Driver Vigilance System.

hello everyone..

Can anyone give me some information on Driver vigilance system in the following areas
1)How I can monitor the driver in real-time and at what rate should i monitor him?
2)where should I save the images in real-time and what should I use to save them?
3)What Technique should I employ to process the obtained Images?
4)And lastly which language should I use for coding purpose?

--Thank you.

stuart gambill said:

hello everyone..

Can anyone give me some information on Driver vigilance system in the following areas
1)How I can monitor the driver in real-time and at what rate should i monitor him?
2)where should I save the images in real-time and what should I use to save them?
3)What Technique should I employ to process the obtained Images?
4)And lastly which language should I use for coding purpose?

--Thank you.

Click to expand...

The answers to your questions are complex an nuanced. Question #1 is a broad field of research called human operator response that goes back some 30-40 years. Question #2 implies that the answer to question #1 is some sort of image capture which would not have been my first choice. Question #3 seeks the proprietary answers to the questions of the machine vision industry which guards their secrets carefully. To answer question #4 we really need the answers to questions #1 through #3.

I suggest that you backup and explain what you are trying to accomplish and what your requirements are. That's the only way that a meaningful response can be constructed.

How many drivers do you need to monitor? What exactly do you mean by vigilance? There are commercial products that log driver's speed, number of stops/duration, and so forth. At my job, the drivers refer to it as a 'tattle-tell', and not too fondly.
If you are refering to images, as in a camera, and in real-time. You are looking at some high end hardware.
Yes, more details about what the problem is with your drivers, and what information you need to monitor. In the United States, D.O.T. requires regular drug testing, so guessing you don't suspect 'special' cigarettes...

By a driver vigilance s/m I mean a s/m that is capable of monitoring the eye movements and the face poses of a driver continuously in real-time.Here I want to keep track of the eye gaze and face position of the driver continuously and store them. And process these in such a way so as to check the fatigue level of the driver.

Regarding this I had asked some questions which are given below

1)How I can monitor the driver(single) in real-time and at what rate should i monitor him?
2)where should I save the images in real-time and what should I use to save them?
3)What Technique should I employ to process the obtained Images?
4)And lastly which language should I use for coding purpose?

I'd suggest search the web, maybe you can stumble on a university research study, and see what's been done. I remember ready about this kind of stuff many years ago. There is a lot of image processing software around, for security cameras, might be adapted to only record changes in the face and eyes. Driving a vehicle would introduce some noise and vibration (potholes, speed bumps, roadkill...), but could reduce the amount of data.

But if you are talking real-time, as in a 6 hour run, will produce 6 hours of video... Maybe hook one up to monitor the fatigue of who ever has to watch and analyse the driver's data... Just kidding.

This would be better suited for a research facility, as you would want a fast computer to crunch your video down to something more useful.

stuart gambill said:

By a driver vigilance s/m I mean a s/m that is capable of monitoring the eye movements and the face poses of a driver continuously in real-time.Here I want to keep track of the eye gaze and face position of the driver continuously and store them. And process these in such a way so as to check the fatigue level of the driver.

Regarding this I had asked some questions which are given below

1)How I can monitor the driver(single) in real-time and at what rate should i monitor him?
2)where should I save the images in real-time and what should I use to save them?
3)What Technique should I employ to process the obtained Images?
4)And lastly which language should I use for coding purpose?

Click to expand...

The nature of this project is still neither simple nor cheap. I'm also not convinced it is the correct approach. The people who know about these issues are in the machine vision industry. I do not know to what extent they will be willing to share their knowledge. I was involved with one company for a brief time some 20 years ago or so. At that time we were considering special purpose computing hardware built with MSI ECL parts that implemented what was called a cyto-computer. I don't know if it ever worked, I bailed when the financing started to look shaky. As you can imagine with hundreds of ECL chips it was a bit of a power hog.

Reverse engineering tech is so advanced nowdays it's impossible to fully protect a design.

Even though it's part of my job to some extent, I have all this interlectual property, pattents bs. I don't think it does technology any good in the long term. I can see why the laws are in place and people should have the right to profit from their ideas but I think it's just gone to far.