resampling, blurring, duffing...just a few terms used in connection with hiding
/masking/censoring information from satellite and aerial imagery. The
end product (apart from frustration), is data which has selective areas at an
abysmally low resolution, hazy patches or worst still zero-fill. If you have (without
checking the preview image) ordered data over a city that houses a number of
'important' and 'strategic' locations, be prepared to admire an image which
reminds you of a badly made chess board - the result of zero-fill. A few years
ago, I procured a high-resolution image covering the entire Mumbai city and was
surprised to see two rectangular areas with zero-fill. In my knowledge, these two
locations had nothing of importance and felt challenged to figure
out what was being masked. Having lived all my life in this city, it hurt
my 'geo-ego' and I did find out what was 'protected'. The entire idealogy of
masking resulted in 'painting a bulls eye on the roof' or could also
mean a 'placemark for a terrorist'.
The recent bill (
AB 255) proposed by a South California assemblyman, if
passed, will make it state law for online mapping services to blur images of
designated "sensitive" locations, whether the pictures were taken at street level,
from the air, or by satellites. We urgently need to address the 'issue' of what
level of detail can be considered to "pose a serious security risk"? - or will it just
mean that the 'highest resolution' data available at that instant will always be
considered '...a security risk...' , just the way 5 m resolution data was considered
a decade ago ?
Incidentally duffing actually means converting something to no-good