 |
|
Parallel HALCON
- What it is:
HALCON can be used in two modes: Standard HALCON is optimized for
single-core computers, Parallel HALCON actively exploits the
speed of multi-processor and multi-core computers.
- Automatic Operator Parallelization (AOP):
Parallel HALCON automatically parallelizes operators if started on a
multi-core computer, by distributing the data, e.g., the images,
to multiple threads, one for each core. You don't even need to
modify your existing HALCON programs to make use of this automatic
parallelization and gain a significant speedup.
 |
When executing the operator sobel_amp on a quad-core computer,
Parallel HALCON splits the image into four parts, which are then
processed in parallel by four threads executing the operator.
|
|
* 37.5 / 19.6 / 13.6 / 10.7 / 8.9 / 8.0 / 7.1 /
6.3 ms on 2x Intel Xeon QuadCore 2.33GHz, image size 1300 ×
1300. Note that the reachable speedup depends on the used HALCON
operator and the image size.
|
|
- Parallel Programming:
You may exploit latest multi-core technology by using Parallel HALCON.
Parallel HALCON supports parallel programming, e.g. multithreaded
programs, by being not only thread-safe but also
reentrant. Thus, multiple threads can call HALCON operators
simultaneously. Using this feature, you can split a machine vision
application into independent parts and let them run in parallel on
different cores.
 |
On a 2-processor computer, different electronic parts in
the image are extraced in parallel by two threads.
|
|
