How do you think the new GigE standards will influence the machine vision industry?
Respond or ask your question now!
Silicon Software’s Noffz adds that “rapidly evolving industries require a fast adaptation of software algorithms. Typical applications can be found in solar inspection, TFT (thin film transistor) or PCB (printed circuit board) inspection. Advanced requirements triggered by new product generations need enhanced algorithms, a faster or high-resolution processing to detect smaller damages and measure very fine structures.”
All high-speed and 3D vision applications are strong, says MVTec’s Kreutzer. He specifies processing of large images, 3D matching, perspective matching, circle and rectangle pose recognition, binocular stereo and multi-grid stereo, depth from focus and photometric stereo and sheet-of-light measurement.
“Modern imaging software now processes large images of more than 32K by 32K. The size of images is not limited,” he says. “Above all, this is interesting for high-resolution line scan camera applications as deployed for print and component inspection. If this technology is combined with fast parallel processing, in spite of the high data amount, the desired real-time will be reached without outsourcing image pre-processing. Thus, programming is significantly easier and faster and the application will run robustly, reliably and accurately.”
He notes that the latest release of MVTec’s Halcon 9.0 software features improved algorithms, extended use of SSE technology, and a comprehensive utilization of multi-core and multi-processor hardware. Halcon’s automatic operator parallelization has been accelerated by more than 20 percent and now runs more than 400 operators in parallel, such as filters, matching, 3D matching, subpixel extraction, and FFT.
The continuing growth of multicore architectures and GPUs seems to be the next great expansion in software in general, and specifically in image processing software.