BYU image algorithm can identify objects without human help

By on January 17, 2014, 1:30 PM
algorithm, image recognition, byu

BYU engineer Dah-Jye Lee has created an object recognition algorithm that can accurately identify objects in images or video sequences all by itself, without any human help. According to Lee, the unique selling proposition of the algorithm is that it defines its own set of parameters with which it performs the task. Not only this, but it also doesn't need to be reset every time a new object is to be identified.

Lee draws inspiration from the idea of teaching children the difference between two animals (say cats and dogs). Instead of manually explaining the difference, if we show children images of the animals, they learn on their own to distinguish between the two.

The algorithm works on the same principle: Instead of explaining it what to look at to distinguish between two objects, it is fed with a set of images, and just like humans, it learns on its own.

Lee's algorithm proved to be better than some of the top object recognition algorithms developed by other universities. When the algorithm was fed four image datasets from CalTech (motorbikes, faces, airplanes and cars), it was able to identify objects with 100 percent accuracy. While other algorithms proved to be 95-98 percent accurate.

According to Lee, the algorithm, published in the December issue of academic journal Pattern Recognition, could be used for various applications like detecting invasive fish species, manufacturing defects, and more.




User Comments: 19

Got something to say? Post a comment
Guest said:

Ohhh, skynet looks so adorable when it was a little baby.

Guest said:

Person of Interest.

1 person liked this | JC713 JC713 said:

Wow this is creepy.

1 person liked this | Guest said:

Skynet is NSA's code name.

cliffordcooley cliffordcooley, TechSpot Paladin, said:

Code Red! Middle finger detected, guy in red and gray sweat shirt, north east corner of campus.

NTAPRO NTAPRO said:

This wont improve the quality of the ufo and bigfoot pictures though sadly

They will be forever grainy no matter how much high definition improves lol

1 person liked this | wiyosaya said:

Now if we could only run them for captcha input.

cliffordcooley cliffordcooley, TechSpot Paladin, said:

Now if we could only run them for captcha input.
LOL

If they did I wouldn't want to witness the next step in stopping automated bots. Thats a step that might stop everyone without a high IQ (that includes yours truly).

Guest said:

Now what we really need is behavioural monitoring software that can spot a paedophile a mile away and call in an airstrike.

Job done, no more Bishop at the alter.

Guest said:

It's still not enough!! NSA needs algorithm that can identify terrorists and their full ID without human help :D

Guest said:

One step closer to taking are rights away and merging with the machine

Guest said:

Absolutely amazing technology...I plan to design a future flying vehicle and needed an object avoidance algorithm, this could be suitable. Kudos to the inventor.

Guest said:

Techspot, please add 'thumbs up' and down, there's some comments I really wanted to thumb up.

ElShotte ElShotte said:

Goddamn it! Thanks for freely publishing the algorithm, +1 NSA.

mailpup mailpup said:

Techspot, please add 'thumbs up' and down, there's some comments I really wanted to thumb up.
We already have a "Like" button for comments but I believe you have to be a registered member to see it and use it.

MilwaukeeMike said:

Ha, I've read about these programs. They have to analyze the whole picture before they can figure out what they're looking at and if you change the angle of the subject it has to start all over again. Kinda funny to think that a computer can recognize a chair, but if you turn it upside down it gets confused for a second while it reprocesses it.

I wish I could remember were I read about that so I could link a source... I think it was WIRED.

Guest said:

But will it blend??

cliffordcooley cliffordcooley, TechSpot Paladin, said:

^^^ Don't see the humor in asking "Will an algorithm blend?".

Emexrulsier said:

100% accuracy, yet in the image example its 99.4%

Load all comments...

Add New Comment

TechSpot Members
Login or sign up for free,
it takes about 30 seconds.
You may also...
Get complete access to the TechSpot community. Join thousands of technology enthusiasts that contribute and share knowledge in our forum. Get a private inbox, upload your own photo gallery and more.