By Brian Dipert
Embedded Vision Alliance
This article was originally published in the September 2011 issue of the INSPECT Buyer's Guide. It is reprinted here with the permission of INSPECT.
Embedded vision technology has the potential to enable electronic products to be more intelligent and responsive, so that they are more valuable to users. It can enable electronic equipment companies to both create valuable new products and add helpful features to existing products. And it can provide significant new markets for hardware, semiconductor and software manufacturers. A unified worldwide alliance of suppliers, system developers and end customers will help transform this potential into reality in a more rapid and efficient manner.
As computer vision technology developers and users, you're already well aware 0f the vibrant potential of this technology and have developed extensive expertise in applying it not only in the lab but also in far more challenging real-life operating environments. Due in no small part to your efforts over the years, the technology required to implement embedded vision is becoming less expensive, and has now crossed a threshold enabling widespread deployment. As such, you're likely excited about the potential for latest-generation chips, sensors, and algorithms to not only make dramatic performance improvements in your applications but also to significantly reduce the cost of creating and manufacturing those designs. And you might even be considering leveraging your technology proficiency in targeting new products for emerging high volume markets.
From High-End to Mainstream
As embedded computer vision technology becomes less expensive, it is proliferating rapidly into a wide range of applications and markets. Odds are high, for example, that the cellular handset in your pocket and the tablet computer in your satchel contain at least one rear-mounted image sensor for photography (perhaps two for 3D image capture capabilities) and/or a front-mounted camera for video chat support (FIgure 1). The same goes for your laptop computer, where bezel-located image sensors are now pervasive and are increasingly also installed in all-in-one desktop computers and standalone displays. Driver-assistance cameras have been resident in high-end cars for several years now, and are now migrating downward into higher-volume mainstream models, along with increasing the number of cameras per car to fill in blind spots, assist in parking and other maneuvers, and provide early warning of impending collisions and other looming catastrophes.
FIgure 1. The embedded vision ecosystem spans hardware, semiconductor and software component suppliers, subsystem developers, systems integrators and end users, along with the fundamental academic research that provides ongoing implementation breakthroughs. A unified worldwide alliance will enable the ecosystem to thrive to the richest possible degree.
Digital still and video cameras have displaced their analog precursors, with latest-generation models going beyond simple image capture and processing functions to incorporate more advanced analysis-and-response features such as face-detection-driven focus and exposure compensation. Advanced cameras will even delay the shutter activation until they discern that the subject is smiling. As such, they not only "see" but also are beginning to "understand" the environments in which they operate. Other consumer electronics systems are also becoming vision-augmented if not vision-centric, with Microsoft's Kinect peripheral for the Xbox 360 game console leading the charge. Medical systems are increasingly supplementing human intelligence with computer vision-fed algorithm analysis to assist in patient diagnosis and treatment. And the ability to assess and react to a subject's emotional state is not only of interest to physicians; imagine the interest in such a capability to a toy manufacturer, for example, or to a retailer.
Open Up New Markets by Decreasing Costs
Embedded computer vision relies on a diverse range of technology. For example, driven by expanding and evolving application demands, image sensors are making notable improvements in key attributes such as resolution, low light performance, frame rate, size, power consumption and cost (Figure 2). Similarly, embedded vision applications require processors with high performance, low prices, low power consumption, and flexible programmability, all ideal attributes that are increasingly becoming a reality in numerous implementation forms: microprocessors and embedded controllers, application-tailored SoCs, DSPs, graphics processors and FPGAs. Similar benefits are being accrued by latest-generation optics systems, lighting modules, volatile and nonvolatile memories, and I/O standards. And algorithms are increasingly up to the challenge, leveraging these hardware improvements to deliver more robust and reliable analysis results.
Figure 2. Computer vision got its start in traditional embedded applications such as assembly line inspection, optical character recognition, robotics, surveillance and military systems. In recent years, however, decreasing costs and increasing capabilities have broadened and accelerated its penetration into key high volume consumer electronics markets.
How Do You Ensure Success in Both Established and New Markets?
The consequent increase in (and improvement in) suppliers, products from those suppliers, and volume shipments of those products is good news for those of you in traditional embedded vision applications such as factory inspection, optical character recognition, robotics, surveillance and military systems. You're able to leverage more cost effective building blocks, hardware, semiconductor and software alike from a burgeoning list of vendors. And you're also able to leverage your existing expertise to target new markets for the technologies you've developed and refined over the years. But how can you identify, track and cultivate effective relationships with the ever-expanding list of suppliers? Similarly, how can you make your company visible to customers in new markets that you may not yet realize exist? And how can you keep abreast of developments elsewhere in the supply chain and distinguish yourself as a leader in your product space?
The Intention of the Embedded Vision Alliance
A unified worldwide forum addressing the full range of embedded vision technology and applications would be the ideal means to regularly assemble ecosystem participants spanning all points of the spectrum. This, in summary, is the intention of the Embedded Vision Alliance, spearheaded by BDTI (Berkeley Design Technology, Inc.) and publicly launched in May 2011. By both formally and informally coordinating the activities of Alliance partners, the Alliance aspires to dramatically accelerate the adoption and progression of embedded vision technology for the benefit of all market participants. The Alliance has already developed a robust website presence, freely accessible to all and including (among other things) technical articles, video demonstrations, and a multi-subject discussion forum staffed by a diversity of technology experts. But the Alliance's vision is much more expansive, including:
- Comprehensive education facilities that will enable new participants in the embedded vision application space to rapidly ramp up their expertise;
- Creating links between research and industry to speed the commercialization of technology breakthroughs;
- The codification of hardware, semiconductor and software standards that will accelerate technology adoption by eliminating the confusion and inefficiency of numerous redundant implementation alternatives; and
- The creation of reference designs and other aids that will streamline the development of embedded vision products.