Advice for imaging

Nick Efford nde at comp.leeds.ac.uk
Wed Aug 22 08:47:40 EDT 2001


On Wed, 22 Aug 2001 11:17:35 +1000, Tom Harris <tomh at optiscan.com> wrote:

> I looked at the Python Imaging Library but unfortunately it does not appear
> to support 16 bit pixel images. 
> Since speed is not really an issue, I was thinking in terms of a simple
> class that is attached to an image that is able to iterate over it in
> various ways, and get/set pixel values.

Tom,

If you absolutely need to preserve 16-bit values during processing,
you should perhaps consider using Numeric.  You could write an
image class with Numeric used for internal representation of the
data.  Algorithms from the Numeric package would do some of
the image processing for you; others, you would probably have
to implement some algorithms in C for speed.  (Pixel-by-pixel
processing implemented in pure Python can be _really_ slow...)

I have an image class that wraps around PIL, with methods to
export data to or import data from a Numeric array.  Let me know
if you want to take a look...


Nick



More information about the Python-list mailing list