This site requires new users to accept that a small amount of member data is captured and held in an attempt to reduce spammers and to manage users. This site also uses cookies to ensure ease of use. In order to comply with new DPR regulations you are required to agree/disagree with this process. If you do not agree then please email the Admins using info@nikondslr.uk after requesting a new account. Thank you. |
Moderated by: chrisbet, |
Author | Post | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
jk
|
Apparently according to LifePixel website. http://www.lifepixel.com/camera-considerations The FAQ and tutorials. http://www.lifepixel.com/tutorials/infrared-diy-tutorials The different filter types. http://www.lifepixel.com/infrared-filters-choices The cameras and why not. Nikon D40, D60, D90, D300, D300s, D3000, D5000, D7000 are not great for IR as they have problems with setting a decent WB Preset when used in IR. The D3 series, D4 and D700 models make use of Nikon's new infrared self-diagnostic shutter monitor circuitry. This function involves an internal infrared LED emitter/detector chip that could fog images. In most cases it will not be an issue unless you shoot in overcast weather using higher ISO and/or longer exposures. The D90, D300, D300s, D3000, D5000, D7000, D600, D610, D800 and D800E these models are random, some will white balance all filters but some not at all but most just won't white balance with our Super Color & Enhanced filters. Will need to shoot in RAW and use Nikon Capture NX2 software to set the white balance. So that leaves as good cameras for conversion the D1, D1X, D70, D70S, D80, D100, D200 and D2x. Then of course there is the Fuji Xpro1, XE1, X100 that shoot IR (unconverted) but with a R72 filter on the lens very nicely but you need a tripod to use this as exposures are typically 2-4secs with ISO 400 at f5.6-f8. It is easy to pre-visualise the IR result with the TFT screen. |
|||||||||
Eric
|
jk wrote:Apparently according to LifePixel website. I did respond to this subject on one of Roberts recent threads. With regard to the first series of WB problem bodies... You don't need to set a white balance if you shoot in raw...I don't even do it on the D200. I established the 'as shot' colour from the D70 files...it was always the same! So now I just open the file and move the sliders in ACR. Suppose I should do a preset...but it only takes a moment to move to the right setting |
|||||||||
Robert
|
D200, maybe D1x?. With IR the composition is usually full frame, so not a lot of cropping to do. The Dxx cameras usually are lower IQ. Unless you have a D2x lying around otherwise unloved, they are a bit 'heavyweight' for IR. If you don't have a D200 I know where there are several, sub £200. The D3XXX and D5XXX have the advantage of Live View, if that IS an advantage? I do all my IR on a tripod, so that could be an advantage to me but I know I am probably alone in that. I have tried the UV on the D3100 and can get some results but it's hugely hampered by the internal filters. I liked the way Bibble handled the post processing, I never bothered too much about WB, it's getting rid of the pink cast from the D1 that caused me most grief. I know that's related to the WB but I don't believe that's the whole story. It's the lack of colour correction the LPF gives that causes the pink cast in my opinion I think it has to do with the number of red receptors. Swapping the colour channels is key to solving it. I have seen some fabulous results from Capture NX and zone control. I think trying to use any unconverted camera is going to produce half hearted results. You are trying to capture IR and the camera is trying to exclude IR with very strong filters. That has to limit the potential IR results, image quality and particularly contrast. |
|||||||||
jk
|
Hmmmm. You mean that all that setting up of the camera (D70) with a PRESET WB was a waste of time as I shoot RAW and I shout just let the camera Auto WB ? Eric, Please can you detail your camera WB settings. I guess you are talking about the Colour Temperature setting in ACR. Where is that Eric's IR settings action ? |
|||||||||
Eric
|
jk wrote: Hmmmm. You mean that all that setting up of the camera (D70) with a PRESET WB was a waste of time as I shoot RAW and I shout just let the camera Auto WB ?When I converted the D70 I set a preset WB by simply pointing the camera at grass in sunlight. I did the same for the D200. But when you open an IR raw in ACR and look at the colour sliders, the blue/yellow slider is ALWAYS maxed at blue end....ie you cannot make it bluer/less yellow. The only adjustment you have available is the green/red slider. The difference between the d70 and the D200 setting is always the same. The D200 is -66and the D70 was -85. If you move the former to -85...you get the same colour. So my logic has always been dont bother with colour in the camera..make each image appear as you want it in CS. Attachment: d70 v d200.jpg (Downloaded 45 times) |
|||||||||
Eric
|
Not exactly the same frames as the raws i grabbed, but from the same location on same day. This is the result of CS processing (different approaches) on those raws. This highlights why getting the colour right in the camera insnt necessary...its what YOU decide to do with the image afterwards that matters. Each image recommends a finish...either by subject content or ...your whim of the moment. Attachment: processed.jpg (Downloaded 46 times) |
|||||||||
Eric
|
OK - I found the original 'Ramparts' nef. This is how it was created 1. Open file in ACR 2. Reduce green red slider to -85 (The D200 has magenta areas that can be annoying) 3. Set the white and black points using the PLOWDEN SHUFFLE. This is a brilliantly simple fast way to set these points shown to me by our own Robert Plowden Esq. As follows... Open Levels move Left (black) input slider to extreme right until everything goes black EXCEPT the whitest points. Using the white dropper click on the point you feel is the best white point. Then push the Right (white) slider all the way to the left so the screen goes white EXCEPT for the darkest points. Using the black dropper click on the point you want to be black. Clearly there is room for adjustment at this stage if you want a more contrast effect. 4. The resulting image has some patches in the 'white' areas that are still reddish Just remember that for now. 5. Using the channel mixer, switch red and blue levels to invert the colours. (eg Red channel: Red -> 0%, Blue ->100% and Blue channel: Blue-> 0%, Red ->100%) 6. You now have inverted the red/yellows to blue/greens. But you can still see that the patches in the white areas (mentioned in 4) are now blue. This is where personal taste comes in...they annoy me! 7. Duplicate layer and desaturate the bottom layer 8. Using an eraser at 50% - rub over the areas with 'offending' blue patches to taste. 9. Reveal the lower layer and flatten 10 Sharpen to taste EDIT - Sorry cant seem to get this size file to be bigger on screen?????? Looking at the ramparts file above ...I clearly toned down the BLUE skies a bit before saving. Attachment: IR processing.jpg (Downloaded 46 times) |
|||||||||
Eric
|
Here's the Plowden Shuffle screens more discernible. Just one warning...this technique works fine for IR or black and white images. Care must be taken with colour images because the 'white' point should really be called the 'light' point. The lightest points in some images may be 'off white' (eg cream) so clicking on them could slew the whole image colour. Similarly the 'dark' point might be dark brown!! Attachment: plowden shuffle.jpg (Downloaded 46 times) |
|||||||||
jk
|
Thanks Eric. That is most helpful. Now I can have a go at testing the Post processing as I certainly havent been doing it this way. When you take the IR images do you adjust the EV compensation. I have seen it recommended that you overexpose by up to +1EV. |
|||||||||
Eric
|
jk wrote:
It's not a fixed formula. The scene, subject and lighting can fool the meter more so than in full colour. In bright sunlight, even lit subjects, the exposure is quite often ok as per the meter....maybe +0.5. But in slightly overcast weather, or with large areas of foliage not exposed to sunlight, the exposure may have to be ....reduced. This is because the meter only sees (and computes) for the visible light component. So in darker conditions it opens up the exposure without reference to the contribution from the foliage IR. These images can as aresult be over exposed .....especially burning out the highlights. I have to say I don't ever recall having to over expose to +1. But I do bracket -0.7 >0<+0.7 on all my shots ...as I know it's easy to misread the scene. Better to have a choice. Having said that, the ramparts image in the above example, from the histo, is probably a good stop underexposed! ( I may have chosen the -0.7 version). But the thing to watch out for is the leading trail of pixels at the front of the curve. Move this exposure up by more than a stop and those pixels, though only few, will be clipped....giving burnt out spots in the image. Now this may not be sufficient concern but I always like to underexpose to save the highlights. After all, in IR, the 'whites' are the supposed most important part! |
|||||||||
jk
|
Well that worked well as a first shot. I have gone from this as a NEF. Attachment: Screen Shot 2014-01-11 at 16.54.20.jpg (Downloaded 43 times) |
|||||||||
jk
|
To a 'finished' effort that looks like this. Not as I want it but it was only a test shot so it doesnt warrant spending loads of time on it. Attachment: Screen Shot 2014-01-11 at 15.42.23.jpg (Downloaded 43 times) |
|||||||||
Eric
|
jk wrote:To a 'finished' effort that looks like this. Not as I want it but it was only a test shot so it doesnt warrant spending loads of time on it. The image seems to have vignetting. What lens was it ? It looks to me as if the highlights has been pushed a tad too far. One of the 'tricks' with IR (IMHO) is to make the foliage secondary to non IR responsive elements. I know this was just a test shot but the image needs a non IR focal point. Apologies if I am stating the obvious. Try getting under one of those trees and shooting 'up the non IR responsive trunk' to the white canopy. |
|||||||||
jk
|
Eric wrote:jk wrote:Sigma 10-20 I guess. Yes that is me being a bit heavy handed. Also the foliage is a little blue but I was concentrating on getting the sky blue.
Will do but the pines tend to be less IR responsive than deciduous trees. |
|||||||||
Robert
|
Some of the Sigma 10-20's don't respond well in the IR range, they seem to hotspot badly which may explain the vignetting, perhaps it isn't vignetting but hotspotting? I have had mixed results with my Sigma 10-20 I think Eric found it hotspotted badly on his D70 but I have had a little success on the D1, although not universal success. I have spent days on a single image and while I may EVENTUALLY have got the effect I visualised when I composed the image, having spent so much time on it the image is badly degraded. I then try to retrace my footsteps and re-create the same image with the minimal of steps from the original NEF In my experience fiddling with an image for days ends up degrading it and making it muddy with a low IQ. Sometime it take several attempts to minimise the steps to the clean image I know It can reveal. Then I wonder why I bothered!!! LOL But then I do enjoy the results in the end when I look back. Too much I am celebrating, I have got my gas cooker self ignition to work!!! Only take me three years to pluck up courage to pull the cooker apart and repair the switch! Yea, I can put my camping ignitor away!!! |
|||||||||
Eric
|
Robert wrote:Some of the Sigma 10-20's don't respond well in the IR range, they seem to hotspot badly which may explain the vignetting, perhaps it isn't vignetting but hotspotting? I have had mixed results with my Sigma 10-20 I think Eric found it hotspotted badly on his D70 but I have had a little success on the D1, although not universal success. I suppose having seen how badly the 24-70 hotspots, I am more forgiving of the hotspot from the Sigma 10-20. Back then it was an irritation and a let down having the centre of an image 'fogged'. But as this failing is a feature in many, if not most lenses to some degree, I guess you have to live with it. That's why I like the 17-35....no hotspot at all (don't ask my why). Sadly it limits mid distance work. But the 80-200 AFS is also fog free. So another keeper! I am not so sure that JKs image is hotspotting though. Usually with hot spots they are associated with high contrast areas near the centre of the image (making the 'fogging' more obvious). I don't see that scenario in JKs image...the centre is all similar tones. Also, the lighter middle does seem symmetrical. I wondered if if was a cloud effect. They do get clouds in Spain dont they...Not sure? |
|||||||||
Eric
|
jk wrote:Eric wrote: Yes, deciduous are best. One of the most interesting time is when they start to bud. You can get little white 'lanterns' against a mostly dark background. This has whetted my apetite...going to get the IR camera out ready. :thumbsup: |
|||||||||
jk
|
Robert wrote: Some of the Sigma 10-20's don't respond well in the IR range, they seem to hotspot badly which may explain the vignetting, perhaps it isn't vignetting but hotspotting? I have had mixed results with my Sigma 10-20 I think Eric found it hotspotted badly on his D70 but I have had a little success on the D1, although not universal success.My Sigma 10-24 is a version1 release of the lens and is exceptionally good for a zoom and shows very little hotspotting. In fact it is so negligible that a couple of passes with the burn tool gets rid of any effects (it is probably about +0.15EV or +0.1 EV in strength). Glad that this IR talk has whetted you appetite Eric. I am looking for a Fuji S3 Pro UV-IR body as I believe that they have the best solution if I want to also do UV. However I am not convinced that it will be better than my Nikon D70 IR converted when used for IR. I blame my interest on Robert and our visit to Westonbirt when we discussed this and Robert showed me his D1 UV converted camera and lights. I really fancy having a go with the UV spectrum as we have so much more sunshine here in Spain also I have a UV torch I can use. |
Current theme is Blue
A small amount of member data is captured and held in an attempt to reduce spammers and to manage users. This site also uses cookies to ensure ease of use. In order to comply with new DPR regulations you are required to agree/disagree with this process. If you do not agree then please email the Admins using info@nikondsl.uk Thank you. |