Toshiba Develops DNN Hardware IP for Image Recognition AI Processor ViscontiTM5 for Automotive Driver Assistance Systems
Toshiba Electronic Devices & Storage Corporation (“Toshiba”) announced the development of Deep Neural Network (DNN ) hardware IP that will help to realize advanced driver assistance systems (ADAS) and autonomous driving functions. The company will integrate the DNN hardware IP with conventional image processing technology and start sample shipments of ViscontiTM5, the next generation of Toshiba’s image-recognition processor, in September 2019.
The DNN hardware IP draws on deep learning to deliver more accurate detection and identification of a wider range of objects than image recognition based on conventional pattern recognition and machine learning. It enables ViscontiTM5 to recognize road traffic signs and road situations at high speed with low power consumption.
Read More: Jumpstart 2019: Interview with Anand Janefalkar, Founder & CEO at UJET
Toshiba will promote ViscontiTM5 equipped with DNN hardware IP as a key component of next-generation advanced driver assistance systems.
Advanced driver assistance systems such as autonomous emergency braking are now being widely adopted, from luxury cars to subcompacts. They are also expected to offer increasingly advanced capabilities – for instance, the 2020 version of the influential European New Car Assessment Programme (Euro NCAP), the EU-backed safety standard, adds testing to avoid collisions at intersections. This trend will increase the demand for more advanced and capable systems.
Read More: 2018 Was a Year of ‘Disruptive HR Transitions’: Tjinsite’s Yearend Conclusion
Toshiba Electronic Devices & Storage Corporation will continue to develop the Visconti™ family and contribute to traffic safety.
Outline of ViscontiTM5 |
||
Product Series Name | TMPV770 series | |
CPU core | Arm® Cortex®-A53 | |
Arm® Cortex®-R4 processor with Floating Point Unit | ||
Image processing DSP | General DSP | |
Image Processing Accelerator | Affine conversion | |
Pyramid Image Generator | ||
Enhanced CoHOG Feature-based Support Vector Machine | ||
Dense Optical Flow | ||
Template Matching | ||
Dense Stereo Matching | ||
Deep Neural Network | ||
Image Signal Processor | ||
Video input interface | MIPI CSI-2 RX | |
Video output interface | MIPI CSI-2 TX | |
Notes
[1] DNN is a machine model using artificial deep neural networks inspired by human brain function.
[2] IP (Intellectual Property) is a function block for SoC development.
* ViscontiTM is a trademark of Toshiba Electronic Device & Storage Corporation
* Arm and Cortex are registered trademarks of Arm Limited (or one of its subsidiaries) in the United States or other countries.
* MIPI is a registered trademark of MIPI Alliance.
* All other company names, product names and service names may be trademarks of their respective companies.
* Information in this document, including product prices and specifications, content of services and contact information, is current on the date of the announcement but is subject to change without prior notice.
Read More: Interview with Megan Kvamme, CEO at FactGem
Copper scrap buying Scrap Copper trading Environmentally friendly scrap metal trading
Copper cable scrap export regulations, Metal scrap sorting, Copper scrap supply
Scrap metal reclamation depot Construction ferrous metal recycling Scrap iron resale
Ferrous material supply chain, Iron scrap brokerage, Metal trade insurance