LifeCam VX-1000 Eyetracking

[tomasmk]

is it possible to do this with java?

[piechulla1966]

@tomasmk

Yes, You can use JavaCV - it contains Java wrapper classes for the full C/C++ API of OpenCV 2.3.0, also works with Android. Personally, I never tried JavaCV. I only know that it exists.

[tomasmk]

@piechulla1966 thanks!

[tomasmk]

hello nice to meet you

i need source code for a similar application for eye detection

because me and my work group are going to make an anti-drowsiness system

we need to do it quickly

thank you very much bye

[piechulla1966]

@tomasmk

PERCLOS or similar indicators of drowsiness are non-trivial. A webcam is too slow to detect eyeblinks and estimate eyeblink duration. Go for a good camera that gives you framerates of 100 Hz or faster.

Cheers,

walter

[tomasmk]

@piechulla1966 yes you are right, but my proyect is intended to detect only when a driver actually falls asleep or closes his eyes for more than 2 seconds. A couple of weeks ago I had a near-death experience in wich i kind of fell asleep for 3 or 4 seconds while I was driving a car with my family in it. Fortunately I woke up when the car was a couple of inches from crashing into a truck. A simple alarm activated by my eyes being closed for too long would have eliminated any danger.

[piechulla1966]

@tomasmk

Active Contour Models (Snakes) seem to be helpful. I never tried it. Zafer Savas gives an introduction in his MSc thesis (just google for his name). Pages 64-68 explain his Snakes approach to eyelid tracking.

Cheers,

walter

[tomasmk]

@piechulla1966 thanks, will do. what did you use for this program?

[piechulla1966]

@tomasmk

Windows boxes (XP/Vista), MS Visual Studio (2008/2010), OpenCV (1.0/1.1/2.0/2.1/2.2) and some years of experience in tweaking high-end eyetracking systems (in order to be able to use these systems for "cognitive ergonomics" aka "man machine interaction" aka "human factors").

[tomasmk]

@piechulla1966 thanks, will do. when will you finish your eye tracking program?

[piechulla1966]

@tomasmk

This is a good question. I bought a Thorlabs DCC1545M (nice camera) in 2010, but I was too busy to try it out. Shall play around with this interesting device next week. Point Grey exhibited a USB3 camera prototype at the Vision 2010 fair (Stuttgart, Germany). In the near future, USB3 could give us high resolution at high frame rates -- exactly what we need :-)

[RobotechMacrossGame]

good job

[hujovida310793]

sorry Freund hat einen Zeitplan für die automatische Erkennung Augen ohne Verwendung der Maus

Anerkennung wird direkt durch eine Webcam tun, wenn ich auf die Idee kommen

[piechulla1966]

@hujovida310793

This is hard. You need very good accuracy and precision to replace the mouse with an eyetracker. As far as i know, this is not possible today for standard applications. You have to use special applications with *BIG* gaze targets.

Cheers,

walter

[hujovida310793]

hola buenos dias

necesito un codigo fuente asi parecido al del video para la deteccion de los ojos

es que yo y mi grupo de proyecto vamso hacer un sistema de antisomnolencia

necesitamos hacerlo menos de 1mes

muchas gracias saludos

[piechulla1966]

@hujovida310793

Sorry, you overestimate my education by far. I only understand German (my first language) and some English (my second language) . Nothing else. Let's try it in English! Looking forward to understand what you wrote,

walter

[savagetwinkie]

How hard is something like this to write, our senior project incorporates tracking eyes and figuring out when they are closed. Well really not just closed so long as they aren't looking forward... we used opencv and found stuff on haardetection but it runs horribly slow, out of the 4 people in the group, i got the image processing, the other 3 are tackling a button and small speaker...

[piechulla1966]

@savagetwinkie

This should be feasible. Maybe you should google for Zafer Savas' thesis. As far as I remember, he used "Snakes" to track the eyelids. Bu sure to check out Zafer's work. I think it can help you a lot.

[savagetwinkie]

@piechulla1966

thank you, i'll check it out. But we were interested in yours too, we actually have to detect sleep but instead we decided to just make sure someone was awake and looking forward. It would be better if it ran in the dark, we just didn't realize how easy it was to get an infrared camera.

[piechulla1966]

@savagetwinkie

In fatigue detection, you typically measure eyelid distance. Not just visibility of the pupil. Problem: Eye blinks are very fast. So, you will run into problems if you use a standard webcam (max. frame rate approx. 30 Hz, i.e. very slow). Classic fatigue detection algorithms like PERCLOS may be hard to implement. So, you need very clever new ideas. But (IMHO) the "looking forward" idea already is quite clever. Be sure to check out Trent W. Victor's papers on this topic.

[piechulla1966]

I experienced app crashes with some of the tests and some of the examples. A rebuild did not solve the problems. I decided to stick to version 1.1 -- using 1.1, you just use the solution files for VS2005. They work like a charm with VS2008.

[piechulla1966]

Google for openEyes and thirtysixthspan.

[piechulla1966]

ROTFLMAO. This reminds me how big and manifold this beautiful world actually is -- I simply took it for granted that Ibrahim would get the joke. To those of you who are well versed in the Bible: This was not blaspemy. This was _only_ a joke!

[05Ibrahim]

hi methew, your wellcome for the advise and i will surley check out the next video clip.

[05Ibrahim]

hi i am just a regular visitor to you Tube, the start up image on you video is roughly described as quite scary, so i think you would get more people to visit your video if you would replace it with a picture of the actual web cam.

[piechulla1966]

Thank you for the hint. The problem is: Machine Vision programming is easy for me, but using YouTube is hard. The "Video Thumbnail" feature seems to force me to choose between just three versions of my ugly face :-(

Good news: The next video is in the pipeline. It's about my AI4CCTV SDK (Artificial Intelligence for Closed Circuit Television Software Development Kit).

Take courage! It is I. Don't be afraid. [Matthew 14:27]  ;-)

[utuber001001]

good video

[piechulla1966]

Please correct me if I'm wrong: All your (very interesting) videos show a head-mounted system. So, you need a camera as small and lightweight as possible. Which one do you use? Another question: Someone told me that IR light with intensities > 4 to 20 mW/sr has to be considered harmful for the eye. I would appreciate Ophthalmologist's comments on this issue, because the OpenEyes project published 'interesting' hardware specifications (as far as this problem is concerned).

[oVISIONo]

Hi Walter,

Until now, I'm just testing my code with a Lifecam VX-1000 and an improvised chin rest. My thesis professor is an electronic engineer and he is helping me with the eyetracker development, but we have not decided wich camera we will use yet.

About the harmfulness of infrared light, you can search for "COGAIN-D5.4.pdf" on google (page 12), or search for "ADA215809" on google too.

I hope it helps.

Regards.

[piechulla1966]

"In other words, the infrared LEDs would have to emit far more power to pose a serious acute hazard to the retina. This is theoretically impossible for current LEDs." (COGAIN-D5.4.pdf p.20). So, this seems to be an all-clear at first glance, but I'll have to read the whole report carefully.

[oVISIONo]

very good Walter! I have a LifeCam VX-1000 too!