Halcon Embedded tested on ARM Cortex-A9

-May 12, 2011

MVTec Software has announced that its engineers have completed an initial port of Halcon 10 to the ARM Cortex-A9 processor-based PandaBoard for testing purposes. The company said it took initial performance measurements with the Halcon 10 machine-vision library running on the PandaBoard under the Ubuntu Linux 10.10 operating system. Typical Halcon image-processing algorithms like those for filtering and blob analysis can be performed on the PandaBoard in milliseconds, the company said, adding that even subpixel-precise measurements are possible in less than one millisecond. These results, the company said, show that applications based on Halcon can successfully make use of multicore ARM platforms.

MVTec said that compared with its performance on the BeagleBoard (based on Texas Instruments' processors featuring the ARM Cortex-A8 core), Halcon Embedded runs up to three times faster on the PandaBoard, because Halcon's automatic operator parallelization (AOP) capability takes full advantage of the dual-core processor--without any need for additional programming. Furthermore, the company reports, Halcon's image-acquisition interface for the LinuxTV project's Video4Linux2 is available for the PandaBoard, thus enabling the use of Video4Linux2 devices for image acquisition.


Loading comments...

Write a Comment

To comment please Log In