Matching Filtered Histograms to an image

Can filtered histograms be from the same image if they looked the same?

 Accepted Answer

Yes, they could. If you take the histogram of an image, and then take the histogram of the same image again, the histograms will look the same, as long as you "filtered" them the same way. After all, it's the same image so why shouldn't they?

5 Comments

More context: Let's say you have an image and you split that image into different spatial regions. Then you took the histogram of each of those spatial regions. Then you filtered it to get rid of the noise from each histogram. Can you match the histograms to their spatial regions assuming you don't know which histogram belongs to which the spatial region?
This is more of what I meant, without context it sounds like a bad question haha. Thanks
No you can't. If you just have two arbitrary histograms, you can't tell which, if any , region they belong to unless you take the histogram of those regions. I mean, how could you? There's no way. You'd need to histogram the two regions and compare your initial histograms to those.
You have an image, and if you have two arbitrary regions from that image, and their histograms are similar, or identical, then can you say that they belong to the same region? If so, how can I put into words how you can compare the histograms?
The histograms alone is not enough to tell if the histograms came from the same region. Imagine you had a ramp coming up from the bottom on the left half of the image and a ramp coming down from the top on the right half of the image, and you took histograms of the left and right halves. The histograms would be identical. Can you then say they came from the same region of the image? No, of course not, and they didn't - you know that because you know for a fact that they were taken from different regions. So comparing histograms is not sufficient to say that the regions are the same region in the same image.
Thanks!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Sign in to comment.

More Answers (0)

Asked:

Neo
on 4 Jan 2015

Commented:

Neo
on 6 Jan 2015

Community Treasure Hunt

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

Start Hunting!