Wednesday, December 10, 2008

A bad case of the blues

After leaving my trusty Canon digital camera in the back of a cab while attending a tradeshow in Las Vegas, I was forced to purchase another. Upon visiting the local Best Buy with the name and model number of the camera that I had inadvertently mislaid, I was informed by the salesman that the camera I had previously purchased was no longer in production. My lost 6-Mpixel camera had been replaced by another, less expensive model, with a lot more features and an 8-Mpixel sensor.


Since I hoped my publishing company would pay for the camera, I decided to purchase said 8-Mpixel product and surprise them with the savings I had made. After using the camera to take numerous images of my son's recent graduation, however, I developed a bad case of the blues. Actually, it was a bad case of the lack of blues. Specifically, the blue response of the sensor was not comparable to that of the previous camera. Luckily for consumers like me, such effects can be compensated by some digital trickery embedded in Adobe Photoshop.


While some planned obsolescence by camera manufacturers may be good for the consumer, for system integrators, theseeffects on industrial cameras, frame grabbers, lighting, and other types of automation products can be much more serious. Imagine, for example, you have developed a machine-vision system designed to examine the color of potato chips and then automatically sort them as either good or bad. From the numerous types of camera vendors, you choose an RGB camera with a simple FireWire interface, interface it to a PC, and write some code to perform the color analysis. Based on this analysis, the PC then initiates a PLC and a remote actuator to sort the chips.


After a year of the machine working perfectly, vibration effects cause the camera to break, requiring a replacement. To yourhorror, you find that the upgraded, lower-cost replacement you install rejects every chip as bad. Needless to say, capturing each image and adjusting it manually in Adobe Photoshop is simply out of the question!


After hours of frustration and wasted production, it's time to call the camera company, where you are informed that it reallyisn't their fault at all. Rather, the blame lies with the CCD vendor that has upgraded its CCD line, making obsolete the original CCD and thus the camera vendor's original camera line.


To shift everyone's share of the blame, the camera vendor suggests that you remap your color space model to compensate for the new RGB spectral curves of the camera and everything should work as before. Not knowing anything about what software you originally used in the application, however, this is all the advice that you receive. What then follows can then only be mitigated by copious amounts of Vicodin and hair pulling.


Unfortunately, this sad state of affairs is not simply the problem of camera vendors. Similar problems can also occur when lenses, lighting, frame grabbers, CPUs, and software need to be replaced. Worse, when smaller manufacturers go out of business, their product lines are often no longer available, making any comparable replacement parts more difficult to find.


Often, I have heard stories from system integrators who have been tasked to replace complete machine-vision systems because many of the components and software of older installed systems can no longer be upgraded. In the mechanical automation industry, mechanical engineers can often replace parts such as CPUs, PLCs, and pneumatic actuators with a number of comparable parts from a host of well-known suppliers. Adding machine vision to such systems, however, is often fraught with danger since machine-vision standards and standard-compatible products still need to be developed.


Perhaps the only solution to this problem will lie in the development of a number of plug-compatible smart cameras from numerous vendors. With integrated optics, sensors, software, and I/O, products like these may become as commonplace as the PLC and just as easy to interface and replace.

Wednesday, November 12, 2008

Will history repeat itself?


Years ago, our publishing company produced a magazine entitled Computer Graphics World. Its premise was rather simple. Editors wrote about a number of different graphics components -- display controllers, monitors, software -- that were used in a number of different markets ranging from architectural and mechanical CAD to 3-D graphics rendering for motion pictures and desktop publishing.


Back then, there were lots of companies that produced different hardware and software. Because of this, there were numerous display controller manufacturers and monitor vendors that were only too willing to spend their advertising dollars to support the magazine.


One thing happened, however, that caused both the downfall of a number of companies in this industry and eventually the magazine. Although part of this was due to market consolidation, technology itself played a larger role.


Years ago, only fixed-frequency monitors were available and to get them to synchronize with a specific horizontal and vertical frequency, a specialized display controller was needed. To meet this demand, companies developed display controllers that supported other companies’ frequency-specific monitors.


In the 1990s, with the development of multisync monitors and multisync circuitry on the PC motherboard, nearly every monitor manufactured could be supported without the need of a specialized display controller. What happed after that is history.


Today, only gamers purchase graphics boards, while many industrial systems simply drive their displays from the PC. The graphics “market” is primarily a software industry replete with companies making games and architectural, mechanical, and electrical CAD and animation packages. While all of these packages together represent a large total market, the industries they serve are radically different.


Like the computer graphics market of the late 1980s, the machine-vision market is still evolving. It will only take time before a combination of technological innovation and consolidation reduces the number of companies that currently participate. For this to happen, a technological change will first need to occur.


At present, thousands of cameras and hundreds of frame grabbers exist that use interfaces such as Camera Link, GigE, USB, and FireWire. Because only two or three of these standards are currently supported on most motherboards, system developers need to purchase specialized frame grabbers to support many of the others. To process images as they are captured from digital cameras, many of these boards incorporate FPGAs that allow image processing functions to be processed in a pipelined fashion before the images are stored in the host PC memory.


Now imagine, however, a high-speed interface that can support a range of different camera clocks, one that is fully deterministic like Camera Link, features all the benefits of PC-based software such as FireWire, and one that is incorporated onto the system’s motherboard. Couple this innovation with the power of multicore CPUs and a graphics processor on the motherboard and the need for multiple cameras supporting different camera interfaces and frame grabbers with on-board processing capability will rapidly diminish.


While the jury remains undecided on what the type of camera interface will be, one thing is for certain. When it is finally developed and accepted, it will herald the demise of a number of smaller companies currently offering both camera and frame grabber products. While this will provide a boon for system developers whose time to integrate machine-vision systems will rapidly diminish, it will also provide an opportunity for vendors of machine-vision software.


Rather than support multiple camera standards, drivers, and frame grabbers, they will be able to concentrate more fully on developing higher-level machine-vision algorithms that run rapidly on multicore hosts. History will repeat itself -- the only question is when. If the computer graphics industry is any benchmark, such advances are at least a decade away.

Thursday, October 16, 2008

Talking about my generation

Talking about my generation

Innovations in machine-vision hardware and software may best be found in unexpected sources and partnerships


On a recent trip to England I was fortunate enough to stay with my brother Dave and his son Paul, who is studying how to program FPGAs at the University of Bristol. As usual, I needed to know about all the books, courses, software programming tools, and development tools he was using in the pursuit of his degree.


During the course of the conversation, Paul mentioned that "on the side" he had coded-up a video game "for a laugh." As I had performed the same task over 30 years earlier (albeit on an Apple Macintosh II), I was intrigued. The game, it transpired, was a first-level rendition of the popular 30-year-old "Space Invaders" game.


While the idea certainly wasn’t original, the way Paul had coded the game certainly was. In fact, he demonstrated to me "his" version running on a mobile phone. It was a marriage of technologies that surely could not fail to win him many fans (and make lots of money). Unfortunately, being 20 years old, Paul had no idea how to go about licensing, marketing, or selling his idea to the likes of Apple, Nokia, or Motorola.


On my return to the United States, I pondered his conundrum and discovered that this was not a phenomenon faced only by the English. During several Web searches, I discovered a piece of code written by Bradley Ward to generate pseudo 3-D images using a webcam. Those of you interested can view a video of the software running at http://www.youtube.com/watch?v=jdeymMz_8HA.


Being intrigued, I e-mailed Bradley and asked him how he planned to offer the code. To my horror, the reply was rather simple. It seems that Bradley will supply the code to "anyone interested." Once again, no licensing, marketing, or sales plan seemed to be in place.


They have an idea

As if things couldn't get much worse, my recent trip to NI Week drove the final nail into this coffin. At National Instruments' own booth, Benjamin Cook, an undergraduate student at Rose-Hulman Institute of Technology and part-time NI intern, had developed an FPGA-based software package that emulated a number of different guitar pedal effects such as distortion, ring modulation, and tremolo effects. With the massive sales of Activision's Guitar Hero on my mind, I asked him if he had thought about plans to embed these effects into such products. He had not, although he was amenable to discussing any ideas with those who approached him.


With so many talented young people and so many good ideas, it is a mighty shame that there is no online brokerage to offer such ideas to manufacturers. So how are poor starving students supposed to make money from their efforts? Some companies involved in machine vision have an idea. Last year, for example, Aqsense, using ideas originally developed at the University of Girona, developed an improved method of determining the center point of reflected laser Gaussian profiles that is more accurate than center-of-gravity-based methods (see "Reading the Shapes," Vision Systems Design, March 2008.).


To market, promote, and sell this concept, the company did not embark on a large-scale marketing, advertising, and public relations campaign. Rather, the software was licensed to other camera and software companies with prior experience in structured lighting and machine vision. By tying together with such companies as Photonfocus and Stemmer Imaging, the company can concentrate on product development while at the same time offering its products to many companies worldwide.


At the November VISION 2008 show in Stuttgart, many of these third-party companies will be demonstrating their products with Aqsense software. Other small companies with a good hardware or software product would do well to follow the example set by Aqsense. By leveraging an installed base of products, software and hardware developers--students included--can add value to existing systems while at the same time making money and funding future machine-vision research. Tapping into the next generation of researchers and students will provide a great boon for the machine-vision industry.

Friday, September 5, 2008

A singularity in time

The future may be an extension of the past, except with androids hard at work



Several years ago, when both my son and I were a lot younger, we decided to watch Tim Holland's movie "The Langoliers" (http://www.imdb.com/title/tt0112040). This made-for-TV movie's basic premise is that people on an aircraft can go back in time. Having arrived in the past, the intrepid heroes discover that time itself--in the shape of the airport at which they have landed--is being eaten away by big teethed balls whose only task is to devour the future. My son and I watched it as New Hampshire storm clouds brought horrendous thunder and lightning to our living room.

After the movie, it was time for bed, and I bid goodnight to my son. As the thundering and lightning increased, I heard a knock on my door. It seems my son did not want to sleep alone during a thunderstorm--especially after watching that movie. Being the wonderful parent that I am, I tucked him safely next to me in a king-sized bed, and we watched the storm rage outside.

He was nearly half-asleep when the next lightning strike occurred, and, trying to be funny, I shouted, "The Langoliers are coming." My son was startled and scared. Unfortunately, with a quick elbow reaction to my face, my son rendered me with a nose that poured so much blood that I had to retreat to the bathroom for half an hour to recover. It was my fault, of course.

But the singularity in time brought by that event and others brings me to the subject of this editorial. Every once in a while a certain singularity in the development of technology brings us to a new point in time. In the 1960s, for example, this was the invention of the transistor and replacement of old tube-based (that is, valve, for our European readers) technology that drastically changed everything from computer science and health care to automation and automotive systems.

Today, there is a new singularity on the horizon. And anyone who has watched the "Terminator" movies will know what I mean. It is not in itself a singular development but a conglomeration of singular developments that will bring about this change. It is an exponential-exponential effect of Moore's Law!

To develop fully functional robots that can speak, hear, listen, touch, lift, and walk in a human manner is a gargantuan task and invokes far more than machine vision. To create such systems demands a certain culture that is, it seems, prevalent in the both the USA and Japan. Leveraging mechanical, electromechanical, electronic, and computer-based systems, developers here and in Japan are looking to develop fully automated robots that will prove beneficial to mankind in a number of ways.

Living proof
In the USA, Vecna Technologies is already working on its Battlefield Extraction-Assist Robot (BEAR), designed to replace the role of the medic on the battlefield. Researchers at Honda in Japan are working on their next generation of Asimo robots that are designed to replace the drudgery currently associated with certain types of work being performed by human beings. In fewer than 20 years, you may be able to purchase an android that looks and feels like a human being.

If you need proof of my argument, go no further than http://www.youtube.com/watch?v=MY8-sJS0W1I&feature=related. There, a researcher demonstrates his female android capable of reading and of face and object recognition. If you don't like the scenario that follows, you are not alone. At the end of the interview, the US interviewer asks, "What happens after the point of the singularity--everything will then change?" And it is there the video ends.

What does indeed happen to mankind after I employ two US or Japanese robots to write my articles, cook for me, and perform my every whim? I only hope that their batteries don't run out halfway through.

Wednesday, July 30, 2008

Six of the best

Not only are rules made to be broken, but breaking them is a good idea for developers of machine-vision products


As my beleaguered colleagues at Vision Systems Design will tell you, I have never really been a person who likes to follow rules. I discovered this when I was just 16 years old and attending a most distinguished public school in England. Before playing football one day, we were all informed that the changing rooms had been locked because someone had stolen some money from one of the other boys. These changing rooms would all be opened at 4:00 PM by one of the teachers to ensure this did not happen again.

On the football field, I purposely scored an “own goal” to get sent home early, and so at 3:00 PM I decided to creep in through one of the open windows of the changing rooms, put on my blue suit (yes, it was a public school, folks!), and go home. Unfortunately, just as I was prepared to leave, the doors were opened and a teacher, with a look of complete disbelief, approached me and told me to report to the headmaster’s office the next day.

Back then, of course, punishments were a little more capital than they are today. After waiting 10 minutes outside the headmaster’s office, I was informed that a good caning was in order and was duly struck very harshly six times on my backside. To this day, I have never understood why this is referred to as receiving “Six of the Best.” Once it was over, I walked back to classes in a rather duck-like manner!

After such a punishment, one would have thought I would have learned a lesson. Unfortunately, later in life the same anarchistic spirit lost me a few jobs. One I remember most particularly. In my 20s, I informed the managing director of a small publishing company that my immediate supervisor was a lazy no-good so-and-so and it was either him or me that had to go. Unfortunately, it was me. But is anarchy such a bad thing? In the machine-vision industry, this doesn’t seem to be the case. Indeed, many of the companies we write about often in this publication are headed and staffed by characters who obviously feel the same way I do about following rules.

Independent thinking
Rather than work for large corporations, these free-spirited independent thinkers have developed their own innovative products, starting companies and selling these products worldwide. Rather than work 9 AM to 5 PM in closeted grey-neon dreamlike cubicles, these people write their own rules, turn up in their “offices” in T-shirts and jeans, and reward their employees with stock options, pension plans, and free medical and dental coverage. While the fruits of their labor may rightfully have bought them private aircraft and second homes in the Caribbean, they are always willing to discuss new products, trends, and technologies. Several books have been written on what makes a good manager.

I should know. In his later years my Dad taught engineering management, after a 30-year stint at designing rolling mills. Many books list the qualities of a good manager as a series of As: one must have A sense of humor and be Approachable, Actionoriented, Able to Plan and Organize, and Able to deal with Ambiguity. I have known many great company directors, editors, and publishers who possess all of these attributes. However, the ones who have developed some of the most outstanding technologies and products have sprinkled these five attributes with a sixth—a little bit of Anarchy.

These folks are not willing to develop me-too products, even though the market for them may be massive. Rather, they look to extend today’s technology to develop more innovative products and systems with unique selling points. But, while a little anarchy may be good, too much can be detrimental to your health and your position. I remember at one publishing company sitting in a meeting that went on for eight hours where we “discussed,” for example, where certain editors should appear on the masthead of the magazine. At the end of the meeting, nothing had been decided, and the subject was conveniently forgotten the next day. For such anarchists, as my boss will heartily agree, six of the best, even verbally, is perhaps the only recourse.

Monday, June 23, 2008

A little knowledge

Asking the right question of a search engine can bring rewarding results, but humans still have an inside track on some important queries
It has been more than a decade since Larry Page and Sergey Brin founded Google. Since that time, the Web search engine has become the company’s most widely known product, and the company itself has become a major American corporation. Google has been used by millions of people worldwide to find information relating to subjects from semiconductors to soybeans.

Google indexes billions of Web pages using keywords and operators that link keywords so that users can search for information. Given a little knowledge of what one is looking for, the search engine can be a powerful tool. If you know that you need to perform an FFT on an FPGA, for example, simply typing “FPGA FFT” into the search engine will return many relevant results. Although a more formal structure such as “field programmable gate array fast Fourier transform” may be more meaningful, such an expression will not return as relevant a list of results. In such cases, a little knowledge about the subject and acronyms may be the fastest way to find the information you need.

For the Google search engine to crawl the billions of pages now available relies on the ability of programmers to encode the data currently on Web sites in a hypertext markup language (HTML). Because this task is time-consuming, Web sites such as Wikipedia (www.wikipedia.org) have developed their own markup language. This so-called “wiki markup” offers a simplified alternative to HTML and is used to describe pages within wiki Web sites. Because it is easy to use, this markup language has attracted great interest from researchers and students.

Perhaps Wikipedia’s most interesting feature is that the details described on each page are intimately linked to other Wikipedia sites. Typing a keyword such as “color” into the site brings you a page on the perceptual property of color, along with links to other Wikipedia pages such as color theory, the spectrum of light, and electromagnetic radiation. Despite offering such a wonderful resource, Wikipedia is still limited.

What is required is a more semantic representation of knowledge such as the Resource Description Framework (RDF) from the World Wide Web Consortium (www.w3.org). This can be used to represent both knowledge and facts contained in a document or resource to alleviate any potential subjectual misconceptions, a fact not overlooked by Powerset (www.powerset.com), a company that has introduced a search tool that uses semantic language representation to present the user with a more natural way of quizzing the Wikipedia database.

Rather than typing selected words such as “FFT” or “field programmable gate array” into the tool, more natural expressions such as “How do I implement an FFT in an FPGA?” returns results that are more pertinent— although limited to the Wikipedia site. Currently, two of Powerset’s competitors in the race toward semantic-based search—Haki (www.hakia.com) and Twine (www.twine. com)—also offer such semantic-based search tools that are not limited to Wikipedia’s site.

Despite these innovations, mining information in an intelligent fashion requires the information database to be well constructed. A visit to the “Ask the Experts” section of the Automated Imaging Association (www.machinevisiononline. org) Web site, for example, revealed that one reader had enquired about the difference between a machine-vision camera and a conventional photographic camera. A simple question, but even when asked, a result specifically answering the question was not returned by Powerset—merely another list of Web sites more relevant than those returned by Google. Luckily, the reader’s question had been answered very accurately by a human being!

While those envious of Google’s success are trying to better the company’s search engine with knowledge-based data-representation tools, only relatively few information databases currently use either the RDF format or RDF query languages such as SPARQL. And, even when such standards are fully adopted, it will be many years before any computerbased program can answer your questions directly. Until then, a little knowledge may not be as dangerous as you think.

Monday, May 19, 2008

Geisterbahn nach Berlin

Travel can be a bit scary these days, but mixing technology and marketing can be far more frightening
When I was young, my parents took my brother and me on holiday every year. Because my Dad was an engineer and underpaid, my parents could not afford the Hilton or visits to Paris. Instead, we stayed in bed-andbreakfasts located in crumbling Victorian seaside towns such as Clacton, Bournemouth, and Plymouth. Every year, my Dad would sit in a deck chair on beaches at one of these resorts and burn his flesh until he turned pink.

Looking back, it was all rather amusing. But the highlight for my brother and me was a visit to the fun-fair (or local carnival) intimately associated with having fun on holiday in England. Unlike America, where public gambling is limited, we could spend hours pouring pennies into numerous slot machines. Often, if my memory serves me correctly, we came back with a profit, which my brother and I greedily stashed away under our beds at the bed-and-breakfast.

At the fun-fair, there were also numerous rides including ghost trains, bumper cars, and a wonderful machine known as a Waltzer. Sitting in a small car, you were spun around chairs situated on a circular platform that added more centripetal force to your turning chair. If you have seen Claude Whatham’s motion picture That’ll be the day (www.imdb.com/title/ tt0070788/), you will know what I mean.

However, it was not this motion picture that reminded me of my ill-spent past at Victorian seaside towns in England. Instead, it was a recent trip to Germany with our illustrious sales representative Johann Bylek. Rather than providing yours truly with an easy walking tour of a German city, Johann decided to set up numerous meetings in cities as far apart as Munich, Radeburg, Jena, Bremen, Hamburg, Dresden, and Berlin.

Company visits
During this time, we visited some interesting companies, including Allied Vision Technologies, Basler, Baumer Optronic, The Imaging Source, and KameraWerk Dresden. Many of these companies described interesting technical developments that you can read about in the pages of this issue. After five train journeys and two airplane trips, exhaustion set in.

Between company visits we did find time to discover what makes a city tick. In Hamburg, Johann insisted on dragging me to the Reeperbahn, where a few young ladies seemed so desperate for company they actually grabbed hold of me! Everything was going well until Johann explained that my health might be in danger.

Feeling disenfranchised, I decided that perhaps the Reeperbahn was not the place to be. But I did notice that a couple of blocks over there was a carnival, reminiscent of those from my youth. It was indeed a place to celebrate.

There was a ghost train (Geisterbahn), dodgems (auto scooters), candy floss (Sucker Wasse), and even penny machines! I immediately bullied Johann onto the Geisterbahn, where I paid four euros for the funniest ghost train ride ever. After disembarking, Johann remarked that it was not that frightening. Do Austrian’s have any humor? Despite his lack of enthusiasm, I rode the dodgems and bought some candy floss to celebrate.

Then we were off to Berlin for the annual European Machine Vision Association business conference. There, I sat motionless for two days listening to talks about the glowing machine-vision markets in Germany, Italy, and Israel. Marketing presentations are really not my cup of tea, I must admit, but it did present an opportunity to meet some very smart people involved in technical aspects of the machine-vision industry.

To defray the cost of the conference, many of those who attended had also combined their trip to Berlin with company visits such as mine. On the way home, however, I was left to wonder whether any of these technical people thought that, like Geisterbahns, marketing presentations never really seem to deliver what they initially promise.

Monday, April 28, 2008

Working for other people

Designers of embedded-vision systems practice an often unrecognized and valuable art--and one they should capitalize on


In Vision Systems Design, we present articles describing how system integrators have developed products using off-the-shelf components. While a variety of lenses, frame grabbers, embedded processors, and machine-vision software are available to build these systems, many applications in medical, military, and industrial applications require the use of custom, embedded designs. Like consumer-based CCD cameras, these imaging systems are designed so that specific tasks can be easily performed by an end user.

While trade shows such as The Vision Show (Boston, MA, USA; June 10–12) allow vendors to exhibit OEM components, these products may not prove useful in small, embedded applications, and the system may have to be designed from scratch. For the designer of these products, it is not easy to identify the components or subassemblies needed.

Specialty embedded events such as the Embedded Systems Conference focus more on components and operating systems; other shows such as Medical Design & Manufacturing aim to bring together both the components and production equipment used in medical applications. Unfortunately, attending every trade show that pertains peripherally to machine vision and image processing is not possible for most engineers or engineering managers.

Despite the power of search engines such as Google, the variety of options or companies that assist in the design of embedded imaging or machine-vision systems is not immediately apparent. On a recent trip to New York, for example, I uncovered two companies involved in embedded system design. One, D3 Engineering (Rochester, NY, USA; www. d3engineering.com), provides DSP engineering services based on the Texas Instruments range of DSPs to companies involved in machine-vision, automotive, and medical markets. With a camera-development kit and SDK, the company currently supports more than 20 different imager types for embedded imaging applications. The other, Imaging Solutions Group (ISG; Fairport, NY, USA; www.isgchips.com) had designed a portable arthroscope based on a custom CMOS color imager that you can read about on page 20 of the May issue of Vision Systems Design.

Unsung heroes
Products such as those from D3 Engineering and ISG thus appear as specialized cameras or products from companies in the medical market. However, there remains a general lack of awareness of these engineering companies by those who may require their services. The reason is twofold: these companies are relatively small and do not have the time or money to spend on self-promotion, and, because no trade show specifically targets embedded imaging, their full capabilities cannot be readily displayed. The very nature of trade shows and the limited budgets of most embedded-systems-design companies means an embedded vision show may not be viable.

To promote their services, these specialized design houses often join partner programs created by vendors of imagers, DSPs, and software. By word of mouth, customers are referred to the custom embedded-design houses. However, those looking for help with their embedded designs are more likely to find resources from press releases, advertising, and trade-show presence from more established vendors. The reason, perhaps, is that companies involved in developing cameras, for example, are more likely to understand how to tailor their product for an embedded application.

While contract work by developers of embedded imaging systems may provide revenue for a few embedded-design companies, their end-user products may prove more lucrative for companies that have commissioned projects. To ensure the future of their companies, many have chosen to spin out the technology developed for others into new products. By doing so, companies previously thought of as specialized design houses will achieve the engineering recognition and financial rewards they deserve.

Monday, March 31, 2008

That joke isn't funny anymore

Machine-vision-system integrators are more than consultants--they actually have to know something about integration


There’s an old joke about consultants that reads something like, "Ask a consultant what time it is and they will borrow your watch and tell you the time." When I first heard the joke, I thought it was quite funny, especially since I have personally been asked questions about how to design machine-vision systems for various applications.

Unfortunately, while the joke may ring true to many considering deploying a machinevision system, it is not as amusing when one considers the disparate disciplines of optics, illumination, image processing, computer science, and mechanical engineering needed to develop such an application. Indeed, this is the main reason that the development of these systems is so challenging and at the same time so frustrating.

Since very few universities and colleges bundle these subjects into a single degree, it is difficult for developers to hire technical staff. Instead, they must rely on pooled knowledge from those with experiences in individual subjects.

At the outset, system development may seem easy. Light the product to be inspected, capture an image of it, and then trigger a reject mechanism should the part fail the inspection. When looked at from 30,000 ft, system development may seem trivial and—to those in management—inexpensive.

When examined from a microscopic level, however, the problem of designing a system becomes more complex. Just choosing a lens to image the subject may result in hours of NRE (nonrecurring engineering) time. Deciding on the type of lens required, optical mount, focal length, and resolution may appear easy, but, because of the lack of detailed specifications offered by many suppliers, an evaluation of a number of lenses may be required—a process that could take days.

Additional complications
This situation is further compounded by the fact that coupling a lens to several different cameras may result in very different images being obtained. As David Dechow, president of Aptúra Machine Vision Systems (Lansing, MI, USA; www.aptura.com) pointed out in our February Webcast, the different formats of imagers employed by camera vendors may result in varying levels of illumination rolloff or vignetting.

With the move to larger-format imagers, this problem is further exacerbated. Worse, if the digital camera you select does not have dead-pixel correction or flat-field correction, the resulting image may not be usable. As can be seen, simply selecting the correct optics and cameras is challenging. But system integrators face other tasks relating to lighting, choosing the correct software package and operating system, and how these are integrated into an industrial automation system controlled by PLCs. While college textbooks may help students understand the basic principles of all of these subjects, deploying machine-vision systems requires more. Luckily, most integrators are fully aware of this situation.

For those considering deploying a machine-vision system, a visit to an engineering facility may be most valuable. If you are led into a conference room and given a sales pitch, beware! Instead, ask for a tour of the engineering department, where you should expect to see workbenches strewn with optics, lighting, cameras, and half-complete computers. If people there appear busy and frustrated, take this as a very positive sign.

Often, however, potential customers visiting these facilities arrive unprepared, handing the company management a few questions and a part that they would like automatically inspected. Hence, the integrator must probe more deeply into exactly what needs to be inspected, the nature of the production line, the type of lighting used in the facility, and the previously installed computer systems— essentially borrowing the potential customer’s “watch” to ascertain the time. In such situations, having your “watch borrowed” is obviously quite a good idea, since it will only lead to the development of a more effective and efficient vision system.

Tuesday, March 18, 2008

The Needs of the Many

Parallel-processor-based systems offer system developers a path to real-time, high-speed image capture and analysis--but only in certain applications As many a Star Trek fan will tell you, one of the most memorable quotes from Star Trek: The Wrath of Kahn (1982) is uttered by Mr. Spock. When asked whether Captain Kirk should assume command, Spock replies that “logic clearly dictates that the needs of the many outweigh the needs of the few.” This same concept was obviously on the minds of developers of the World Community Grid (WCG; www.worldcommunitygrid.org), an organization intent on creating the largest public computing grid to benefit humanity.

The idea of the network is very simple. Researchers sign up with highly parallel computational tasks such x-ray crystallography and protein analysis. To perform this analysis requires large numbers of data sets to be analyzed, a procedure that can easily be accomplished on a distributed network of computers. Luckily, with more tahn 340,000 members and 840,000 processors networked online, the WCG is providing much of the computing power required.

But even with this number of distributed processors, the research tasks that need to be accomplished require an even larger number of computers. With this in mind, WCG developers are asking for donations—but not in the form of money. WCG wants to harness the power of your computer at home or at work to help speed this research. Basically, the idea is rather simple and resembles a peer-to-peer network.

To become a member of WCG, simply download a small program from the WCG Web site onto your computer. When your computer is idle, it requests data on a specific project from the WCG server. Your computer then performs computations on these data, sends the results back to the server, and asks the server for a new piece of work. Since each data set is only approximately 50 Mbytes, all of today’s PCs can easily handle the task.

The software also allows you to configure your system so that it can be set to perform these tasks during midnight hours or at weekends. To make this more interesting, you can set up your own “team”—get your friends and colleagues to join and accumulate “points” that, to be honest, are worth about as much as my frequent-flyer miles! So, instead of turning your computer in the office off when you leave for home you can leave it on knowing that you are contributing to invaluable research on cancer, climate change, and human proteome folding.

Avoiding gridlock
While many research projects such as these lend themselves naturally to parallel distributed processing, so do many machine-vision and image-processing tasks. In stereo image processing, for example, two processors can be used to simultaneously process image data from two independent cameras.

Indeed, in this issue, Homer Liu of QuantaView (Tokyo, Japan) describes how two Intel Xeon processors have been used for this very task (see “Vision-based robot plays ping-pong against human opponent,” p. xx). With the advent of dual- and quad-core processors, this trend is likely to continue as software vendors rework their code to take advantage of parallel-processing concepts.

To achieve the optimum performance for parallel-processor-based systems, however, developers will need to closely match the I/O, processing, and storage capabilities of such systems. Today’s Camera Link-based systems, for example, can be used to transfer data from a single camera to a host-based frame grabber at rates of up to 850 Mbytes/s using 85-MHz transceivers.

However, there is no single or multiprocessor von Neumann-based CPU commercially available that at present could possibly process individual images at this data rate, relegating such camera-based systems to high-speed image analysis where image data are captured and then later played back for image analysis. Because of this, it is likely that for the foreseeable future, heterogeneous networks of distributed computers may remain useful only for large-scale algorithmic research projects such as those currently running on the World Computing Grid.

Monday, January 21, 2008

Sicko

VISION 2007 and its new venue reveal the perils and the glory of machine-vision exhibitions
VISION 2007, held in Stuttgart, Germany, 6–8 November 2007, will be remembered as the largest, most impressive machine-vision show in the world by most attendees. Certainly, the nearly 300 exhibitors, a record visitor turnout, and a brand new location were cause for celebration. But not by all those who attended the show. One man fell head first down one of the escalators that linked the two halls where exhibitors demonstrated the latest hardware and software products. Another performed a rather large incision on his leg while attempting to open one of the packages he had shipped to the show. He was hospitalized.

And then, of course, there was yours truly. After the first day of the show, pounding along the uncarpeted concrete of the show halls, my right leg decided it had had enough. Back at the hotel, my right knee became rather large, probably due to an injury sustained while running after a young lady outside a fish-and-chip shop in England when I was 18 years old. After a 3 a.m. (0300 h) visit to the local “Krankenhaus” to see Dr. Schnuck, I was prescribed Ibuprofin. The doctor also thought that a certain amount of “foot tapping” exercise might be beneficial. It wasn’t quite what I had expected after paying out €137.94.

Michael Moore, of Sicko fame, is no longer one of my favorite film directors. For the next two days, I was confined to a wheelchair where I discovered the benefits and short falls of being temporarily handicapped in a foreign country. These benefits included high-speed traversing of the trade show, the lack of a need for excuses for being tardy for appointments, and the large amount of sympathy heaped upon me by my associates in the machine-vision community.

Unfortunately, especially for those in wheelchairs, the handicapped bathrooms on all the floors of the “Neue Messe” were locked. As you can imagine, a 50-something journalist hobbling into a bathroom in Germany and performing “foot-tapping” exercises was out of the question.

Despite my condition, I did manage to visit a large number of companies that exhibited in the two vast halls that comprised VISION 2007. For those of you that could not attend, let me describe the show.

The first hall was replete with rather smaller booths and newer, less-established companies that exhibited some very interesting technologies and applications. Many of these you can read about in this issue.

In the larger hall, many established companies seemed to be playing a booth game of “mine is bigger than yours.” Some, with really nothing new to introduce, had purchased voluminous booths with which to impress the attendees. In fact, some of these booths were so large that one felt lost in a “booth canyon” trying to see any other companies exhibiting. According to a spokesperson for the Messe, this will change next year when the show is moved into another hall the size of five baseball fields.

On the last day of the show, after receiving a rather long lecture about caring for myself from my publisher, Kathy Bush, it was time to leave. To do so, the trusty chariot that I had ridden for two days needed to be returned to the organizers. Before securing my wheelchair, my boss, Conard Holton, had left his American driving license as a “deposit.” Unfortunately, upon returning the chair, he found that the organizers had assumed that since it had not been returned on the first day, it had been stolen. Luckily, other members of our organization speak German. Otherwise, it seems, my boss would have been imprisoned for theft. Although some of you may think that hospital or prison is the best place for editors of trade magazines such as this, my publisher, like Queen Victoria, was not amused.

Hopefully, next year VISION 2008 will feature lushly carpeted hallways, bathrooms that open, and less bureaucratic staff. Other than that, VISION 2007 can only be rated 10/10 by this reviewer.

Friday, January 18, 2008

The Needs of the Many

Parallel-processor-based systems offer system developers a path to real-time, high-speed image capture and analysis--but only in certain applications As many a Star Trek fan will tell you, one of the most memorable quotes from Star Trek: The Wrath of Kahn (1982) is uttered by Mr. Spock. When asked whether Captain Kirk should assume command, Spock replies that “logic clearly dictates that the needs of the many outweigh the needs of the few.” This same concept was obviously on the minds of developers of the World Community Grid (WCG; www.worldcommunitygrid.org), an organization intent on creating the largest public computing grid to benefit humanity.

The idea of the network is very simple. Researchers sign up with highly parallel computational tasks such x-ray crystallography and protein analysis. To perform this analysis requires large numbers of data sets to be analyzed, a procedure that can easily be accomplished on a distributed network of computers. Luckily, with more tahn 340,000 members and 840,000 processors networked online, the WCG is providing much of the computing power required.

But even with this number of distributed processors, the research tasks that need to be accomplished require an even larger number of computers. With this in mind, WCG developers are asking for donations—but not in the form of money. WCG wants to harness the power of your computer at home or at work to help speed this research. Basically, the idea is rather simple and resembles a peer-to-peer network.

To become a member of WCG, simply download a small program from the WCG Web site onto your computer. When your computer is idle, it requests data on a specific project from the WCG server. Your computer then performs computations on these data, sends the results back to the server, and asks the server for a new piece of work. Since each data set is only approximately 50 Mbytes, all of today’s PCs can easily handle the task.

The software also allows you to configure your system so that it can be set to perform these tasks during midnight hours or at weekends. To make this more interesting, you can set up your own “team”—get your friends and colleagues to join and accumulate “points” that, to be honest, are worth about as much as my frequent-flyer miles! So, instead of turning your computer in the office off when you leave for home you can leave it on knowing that you are contributing to invaluable research on cancer, climate change, and human proteome folding.

Avoiding gridlock
While many research projects such as these lend themselves naturally to parallel distributed processing, so do many machine-vision and image-processing tasks. In stereo image processing, for example, two processors can be used to simultaneously process image data from two independent cameras.

Indeed, in this issue, Homer Liu of QuantaView (Tokyo, Japan) describes how two Intel Xeon processors have been used for this very task (see “Vision-based robot plays ping-pong against human opponent,” p. xx). With the advent of dual- and quad-core processors, this trend is likely to continue as software vendors rework their code to take advantage of parallel-processing concepts.

To achieve the optimum performance for parallel-processor-based systems, however, developers will need to closely match the I/O, processing, and storage capabilities of such systems. Today’s Camera Link-based systems, for example, can be used to transfer data from a single camera to a host-based frame grabber at rates of up to 850 Mbytes/s using 85-MHz transceivers.

However, there is no single or multiprocessor von Neumann-based CPU commercially available that at present could possibly process individual images at this data rate, relegating such camera-based systems to high-speed image analysis where image data are captured and then later played back for image analysis. Because of this, it is likely that for the foreseeable future, heterogeneous networks of distributed computers may remain useful only for large-scale algorithmic research projects such as those currently running on the World Computing Grid.