Today, digital fraud has increased in proportion to the increase in the digital world in our lives. In this world of technology, fraudsters; both in digital evidence to distort scientific facts, intellectuals, artists, politicians, etc. they aim to make psychological guidance on societies by changing their pictures. Because we know that today, wars are mostly not with weapons, but with psychological warfare methods, as in the past, and because we know that insi-devils who use this job for their own benefit, for the evil of humanity, mostly do it with digital fraud; With this training, it is obvious how important it is for volunteers of truth (rights) to reach the truth, reveal it, and prevent fraud, lies, slander, chaos and evil from being victorious.
First of all, since the subject of digital crimes is a new field, crimes with clear boundaries are not clear in the United States. In Turkey, since even the existence of the justice mechanism is not clear, it is almost an educational site that explains the subject properly and clarifies everything, let alone a study in this field. It doesn’t even have a video.
This need for us actually arose while we were preparing the WORLD’S BLACK BOX Documentary series. While preparing this documentary series in order to separate the lies from the truth and show people the big picture that is compatible with each other, we have trained ourselves in this field to understand whether many of the images we come across are fake or real, and now we say why only we should analyze the truth. There are many people who are right-friendly and truth-lovers, and volunteers on this subject should learn and research this subject themselves, so they can distinguish the truth from the lie. If you want to learn the authenticity of a picture for your own benefit, you can of course benefit from our article. Maybe someone set you up with photoshop and you want to prove it…
First of all, people who understand this training from photoshop will understand it much easier, but do not be afraid, if you do not know photoshop, you will understand it because we explain its logic.
This is the free program we use to analyze images;
https://29a.ch/photo-forensics/#forensic-magnifier
Actually, http://fotoforensics.com/ site is based on the same subject that we will explain in this article, ELA (error level analysis), but since the first site (29a.ch-forensically beta) program is a more detailed program, we will continue through this program.
The parts analyzed in the picture are as follows.
>The “clone detection” section in the program shows the clones made with Photoshop. The red stripes show where it was taken, where it was taken and pasted.
>”ELA error level analysis” shows cut-paste operations. If it is taken from another picture and pasted on that picture, it will usually be glossy white. If the edges of the glued or played part are heterogeneous, it is understood that way.
>”Luminance gradient” shows the parts where the light hits and reflects. in this way; Since the angle of receiving and reflecting the light of a picture that is added-pasted later on the picture is different, it is understood that there is a play in the picture and in which part there is a play.
For now, we will only touch on the ELA analysis, which will be enough to learn the ELA analysis for you to understand whether a picture is real or fake. It is very rare to have an image that cannot be identified as fake from the ELA analysis. The existence of a ptohoshop artist of this skill is very rare, according to scientists writing a thesis on the subject. In order not to be understood from the ELA analysis, an image must be a very small size or a screenshot taken from a video. In other words, when you stop the godzilla movie, copy the screenshot and paste it into photoshop and get a new picture; It will not be understood from the ELA analysis that godzilla is fake.
Simply the ELA analysis shows the intensity of the image and is a black and white result as you will see plenty of examples below. If an image has close and homogeneously distributed colors as a result of ELA, the image is most likely real. However, if there is dark white-dark blackness or a different color tone gathered in a certain region, that region has been changed.
This picture is a picture taken with my personal camera and uploaded here (from my hometown) without being exposed to any image program. Below you will see the ELA analysis image I got from the “photoforensically beta” site of this unplayed image. We drag the picture to the site and click on “ERROR LEVEL ANALYZ” from the tabs on the right. It already shows us the result from the automatic settings.
In the first picture above, you can see how normal the ELA analysis is and how the colors are compatible. Now, let’s save this picture once again with Photoshop without playing and get an ELA analysis. Normally, according to experts, even if Photoshop has never been played with, when it is recorded 60-odd times in a row, the colors will change, so the picture will be very dark white rather than dark black. In the picture below, you see the ELA analysis of the first record;
I did not put it because there was no visible change in the original of the picture. There is almost no change in the ELA analysis. Now (below) I cut and add a “transparent arrow” image and an “airplane image” from another image with care (below) and I get an ELA result.
In the ELA analysis you see the different and intense hue around the Arrow mark and the different intense hue on the plane. Here you understand whether a picture is photosoped. ELA settings we use in the program (ELA analysis site); JPEG quality:75 is Error scale:20. This is the most common setting. You can increase or decrease these settings so that you can clearly see the problem. If you notice a problem at any level (in quality and error scale settings), it means that the picture has been changed.
Now (below) I’m copying 2 more from the plane and adding it to the picture and getting an ELA result.
