Register to reply

Digital video image stacking

by Bostonpyramidbu
Tags: digital, image, stacking, video
Share this thread:
Bostonpyramidbu
#1
Mar2-12, 05:09 PM
P: 2
Hi folks, its been a while since I've visited.

I had an idea a few years ago I never pursued and I thought I'd run it past this group simply because not only is it beyond my education to develop but its combining both optical physics and some techniques used in cosmology.

OK so here goes

I want to build a camera that used multiple video feeds, stack the images, remove the dissimilar pixels and replace them with an average of all pixels removed from that location

Call it focus stacking but with multiple frames in real time.

Its sorta like stacking images in astronomy where your looking for anomalous movement in the frames, except this things gotta be fast.

What I'm trying to do is remove ground flutter so I can image across the atmosphere and get clear moving pictures.

I know there is a relatively simple program that stacks images and I suppose you'd not even have to average pixels if you could just get them to be translucent, but I don't really know how to reprogram the data stream.

Anyone feel like helping out a little. Could be a cool project and several peeps tell me if I can make it work it might be worth something.

cheers
Dan
Phys.Org News Partner Engineering news on Phys.org
UConn makes 3-D copies of antique instrument parts
Amazon worker piloted drone around Space Needle
Five next-generation technologies for positioning, navigation and timing
Floid
#2
Mar3-12, 12:12 PM
P: 235
Your post is kind of unclear. From what you are describing you want to do exactly what they do in astronomy but you just want it done in real time built into a camera. This would be fairly simple to do if you had access to the firmware inside of a digital camera so you could reprogram it to do that.

The bigger problem is that it sounds like you want to use it to scan across something (moving the camera). The problem here is as you move the camera every pixel is going to change so you are just going to end up bluring your video. Well, the only case in which every pixel would not change would be if you were perfectly tracking a moving object with the camera, but in this case why do you need to average the background?

I might can give more help if you clarify the following:

"What I'm trying to do is remove ground flutter so I can image across the atmosphere and get clear moving pictures."

What is ground flutter? And what is "image across the atmosphere"? Is that scanning across the horizon?

An example might help.
Bostonpyramidbu
#3
Mar3-12, 12:55 PM
P: 2
Ground flutter is the heat shimmering effect you get as heat waves rise from the a surface that is warmer or cooler than the surrounding atmosphere.

the camera would be stationary. Although it would be nice to be able to pan and tilt to follow a live target. Although I didn't think the focus stacking function would work on the go either.

what I'm trying to do is view the yellowstone wolves in real time video from great distance. Like a mile or so. With greater clarity than the manual spotting scopes used now. The limiting factor to the spotting scopes is the ground effects, IE that ground flutter blurs the image to the point were its difficult to make out what exactly the animals are doing. It would be fantastic to set up a remote binocular camera base and be able to scan the area without the system on. ( stacking function off ) Then with the target subject spotted, turn the stacking function on and get some nice clear video of what the animals are doing.

By way of explanation I've spent years guiding in Yellowstone, I've watched dozens of dedicated wildlife researchers ruin there eyes looking though a lens for twenty years, day in day out. Typical video camera's just don't have the range. So I thought of using some of the techniques found in the optical physics of astronomy.

a crude explanation of the camera system would be as follows.

two smaller telescopes set together in a frame and calibrated for parallel line of site.
two digital camera's prime focus coupled one each telescope
both video feeds going into my lap top which will run the focus stacking software.

If that basic set up works then I'd move on to include a pan and tilt base as well as maybe a laser range finder and binocular focus. Which might be the more difficult part. Not sure if binocular focus wouldn't be already taken care of with the software program.

I'd use one camera and one scope to "search" for my target, then once found, I am hoping I could turn on the other, engage the focus stacking program, and have my subjects come up real time on a nice big screen. Sit back and enjoy.

Anyway I'm just beginning this journey and any help or advice I can get would be most appreciated, Unfortunately although I have a good conceptual view of what I'd like, I simply don't posses the technical skill to pull this one off.

Cheers
Dan


Register to reply

Related Discussions
Is Digital Image Processing practical outside academia Academic Guidance 6
How to capture live video/image using Matlab without Image acq toolbox Math & Science Software 1
Advice on enhancing digital image General Discussion 10
Have digital video product idea, need EE help Electrical Engineering 6