We talked a little about the revolutionary power of robotic machine vision cameras in a previous post about stock car racing. Without high-performance hood-mounted cameras that adjust themselves in nearly instantaneous fashion, millions of NASCAR and IndyCar fans around the country would have to content themselves with stationary shots from sluggish aerial cams. The TV networks that rely on these high-action shots to keep fans interested owe machine vision developers a serious debt of gratitude.It's an exciting time for the machine vision industry. Although car-mounted cameras continue to offer the most dramatic illustration of this technology's power, the landscape has undergone radical change over the past few years. New smart cameras boast processors that rival those of modern PCs in speed and processing power. Meanwhile, an ever more efficient design framework has measurably reduced their size and cost while improving their adaptability and increasing the number of functions that they can perform.
In other words, smart cameras are more powerful and user-friendly than ever before. It shouldn't be surprising that they're quickly becoming the preferred machine vision platform for industrial concerns that demand adaptable, cost-effective vision technologies.
A recent Markets and Markets research report highlights the segment's rapid growth. Out of the three major machine vision setups, smart cameras grew at a far faster rate than either traditional PC-based systems or embedded systems. The report predicted that smart cams would comprise more than one-third of the entire machine vision market by 2018. If projected sales growth rates hold, that's good enough for a $1.75 billion share of a $5 billion-plus market. It's worth remembering that the segment accounted for little more than a rounding error as recently as the middle of the last decade.
While embedded machine vision systems will continue to prove useful in certain highly specialized industrial and non-industrial applications, PC-based systems will have limited relevance in a landscape that's dominated by powerful, user-friendly and ever more compact smart cameras. Why settle for a gas-guzzling diesel station wagon when a gas-sipping, functionally equivalent hybrid hatchback will do?
Modern smart cameras cut out the PC middleman and process images on-site. Although they're network-enabled for distribution purposes, all of the requisite processing equipment sits within their housing.
Sony's groundbreaking XCI-SX1 camera captures images in a high-resolution display format of 1280 x 1024 pixels and lets its users toggle between frame rates of 15 frames per second or 24 frames per second. The built-in processor uses x86 chips that run on Linux or Windows XPe operating systems, and its customized 258 megabyte SDRAM card connects to the outside world via Ethernet and standard USB ports. Simply put, this camera has the ability to simultaneously capture and process high-resolution images while maintaining the user-friendly interface of a high-performance, Internet-enabled PC. It's no wonder that machine vision users from Shenzhen to Cleveland are switching to smart cameras en masse.
In fact, Smart cameras are already employed in a bewildering array of fields and look certain to be tapped for additional uses. In addition to quality assurance, product sorting and rudimentary robot guidance applications, machine vision smart cameras are certain to play a growing role in more advanced applications like biometrics, continuous-part inspection and remote measurement. PC-based machine vision systems might have some nostalgic value, but the all-in-one appeal of smart cameras may soon prove too powerful to ignore.