|
Tech Focus: The future of embedded vision
HIGHLIGHTS
My vision for embedded vision
Building mobile and embedded consumer devices that can "see"
Creating video that mimics human visual perception
Editor's Note
The Embedded Vision Alliance (EVA) is a new group formed with the objective of taking 30 years of signal processing hardware and algorithm knowledge and using it to transform the way we interact with our mobile, personal and consumer computing devices. Its members include the likes of Analog Devices, CEVA, Freescale, and Texas Instruments.
According to Jeff Biers, founder of the group, the aim is the creation of consumer products that not only allow us to view content better, but to allow devices to “see” us, recognize our touch or voice and respond appropriately through new types of user interfaces that combine gesture, voice, and face recognition.
Other vision applications that EVA wants to make a part of the mainstream of consumer computing include optical character recognition, pattern recognition, visual search, augmented reality, and creating 3-D models from collections of photographs, as well as more subtle tasks such as measuring respiration and heart rate from relatively low resolution videos. For more about what Jeff Biers thinks about the challenges read "My vision for embedded vision," and view the video with Biers at DAC recently.
The articles selected for this Embedded Tech Focus on embedded vision technology illustrate that a wealth of DSP technology and algorithm knowledge already exists, developed originally for industrial, automotive, medical and military/aerospace designs. My Editor’s Top Picks include:
Building mobile & embedded consumer devices that can “see”
Dealing with the minutiae of fingerprint analysis
Sorting image data in two clock cycles
Video cameras and machine vision
The challenge facing developers of embedded vision designs will be to translate this knowledge base into applications that meet the demanding cost, size, and power constraints of consumer and mobile devices. I look forward to hearing from you about these challenges and how you are addressing them in the form of design articles for use on Embedded.com and in ESD Magazine.
Design How Tos
Building mobile and embedded consumer devices that can "see"
Eldad Melamed of CEVA provides some general guidelines for developing signal processing algorithms that will allow the use of real-time face detection applications on any mobile device.
Dealing with the minutiae of fingerprint analysis
Fingerprint analysis involves using fingerprint sensors and sophisticated DSP-based image-processing algorithms. Here's a description of the minutiae-based system for fingerprint analysis.
Sorting data in two clock cycles
Image recognition can be a tricky and math-intensive problem. This research uncovered a way to do much of the work using an FPGA ... and just the right sort of algorithm.
Video Cameras and Machine Vision: A Technology Overview
Video technology is playing an increasing role in a number of real-time systems. While many early systems were tied in to standards developed for the broadcast industry, video is now moving out on its own. One of the areas where this is most evident is in machine vision, where systems today use frame sizes and frame rates totally distinct from common broadcast industry standards. This article gives an overview of standard and non-standard video cameras.
Picking the right MPSoC-based video architecture: Part 1
A look at the design of multiprocessor systems-on-chips (MPSoCs) for video applications and how to optimize them for computational power and real-time performance as well as flexibility.
Taking a multicore DSP approach to medical ultrasound beamforming
A Product How-to on using Freescale MSC8156 multicore DSP to produce diagnostically useful medical ultrasound imaging results, using no more than about 38% of the resources of the DSP, leaving enough room to also perform Doppler imaging.
Implementation of A 2D Real-Time Correlator Based on a DSP
Image correlation plays an important roll in vision systems and it can be used for autonomous navigation, tracking systems, detection of environment changes, and automatic pattern recognition. The correlation between images must usually be calculated in a very short period of time to be integrated in real-world applications. In this project, the operation of a 2D correlator is tested by implementing a pattern recognition system based on the real-time location of a vehicle model moving along a square-shaped area.
Fingerprint identification with Blackfin
This article explains the basics of biometrics, and shows how to create an fingerprint identification system with the Blackfin processor and Atmel's FingerChip sensor.
Application dictates union between robotics and machine vision
Pairing robotics with machine vision marries industrial muscle to electronic intelligence, creating one of the most powerful automation solutions available to industry today.
Automated Visual Inspection - Part 1: Introduction and Review of Object Types
An excerpt from the book "Machine Vision: Theory, Algorithms, Practicalities" begins with a look at the process of inspection and the various types of objects that visual inspection systems have to cope with.
Camera design for machine vision
The fundamental components of machine vision and a close look at camera requirements, including image sensor technology, architecture requirements, interface standards, resolutions, frame rates and signal processing.
Parallel processing advances vision applications in cars
Safety tops the list of key concerns in automotive design. But, while they primarily focus on keeping drivers and passengers safe, manufacturers also look for ways to protect people outside the car--such as bicyclists and pedestrians. Accident-avoidance equipment, seen in applications such as vision-based warning systems, could provide the ultimate solution for protecting lives both inside and outside of vehicles.
Vision Guided Robotics
Vision and robotics systems are able to be programmed within one application with LabVIEW and the ImagingLab Robotics Library for DENSO.
Automotive vision system recognizes road signs: Part 1 - Basic functions
Pattern recognition and processors deal with computational and I/O challenges of handling video data stream.
Embedded Systems Bookshelf
Excerpts
Embedded Books Reading Room
Bernard Cole's favorite links to book excerpts.
Reviews
Engineer's Bookshelf
Airport fiction blows. A look at books other engineers are reading and why you should read them, too. Recommend and write a review yourself. E-mail Brian Fuller.
Jack Ganssle's Bookshelf
A list of book reviews by Jack Ganssle, contributing technical editor of Embedded Systems Design and Embedded.com.
Max's Cool Beans
Clive "Max" Maxfield, the editor on Programmable Logic DesignLine, often writes about interesting books.
Products
Renesas : Automotive dual-core SoC includes image recognition
Renesas Technology has announced the SH7776 (SH-Navi3), a dual-core system-on-chip (SoC) device with on-chip enhanced graphics functions and a high-performance image recognition processing function for the next generation high-performance car information terminals.
SoC enhances image recognition algorithms
Developed for biometric applications, Pixelplus Co. Ltd. unveiled the PM1002 SoC that integrates a 32-bit RISC microprocessor and on-chip memories to enhance image recognition algorithms.
Face recognition SDK with high accuracy for security and access control systems
Luxand Face SDK enables highly accurate recognition from a digital photograph or a video frame.
TI DSP-based dev kit puts fingerprint functions into portables
Texas Instruments recently launched the TMS320C5515 Fingerprint Development Kit, enabling designers to quickly and easily integrate fingerprint biometric features in portable applications.
|
