How do you randomly choose a grey bar in picture to turn white?

1 visualización (últimos 30 días)
I have the task to randomly choose a bar to be the target in a stimulus experiment with grey bars. The code for the bars thus far is stated beneath. However, it turns none of the bars in white. In the end, I have to have two pictures: one with a black background and 16 grey bars and the other one with just one randomly chosen bar white.
Can somebody help me how to randomly choose one of the bars to become white?
picture=zeros(500,500);
picture(100,96:104)=100;
picture(200,96:104)=100;
picture(300,96:104)=100;
picture(400,96:104)=100;
picture(100,196:204)=100;
picture(200,196:204)=100;
picture(300,196:204)=100;
picture(400,196:204)=100;
picture(100,296:304)=100;
picture(200,296:304)=100;
picture(300,296:304)=100;
picture(400,296:304)=100;
picture(100,396:404)=100;
picture(200,396:404)=100;
picture(300,396:404)=100;
picture(400,396:404)=100;
% Saving as image
imwrite(picture,'stimulusimage.tiff','tiff')
% Replacing a random bar with colour white ????
x=picture==100;
y=length(x)==9; % =9 because I thought that when x has a length of 9 times 1 (1 1 1 1 1 1 1 1 1) that represents a grey bar
picture(y)=255;
% Saving as image
imwrite(picture,'stimulusimagewithwhitebar.tiff','tiff')

Respuesta aceptada

Adam Danz
Adam Danz el 26 de En. de 2020
Editada: Adam Danz el 27 de En. de 2020
For grayscale images, imwrite() assumes that the dynamic range of the first input is [0,1] and automatically scales the data by 255 before writing it to the file as 8-bit values.
gray = 100/255;
picture=zeros(500,500);
picture(100,96:104)=gray;
picture(200,96:104)=gray;
% etc....
To randomly select a bar in the image and change its color to white, you can use bwlabel (requires Image Processing Toolbox) to identify the bar-groups, then randomly select one and change its value to 1.
% Group the picture values
bwl = bwlabel(picture > 0);
% Select a random segment
randSection = randi(max(bwl(:)),1,1);
% Set that bar to white
picture(bwl==randSection) = 1;
% Show image
imshow(picture) % or, in your case, imwrite(...)

Más respuestas (2)

Stijn Haenen
Stijn Haenen el 26 de En. de 2020
This is most likely not the best way to do it but it works:
picture=zeros(50,50);
picture(10,10:12)=100;
picture(20,10:12)=100;
picture(30,10:12)=100;
picture(40,10:12)=100;
picture(10,20:22)=100;
picture(20,20:22)=100;
picture(30,20:22)=100;
picture(40,20:22)=100;
picture(10,30:32)=100;
picture(20,30:32)=100;
picture(30,30:32)=100;
picture(40,30:32)=100;
picture(10,30:32)=100;
picture(20,30:32)=100;
picture(30,30:32)=100;
picture(40,30:32)=100;
% Saving as image
%imwrite(picture,'stimulusimage.tiff','tiff')
% Replacing a random bar with colour white ????
imagesc(picture)
[x,y]=find(picture==100);
x0=datasample(x,1);
y0=datasample(y,1);
picture(x0-3:x0+3,y0-3:y0+3)=picture(x0-3:x0+3,y0-3:y0+3)*2.5;
figure()
imagesc(picture)

Eveline Kallenberg
Eveline Kallenberg el 29 de En. de 2020
Thanks for your help guys.
I ultimately did it this way, since I thought it had to be a little different.
picture=zeros(500,500);
x = [100 100 100 100 200 200 200 200 300 300 300 300 400 400 400 400];
y = [100 200 300 400 100 200 300 400 100 200 300 400 100 200 300 400];
for i = 1:16
picture(x(i),y(i)-4:y(i)+4)=100;
end
% Saving as image
imwrite(picture,'stimulusimage.tiff','tiff')
% Random bar turn white
randNumber = randi(16);
picture(x(randNumber),y(randNumber)-4:y(randNumber)+4)=255;
% Saving as image
imwrite(picture,'stimulusimagewithwhitebar.tiff','tiff')
% Random bar turn black
randNumber = randi(16);
picture(x(randNumber),y(randNumber)-4:y(randNumber)+4)=0;
% Saving as image
imwrite(picture,'stimulusimagewithblackbar.tiff','tiff')
  2 comentarios
Adam Danz
Adam Danz el 29 de En. de 2020
Editada: Adam Danz el 29 de En. de 2020
Have you viewed the tiff images being produced? There are no gray bars. The stimulusimage and the stimulusimagewithwhitebar are the same (all white). The stimulusimagewithblackbar image is all white and one black.
See the first sentence of my answer to understand why.
Also, the method I shared to identify pixels of the randomly selected bar is more robust than the one you're using. For example, if you ever decide to change the bar size or orientation, my method will deal with it but yours will need changed.
Last bit of advice; if you want to see the images without having to open a file, use
imshow(picture)
Eveline Kallenberg
Eveline Kallenberg el 29 de En. de 2020
You are actually very right now I look at it again. Thank you very much Adam!

Iniciar sesión para comentar.

Categorías

Más información sobre Printing and Saving en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by