Ed Applying luma curve to photos and footage

[JayUtah]

Picture i supplied.

No.

The picture you supplied is far too pristine. It's sharp, well-focused, and probably first or second generation.

Oversampling is part of what makes your method fail. All the images you've used have shown themselves to be egregiously oversampled. This matters because the oversampling requires some sort of interpolation. If you have a 512 x 512 pixel image and you enlarge it to 2048 x 2048, where does the image data come from to fill in all those new pixels? There are a dozen or so methods for interpolating image data, some of them aimed at smoothing the image and others aimed at trying to create a false sense of sharpness in the invented pixels. @Oystein described that to you. Further, if you have a slanted line in an image of low resolution, the antialiasing doesn't scale well. The well-worn Bresenham algorithm for line smoothing borrows from adjacent pixels to minimize the visual effect of the otherwise inevitable stair-stepping. But when you put that through interpolation for resolution changes and then reduce the quantization via a cocked-up luminance map, those stair-steps come back.

No, we need a more rough-and-tumble image, closer to what you are using from 9/11 news coverage. Here's the classic Tiananmen Square photo in the banner of this web page.

Photojournalism of the 80’s and 90’s

Jacob hales discusses why we should appreciate 80’s and 90’s photojournalism…

heritageherald.com

It's only 72 DPI, so it has a nice garbage resolution for us to play with. I'm using Adobe Photoshop 2025. You're free to determine for yourself whether there is an alpha channel in that image. For those of you playing along at home: there isn't.

Step 1 is to ignore the chrominance information. We care only about luminance, because that's all you're changing with your "Luma" adjustment. To avoid bogging down, I'll just represent that I'm using a perceptual weighting to go from the RGB color space to a pure Y-space image. Since there is very little chromaticity in that photo, it won't matter. Your toy tool is doing this implicitly under the hood, where you can't see it. I'm doing it explicitly.

Let's focus on the brave student. Here he is, in all his 72 dpi pixelated glory...

In order to replicate the oversampling in your 9/11 images, we have to oversample here. I'm oversampling by about 4X using bicubic smoothing.

He's not any clearer or fuzzier visually, but when you look closely you see that there are interpolated pixels. Each pixel in the original image has been replaced by about 16 (4²) new pixels. The luminosity has been interpolated for the new pixels by means of a particular algorithm. Again, your convenience sources have done this implicitly. I'm doing it explicitly.

Why? Because your method simply amplifies transitional pixels. It turns midtones into white bands of selectively, artificially amplified luminosity. In order for that effect to emerge as an artifact from your images, there have to be a lot of transitional pixels such as those created in oversampling. All those invented pixels from the oversampling are transitional pixels. Your images supply that oversampling by virtue of all the violence that has been done them in their journey to your computer.

Now we apply your inverted-paraboloid luminance map.



And there are your "halos."

I've explained the mathematics to you. I've shown empirically that your "alpha halos" are conclusively and exclusively the artifacts of your misguided application of a luminance map.

Will you kindly concede now?

 

[arthwollipot]

We've been ready for you to make some kind of claim for ages now.

 

[nt1]

@nt1, have you tried falsifying your result? Take a still from.a genuine videoed image from the same time frame (around 2001) which has been digitized and downloaded through YouTube, then run it through your editor, and see if the masking effect shows? You claimed this earlier with an airplane image but im.not sure if the original image was digitized from 25 yr old video et al.

Alternatively, could anyone else here familiar with nt1's process duplicate it?

This is one of the first genuine questions. I responded earlier. Yes. I have. Yes. They are NOT conclusive. Some are conclusive that they are not natural and would need the absolute direct files and metadata. That is rarely available nor do i want it. I would want a professi

Address the explanation in the second paragraph.

Jayutah. I want to give you a chance to regroup and center your happiness. It seems you are frustrated at more than just one thread in a forum.

Are you heavily invested into this topic? If so. Me too. Maybe we could be internet buddies and be helpful to each other. But your level of frustration has exceeded anything i have brought to a thread.

What can i do for you to help you be a little less aggressive?

 

[arthwollipot]

Okay, so if I'm reading it right - and I'm not sure I am, so I'm open to being corrected - nt1 has manipulated an image and used that as proof that the image has been manipulated.

 

[nt1]

So you've manipulated or tweaked images, however you want to describe it. What is this supposed to prove?

#1

Curve manipulation is the term.

It does sound bad. Agreed 100%

Manipulation does not mean falsly altering.

it shifts the alpha and RGB red green blue spectrum

together or independently. It can show unnatural lighting. It can show composite images. It can expose altered photos or footage.

It can be replicated by anyone with any app or program with a curve function.

I am sure that a cheap or free app may not be the same as a professional product.

I have tried both and not seen a difference that is important.

When i say important. Wheter it is a frwe app or a paid professional product. Without the actual original file with metadata, it is not scientifically conclusive. But scientifically conclusive should be difficult. And it should be left for professionals to finalize.

But a free or cheap photoshop app in 2025 is sufficient to investigate.

There are things found that you cannot replicate with a photo of someone you know compared to media supplied photos.

There are reasons. Social influencers will show the same red flags as AI photos. Usually not identical though. It just requires constant observation.

I am not a professional and i am not an expert. I have researched what is possible and not possible. It is not possible to prove conclusively anything without the original file, with metadata and a trained knowledge.

What it can do. Even at a novice level. Is expose artifacts through unnatural lighting. Or. Through digital markers that were not thoroughly removed.

Im breaking here because it is no fun reading a massive text. I will continue with #2

 

[nt1]

So you've manipulated or tweaked images, however you want to describe it. What is this supposed to prove?

 

[nt1]

Okay, so if I'm reading it right

Hi Art.

I am bending the light of footage and photos to show unnatural light and composite (layered) images.

I am not showing. I said that make a direct answer.

I am trying

I am investigating

I am double checking

I do not have the authority to say it is conclusive.

So yes. I am manipulating them by bending the alpha or R,G,B spectrum.

It is only conclusive with an original file containing original metadata in a professionals hands. But a novice can reveal and see obvious cases and also can bring up cause for concern on photos that dont react properly.

*I deleted the middle for being long winded

 

[nt1]

We've been ready for you to make some kind of claim for ages now.

I like you art.

To be funny. Im not a gold digger so ive never tried to make a claim in my life. But here we go.

Smooth line. Zig zag from computer graphics.


Black line. Smooth. Year 2000. Home video. Wtc7

White line. Zig zag. Nist video. N64 graphics.


I am looking to reveal unnatural lighting and lines. Masking lines. Composite (layered) photos

and think this is an example.

Please remember

This example with the white line comes from the one video provided by a member here. I could also make it black. By bending it in the opposite way. It doesnt change the zig zag. Which is the focus hede

Yes i think i could do better.

Yes i have provided a couple

No. I have not shown 5% of the stuff i have.

Because
1. I have been recieved like a leper.
2. This one shows just enough that if no one cares whats the point of moving on
3. Id be more than happy to flood the place with examples but it seems to be unwelcome or not understood yet. Which both are fair and appreciated.


Claim.

Cause for concern that
Pre or post footage manipulation of footage.

Reason obviously unknown.

If manipulation of any kind can be found

A re-examination needs to take place.


Art.

I have a ton of public facts that are cause for concern when compared to the level of investigation that took place. Public facts and documents that anyone can see and access.

Thanks art

 

[Oystein]

Jayutah. I want to give you a chance to regroup and center your happiness. It seems you are frustrated at more than just one thread in a forum.

Are you heavily invested into this topic? If so. Me too. Maybe we could be internet buddies and be helpful to each other. But your level of frustration has exceeded anything i have brought to a thread.

What can i do for you to help you be a little less aggressive?

You still did not "Address the explanation in [JayUtah's] second paragraph."

#1

Curve manipulation ... can show unnatural lighting. It can show composite images. It can expose altered photos or footage.

You have not shown this to be true.
Nor have you, assuming this is true, shown that you are doing it right.

Without the actual original file with metadata, it is not scientifically conclusive.

In other words: No matter which card we pick (which video uploaded to the internet), since we will never look at the original file with metadata, the method will NEVER yield a scientifically conclusive result.
AND you knew it, I suppose, before you even asked us to pick a video.

There are things found that you cannot replicate with a photo of someone you know compared to media supplied photos.

You have not demonstrated that this is true.
You make a global negative claim here. You can only prove a global negative claim if you can prove it for all possible instances, while, to falsify a global negative claim, it suffices to present a single example to the contrary.
Which JayUtah did in #361.

You need to respond to #361.

I am not a professional and i am not an expert.

It shows.

I have researched what is possible and not possible. It is not possible to prove conclusively anything without the original file, with metadata and a trained knowledge.

In other words, your entire endeavour here - have us pick a video for you to apply an alleged "scientific method" that is virstually guaranteed to detect edges and make them bright and obvious, and then declare them to be "suspicious" - is sure to FAIL, because "It is not possible to prove conclusively".
But you also have not shown that your fiddling with luminance curves is of ANY value to assess the likelihood of tampering other than the usual conversions as a video is copied and copied and shrunk and enlarged and compressed and enhanced and copied again and shrunk again and enhanced again.

What it can do. Even at a novice level. Is expose artifacts through unnatural lighting.

First and foremost, you CREATE artefacts.

Or. Through digital markers that were not thoroughly removed.

You have not shown that this is what you claim you detect. It's a bare-assed claim.

Im breaking here because it is no fun reading a massive text.

Being able to read, comprehend and respond to longish texts is, alas, a basic required skill for people learning and/or communicating science. It is troubling if you lack that skill.

---

The next reply after this "#1" I just responded to contains nothing but attachments. You did not even label it "#2".
You have been advised before by moderators that this is a text-based forums, that you should not just post images. You MUST say in actual text what it is you wish to communicate.

 

[Oystein]

Hi Art.

I am bending the light ...

That is a nonsense claim.

 

[Andy_Ross]

Toy image apps.

Ok. Please name a non toy. Ill try that.

Ill go try others now. Thank you for the advice

What images or videos are you using as your control?

 

[theprestige]

I dont know zeno or their theory. I am not a content creator. My apologies if your genuine questions were missed.

It's a play on Zeno's Paradox. It would be better if you were a content creator. You didn't miss my genuine questions, you just acknowledged them and then ignored them.

Edited by Agatha: 

Edited to make quote visible. Please check to ensure no extraneous {color} tags have appeared when quoting.

 

[theprestige]

Okay, so if I'm reading it right - and I'm not sure I am, so I'm open to being corrected - nt1 has manipulated an image and used that as proof that the image has been manipulated.

He's used that as proof that there should be "cause for concern".

I agree, incidentally. There is cause for concern. But not about the image.

 

[nt1]

You still did not "Address the explanation in [JayUtah's] second paragraph."


You have not shown this to be true.
Nor have you, assuming this is true, shown that you are doing it right.


In other words: No matter which card we pick (which video uploaded to the internet), since we will never look at the original file with metadata, the method will NEVER yield a scientifically conclusive result.
AND you knew it, I suppose, before you even asked us to pick a video.


You have not demonstrated that this is true.
You make a global negative claim here. You can only prove a global negative claim if you can prove it for all possible instances, while, to falsify a global negative claim, it suffices to present a single example to the contrary.
Which JayUtah did in #361.

You need to respond to #361.


It shows.


In other words, your entire endeavour here - have us pick a video for you to apply an alleged "scientific method" that is virstually guaranteed to detect edges and make them bright and obvious, and then declare them to be "suspicious" - is sure to FAIL, because "It is not possible to prove conclusively".
But you also have not shown that your fiddling with luminance curves is of ANY value to assess the likelihood of tampering other than the usual conversions as a video is copied and copied and shrunk and enlarged and compressed and enhanced and copied again and shrunk again and enhanced again.


First and foremost, you CREATE artefacts.


You have not shown that this is what you claim you detect. It's a bare-assed claim.


Being able to read, comprehend and respond to longish texts is, alas, a basic required skill for people learning and/or communicating science. It is troubling if you lack that skill.

---

The next reply after this "#1" I just responded to contains nothing but attachments. You did not even label it "#2".
You have been advised before by moderators that this is a text-based forums, that you should not just post images. You MUST say in actual text what it is you wish to communicate.

Hi there. Would

You still did not "Address the explanation in [JayUtah's] second paragraph."


You have not shown this to be true.
Nor have you, assuming this is true, shown that you are doing it right.


In other words: No matter which card we pick (which video uploaded to the internet), since we will never look at the original file with metadata, the method will NEVER yield a scientifically conclusive result.
AND you knew it, I suppose, before you even asked us to pick a video.


You have not demonstrated that this is true.
You make a global negative claim here. You can only prove a global negative claim if you can prove it for all possible instances, while, to falsify a global negative claim, it suffices to present a single example to the contrary.
Which JayUtah did in #361.

You need to respond to #361.


It shows.


In other words, your entire endeavour here - have us pick a video for you to apply an alleged "scientific method" that is virstually guaranteed to detect edges and make them bright and obvious, and then declare them to be "suspicious" - is sure to FAIL, because "It is not possible to prove conclusively".
But you also have not shown that your fiddling with luminance curves is of ANY value to assess the likelihood of tampering other than the usual conversions as a video is copied and copied and shrunk and enlarged and compressed and enhanced and copied again and shrunk again and enhanced again.


First and foremost, you CREATE artefacts.


You have not shown that this is what you claim you detect. It's a bare-assed claim.


Being able to read, comprehend and respond to longish texts is, alas, a basic required skill for people learning and/or communicating science. It is troubling if you lack that skill.

---

The next reply after this "#1" I just responded to contains nothing but attachments. You did not even label it "#2".
You have been advised before by moderators that this is a text-based forums, that you should not just post images. You MUST say in actual text what it is you wish to communicate.

Im sorry. This reaponse is too difficult tk

You still did not "Address the explanation in [JayUtah's] second paragraph."


You have not shown this to be true.
Nor have you, assuming this is true, shown that you are doing it right.


In other words: No matter which card we pick (which video uploaded to the internet), since we will never look at the original file with metadata, the method will NEVER yield a scientifically conclusive result.
AND you knew it, I suppose, before you even asked us to pick a video.


You have not demonstrated that this is true.
You make a global negative claim here. You can only prove a global negative claim if you can prove it for all possible instances, while, to falsify a global negative claim, it suffices to present a single example to the contrary.
Which JayUtah did in #361.

You need to respond to #361.


It shows.


In other words, your entire endeavour here - have us pick a video for you to apply an alleged "scientific method" that is virstually guaranteed to detect edges and make them bright and obvious, and then declare them to be "suspicious" - is sure to FAIL, because "It is not possible to prove conclusively".
But you also have not shown that your fiddling with luminance curves is of ANY value to assess the likelihood of tampering other than the usual conversions as a video is copied and copied and shrunk and enlarged and compressed and enhanced and copied again and shrunk again and enhanced again.


First and foremost, you CREATE artefacts.


You have not shown that this is what you claim you detect. It's a bare-assed claim.


Being able to read, comprehend and respond to longish texts is, alas, a basic required skill for people learning and/or communicating science. It is troubling if you lack that skill.

---

The next reply after this "#1" I just responded to contains nothing but attachments. You did not even label it "#2".
You have been advised before by moderators that this is a text-based forums, that you should not just post images. You MUST say in actual text what it is you wish to communicate.

I appreciate your effort in responses. Thank you for your time. Ill will dissect that and hope to satisfy you with a reaponse.

 

[nt1]

That is a nonsense claim.

I do apologize if my terminology is not accurate.


Curve manipulation in pictures is not bending light. It’s a digital adjustment of pixel brightness values.


Loosely, you could say curve manipulation simulates how light might look if it were changed—like making an image brighter or darker—but it’s not actually bending light. It’s just adjusting pixel values digitally. So as a very rough analogy, it’s okay, but not physically accurate.

 

[nt1]

You still did not "Address the explanation in [JayUtah's] second paragraph."


You have not shown this to be true.
Nor have you, assuming this is true, shown that you are doing it right.


In other words: No matter which card we pick (which video uploaded to the internet), since we will never look at the original file with metadata, the method will NEVER yield a scientifically conclusive result.
AND you knew it, I suppose, before you even asked us to pick a video.


You have not demonstrated that this is true.
You make a global negative claim here. You can only prove a global negative claim if you can prove it for all possible instances, while, to falsify a global negative claim, it suffices to present a single example to the contrary.
Which JayUtah did in #361.

You need to respond to #361.


It shows.


In other words, your entire endeavour here - have us pick a video for you to apply an alleged "scientific method" that is virstually guaranteed to detect edges and make them bright and obvious, and then declare them to be "suspicious" - is sure to FAIL, because "It is not possible to prove conclusively".
But you also have not shown that your fiddling with luminance curves is of ANY value to assess the likelihood of tampering other than the usual conversions as a video is copied and copied and shrunk and enlarged and compressed and enhanced and copied again and shrunk again and enhanced again.


First and foremost, you CREATE artefacts.


You have not shown that this is what you claim you detect. It's a bare-assed claim.


Being able to read, comprehend and respond to longish texts is, alas, a basic required skill for people learning and/or communicating science. It is troubling if you lack that skill.

---

The next reply after this "#1" I just responded to contains nothing but attachments. You did not even label it "#2".
You have been advised before by moderators that this is a text-based forums, that you should not just post images. You MUST say in actual text what it is you wish to communicate.

I apologize you feel so strongly that it required the length and effort you applied.

 

[nt1]

It's a play on Zeno's Paradox. It would be better if you were a content creator. You didn't miss my genuine questions, you just acknowledged them and then ignored them.

Edited by Agatha: 

Edited to make quote visible. Please check to ensure no extraneous {color} tags have appeared when quoting.

I will go back specifically to all of your replies.bIf you feel i ignored your questions that is now my top priority.

 

[nt1]

Help me out here. What's the gist of your theory?

Bin Laden did it, but the US government was behind it and covered up the connection?

The US government did it, covered it up, and blamed bin Laden?

The US government did it, without planes, then faked the planes, covered it up, and blamed bin Laden?

I am not currently claiming a full theory with this presentation.

I only wanted to show the possibility that footage manipulation can be revealed by utilizing tools to display artifacts and composites.

The theory behind that would be.
Someone would be in on it.


To answer you text directly some of what you asked i have alternative theories on personally. I would expand on that later after i explore this photo analysis with the group.



I have a body of research. I have several
ideas and theories based in facts. I do not subscribe to any content creators theory and believe they are all wrong and possible created on purpose.

I hope this was a decent response for you. Let me know anything you want more out of on this.

 

[nt1]

You also said you weren't going with the more fantastical theories (space lasers, etc.) And you brought up the government investigations. So I figured it must be something along the lines of the "official" story of what happened, but with a different mastermind behind it.

So it's Zeno's Conspiracy Theory.

The official story is corrupt. Not wildly. Its documented the push by familes to even get the commision. Its documented the time frame being pleaded for. The start date is much farther out from the event than one would expect.

its not one of these laser beam theories. There just simply is more to be found. If not from the story then from the obstruction of it. I believe both can be found.

 

[nt1]

So what's your claim? People who physically present at the collapse saw something different than what is seen in the video?

What do you think they saw?

That is quite accurate to my main theory.

I believe that there was prefabricated footage to match the event. To allow any errors to be masked.

What people saw very closely simulated the footage seen.







I could expand on this at any length at your request. With references and citations from info in the public domain. Some would be direct facts. Some would be inference supplied with multiple examples to show cause for concern.