Approve the Cookies
This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies. Read More.
OK
Index  •   • New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
New posts  •   • RTAT  •   • 'Best of'  •   • Gallery  •   • Gear  •   • Reviews
Register to forums    Log in

 
FORUMS Post Processing, Marketing & Presenting Photos RAW, Post Processing & Printing
Thread started 08 Jun 2017 (Thursday) 18:35
Prev/next
sponsored links
(this ad will go away when you log in as registered member)

PP Heat Haze Correction Software

 
Pagman
I just hold the thing :-)
Pagman's Avatar
Joined Dec 2011
Jun 08, 2017 18:35 |  #1

Its a shame someone cant come up with an ap to run inside something like LR that can correct heat haze distortion, a kind of pixal neighbour predictive adjustment control.


P.


Nikon D7100, Nikkor 300 f4 IF ED :-)

LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as registered member)
BigAl007
Cream of the Crop
Joined Dec 2010
Repps cum Bastwick, Gt Yarmouth, Norfolk, UK.
Jun 08, 2017 19:25 |  #2

I cannot see how it could possibly be done, unless you can somehow fire a shed load of laser beams at the target first, to measure the refraction effects directly. Actually I would be better still if the lasers could be mounted on the subject, and just fired at you, since then they would only be loaded with a one way refraction map. You would also want to use at least three colours, matched to the peak wavelength of each of the three colours of the Bayer CFA. The real issue for this to work effectively is that you need to be running the lasers during the exposure.

If you had the lasers coming into the camera you would know the Modulation Transfer Function of the camera system, and you would know the location of the lasers, so any difference in the recorded image of the lasers is down to the Modulation Transfer Function of the atmosphere, and this should give you the means of calculating the MTF.

Unfortunately the MTF of the atmosphere is constantly changing, so it would have to be measured during the exposure time itself. So not really practical from a photography point of view.

I know that at one of the large optical telescopes in the high Andes mountains they are using a green (I think) laser based system to do this sort of thing to help reduce star shimmer. Of course being over 12000 feet above sea level they do have quite an advantage, since they are above quite a bit of the atmosphere.

The other option would be to know exactly what the image should look like, so that again you can build an instant MTF of the atmosphere. Of course that would make the system pointless, since you would already have the corrected image.

Alan


My Flickr (external link)
My new Aviation images blog site (external link)

LOG IN TO REPLY
Bassat
"I am still in my underwear."
Bassat's Avatar
6,544 posts
Joined Oct 2015
Bourbon, Indiana - USA
Post has been edited 5 months ago by Bassat.
Jun 08, 2017 19:34 |  #3

Technologically, it would be simple. Shoot two frames, preferably from a tripod. Everything will be the same EXCEPT the heat/distortion waves. As the only difference will be the heat distortions, software could easily subtract it from the processed product, leaving a proper photo. This is exactly how dark-frame subtraction works for long exposure noise removal. In a sense, it has already been done.

If you write the software, and make $$$-millions-$$$, come to Indiana and buy me a beer, on a really hot summer day. :)


Tom

LOG IN TO REPLY
gjl711
They have pills for that now you know.
gjl711's Avatar
53,203 posts
Joined Aug 2006
Deep in the heart of Texas
Jun 08, 2017 19:34 |  #4

I wonder if you used the astro software that combines many images into one stack while removing distortion. That would be an interesting experiment.


Not sure why, but call me JJ.
I used to hate math but then I realised decimals have a point.
.
::Flickr:: (external link)
::Gear::

LOG IN TO REPLY
Pagman
THREAD ­ STARTER
I just hold the thing :-)
Pagman's Avatar
Joined Dec 2011
Post has been edited 5 months ago by Pagman.
Jun 08, 2017 19:51 |  #5

What prompted me to ask is I recently took some shots of static aircraft at ground level across an averagly long distance - about 2000ft not excessivly hot, but there was a lot of heat distortion even affecting the focus, and in reality the photos are only bin worthy, but Im keeping them as a reminder of the day out.

I have tried to do some work to them in LR but as to be expected I can not do anything with the dsitortion.


P.


Nikon D7100, Nikkor 300 f4 IF ED :-)

LOG IN TO REPLY
Pagman
THREAD ­ STARTER
I just hold the thing :-)
Pagman's Avatar
Joined Dec 2011
Post has been edited 5 months ago by Pagman.
Jun 08, 2017 20:02 |  #6

For example ok mid afternoon and subject matter for that shot was positioned across the runway and this was the only vantage point(fire gate) to get a shot of the full B1 Bomber, the distance between me and the plane was something like 2000feet.


P.

HOSTED PHOTO
please log in to view hosted photos in full size.

Nikon D7100, Nikkor 300 f4 IF ED :-)

LOG IN TO REPLY
kirkt
Cream of the Crop
5,394 posts
Joined Feb 2008
Philadelphia, PA USA
Post has been edited 5 months ago by kirkt.
Jun 08, 2017 20:15 |  #7

I think I know to what you are referring - refraction due to heat rising from the ground? I'm not sure how you would go about correcting it - the refraction occurs from local fluctuation in air temperature, locally changing the air's index of refraction - as the heated air rises and mixes with the surrounding air, it is constantly changing its local temperature field and optical properties. It is not random, like noise, so I would doubt you could "average" it out. I'm not sure how you would "subtract" it - how do you know what to subtract? If you compared two successive images and subtracted one from the other, you would potentially be able to identify the areas that are affected by changes like this (they would show up in the difference image compared to black pixels from identical areas in each image). But what would you replace the pixels in those affected areas with?

Maybe, if the plane was stationary, you could shoot with a 10 stop ND filter (or even stronger) and use a long shutter speed to "smooth" out the fluctuating air. It would probably produce a slightly blurred image in the areas where the shimmer is most obvious, but it may "remove" the distracting shimmer distortion that is frozen in the image at higher shutter speeds - sort of like shooting flowing water.

That's an interesting problem to solve!

kirk


Kirk
---
images: http://kirkt.smugmug.c​omexternal link

LOG IN TO REPLY
Pagman
THREAD ­ STARTER
I just hold the thing :-)
Pagman's Avatar
Joined Dec 2011
Jun 08, 2017 20:33 |  #8

kirkt wrote in post #18374274 (external link)
I think I know to what you are referring - refraction due to heat rising from the ground? I'm not sure how you would go about correcting it - the refraction occurs from local fluctuation in air temperature, locally changing the air's index of refraction - as the heated air rises and mixes with the surrounding air, it is constantly changing its local temperature field and optical properties. It is not random, like noise, so I would doubt you could "average" it out. I'm not sure how you would "subtract" it - how do you know what to subtract? If you compared two successive images and subtracted one from the other, you would potentially be able to identify the areas that are affected by changes like this (they would show up in the difference image compared to black pixels from identical areas in each image). But what would you replace the pixels in those affected areas with?

Maybe, if the plane was stationary, you could shoot with a 10 stop ND filter (or even stronger) and use a long shutter speed to "smooth" out the fluctuating air. It would probably produce a slightly blurred image in the areas where the shimmer is most obvious, but it may "remove" the distracting shimmer distortion that is frozen in the image at higher shutter speeds - sort of like shooting flowing water.

That's an interesting problem to solve!

kirk


Thanks kirk, yes it is interesting and probably a problem for many who shoot planes across distances with thermal heat haze problems, we have to use long-ish lenses (mine is fairly decent nikkor 300 f4 prime).


P.


Nikon D7100, Nikkor 300 f4 IF ED :-)

LOG IN TO REPLY
Dan ­ Marchant
Goldmember
Dan Marchant's Avatar
Joined Oct 2011
Where I'm from is unimportant, it's where I'm going that counts.
Jun 08, 2017 21:51 |  #9

Bassat wrote in post #18374247 (external link)
As the only difference will be the heat distortions, software could easily subtract it from the processed product, leaving a proper photo. This is exactly how dark-frame subtraction works for long exposure noise removal. In a sense, it has already been done.

But it's not the same at all and wouldn't work. Dark-frame subtraction captures an accurate image with sensor noise and a blank image which contains no image data, only sensor noise. This works exactly because the test image is a dark-frame - it has no image data except the noise so the software knows to remove that same noise from the image file.

With heat haze you have a scene that is distorted (so no accurate frame). Taking two shots will just give you two shots that are distorted (and different). They are in effect just two images - software won't know they are distorted because there is no accurate frame and no frame with just distortion. Even if software could analyse one frame to work out which area is distortion (which it can't) it wouldn't be any use because the other image is differently distorted so applying the data from the other frame wouldn't help. It would simply give you mush.


Dan Marchant
Website/blog: danmarchant.com (external link)
Instagram: @dan_marchant (external link)
Gear Canon 5DIII + Fuji X-T2 + lenses + a plastic widget I found in the camera box.

LOG IN TO REPLY
gjl711
They have pills for that now you know.
gjl711's Avatar
53,203 posts
Joined Aug 2006
Deep in the heart of Texas
Jun 08, 2017 22:03 |  #10

Dan Marchant wrote in post #18374310 (external link)
With heat haze you have a scene that is distorted (so no accurate frame). Taking two shots will just give you two shots that are distorted (and different). They are in effect just two images - software won't know they are distorted because there is no accurate frame and no frame with just distortion. Even if software could analyse one frame to work out which area is distortion (which it can't) it wouldn't be any use because the other image is differently distorted so applying the data from the other frame wouldn't help. It would simply give you mush.

I believe that the astro software uses dozens or hundreds of frames usually taken from video and does it's magic. I've never tried using it on a non astro image but the technology should be similar. It would be an interesting experiment.


Not sure why, but call me JJ.
I used to hate math but then I realised decimals have a point.
.
::Flickr:: (external link)
::Gear::

LOG IN TO REPLY
Dan ­ Marchant
Goldmember
Dan Marchant's Avatar
Joined Oct 2011
Where I'm from is unimportant, it's where I'm going that counts.
Jun 09, 2017 01:36 |  #11

gjl711 wrote in post #18374313 (external link)
...I've never tried using it on a non astro image but the technology should be similar.

No, it's not. What is being done in Astrophotography is completely different to the scenario the OP is facing.

To explain we substitute a truck for the heat haze. Take a photo of your scene but there is a truck (#1) in front of the aircraft. There is no data to tell you what is behind the truck so software can't reconstruct the image to show the aircraft without the truck. The truck moves but as you take a second picture truck #2 passes in front of the aircraft. Third shot the same with truck #3.... in each frame the image of the aircraft is "distorted" by a different truck. It doesn't matter how many frames you take because every one has a different truck. The software can't combine all the images of trucks together to calculate what the aircraft behind looks like because there is no data.

Replace the trucks with heat haze. It doesn't matter how many images you take because the haze will be there distorting the image. In each shot it may be different but just like multiple trucks, combining multiple distortions just gives you distortion. The only way for software to calculate what is behind the distortion would be to accurately map the distortion. For an explanation of why that can't be done see Alan's excellent post above.


Dan Marchant
Website/blog: danmarchant.com (external link)
Instagram: @dan_marchant (external link)
Gear Canon 5DIII + Fuji X-T2 + lenses + a plastic widget I found in the camera box.

LOG IN TO REPLY
evelakes
Goldmember
evelakes's Avatar
Joined Feb 2013
Netherlands Wijchen
Jun 09, 2017 02:46 |  #12

Pagman wrote in post #18374268 (external link)
For example ok mid afternoon(couldn't do anything about that I wasn't driving) and subject matter for that shot was positioned across the runway and this was the only vantage point(fire gate) to get a shot of the full B1 Bomber, the distance between me and the plane was something like 2000feet.


P.


thumbnailHosted photo: posted by Pagman in
./showthread.php?p=183​74268&i=i219566786
forum: RAW, Post Processing & Printing


Natural effect on the lens and sensor those heat waves, because our eyes are not so effected by this and we see a much clearer image in real time.
The only thing to avoid this is to wait on better light/weather conditions. No software can penetrate those layers.


A camera is a tool for learning how to see without a camera.
flickr (external link)https://www.flickr.com​/photos/97611076@N06/ (external link)

LOG IN TO REPLY
BigAl007
Cream of the Crop
Joined Dec 2010
Repps cum Bastwick, Gt Yarmouth, Norfolk, UK.
Jun 09, 2017 03:33 |  #13

evelakes wrote in post #18374390 (external link)
Natural effect on the lens and sensor those heat waves, because our eyes are not so effected by this and we see a much clearer image in real time.
The only thing to avoid this is to wait on better light/weather conditions. No software can penetrate those layers.


Actually our eyes see just as much distortion, or would if it were possible to have both the naked eye and optical magnification super imposed in one location. Adding magnification simply magnifies the distortions by the same amount.

As a competetive target rifle shooter and coach, as well as a photographer, for many years I have spent a lot of time studying and thinking about mirage, as the effect is known.

The main issue with mirage as has been repeatedly pointed out is that it is completely random, and constantly changing. Although it is totally random in its nature, it also picks up larger scale fluctuations from the atmosphere.

For example wind direction and strength will have a very strong effect on the direction the ripples will seem to flow in. Generally the ripples will flow in the direction of the breeze. If the wind is directly towards or away from you, or there is no wind it will tend to "boil" upwards. Wind flow can move the apparent location of an image by over 1/4" at only 25 yards. Unfortunatly this is not a fixed angle, it grows with distance. So at 25 yards it might be 1 Minute of Arc, while at 200 yards it gets you closer to 2 MoA movement.

When rifle shooting you have to add this optical offset to the amount the wind will also move the bullet as it travels.

At least there is a wind speed limit to when mirage will usually form. Over a dry grass surface mirage will not usually form at wind speeds above about 12 mph. This may change a bit over different types of surface, as surface conditions have a big influence over mirage strength.

One other really important observation I have made about mirage is that the higher the quality of your optics, the sooner you will actually pick up the precence of mirage. Really good high quality high magnification optics can pick up mirage on even a cold day, if it is sunny.

Alan


My Flickr (external link)
My new Aviation images blog site (external link)

LOG IN TO REPLY
Pagman
THREAD ­ STARTER
I just hold the thing :-)
Pagman's Avatar
Joined Dec 2011
Jun 09, 2017 09:54 |  #14

BigAl007 wrote in post #18374402 (external link)
One other really important observation I have made about mirage is that the higher the quality of your optics, the sooner you will actually pick up the precence of mirage. Really good high quality high magnification optics can pick up mirage on even a cold day, if it is sunny.

Alan


I guess the lens I use could fall into that category.


P.


Nikon D7100, Nikkor 300 f4 IF ED :-)

LOG IN TO REPLY
davesrose
Goldmember
2,854 posts
Joined Apr 2007
Atlanta, GA
Jun 09, 2017 16:20 |  #15

It's always amazing to me how many computer graphics research projects never become mainstream (or are picked up by Adobe, Google, etc). I have seen some interesting presentations at Siggraph from CG PhDs showing some amazing selection tools or algorithms that automatically change the aspect ratio of an image without distorting the subject. Looks like there currently is quite a few papers on eliminating haze from photos. Of course, there's only so much detail the algorithms may be able to provide:

Method and system for removal of fog, mist, or haze from images and videos (external link)

Haze effect removal from image via haze density estimation in optical model (external link)

Something different, but also a fairly recent cool photo filter presented at Siggraph:

This Algorithm Magically Removes Reflections from Images (external link)


Canon 5D mk III , 7D mk II
EF 135mm 2.0L, EF 70-200mm 2.8L IS II, EF 24-70 2.8L II, EF 50mm 1.4, EF 100mm 2.8L Macro, EF 16-35mm 4L IS, Sigma 150-600mm C, 580EX, 600EX-RT, MeFoto Globetrotter tripod, grips, Black Rapid RS-7, CAMS plate and strap system, Lowepro Flipside 500 AW, and a few other things...
smugmug (external link)

LOG IN TO REPLY
sponsored links
(this ad will go away when you log in as registered member)

2,654 views & 2 likes for this thread
PP Heat Haze Correction Software
FORUMS Post Processing, Marketing & Presenting Photos RAW, Post Processing & Printing


Not a member yet? Click here to register to the forums.
Registered members get all the features: search, following threads, own gear list and ratings, likes, more forums, private messaging, thread follow, notifications, own gallery, settings, view hosted photos, own reviews and more...


AAA

Send feedback to staff    •   Jump to forum...    •   Rules    •   Index    •   New posts    •   RTAT    •   'Best of'    •   Gallery    •   Gear    •   Reviews    •   Polls

COOKIES DISCLAIMER: This website uses cookies to improve your user experience. By using this site, you agree to our use of cookies. Privacy policy and cookie usage info.

POWERED BY AMASS 1.4version 1.4
made in Finland
by Pekka Saarinen
for photography-on-the.net
Spent 0.00117 for 6 database queries.
PAGE COMPLETED IN 0.04s
Latest registered member is J4T4lyfe
716 guests, 272 members online
Simultaneous users record so far is 6106, that happened on Jun 09, 2016