Showing posts with label vision based algorithms. Show all posts
Showing posts with label vision based algorithms. Show all posts

Friday, November 4, 2011

Mistaken identity

Next week marks the start of one of the biggest event in Vision Systems Design's calendar -- the VISION 2011 show in Stuttgart Germany.

Accompanying me to the show this year will be Susan Smith, our publisher; Judy Leger, our national sales manager; and the latest addition to our editorial team, Dave Wilson.

As many of you may know, Dave joined the magazine just last month to increase our presence in the important European market. But what some of you may not know that Dave is also my identical twin brother, a fact I thought I'd make perfectly clear before the show begins, in order to diminish the confusion that will inevitably arise as a case of mistaken identity on the show floor.

You see, although numerous vision system algorithms have been developed over the years to differentiate between products of a similar nature, I'm sorry to say that most human beings' visual systems -- even those in the machine vision industry -- seem to be incapable of differentiating between the two of us, despite the fact that I clearly inherited all the brains and good looks.

For that reason, I have programmed my brother's central processing unit to respond to the greeting "Hi, Andy" whenever he hears it, after which he will instigate a verbal subroutine, which will explain that he is simply a poor imitation of the real thing.

However, if you do run into the man instead of me, you will find that he is equally as willing to learn what new technologies are being discussed at the show.

He would be especially delighted to discuss any applications of machine vision related to the use of smart cameras and hyperspectral imaging. Please be sure to bend the man's ear if you see him!

Tuesday, November 1, 2011

Win $50,000 courtesy of the US Government!

Today's troops often confiscate remnants of destroyed documents from war zones, but reconstructing entire documents from them is a daunting task.

To discover if they can unearth a more effective means to do just that, the folks at DARPA have come up with a challenge that they hope will encourage individuals to develop a more automated solution.

That's right. The defense organization is hoping that by offering a whopping $50,000 in prize money, entrants to its so-called "Shredder Challenge" will generate some ideas that it might be able to make use of.

The Challenge itself consists of solving five individual puzzles embedded in the content of documents that have all been shredded by different means. To participate in the challenge, participants must download the images of the shredded documents from the challenge web site, reconstruct the documents, solve the puzzles, and submit the correct answers before Dec. 4, 2011.

Points will be awarded to those who provide the correct answers to the mandatory questions associated with each puzzle. $1,000 will be awarded for each point scored up to $50,000 for a perfect score. DARPA will then award one cash prize of up to $50,000 to the participant who scored the highest total number of points by the deadline.

Registration is open to all eligible parties at www.shredderchallenge.com, which provides detailed rules and images of the shredded documents for the five problems.





Clearly, this is an application that would benefit from the expert knowledge of those in the image processing field who might be able to develop -- or deploy -- a set of vision-based algorithms to reconstruct the documents and hence solve the puzzles.

Interestingly enough, of course, several individuals contributing to the discussion forums on the Shredder Challenge web site are taking exactly that approach...