Deep learning frame grabber from Silicon Software to be showcased at VISION 2018

At VISION 2018 in Stuttgart, Germany, Silicon Software will showcase its deep VCL Camera Link frame grabber, which is a microEnable 5 VCLx frame grabber with a convolutional neutral network (CNN) runtime license for VisualApplets. The frame grabber features an input bandwidth of 850 MB/s, support for all Camera Link formats, and a PCI Express x4 bus interface.

Using graphical FPGA (Xilinx Kintex) programming with VisualApplets, suitable network architectures of different sizes and complexity can be integrated and pre-trained configuration parameters for the weights of the networks can be loaded for a variety of image processing applications, according to the company. Silicon Software also notes that the deep VCL board has a high-prediction accuracy of results above 98% and is more energy efficient and faster than a comparable platform with an industrial GPU.

VISION 2018 booth number: Hall 1, Stand 1C72

To Learn More:

Contact: Silicon Software
Headquarters: Manheimm, Germany
Product:
Key Features: microEnable 5 VCLx Camera Link frame grabber with a convolutional neutral network (CNN) runtime license for VisualApplets.

What Silicon Software says:
View a blog post on the frame grabber.
View more information on the frame grabber.

View More Products| Locate a vendor or system integrator | Receive e-mail updates

Share your vision-related news by contacting James Carroll, Senior Web Editor, Vision Systems Design

To receive news like this in your inbox, click here.

Join our LinkedIn group | Like us on Facebook | Follow us on Twitter