLAB Online 中打开

rgbImage=imread('image_file_name');
[r,c]=size(rgbImage);
grayImage=rgb2gray(rgbImage);
bwImage=~bwareafilt(~imbinarize(grayImage),1);
figure,imshow(bwImage); title('Region of Interest');
%% Pixels Area Calculation in Percentage
data1=bwImage(:);
pix_counts=sum(data1);
fprintf('The Approx Percentage area (Pixel Counts): %f',pix_counts/length(data1)*100);
disp('%');

Result:

The Approx Percentage area (Pixel Counts): 70.739068%

2 个评论
显示无隐藏无

Dinuka Ratnasinghe 2021-2-24

I'm very thankful for your comment and support.

It seems like this counts the circular area? What I wanted is detect the darken patch inside the circle and calculate the percentage by comparing it with the area of inner circle. I'm new to matlab and sorry If I'm wrong.

KALYAN ACHARJYA 2021-2-24

@Dinuka Ratnasinghe This is also possible with the addition of just one or 2 lines in my code. I have answered the question based on your original question "How do I detect and find the discolored area as a percentage of selected area"

Steps:

The Larger Section is already extracted (Done)
Get the inner circle with considering different threshold value, here you need to segment the inner circle only (Or you may consider the edges and consider largest blobs only)
Once done calculate the percentage of pixels as you want.

Good Luck!

请先登录，再进行评论。