How can I represent my image pixel values inside a HSV cone

Hello,
I'm trying to develop a colour model for fire in HSV colour space.I have a million sample pixels per channel for the HSV colour space. I want to represent these samples inside the HSV cone to get an idea of how they are distributed. Any thoughts or suggestions on how i could possibly proceed with this? Thanks in anticipation.

 Respuesta aceptada

Image Analyst
Image Analyst el 22 de Nov. de 2013
See my attempt at doing it in MATLAB attached below in blue.
A better option is to use the Java code that was originally written for ImageJ but can be called from MATLAB. See my code samples in http://www.mathworks.com/matlabcentral/answers/?search_submit=answers&query=color-inspector&term=color-inspector

3 comentarios

I've visualized and managed to 'carve' out the gamut in RGB color space. However as part of my analysis I want to visualize it in the HSV cone as well.
I initially used polar plot command to plot Hue and Saturation. It seems to make sense to me but I'm not sure if it is correct.
I followed this post ( http://www.mathworks.de/matlabcentral/answers/84068-lab-vs-rgb-vs-hsv-color-spaces-for-color-detection ) earlier and installed ImageJ but i can't figure out how to import my data set into ImageJ. It's not an image, just a million by 3 matrix of HSV values.
Just create any image. Like (untested)
k = 1;
for row = 1 : 1000
for col = 1 : 1000
hsvImage(row, col, 1) = hsv(k, 1);
hsvImage(row, col, 2) = hsv(k, 2);
hsvImage(row, col, 3) = hsv(k, 3);
k = k + 1;
end
end
% Now convert that to an RGB image
rgbImage = hsv2rgb(hsvImage);
imwrite(rgbImage, fullFileName);
Then import it into the Color Inspector and view the gamut as HSV.
Thanks. That worked like a charm. I should have thought of that sooner.

Iniciar sesión para comentar.

Más respuestas (1)

Bjorn Gustavsson
Bjorn Gustavsson el 22 de Nov. de 2013
Simply use scatter3?

1 comentario

I tried scatter3 but since the circular nature of the HSV colour space isn't taken into account the results are not adequate for my needs.

Iniciar sesión para comentar.

Categorías

Preguntada:

el 22 de Nov. de 2013

Comentada:

el 22 de Nov. de 2013

Community Treasure Hunt

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

Start Hunting!

Translated by