gestaltexp
This script reproduces the partially occluded pictures described in Doniger et al, 2000, Journal of Cognitive Neuroscience, "Activation timecourse of ventral visual stream processes high density mapping of perceptual closure processes." As per the author's specification, the input images should be 256x256. The image is divided in a 16x16 grid. A (1 - 0.7^(LEVEL-1)) share of the grid elements that are not totally white is then occluded with white (as the sample image appears to have a white background), and the results are stored in BMP files.
Prerequisites:
- Python 2.7 (direct downlaod for windows 32 bit: http://www.python.org/ftp/python/2.7.3/python-2.7.3.msi)
- The Python Imaging Library (PIL) for Python 2.7 (direct download for windows: http://effbot.org/downloads/PIL-1.1.7.win32-py2.7.exe)
In this repo:
-
the file elephant_256x256_clean.bmp, which is a sample input from the original paper.
-
the script "script.py", to be run from your operating system's command-line. Usage information (also available by running script.py -h):
Usage: script.py [-l LEVEL] [-o OUTPUT] filename
Options: -h, --help show this help message and exit -l LEVEL, --level=LEVEL desired fragmentation level between 1 and 8 -o OUTPUT, --output=OUTPUT base output file name
Example:
python2 script.py elephant_256x256_clean.bmp
I tested the script on Windows 7 64bit, Windows XP 32bit, and Linux 32 bit, all worked as expected. I do not expect any difficulty on a Mac either. However, if you find a bug, please do tell me.