ATOM one is the smallest Full HD camera with Dual 3G-SDI output.
The camera is based on Sony’s IMX174 image sensor that provides exceptional image performance with multi matrix support.
ASIC, FPGA and SoC development
System on Chip (SoC), FPGA and embedded software designs for consumer electronics, telecommunications and automotive industries
ASIC, FPGA und SoC Entwicklung
Ihr deutsches Systemhaus für die Chip und FPGA Entwicklung im deutschspraching Raum. Embedded Software Entwicklung und Leiterkartenentwicklung auf höchstem Niveau.
Cutting-edge Image Processing for ADAS
Experience the future of driving with our state-of-the-art components for automotive safety!
Bildverarbeitung für Fahrerassistenzsysteme
Algorithmen und Komponenten für das autonome Fahren. Embedded Software Entwicklung nach ISO 26262. Autosar Entwicklung auf Infineon Aurix Bausteinen.
Arria 10 System on Module
The Arria 10 SOM is an Arria 10 SoC System on Module with an emphasis on embedded and automotive vision applications. Using Alteras Arria 10 SoC Devices in the 29x29 mm package, the module off ers a multitude of interfaces in a small 10 cm by 6 cm form factor.
Dream Chip Technologies (DCT) will participate with several sessions and life demonstrations with the recently announced 22nm FDSOI ADAS Chip.
The key sessions are:
- A New Computer Vision Processor Chip Design for Automotive ADAS CNN Applications in 22nm FDSOI
- Using the Cadence Tool Chain and Vision P6 Processors to Design a 22nm FDSOI Automotive ADAS CNN SoC
- Object Detection for Mobile and Automotive - Convolutional Neural Networks (CNNs) on Tensilica Vision DSPs
The presentations explores an European collaboration on a 22nm System-on-Chip development for Advanced Driver Assistance System (ADAS) in FD-SOI technology
It will highlight partner co-operation and key trade-offs necessary to deliver the target architecture and performance. The scope includes:
- Project setup and target applications in the automotive market
- Chip architecture and performance achievements, e.g. for Convolutional Neural Networks (CNN)
- Physical chip presentation and reference platform for evaluation
Dream Chip Technologies, based in Garbsen and Hamburg, Germany designed together with its partners Cadence, Global Foundries, ARM, Arteris, Invecas, Fraunhofer and Mentor Graphics a the first 22nm MPSoC design on the brand new 22nm FD-SOI technology from Global Foundries. The SoC is based on a Multi-Processor Architecture containing four cores of the recently released Vision P6 Tensilica VLIW processor, an ARM Cortex A53 quad and an ARM Cortex R5 lock step architecture for functional safety.
The design is targeted on Advanced Driver Assistance Systems (ADAS) requiring a massive image and video processing performance as well as on Convolutional Neural Network (CNN) applications in the automotive ADAS market. Typical applications for such an architecture are high quality 360° Top View, Traffic Sign recognition or pedestrian detection.
Based on an analysis of these applications, all data plane tasks have been divided into tasks on a highly vectorizable pixel level and on tasks on scalar (or small vector) object level.