-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
A few new IR images
- Thread starter Diane B
- Start date
Meant to mention that the first image is done to simulate HIE IR film.
Dogleader35037
Veteran Member
Diane I really like the third one, starting to see what the IR does to trees and clouds, that effect I like. DL
H
hugh
Guest
In the first, the leaves seem over saturated, or something. They don't have that clean, crisp IR whiteness.
The second might have worked better without the tree.
I'm with DL, I like the IR effect in the third picture.
hs
Hugh, the first is a simulation of the Kodak HIE film--which has a 'glow' and a 'softness' to it. Larry Sabo, who uses HIE film, has the technical data as to how to achieve this on his site. Its sort of the opposite, it seems to me, of the other 2 B/W films available. I also like the images John Paduano has gotten with the Kodak film (he refers to 'halos' and 'glow') in "The Art of Infrared PHotography". My original is quite crisp--but I altered the effect in post.
Thanks for the comments.
Diane
John Bower
Active member
Diane,
I've always been a fan of B&W. Very nice images, reminds me of some of my grand father's photos. When I see B&W digital images I can't help but think of how much better they might look if they weren't filtering out 67% to 75% of the photons, then "faking" them back in with interpolation. Of course then you turn around and throw out a bunch of that information to get back to black and white... Just think, a 3MP color camera would be more like a 12 MP B&W camera if you took away the Bayer filter.
John Bower
I've always been a fan of B&W. Very nice images, reminds me of some of my grand father's photos. When I see B&W digital images I can't help but think of how much better they might look if they weren't filtering out 67% to 75% of the photons, then "faking" them back in with interpolation. Of course then you turn around and throw out a bunch of that information to get back to black and white... Just think, a 3MP color camera would be more like a 12 MP B&W camera if you took away the Bayer filter.
John Bower
I don't honestly follow you. I don't shoot in b/w digital, but in color using an IR filter and not sure where the interpolation comes in. Perhaps I'm just not understanding you.
LMC54
Forum Pro
His saying there isn't enough tonal range in IR pics. I'm with him on this. IR has its place if its done right.
adm #56117
Member
- Messages
- 10
- Reaction score
- 0
Nice pics Diane !Meant to mention that the first image is done to simulate HIE IR film.
I'm an almost IR photographer ! (Just waiting for my Hoya R72 to arrive) As soon as I get it, I should be posting IR pics too....
Al.
John Bower
Active member
Your G1 is nominally listed as a 3MP camera, or 3 million pixels.
Assuming it has an RGB Bayer CFA (color filter array), on the sensor there are 750,000 red pixels, 750,000 blue pixels and 1,500,000 green pixels. When you shoot in "color" with an IR filter, then later "convert" it to B&W, several things have happened.
Disregarding some of the "information" filtered out by the IR filter, any given pixel in the final image consists of the "averaging" of a group of 4 pixels (1 R 2 G and 1 B ). That single full color pixel has been interpolated or averaged from 4 other pixels. This interpolation process blurs the boundaries of 4 adjacent pixels, making ones actual resolution and sharpness less than one could achieve if one truly measured all colors at each pixel.
When you convert to black and white, you've tossed away a bunch of color information. If you were only going to shoot black and white, the ideal camera would not have a CFA in front of the sensors and would not need to interpolate any points in between. The resulting B&W image would have a sharpness more like (assuming it was a 3MP sensor) a 6 or 12 MP sensor.
That is what I meant.
For fun, could you post a small .jpg of one of your color IR pictures, BEFORE
you convert it to B&W. I am curious what they look like. If it is as I suspect, a black and red image, then your camera has only used 1/4 of it's available pixels, to create its image (again assuming an RGB CFA).
If you don't believe that interpolation doesn't signifcantly enhance an image, you should read Phil's reviews of the Fuji cameras with its Super CCD, those sensors try to interpolate even more pixels, and there is negligible improvement in the image.
Regardless, your IR photos are very niec.
John Bower
Carl Schofield
Veteran Member
Careful Diane - IR can be very addictive. I shot almost nothing but IR for about a year with a coolpix 950:
http://homepage.mac.com/scho/MySlideshow/slideset.html
I no longer have the 950, but still find myself visualizing landscape scenes in IR. Your IR images are excellent.
Thanks John, for the primer on this S . I understand what you are saying, but if I continue to do IR or B/W, then there's little I can do about it, right?!?
Here's a 'right out of the camera converted to tiff from RAW' image. I bracketed this, so this may not be the one I used for the 'finished' image, but its close.
and here's my final image.
Diane
Thanks all.
Yes Carl, they are addictive S . I work with color all day long and am a bit surprised that I'm drawn to b/w almost exclusively--but then again, maybe not. Shadows, light, pattern--all appeal to me--and b/w is very meditative and calming.
ARggggggghhhhhhhhhhh--I didn't look close enough quickly--def. not the same one--but taken just minutes apart I know.Here's a 'right out of the camera converted to tiff from RAW'
image. I bracketed this, so this may not be the one I used for the
'finished' image, but its close.
and here's my final image.
Diane
Diane
H
hugh
Guest
I looked at these quickly this morning and I guess I wasn't totally awake--I was looking at the larger finished image and the other and the clouds didn't match so I assumed they weren't the same. Well, right after I posted the reply saying they weren't the same I realized my mistake and clicked on the 50% making them the same size and saw that they were--the same. Then I checked the numbers (the CRW numbers) and they are ONE AND THE SAME. Just ask me if I'm embarrassed or not--- LOL (very red face).
Guess I was so excited to be back online. My cable company went off last night and expected to be off a good part of today--changing to a new faster backbone connection. Much to my surprise, after I repowered my modem and rebooted---my new ISP number is right, I'm connected and I was able to come on and read all about the new G2--and---just got totally confused LOL.
Diane
H
hugh
Guest
i usually blame getting older for everything. In your case, being "disconnected" is excuse enough
.hs
John Bower
Active member
Diane
Thanks, but now I am confused.
Is your IR filter dark red if you look through it?
Are you shooting in color or B&W when using the filter?
(I assume color, since there is a light blue cast)
Your final images are dead ringers for the B&W IR images I've seen (dark sky, nearly white foliage).
What I don't get is the pastel blue in the raw image. I would think the image (if shot in color would be primarily red). The only thing I can think of is that IR is getting through to all of the color channels, and given the roughly neutral tone of the image, in nearly equal amounts, interesting.....
This would imply that IR can get through each of the color filters (which, in the case of the G1 is a CYMG array Cyan, Yellow, Magenta, Green).
That being the case, you could shoot in B&W mode in the camera when using the filter, you might get similar results, depending on what type of image manipulation you are doing.
As far as being able to do anything about it the loss of image information, there is not much you can do. (referring to my previous posts)
However, if it is true that IR is getting through to each color pixel in almost equal amounts, I think, in theory, if you shot in RAW mode and had a custom TWAIN driver written, you could possibly create a high definition image, by NOT interpolating each point. IE, each pixel would represent the tonal value at THAT point, and not be interpolated from the 4 or so surounding pixels. It would be similar in effect to sharpening, but without the halo effects. IN this case you would not be "throwing away" pixel information and you would have stumbled across a way to circumvent the CFA (at least for IR photography). Hmmmm......
John Bower
Thanks, but now I am confused.
Is your IR filter dark red if you look through it?
Are you shooting in color or B&W when using the filter?
(I assume color, since there is a light blue cast)
Your final images are dead ringers for the B&W IR images I've seen (dark sky, nearly white foliage).
What I don't get is the pastel blue in the raw image. I would think the image (if shot in color would be primarily red). The only thing I can think of is that IR is getting through to all of the color channels, and given the roughly neutral tone of the image, in nearly equal amounts, interesting.....
This would imply that IR can get through each of the color filters (which, in the case of the G1 is a CYMG array Cyan, Yellow, Magenta, Green).
That being the case, you could shoot in B&W mode in the camera when using the filter, you might get similar results, depending on what type of image manipulation you are doing.
As far as being able to do anything about it the loss of image information, there is not much you can do. (referring to my previous posts)
However, if it is true that IR is getting through to each color pixel in almost equal amounts, I think, in theory, if you shot in RAW mode and had a custom TWAIN driver written, you could possibly create a high definition image, by NOT interpolating each point. IE, each pixel would represent the tonal value at THAT point, and not be interpolated from the 4 or so surounding pixels. It would be similar in effect to sharpening, but without the halo effects. IN this case you would not be "throwing away" pixel information and you would have stumbled across a way to circumvent the CFA (at least for IR photography). Hmmmm......
John Bower
LOL--well, I could have used that excuse too.i usually blame getting older for everything. In your case, being
"disconnected" is excuse enough
hs
Diane
Okay, this is what I am doing. I shoot in color with a Hoya R72--which is black red (VERY dark--but you already know that). IF--I put the white point at 'auto' (I'm using a G1) I will get a very red/black image. However, I use a custom white point--I usually use a white card or sometimes choose a very light area in the LCD to select for the custom white point and set it. Occasionally I just use auto WP and edit in RAW--but not too often. I like having some idea of what I'm getting while I'm shooting--though I know it won't be my final image. I generally use a small aperture and relatively slow shutter speed--my understanding is that with IR, the focus is somewhat different so I usually take advantage of the digicam's deep DOF. Don't know if this is totally correct for digicams--most of the info I have is for IR film, so I'm sort of flying by the seat of my pants. I'll post one of the shots w/o choosing a custom white point. I have actually shot in P mode in very bright sunlight and gotten a decent shot--I usually take note of what it meters--more often a wider aperture and faster shutter.
The foloowing is as it came out of the camera, auto white point
This is not the same shot, but shot within same period.
Hope this gives you the info that will clear up what's occurring.
Diane
Keyboard shortcuts
- f
- Forum