Analog Devices processor enables lower cost vision systems

  • 30-Apr-2012 05:41 EDT
aetrADIbfin.jpg

ADI’s latest Blackfin chips use a Pipelined Vision Processor to speed up image processing without pushing up costs.


In response to the soaring use of cameras in vehicles, Analog Devices Inc. (ADI) has unveiled a digital signal processor with a video analytics accelerator targeted at advanced driver assistance systems. The Blackfin processors can help vision systems do multiple tasks such as pedestrian detection and traffic-sign recognition.

The ADSP-BF608 and ADSP-BF609 dual-core devices, which run at 1 GHz, feature a Pipelined Vision Processor (PVP) that can accelerate up to five concurrent image algorithms. That lets a single camera run several tasks such as high-beam control, front-end crash avoidance/mitigation, lane-departure warning, pedestrian detection, and traffic-sign recognition. Handling many tasks with one device can help cut costs.

“The PVP helps us provide pricing and power consumption points that will enable wider adoption of embedded vision systems,” said Colin Duggan, Director of Product Marketing, at ADI's Processors-DSP Core Technology Group. “That’s important now as safety ratings agencies start looking at vision’s impact when they award safety ratings.”

He noted that power consumption can be as important as pricing, particularly in environments where the camera and processors are placed near the rearview mirror; it can be very hot there, and there’s no room for heat sinks. The new Blackfin devices only draw 400 milliwatts, Duggan added.

One reason for the low power consumption is that the large SRAM bank on the chip of 4.3 Mb can store incoming images so that the PVP can do a lot of image processing before data are moved to external memory.

It takes power to move data off chip, so minimizing the number of transfers can bring a significant power savings. Memory requirement must be high when high-definition video means moving 1280 x 960 pixel images at rates of 30 frames/second. The DSP family also focuses on high reliability, which is central in safety devices because false positives can be as problematic as not detecting a problem in time to alert a driver.

“Memory parity, error correction coding, and system protection for detecting faults are all important parts of safety system designs,” said Richard Murphy, Business Development Manager of ADI's Processors-DSP Core Technology Group.

Share
HTML for Linking to Page
Page URL
Grade
Rate It
4.50 Avg. Rating

Read More Articles On

2017-03-04
Euro NCAP will establish a separate category for autonomous vehicles, but there is not likely to be one for cars that are claimed to protect all occupants from serious injury or death.
2017-03-06
Unveiled at the 2017 Geneva Motor Show, this uber-performance hybrid uses F1 technology.
2017-01-11
Toyota tests ‘relationship-based driving’ along with your friendly, AI-enabled, net-connected clone companion.
2017-03-17
Experts serve cautionary tales for connected cars while discussing potential cyber solutions.

Related Items

Technical Paper / Journal Article
2010-10-19
Training / Education
2010-03-15
Training / Education
2007-03-01
Technical Paper / Journal Article
2010-10-25
Training / Education
2017-11-01