Decision time? Check out our latest product comparisons

New object recognition algorithm learns on the fly

By

January 19, 2014

The Evolution-Constructed Features algorithm can identify and learn new objects without hu...

The Evolution-Constructed Features algorithm can identify and learn new objects without human involvement

Image Gallery (2 images)

Scientists at Brigham Young University (BYU) have developed an algorithm that can accurately identify objects in images or videos and can learn to recognize new objects on its own.

Although other object recognition systems exist, the Evolution-Constructed Features algorithm is notable in that it decides for itself what features of an object are significant for identifying the object and is able to learn new objects without human intervention. The researchers say that unlike other methods, it does not require retuning or reworking for different tasks.

"In most cases, people are in charge of deciding what features to focus on and they then write the algorithm based off that,” says Dr. Dah-Jye Lee, professor of electrical and computer engineering at BYU and author of the paper. “With our algorithm, we give it a set of images and let the computer decide which features are important."

“With our algorithm, we give it a set of images and let the computer decide which features...

According to Dr. Lee, most other algorithms require a lot of fine-tuning of parameters and methods to achieve their best accuracy, whereas the Evolution-Constructed Features algorithm does not. Despite this, the researchers ay the algorithm has performed as well or better in object recognition tests than other leading object recognition algorithms.

For example, the Evolution-Constructed Features algorithm achieved 100 percent accuracy on motorbike, face, airplane and car image datasets from Caltech.

Caltech's database is used to benchmark the algorithm against other similar research, with other "published well-performing object recognition systems" scoring 95-98 percent accuracy in the same tests.

Lee and his team suggest that the algorithm could be used applications such as detecting invasive fish species or identifying flaws in produce such as apples. To this end, it was also shown to have 99.4 percent accuracy on fish species image datasets from BYU’s own biology department.

"Within some predefined criteria and situation, object recognition will continue to show its progress," predicts Lee. "Who knows? Maybe one day, when the computation power of computing platforms increases to be close to human brain, we could see some real breakthrough."

Source: Brigham Young University

About the Author
Stu Robarts Stu is a tech writer based in Liverpool, UK. He has previously worked on global digital estate management at Amaze and headed up digital strategy for FACT (Foundation for Art and Creative Technology). He likes cups of tea, bacon sandwiches and RSS feeds.   All articles by Stu Robarts
8 Comments

"You are being watched. The government has a secret system: a machine that spies on you every hour of every day. I know, because I built it. I designed the machine to detect acts of terror, but it sees everything. Violent crimes involving ordinary people; people like you. Crimes the government considered 'irrelevant'. They wouldn't act, so I decided I would. But I needed a partner, someone with the skills to intervene. Hunted by the authorities, we work in secret. You'll never find us, but victim or perpetrator, if your number's up... we'll find you".

-- Opening to Person of Interest

www.youtube.com/watch?v=IQKsYXH2LCM

Kevin Bingham
19th January, 2014 @ 06:27 pm PST

Combine this with some nice AI and we can make a terminator.

BT
19th January, 2014 @ 07:12 pm PST

This is the kind of thing that will lead to computers that can think like humans. After all, our brains have image-recognition algorithms too, the main difference is they're analog and organic, constantly tweaking and updating to best fit the inputs.

Joel Detrow
19th January, 2014 @ 09:52 pm PST

Eden of the East mobile phone search scanners here we come! :-D

Dennis Jay Dole
19th January, 2014 @ 10:26 pm PST

Question: how long before "human" is reclassified as "meatbags"?

digi_owl
20th January, 2014 @ 01:01 am PST

What with all the NSA and GCHQ prying into our privacy and software like this, one is forced to wonder where the checks and balances are to ensure technology doesn't get out of control.

Even if we froze technological development at the stage it is currently at, we have probably reached a point where we have sufficient ingredients to make a technological primeval soup from which could grow a non-human or at least a minimal human military. It is not as though numbers are a problem. Unlike a human, once put into storage these automatons do not need much in the way of upkeep and even that could be automated.

Of course we will not freeze technological development. It will continue to follow the exponential trajectory that it is currently on, with all that that implies.

There is a move to name the age we are living in, the anthropocene. I wonder how long it will last.

Mel Tisdale
20th January, 2014 @ 06:37 am PST

Bt & Others, the Skynet reference is obvious. My concern is how quickly pinheads like Ed Snowden collaborate with those nice people from the People's Liberation Army, or the Federal Security Bureau, (Russians), or the devout folks from any number of Jihadist groups, or the Persians, etc., to liberate this algorithm and share it anywhere.

StWils
20th January, 2014 @ 11:06 am PST

If its affordable, it would be a great addition to home security systems.

With a couple more capabilities, it could tell the difference between a real hummingbird and a RC version.

tomt
20th January, 2014 @ 08:07 pm PST
Post a Comment

Login with your gizmag account:

Or Login with Facebook:


Related Articles
Looking for something? Search our 29,151 articles