noviceCVer's profile - activity

awesome idea... might be highly dependent on light conditions, but I'd be keen to learn more about your work.

hi just wondering if anyone was in Boulder, CO tat could help with OPEN CV stuff and length measurements from Stereo cameras.

Cheers

Yes sorry for my bad wording. I mean that using two video cameras recording simultaneously (obviously) if the calibration object moves (which it needs to do calibration right, different orientations) then the left and right cameras have to be synchronized to the same point in time? I guess Ill just do that as it cant hurt to anyway. Thanks for your help.

Hi Mathieu,

Quick question when running a stereo calibraiton after returning intrinsic properties from calibration. Do the images for stereo need to be from the same point in time in space?

Hi.

So my advice is coming from a photography/ filming standpoint, so not sure if this works with what you need.

But the best cameras are built by those that build cameras and have done for a long time. Of course right? The reason being is that a lot of your image quality comes form the glass that is in front of the ccd/cmos or in the old days film. Terrible glass, terrible image. Glass also has its resolution capabilities and I believe they call it lines pairs per mm. Maybe you;ve heard of Carl zeiis, Leica or Hasselblad? No doubt you've heard of Canon and Nikon. All these guys know how to make glass. In regards to video cameras, then canon have a great combination of glass knowledge and image processing knowledge.

The next thing to consider is your ccd/cmos. The world got all caught up in the megapixel race, and making cameras really small. Which meant ccd/cmos's got really small, and also they crammed a lot of pixels on them. That is a false economy in regards to your image quality. What we need, and again Canon has turn 180 on the industry for the better, at least for video where our TV's and output devices are only 1080, they have built bigger sensors with less but larger pixels, in fact just as many as we need 2 megapixels, 1920 x 1080. This means that an individual pixel is more sensitive to light, which means less noise, which means sharper images even as the light drops. Flash back to film. Film speed was based on the size of the silver particles in the film. 800 film had larger silver particles which react quicker and sooner to less light than did say 100 speed film which has small grains of silver in them.

So, if you have a budget and OpenCV works with them in terms of live uploads etc, then look at Canons new HFG range. Starting around 1k. Not sure if some of the lower models have the same sensor, I think they might? But here you get great glass, amazing and big cmos sensors, and also HDMI out. Guess they will be heavier than webcams and things like that so they may not suit for many robotic projects.

Otherwise certain things to look for.

Glass lenses not plastic. If the lens is designed by a well known lens manufacturer even better. The bigger the glass the more light is let in, which means better images Just enough pixels for the job. On bigger sized sensors. with bigger pixels (sometimes called pixel pitch measured in um).

Guess its all food for thought, not sure if its helpful at all.

Hi Mathieu and Steve. Thanks for your input. I understand that mm accuracy may not be possible, that is OK, as the objects will range from a few cm to up to +10 m. More important is consistency.

I noticed that in the code there is no place for inputting pixel size of the sensor. is this roughly calculated by the censor length x width inputs and resolution dimensions (ie. 1920 x 1080) for the calibration code? Seems as if this is a crucial part of obtaining accurate legths as it is your only known dimension. I guess aside form the checker board. But if pixel size is wrong it would also change focal length values.

Ill see how i go this week, and hopefully my non-programmers brain will hack its way through the OpenCV language.

really appreciate your help.

Hi All,

I apologize in advance for my complete inability to really use this stuff so bare with me.

I think I have a simple things compared to all the really amazing things you all are doing in here. Alli want to do is measure the length of objects using stereo-video cameras. This is for use for wildlife so the animals and cameras are oftening moving. I know I need to calibrate the cameras, both as individuals (distortion, focal length, etc) and as a pair (i.e, distance between etc, rotation and angle). I assume a checker board is ok for that? By the way these need to be really accurate measurements within a few mm if not one or two mm.

Then i need to figure out how clicking on the ends of the objects on the left and right cameras on a computer image in 2d space is converted to 3d real-world space.

Is this easy to do? There is no automation required, no tracking, no auto identification of the ends of the objects etc. The user will play the footage, then pause it and click on the two ends of the object/animal in the left and right camera, and i want its length returned to me. I guess also with that it would give me distance from cameras, and position relative to cameras given that I think the math is based around trigonometry?

If anyone can help point me in the right direction that would be awesome. I plan on using some high level cameras such as sony or canon consumer level HD cams. I've been told that gopros wouldn't be any good because they are so wide? Thats a different issue however.

Thanks all.

Your friendly novice

RD