Darktable to Uncover NEF Fraud. Using 3,6,1 + 6

Start a topic the way you did for your first post and upload your photo’s. Clearly state your query. Its a touchy topic but I suppose you could ask for ways to try and detect retouching or compare two images otherwise not really sure this forum would be a place to continue much beyond that …
Darktable is a raw processor not a forensic tool…simply looking at the waveform scope should show gross differences of the type you have described…

Negadoctor is just inverting your image…you could just invert a tone curve…

original

After negadoctor What is black becomes white what is red becomes green

Perspective is to help with distortion…see this recent play raw image. The building has distortion. The algorithm tried to identify horizontal and vertical elements that it can use to correct

Before…

Verticals corrected and the image cropped to clean it up

Even if it’s not going to court, be aware that using these tools may not actually “prove” anything, especially since you have already decided that the image is fraudulent?

2 Likes

There’s clearly a language barrier issue here, but for the purposes of proving that something has been altered or forged, darktable is not the correct tool.

A decade or so ago, some Nikon cameras shipped with a feature that allowed an image to be authenticated cryptographically such that you could be sure it was unaltered since it was shot by the camera. This was intended to be used by photographers providing court evidence/etc.

The feature was discontinued because it simply did not work - it got pwnt over a decade ago - Nikon Image Authentication System: Compromised | ElcomSoft blog

It would be really cool if you could put a gpg key on your sd card and the camera would sign the files it produced with the key.

1 Like

I think Canon had the same…funny you can still see it on the Nikon page…Nikon | Imaging Products | Image Authentication Software

True, although it is in the Nikon “Product Archive”, whatever that is.

2 Likes

Andy,
You are absolutely right! there is a language barrier due to some not actually listening.
In this matter, the images were (1st) edited and (2nd) edits reshot with the D200. What is apparent to me is that the responders don’t seem to get that the image has been reshot. Once the image has been reshot there would be ‘No Inconsistencies’ between the (2) sets of photos generated by the D200 so the jpg will match the raw file exactly. This was told to me directly by Nikon. I was also told that as result of the latter, that I would have to look for other obvious edits in relation to how the edits were performed and to think outside the box.

I have the absolute proof in some of the pictures and the Police has seen them as well and have given me a set instructions to follow if I want to move ahead with a case. I don’t want that. I would rather have a live conversation with a photo expert who can help me understand (1) Tilt Shifting (2) Exposure Shifting. This is how I got to DT. *It’s working for me!

In this matter, the images were (1st) edited and (2nd) edits reshot with the D200. What is apparent to me is that the responders don’t seem to get that the image has been reshot.

How was the 1st edit “reshot with the D200”. What exactly was “reshot”? Perhaps the 1st image was displayed on a screen, and that was photographed with the D200. Or perhaps it was printed, and that was photographed with the D200. But why does the result show negated colour? And why are the top row and bottom two rows an exact shade of gray?

In my opinion, we are being given too little information about what has happened, or is alleged to have happened. (Perhaps there are good reasons for this.)

We hear you I think but you are not making much sense and you are a moving target…I think we all get that someone took the jpg of your raw file , doctored it and then reshot it passing it off as the original…what you don’t seem to want to hear is that finding DT however you did is not really any solution to your problem or at least not one that most of us have envisioned so I don’t think that we don’t understand we are just not sure what you are expecting in the way of forensics from DT when you can apparently look and images and see doctored or missing features…If you need something better than your eyes you need a forensic image analysis and not poking around with DT…sorry to be blunt…I think you stumbled on to the perspective correction indicators and something you see when the image is inverted that makes you think you can use DT to reverse engineer or define the retouching that happened and I don’t really see that happening…good luck perhaps someone will give you a better answer…

1 Like

The reason why this request for help went so far left is (1st) I am inexperienced with forums and I took it for granted that the viewers would know that it was not the primary image (as shot) so it is my bad! (2nd) I was unable to figure out on my own what the correct process was to get through to forum viewers.

Now the image I placed online here was a result of using NegaDoctor and no originals have been placed yet. I have 95 Jpg and 95 nef’s and I know that would be acceptable to send all these pics to a forum. What I was hoping for was someone to offer to accept a jpg which is fraud and its nef… and then give me their views… But instead I get many reminding of my misgivings and such.

My initial request was to hire an expert who is knowledgable in DT because I found some that brought all of my work together.

David I hope you get a better answer. You are determined that DT is somehow useful and others are telling you it’s not. Why is it that you think there is some aspect of DT that is the magic bullet over Gimp and all the other software you said you tried. I sent you some links and some papers on the types of analysis that are done when looking in to these things and none of this is offered in DT.

1 Like

Okay, so you negated the image. But what image did you negate? Was it one of the JPEG files or a JPEG extracted from a NEF file, or did you process the raw image in the NEF to make an image that you then negated? Did you do anything else such as resizing? Did you add the rows at the top and bottom?

I’ve had an inspiration about the gray border at top and bottom…

I guess someone has used some program to display an image on their screen, and done a screen-grab, and been slightly careless about which pixels were grabbed, so the gray lines come from the program. Hey, we’ve all been there, done that.

I’ll go further, and guess that @ELA did that, using Negadoctor to show an image negated, and did a screen grab from that. And I’ll guess further, that Negadoctor didn’t display the image full size, but somewhat smaller.

Did I guess right?

Do you not find this strange that someone with a business with this description is asking these questions…

About Us

Eugene Lamb Design is a small business that creates high-end websites with flare. I am a home business with clients around Sechelt and as far away as Ontario. I specialize in web design and video editing not only for the web but also editing old videos to new digita formatsl. Other services include Photo Restoration and general Computer training (Mac Only) I set-up and network computers to work with Smart Tv’s, Apple TV and internet to name a few.

Website: eugenelambdesign.com
Well known client showcased: kathleengalleystudio.com

Then there is this website with really no content that is a year old??

https://www.eugenelambdesign.com/

offering photo restoration services…

image

Contact email at the bottom shows one thing directs to another … dyfacto.com ??

Maybe its supposed to be a sample website no Idea but this is why I was initially skeptical but tried to help …

But how does tilt shifting and exposure shifting prove the image was fake?

I’d strongly advise going by what the police are advising. By using software not designed for photo forensics, to “prove” the fraudulent image, you could find this “evidence” gets ignored / thrown out. And would that help the case any?

1 Like

Wikipedia explains the mechanics of tilt and shift, both for physical T&S lenses/cameras, and simulation in software. See Tilt–shift photography - Wikipedia

In a nutshell: tilt changes the plane of focus, where shift changes the angle of view without needing to tilt the plane of the sensor (or film) hence it changes the geometry of an image compared to tilting the camera. Shifting (compared to tilting the camera) also changes the plane of focus.

If either or both of these effects have been simulated in software, this might be readily detectable, or the effect may be very subtle. For example, if the camera was pointed upwards at a skyscraper and the photo adjusted so the verticals no longer converge, pixels at the top of the image would be stretched (blurred) horizontally.

In the “negative room” image posted, the camera was apparently pointed somewhat downwards, so lines that were vertical and parallel in the real world have become tilted in the image, converging somewhere below the bottom of the image. It is possible that the camera wasn’t really tilted, and the effect has been created in software. The photo shown is very poor quality, and proving or disproving this would be very difficult or impossible.

1 Like

So your skeptical!
Your comments seem bent on warning people away from assisting me … to the tune of placing some of my private information on this forum in the manner shown without my consent. I have thanked you for the comments which have made sense to me but that doesn’t seem to be enough. You are personally attacking a (New Member) without provocation. Perhaps ‘Stopping would be best!’

Also for those reading this response!
I just about went blind while working on my websites: (www.eugenelambdesign.com) and starting in February 2021 I had both eyes removed and replaced in June. I was unable to sit in front of a computer without damaging my sight until July of this year. It was not possible to see the shadows and lines in photos due to near blindness. I primarily work with Final Cut and produce movies for the web. It is unfortunate that some people continue in our time to be skeptical, and mistrusting of people they don’t know.

I ask publicly here on this forum that you ‘Priort’ refrain from further comments in regards to my posts on any level whatsoever. I wish you kindness and safety.

David

I think we have had enough. If you need further assistance, please start a new thread and include as many details as possible.

5 Likes