| Re: Wanted: Hough transform [message #3776] |
Fri, 10 March 1995 15:03 |
sterner
Messages: 106
Registered: February 1991
|
Senior Member |
|
|
Wayne E. King <weking@llnl.gov> writes:
> I seek a high speed Hough transform algorithm for detection of lines in
> images. IDL code preferred.
My IDL library has a routine called radon which I made to do exactly
what you state above. The built-in help for this routine is:
IDL> help,radon(/help)
Compute the Radon Transform using the FFT method.
t = radon(img)
img = input image. Must be square. in
t = Radon Transform of img. out
Keywords:
EMBED=n size of zero image to embed given image in.
Def=no embed.
START=a1 start angle in degrees, default=0.
STOP=a2 stop angle in degrees, default=179.
STEP=da angle step in degrees, default=1.
ANGLES=ang returned list of angles used.
/DEBUG does a debug stop.
Notes: Images must be byte.
No preprocessing is done.
It may be useful to subtract the mean.
Ref: Linear feature detection and enhancement in noisy
images via the Radon transform,
Lesley M. Murphy, Patt. Rec. Letters 4 (1986) 279.
Make sure you use the EMBED keyword to embed your image in a
larger image of zeros, it greatly improves the transform. I use this
routine to successfully detect linear features in ocean images.
I only did two cases. You are on your own, I can do much support
for this routine, other than make it available.
The JHU/APL/S1R IDL library is available by anonymous ftp as follows.
ftp fermi.jhuapl.edu
login: anonymous
password: enter your email address
cd pub/idl
get README
bye
Follow the instructions in the README (~6.6k bytes) file to get the
actual library. You may also
want to get the one line description file cat.one (~30kb).
Ray Sterner sterner@tesla.jhuapl.edu
The Johns Hopkins University North latitude 39.16 degrees.
Applied Physics Laboratory West longitude 76.90 degrees.
Laurel, MD 20723-6099
WWW Home page: ftp://fermi.jhuapl.edu/www/s1r/people/res/res.html
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
