jfake

Python CLI tool to visually detect photoshopped pictures using ELA

usage:

jfake.py [-h] --input INPUT [--output OUTPUT] [--verbose] [--debug] [--quality QUALITY] [--multiplier MULTIPLIER] [--benchmark] [--entropy] [--psnr] [--numba]

Required arguments:

--input STR, -i STR Input image file [BMP], [GIF], [JPEG], [PNG], [PPM], [TIFF]

Optional arguments:

-h, --help show this help message and exit
--output PATH, -o PATH Define output folder jfake (default: output)
--verbose, -v Write all steps to terminal (default: False)
--debug, -d Write all steps to output folder (default: False)
--quality INT, -q INT JPEG-Quality [1-99] jfake (default: 50)
--multiplier INT, -m INT Multiplier jfake (default: Automatic)
--benchmark, -b Write needed time per step in file (default: False)
--entropy, -e Calculate entropy in each processing step (default: False)
--psnr, -p Calculate signal-to-noise-ratio (PSNR) (default: False)
--numba, -n Use numba jit compiler for better performance (default: False)

Examples:

jfake.py -i lenna.png
jfake.py --input lenna.png --entropy
jfake.py -i lenna.png -o lenna -vdbepn