How to convert gray image to HSV plane V - value image?

Hi,

Given my image is a gray image and i don't have the rgb image of this gray image, can i know how can i convert this gray image to the HSV colour plane V - value plane image only?

The gray image and the value plane image looks quite similar but actually their values are quite different so if anyone have any idea on this, please let me know about the possibility.

Thank you in advance.

Answer 1

[Walter Roberson]

I used

cam = imread('cameraman.tif');
HSV = cat(3, zeros(size(cam)), zeros(size(cam)), double(cam)./255);
RGB = hsv2rgb(HSV);
imshow(RGB)

It seems to have worked

[MR. W]

Hi,

Thanks and appreciate it, it looks to be working , let me test with a real image to see is there any different in the value of V and gray image by performing this method.

I will let you know after this.

[Thorsten]

It depends on how you compute the gray image. In HSV the V is computed as

V = max(I, [], 3);

From the help of rgb2gray:

rgb2gray converts RGB values to grayscale values by forming a weighted sum of the R, G, and B components:

    0.2989 * R + 0.5870 * G + 0.1140 * B

The coefficients used to calculate grayscale values in rgb2gray are identical to those used to calculate luminance (E'y) in Rec.ITU-R BT.601-7 after rounding to 3 decimal places.

So unless you compute the gray image like HSV as

V = max(I, [], 3);

you will find that your V and the gray image differ.

[MR. W]

Thank you both for the info. Looks like the rgb2gray method result and the max(I,[],3) result get will always not be the same. Therefore, given a gray image, is there any method we can use to detect which method they have been using to convert the image into gray format?

Or is there any formula/equation can be used to convert a rgb2gray method result into max(I,[],3) result?

[Walter Roberson]

Good points, Thorsten.

[Walter Roberson]

I am certain that there is no formula to convert between the two methods.

As for testing which method had been used: good question. I do not have an answer for that at this time. I notice that they have different distribution statistics, but I do not know yet how that could be taken advantage of.

If you happened to know that an area was originally bright blue, then if the area shows up bright then the HSV method has been used; if the area shows up dark then the rgb2gray method has been used.

[MR. W]

Thanks again for the information and appreciate it. I will try to read through some related materials first to check on the possibility.

Have a nice day.