Message Boards Message Boards

0
|
5737 Views
|
17 Replies
|
4 Total Likes
View groups...
Share
Share this post:

Double check if an image transform function has the appropriate colors?

Attached is a file whith what the imagetransform is making , i would like to know if this what the imagetrasnform would make? anyone? Sorry I had to edit imagehistogram is doing this but the picture does not have red, i modieifed the pixels values i just would like to know if somonelse could try the histogramtrasnform to see if the colors are the same.

Attachments:
POSTED BY: lfelipemassena misiec
17 Replies
Sort By:

things are running well...

Attachments:
POSTED BY: lfelipemassena misiec

Thanks for sharing! It looks like a good processing for enhanced visualization of some medical images.
POSTED BY: Matthias Odisio

Are you peoplesleeping? I have not fall in reality yet ...i am going to send the file with picture augmented...bigger...just people realize this show to a radiologist a breast caner mastologist you can not believe what this is for real.... any business be it on the it mine also you make it ....everyone who checked this page to see it it is on on the contract to take their also sharing of pie...for real... i can not believe!!!1

Attachments:
POSTED BY: lfelipemassena misiec

It is working it is working i ca not belive it is working .....wolfram let´s talk business huhuhuhuhuhuhuhuhu
POSTED BY: lfelipemassena misiec

Mathias I did it in 2 ways it is working look at the resolutionof the mamography the other way is a little bit incorrect but it also works and has different aplications. look at anexed file women be happy!!!!!!

Attachments:
POSTED BY: lfelipemassena misiec

I'm glad I could be of some help. You seem to be onto something interesting. Hopefully you can share code and results here when you feel ready.

POSTED BY: Matthias Odisio

Hi mathias... sorry to bother you i va checked for histogram interpolation and for thistogramtransform fucntion and as I undeerstand it does not do what i am doing...but since the transformation is very sensible, i have no means to know if the positions are changing for real running the same program on images different images it seems to work like the histogram transform for the birds but with different pictures as it is supposed to work because by flattening the values of the colors histogram it kind maintain the original colors, but what i am doing is stablishing an equation with a pattern to follow so that when the numbers representing pixels and positions are considered as image levels it is mathmatically transformed qaccording to that equation , so i do not know the outcome , and i do not want to know the outcome i just want for the mathmatical equation to reveal me the behavior, and it does.. but by flattening the hystogram for yellow and blue for instance the result should not be a gray bar but a green bar instead... then i would be considering that the hystogram proposed application would be working according the histogram trasnform command...but i do not want that i want that the image levels values according to their positions be modified according to the equation which is a imaginary complementary set of numbers that can be applied to any digital number... as i run the program i can see the changng in the pixel values because instead of showing for instance 360 499 345 it separates and it shows 3 604 993 45() and it works fine it changes the colors for hystological tissues images without me manipulating or interfereing in the process but it reveals the natural original numbers of the clusters of the before image and it enhances some images but not all of them and this is exactly what i want .. so it would be working for me.. but i just can not proove it is working..... i am not carrying if it works all the time .... because the relations between the expression of numbers in whatever greater set it belongs represents reality in a certain way ... i do not want to manipulate reality i want it to reveal for me as it should... do not take it personally.. but i really thank you... you see the picture of the christ redentor is very old and it was taken in the summer... so the infrared ultraviolet is hidden in the picture it allows to verify how was the enviroment aat the time of the taking of the picture so if i take another picutre today with different light or climate conditions i can know if there was a climate change on the enviroment.. of the surroundings...but..if there has been any incorporation on the histogram function without me knowing i can not tell if my program is working or if the others are behaving just like mine...oam not coplaining about the mathematica program or being said as a customer i am just arguing how I can proove that this is woring .. if it works for me...ok...it works...are you going to beleive in me just for saying that there was no enviromental or there was a tremendous bterring of the enviroment after years jsut by looking at a more red or greenish picutre...? no off course so this is the same here
POSTED BY: lfelipemassena misiec

I have not understood all of your message. Below I describe how you can find the functions that HistogramTransform uses.

image = ExampleData[{"TestImage", "Lena"}];
HistogramTransform[image]

Find the corresponding functions, plot them, and apply them to the image, and get the same result as HistogramTransform:

{rfunc, gfunc, bfunc} = HistogramTransformInterpolation[image]
Plot[{Quiet@rfunc[t], Quiet@gfunc[t], Quiet@bfunc[t]}, {t, 0, 1}, PlotLegends -> {"R", "G", "B"}]
MapThread[ ImageApply, {{rfunc, gfunc, bfunc}, ColorSeparate[image]}] // ColorCombine
POSTED BY: Matthias Odisio

HistogramTransform[grayscaleimage] applies a function to each pixel value so that in the output the set of pixel values is following a uniform distribution between 0 and 1. For color images, this process is repeated independently for reach color channel.

Note that in practice the resulting histogram is typically not perfectly flat because of the quantization of pixel values in the input image.

What are you trying to achieve by the way?
POSTED BY: Matthias Odisio

That between zero and 1 thing, i am modifiyng the relation using histogram so that it folows a new imaginary rule of complementary numbers, it relates to the position also instead, so that color oriented to position vanishes, and becomes more random than the histogram, but it works for me , but since the built in function does not have a line definition of the program, how it is made , i can not affirm that i ve done somehting new even thought the mathmatical formula , equation is, mr highness wolfram should know that the imaginary numbers rules that he uses is limited to decimal places but it goes far beyond basara formula or any other complex rule number before established., but wihtout the pattern it is hard to say is something new and i do not know how far my cloud has been used by other s without me knowing that they are improving a program similar to mine... so it would help to know exactly what comands it is used for the purpose of the histogramtransform comand at least I could know if it is working for real or if the histogram transform command is bothering me.
POSTED BY: lfelipemassena misiec

I get the same result. You're getting a lot of red in the output because HistogramTransform operates channel by channel, and after equalization the red channel will have large values in some areas of the image.

As for your follow up question, I think you're looking for ImageApply.
POSTED BY: Matthias Odisio

What should I expect to happen using histogramtransform...? should it allways enhance the original colors of the pictures shouldn´t it?
POSTED BY: lfelipemassena misiec

I would like to know how can I index a change of pixel according to a mathmatical value, these buit in functions kinda restricts you... is there any place where they show the lines of the buit infunctions to see if it is the same plines that someone is using, it really would save people´s time...
POSTED BY: lfelipemassena misiec

Sorry Valeriu! do you mean the colors are the same as mine? red in the mountain?
POSTED BY: lfelipemassena misiec

Yes! You can verify this in the attached file!

Attachments:
POSTED BY: Valeriu Ungureanu

thanks valeriu!
POSTED BY: lfelipemassena misiec

Yes! I tried and the colors are the same!
POSTED BY: Valeriu Ungureanu
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Tag limit exceeded
Note: Only the first five people you tag will receive an email notification; the other tagged names will appear as links to their profiles.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract