beautybelow
Contributor
I would use a filter for video too but I didn't think he was asking/debating about video.
Raw, Red filters & WB...oh my!
- Thread starter cpanddp
- Start date
Please register or login
Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.
Benefits of registering include
- Ability to post and comment on topics and discussions.
- A Free photo gallery to share your dive photos with the world.
- You can make this box go away
spoolin01
Contributor
I'm just coming at this from the simple spectroscopy/histogram side, and I don't know much about camera physics or post-processing. I appreciate the comments and indulgence while I try to understand this.
The water is a blue filter that cuts out red light, so you take the red light that you let in through the lens and differentially amplify it with software. You've let in less red light than you possibly could, because if you let in more red, you'd have too much blue. You have to make that trade-off because the exposure settings can't differentiate red from blue, and there's no option on the hardware side (except for filters) to do that either - no red ISO vs blue ISO.
So you've got a low intensity image (the red one) that you bump up with software - that increases the noise, because detectors have sensitivity threshholds, and you've underexposed the red ones.
If on the other hand, you choose to filter the blue light similarly to what the water has done to the red, such that the incident light histogram is now (largely) restored to above-water spectral distribution, then increase the exposure to restore intensity lost to the additional filtering - now your red and blue detectors are getting more balanced input. Other considerations aside (depth of field from opening the iris, say), the detectors don't know what machinations created their light, just that the spectral distribution is better red-blue balanced, and that overall exposure is correct.
Given the noise issue (using that term broadly to mean the quality of the data), how can than not be a better starting point, that's the part I haven't yet gotten. I don't think it matters whether the [-]blue[/-] redfilter exactly complements the red filtration done by the water - it only matters which incident histogram more closely matches what you want (since software will do the rest), and whether that makes for a better image in the end.
The water is a blue filter that cuts out red light, so you take the red light that you let in through the lens and differentially amplify it with software. You've let in less red light than you possibly could, because if you let in more red, you'd have too much blue. You have to make that trade-off because the exposure settings can't differentiate red from blue, and there's no option on the hardware side (except for filters) to do that either - no red ISO vs blue ISO.
So you've got a low intensity image (the red one) that you bump up with software - that increases the noise, because detectors have sensitivity threshholds, and you've underexposed the red ones.
If on the other hand, you choose to filter the blue light similarly to what the water has done to the red, such that the incident light histogram is now (largely) restored to above-water spectral distribution, then increase the exposure to restore intensity lost to the additional filtering - now your red and blue detectors are getting more balanced input. Other considerations aside (depth of field from opening the iris, say), the detectors don't know what machinations created their light, just that the spectral distribution is better red-blue balanced, and that overall exposure is correct.
Given the noise issue (using that term broadly to mean the quality of the data), how can than not be a better starting point, that's the part I haven't yet gotten. I don't think it matters whether the [-]blue[/-] redfilter exactly complements the red filtration done by the water - it only matters which incident histogram more closely matches what you want (since software will do the rest), and whether that makes for a better image in the end.
Last edited:
spoolin01
Contributor
If you're too shallow or the incident light is spectrally balanced, like using a strobe and red filter at the same time, the filter reduces the balance rather than making it better, I've seen that happen a lot with shallow water video.
beautybelow
Contributor
Do you just want to debate the physics of using a filter or do you want to figure out the practicality of using a filter in still photography underwater?
You lost me on the water is blue...
What kind of camera are you talking about using? If it is one you can slip a wet filter over the lens port then go for it. If you like it yippee and if you don't, take it off.
If you are talking about using a threaded filter on the end of your lens then I would feel too restricted, not being able to take it off during the dive.
You lost me on the water is blue...
What kind of camera are you talking about using? If it is one you can slip a wet filter over the lens port then go for it. If you like it yippee and if you don't, take it off.
If you are talking about using a threaded filter on the end of your lens then I would feel too restricted, not being able to take it off during the dive.
under water
Contributor
I am of the opinion that MY videos are optimized when I use a red filter. I have however seen some wonderfully balanced videos that apparently only relied on proper White Balance.
OK. I realize that was about video, not stills.
If I am shooting stills as well and cannot remove the red filter, I will use RAW and the same argument that says "you don't need a filter" lets you adjust the red out in post processing.
From a densitometric perspective, it does make sense that you can get a better image if you use a red filter, White balance and RAW together for ambient light pix. If you can minimize the delta between the sensors and get to a more neutral area via a red filter, then you are not going to need to boost red (actually attenuate blue & green)as much in post processing (whether on the PC or in the camera).
It is important that you white balance if you are using a red filter if you are saving JPEGs. You can adjust RAW later, but I think it is way better to optimize during shooting so that your JPEG is reasonable. If you don't save JPEGs then you don't need to White Balance.
In practice though, most use a strobe that is close to daylight and the foreground will come out too red if you use a strobe and a red filter, unless you white balance for the combo (again, WB only useful for JPEGs).
Some folks have used blue filters on their strobes and red filters to try and balance everything, but that is a pain and I doubt many folks do it.
An example of a strobe picture and a red filter (using RAW)
After color balance in Lightroom.
Another thread that discussed this topic
http://www.scubaboard.com/forums/tips-techniques/366567-red-filters.html
OK. I realize that was about video, not stills.
If I am shooting stills as well and cannot remove the red filter, I will use RAW and the same argument that says "you don't need a filter" lets you adjust the red out in post processing.
From a densitometric perspective, it does make sense that you can get a better image if you use a red filter, White balance and RAW together for ambient light pix. If you can minimize the delta between the sensors and get to a more neutral area via a red filter, then you are not going to need to boost red (actually attenuate blue & green)as much in post processing (whether on the PC or in the camera).
It is important that you white balance if you are using a red filter if you are saving JPEGs. You can adjust RAW later, but I think it is way better to optimize during shooting so that your JPEG is reasonable. If you don't save JPEGs then you don't need to White Balance.
In practice though, most use a strobe that is close to daylight and the foreground will come out too red if you use a strobe and a red filter, unless you white balance for the combo (again, WB only useful for JPEGs).
Some folks have used blue filters on their strobes and red filters to try and balance everything, but that is a pain and I doubt many folks do it.
An example of a strobe picture and a red filter (using RAW)
After color balance in Lightroom.
Another thread that discussed this topic
http://www.scubaboard.com/forums/tips-techniques/366567-red-filters.html
Last edited:
Gilligan
Contributor
spoolin01
Contributor
I wasn't thinking about filters inside the housing, just the point and shoot type where the filter is on the wet side. I'm interested in the physics, optics, sensor-response, and post-production of underwater stills, looking for a technical explanation of why those would all work better jointly - image quality-wise - w/o use of filters, when it seems the raw material would start out in better balance with use of a red filter. If it's just a matter of convenience, I can see that. But I'd also like to understand the quality effect.
Last edited:
spoolin01
Contributor
Here's a discussion of filters and UW photo I googled up.
Filters and Ambient Light Photography :: Wetpixel.com
under water - you and the wet pixel guys mention white balance. Does that mean you're
a- shooting jpeg
b- talking about white balancing during post-processing (do you use the grey card in that way also?)
c- referring to some use of white balance in the context of RAW that I haven't heard about
Filters and Ambient Light Photography :: Wetpixel.com
under water - you and the wet pixel guys mention white balance. Does that mean you're
a- shooting jpeg
b- talking about white balancing during post-processing (do you use the grey card in that way also?)
c- referring to some use of white balance in the context of RAW that I haven't heard about
under water
Contributor
a. White Balance while shooting is needed if you shoot JPEG combined with RAW or just JPEG, OR (especially) if you are shooting video.
b. Post processing color balance using the dropper to adjust the color is also referred to as White Balance and is what you do to balance those RAW pictures. Usually it is a first step, sometime combined with sharpening and other changes to the curves.
c. No need to use WB while shooting RAW only. It doesn't change what the sensor gets and stores, just how it processes it for JPEGs. Which is why a filter can be useful for ambient conditions for the reasons stated before.
Gilligan
Contributor
I guess the only way to know for sure is to take photos at varying depths with and without a red filter in RAW mode for comparison. I would gladly do it but I don't have a red filter anymore.
Similar threads
