minimising noise in digital cameras

[pellicle]

PS

Your work isn't extendable, because it's totally wrong/faulty.

this assertion is looking less likely all the time

 

[Loris Medici]

I can't comment any further w/o having the original RAW file, or the screen capture of the histogram of the unmanipulated RAW data. But, again, to me it seems that you've definitely underexposed that...

OK. I hope to do my test tonight, I guess I will have something ready by Thursday evening. I work, therefore can only do tests in the evenings.

BTW, perhaps it's a good idea provide the RAW files of my tests; I think that will satisfy even the most skeptic. (P.S. Of course, I'll request yours in return...)

P.S.2. I'm a member of the Roman Catholic Church... Culturally at least!

 

[cupcake_ham]

Loris

thanks for clearing that up, previously in another thread you'd insinuated that you teach photography.

So now we know your fields of expertise .. thanks for that.

While you are preparing your comeback experiment I've put this quick experiment together to cover that topic you alluded to above.

This is a white toilet paper roll, photographed (of course) in tungsten light (not a fluro or energy saver). Its not a photoflood, so the colour temperature will be below the proper rating of Tungsten ... still, its what most of us work with



I left the camera on Av to determine the exposure, took the one on the left with it set to tungsten, and the one on the right with it set to daylight and an 80A filter in front of the lens.

RED


GREEN


BLUE


I don't know about you, but I'm seeing more noise in that left hand side, particularly obviously in the shadows but if you look around its even in the areas which are up around the level of 200 on the white of the roll.




I agree that this is, hardly a rigorous scientific test, but it does further lend evidence to support my initial suggestion that using a filter helps minimize noise in digital cameras. But let me know if you want to borrow that toilet roll up there ... it could be helpful for you

I'm just waiting for the "whatever" to appear. Funny how it doesn't when the proof's in the image.

 

[Loris Medici]

David,

Be patient!

I won't comment further about pellicle's "unexceptionable" test. (* See the reasons below) Let's talk about mine later... (Will let you play with my RAW files too -> not for publication but your own personal examinations.)

(*) I find that test flawed because:
1. It's clear the image is was underexposed during capture, whereas all that I suggest since the beginning is "to give adequate exposure".
2. I don't believe that it was shot under usual indoor conditions. (We don't have any information about the lightsource other than it is an ordinary incandescent bulb: What's the wattage or the lumen value? How far it is from the main subject? ...) I just don't see that.

Loris

I'm just waiting for the "whatever" to appear. Funny how it doesn't when the proof's in the image.

 

[Mike1234]

Mike1234, nobody has posted "comparative" tests including our assertor... Do you often have comprehension problems?

Yes, Loris, I do have comprehensive problems due to progressive brain damage. I'll refrain from retaliatory comment.

This aside, it doesn't change the fact that one has produced at least some proof of arguement and no others. Yes, this includes me. Time is a factor here.

I look forward to seeing your examples of proof and hope they have wide dynamic range. This is my arguement... that many images have too much DR to compensate with exposure alone to preclude underexposing certain channels. Either the brightest areas in overexposed channels will clip or noise is created in underexposed channels. It's a matter of physical limitations of the sensor.

 

[pellicle]

Loris

(*) I find that test flawed because:
...
2. I don't believe that it was shot under usual indoor conditions. (We don't have any information about the lightsource other than it is an ordinary incandescent bulb: What's the wattage or the lumen value? How far it is from the main subject? ...) I just don't see that.

I would ask you to be so kind as to explain how the lumen value or the wattage of the light source makes any significant difference?

The purpose of the shutter in a camera is to allow sufficient light in to expose the sensor / film according to the light and sensitivity of the camera.

It seems to me that you are proposing that a longer shutter period would result in more noise than a shorter one? At least that is the only effect I can think of between a 10W bulb and a 1000W bulb or placing the 1000W bulb more distant from the subject.

could you please bring some clarity to this request and its basis? Without such knowledge it may be important for me to report other things which I know not to report.

PS I'm getting mighty tired of your making slanted accusations that I've stacked the test for some purpose ... without even suggesting what that purpose is.

or are you simply a troll? My bullshit meter has been going off like a gieger counter in the north of the Ukraine

reading through this thread you've done nothing other than bounce up and down like a parrot repeating evidence evidence (raaak) from one side of the perch to another, you've accused me of stacking the books and rigging what I do but through out all of this you have not put a single appropriate bit of evidence where your mouth is.

you seem to expect that I jump around when you call and then when I do something you claim that's nonsence too.

but still you do nothing more than toss around sarcasm, insults and innuendo ... if the next post from you does not contain some substance then fella .. your a sham

 

[Mike1234]

Pellicle, while he didn't specifically say so, I'm sure Loris is alluding to the "color balance" of the light from the specific bulb you used for your test(s). Usually, the dimmer the bulb the lower the color temperature which will, as you you know, makes matters worse. Color temperature, as you know by your experiments, plays a significant role in image quality and a typical 25 watt incandescant lamp vs. a similiar 500 watt will be far more difficult to correct. In effect, the brighter bulb is, generally speaking, a much higher Kelvin temperatue and closer to the sensor's best effective color range. The truly awful blue-green flourescent bulbs cause issues in the red channel.

In other words, Loris is about to shoot down his own theory.

 

[pellicle]

Hi

while he didn't specifically say so, I'm sure Loris is alluding to the "color balance" of the light from the specific bulb you used for your test(s). Usually, the dimmer the bulb the lower the color temperature which will, as you you know, makes matters worse.

good point ... so I guess that one needs a stronger filter, or blow those reds so much further out.

this is the first time I've ever written ROTFL and really mean it

sniff ... oh that's funny

 

[Bob Carnie]

Blue channel is looking pretty messy

From most work coming my way,,,,, well exposed, over exposed, under exposed, the blue is usually culprit.

I am seeing files from all types of cameras, phase , better light right down to camera phones. While the worst noise is from the lower end cameras I still see it in the big boys.

I have kind of lost interest/focus in this thread due to the constant bickering , but maybe in an attempt to put it back on track, let me see if I am right about what is being said here or being debated.

Loris is saying underexposure is causing the noise and filtering the source image to its proper colour balance at time of capture will not reduce noise.

Pellicle is saying filtering the capture will indeed lessen the noise .


Since I personally do not use a digital capture, I feel I am not competent to comment on capture techniques to reduce noise using digital cameras.
As a recipient of a lot of supplied files from various sources, I would love to know how to advise clients methods of reducing capture noise.

Personally I believe that good exposure , good colour balance with the RGB channels well nuetralized at time of capture is a good thing. Lots of experts advise Macbeth colour checkers white balance every time an setup is being set up.

From my personal experience , after quite a few images here is a simple test that know shows what grey balance is important for.

Do a test in PS from a colour neg scan that has a colour inbalance.

1. do not nuetralize at the beginning but curve shape, sharpen then enhance colours to your taste then make a large print at high resolution.
2 Do nuetralize at the beginning and do the same adjustments and same print.

You will find artifacts/noise in the non nuetralized method, you will not find noise in the properly nuetralized method , or you will find much less

So from an noobiee who does not use a digital camera I would think that getting expousure / white balance/ rgb grey point nuetralized at time of capture is a good thing.
I believe that some cameras / sensors are much better than others and when selecting a digital camera I would be more concerned about the sensors ability to see each channel correctly.
It probably is something that all the manufactures are aware of and over time will solve .

 

[pellicle]

Lois

In post #40 you say

I think it's more like you're pretty much aware of the consequence, therefore avoid it like hell...

which is interesting as in post #48 I provide exactly that supplimentary evidence

but as you go on to say:

OTOH, I would expect that you had already tested that before publishing the article!?

you're clearly implying I'm cooking the books ... buddy I bet you have not got the guts to say "I'm sorry" when you're wrong.

but then, in your mind you won't be ... so that's solved already.

meanwile in post #46 you take my words of:

so why did you bring them up, why not just dispassionately post information and supporting evidence rather than jumping up and down?

and reply to that:

That's rather something that you do way better than me; would you really like me to quote each and every instance you did that in this thread?

which is interesting, as so far I've been doing only the reasonable thing and suppling information, you on the other hand have input nothing aside from denial and sarcasm

Its classic used car salesman technique to shift the blame when in fact it has been you who have been making the slurs on me.

the funny part is that while you were whinging for:

The evidence "boomerang" hits right back to you; would you please kind enough to expand (or better complete) your "scientific" article with some "evidence"?

I was taking photographs for you (you ingrate)

which I posted in post #48 without any such boomeranging

... now that I see this relationship (as I had missed your post till reviewing now) I can understand your squealing about "cooking the books"

so in post#54

you begin setting up a denial

2. I don't believe that it was shot under usual indoor conditions. (We don't have any information about the lightsource other than it is an ordinary incandescent bulb: What's the wattage or the lumen value? How far it is from the main subject? ...) I just don't see that.

without supplying a single fact or even basis for criticism.

Lois, I advise you to quit now, you're making a spectacle of your self

 

[pellicle]

Hi Bob

Blue channel is looking pretty messy

just a bit

I have kind of lost interest/focus in this thread due to the constant bickering , but maybe in an attempt to put it back on track, let me see if I am right about what is being said here or being debated.

Loris is saying underexposure is causing the noise and filtering the source image to its proper colour balance at time of capture will not reduce noise.

Pellicle is saying filtering the capture will indeed lessen the noise .

yep ... that's about it.

 

[Loris Medici]

I'm "alluding" that you've have arranged a test(!) with tuned parameters so that it gives biased results conforming your assertion. In the examples you've provided, I don't see anything that remotely looks like being shot in usual indoor lighting conditions for instance.(1)

And it's not me who makes spectacle of himself, it's you: you've given clear evidence of lack of "elementady knowledge" of how to properly expose when shooting with digital cameras...(2)

Anyway, pellicle (and his howlers) you'll see what I mean later, a little more patience...

 

[Mike1234]

Hi Loris,

I guess you're calling me one of pellicle's "howlers". Let's try to keep this civil, okay?

The facts are: 1. The further the light source's color temperature is from what a sensor is designed for and, 2. The more improper the exposure is... the worse the noise and/or clipping will be in certain channels. It's a scientific fact. For those with a lot of experience correcting (or attempting to correct) these unwanted artifacts it's accepted as common knowledge.

 

[donbga]

Hi Loris,

I guess you're calling me one of pellicle's "howlers". Let's try to keep this civil, okay?

The facts are: 1. The further the light source's color temperature is from what a sensor is designed for and, 2. The more improper the exposure is... the worse the noise and/or clipping will be in certain channels. It's a scientific fact. For those with a lot of experience correcting (or attempting to correct) these unwanted artifacts it's accepted as common knowledge.

http://www.flickr.com/photos/canbalci/1371910755/

 

[Loris Medici]

The culprit in all of the images we've seen so far is improper exposure / lack of knowledge of giving proper exposure.

I say, (1) when proper exposure is given there shouldn't be any perceptible difference between the image taken w/o CC filter and the one taken w/ CC filter, (2) and with CC filter you loose 2 stops of light which makes your shooting more difficult (forcing to use a tripod -> where you could shoot handholding the camera or forcing you to increase ISO -> where using a tripod isn't possible, which also is something totally against the original purpose of keeping noise levels down) and finally (3) using a USD 10 CC filter (pellicle's definition, not mine...) over a good lens will probably turn it into a mediocre one.

Will prove all of above with test images...

...
Loris is saying underexposure is causing the noise and filtering the source image to its proper colour balance at time of capture will not reduce noise.
....

 

[Mike1234]

http://www.flickr.com/photos/canbalci/1371910755/

Nice work there, Don. But is that person a member here? And is there some discussion there regarding before-capture color correction and avoiding noise and clipping that I missed?

EDIT: Is the point regarding the double exposure in order to capture the overall scene including the bright red burner? Okay, now I understand the connection. Your point is that, given the photographer's available lighting (flash + the burner) there was no way to capture that image in a single exposure because the DR was too great? Good point and good example. He could have accompished the same effect in a darkened room with the modeling lights switched off and used a slower shutter speed but... who wants to work in the dark?

 

[pellicle]

(3) using a USD 10 CC filter (pellicle's definition, not mine...) over a good lens will probably turn it into a mediocre one.

gee ... decades of photographers using Cokin and Hoya filters turning their good lenses into mediocre ones. If only they knew what they were doing...

well I guess we all learn from the participation we are privileged to share here.

 

[Mike1234]

gee ... decades of photographers using Cokin and Hoya filters turning their good lenses into mediocre ones. If only they knew what they were doing...

well I guess we all learn from the participation we are privileged to share here.

And let's not forget good ol' Kodak Wratten gels.

 

[pellicle]

don, thats brilliant

deserves to be put in

 

[pellicle]

EDIT: Is the point regarding the double exposure in

I thought it was about blowing steam

 

[donbga]

And let's not forget good ol' Kodak Wratten gels.

Wratten gels in proper condition are fine. And they are no longer Kodak, BTW - Tiffen or Lee.

Cokin filters are crap and I'm not saying that to be contrary.

Hoya is a mixed bag.

Basically you get what you pay for. Avoid the use of filters unless they are high quality. There is a reason I spent almost $160 for a UV filter that I use for my Canon 70-200 f/2.8 L zoom - QUALITY! Why spend a lot of money on a lens just to put a cheap piece of crap filter in front of it.

And yes I do keep UV 0 filters on my lenses when shooting digital. We can open that can of worms up if people want to since some of you all think I'm such a contentious bastard moderator.

Don

 

[donbga]

don, thats brilliant

deserves to be put in

I'm really just trying to add a little humor! Now I have to go and do real work.

 

[Mike1234]

Wratten gels in proper condition are fine. And they are no longer Kodak, BTW - Tiffen or Lee.

What about these? http://www.bhphotovideo.com/c/search?Ntt=gelatin+filters&N=4291384683

We can open that can of worms up if people want to since some of you all think I'm such a contentious bastard moderator.

I never wrote anything like that. I've seen a couple of replies from others stating or insinuating that though. I'll say this last post seems a bit confrontational though. Why? :confused:

 

[pellicle]

Don

Wratten gels in proper condition are fine. And they are no longer Kodak, BTW - Tiffen or Lee.

Cokin filters are crap and I'm not saying that to be contrary.

certainly Lee are better regarded

 

[pellicle]

Don

I know you have work and everything, but :

Basically you get what you pay for. Avoid the use of filters unless they are high quality. There is a reason I spent almost $160 for a UV filter that I use for my Canon 70-200 f/2.8 L zoom - QUALITY! Why spend a lot of money on a lens just to put a cheap piece of crap filter in front of it.

And yes I do keep UV 0 filters on my lenses when shooting digital. We can open that can of worms up if people want to

if you care to start another thread on that I'd be keen to read it. I've never done any significant testing on the subject and try not to use filters unless they're needed (like when I use HIE on my SLR).