Comments (13)
Please send me the image and i will look into it.
from forensic.
I have tried with the forgery threshold (-ft
) set to 430 and it's not showing as forged. This flag is used to set a forgery threshold above an image is considered forged if forged blocks are detected. On images with very identical parts it's better to increase this value to reduce the false positive detection rates.
Generate: 135894 / 135894 [=====================================================================================================================================================================] 100.00% 0s
Analyze: 1223045 / 1223045 [====================================================================================================================================================================] 100.00% 0s
Detect: 16765 / 16765 [=========================================================================================================================================================================] 100.00% 0s
Filter: 16692 / 16693 [=========================================================================================================================================================================] 99.99% 0s
Number of forged blocks detected: 0
100% the image is NOT forged!
Done in: 1.90s
from forensic.
This app is capable of recognizing identical image parts, which does not means that is capable recognizing any kind of forged images. So it depends of the use case you wish to use. Also as it is mentioned in the documentation does not guarantee 100% accuracy.
from forensic.
This is an id card i've forged by copy-moving parts.
i've tried the app with several such images with little results (many false positives, results seem pretty random). Wondering if the app is just not accurate with documents or if I could improve the results with proper settings.
from forensic.
Sorry but the provided link is broken. You have to play with the settings to get a better accuracy. The accuracy depends on the image composition, the forged region size etc.
from forensic.
ok thanks
from forensic.
the link does work on my end when copy pasted in the url bar
from forensic.
Running the library over this image with the following settings is showing that is forged.
$ forensic -in ~/Desktop/sample-id-card.jpg -out ~/Desktop/out.png -bs=6`.
Number of forged blocks detected: 34
100% the image is forged!
The more condensed/intense (read as red) a detected forged region is, the bigger is the probability the image is forged, but this image in particular is not really suitable for forge detection.
from forensic.
Yes the app recognizes the image is forged, but this same goes for non-forged similar documents 100% the image is forged!
and the number / ratio of blocks detected as forged does not differ by much.
from forensic.
This is a similar document (not forged) and the corresponding heat map.
https://ibb.co/ZVSy9vd
https://ibb.co/LYX1Ymf
Number of forged blocks detected: 1514
87% the image is forged!
I took the pic myself the image is genuine.
from forensic.
Nice! I'll take this direction, make a few test with different images and revert to you here with the results!
Thanks for the guidance
from forensic.
@TheDApact any update on your test?
from forensic.
@PierreMarieRiviere What happened with the above?
from forensic.
Related Issues (2)
- Running code on Windows HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from forensic.