dhoelscher Posted February 18, 2007 Share #1 Posted February 18, 2007 Advertisement (gone after registration) Just a quick post to give people some ideas for an alternative way to deal with cyan drift that occurs with wide angle lenses and the 486 filters. This is important, because even when firmware 1.10 comes out, there is still a need to deal with this problem in post-processing when dealing with non-Leica lenses that: 1) have no equivalent Leica focal length - like the CV 12mm f/4.5 Heliar 2) cannot easily be hand coded - like the Zeiss 15mm f/2.8 Biogon, because of a bayonet screw that's in the coding area. I have been playing with PanoTools to correct cyan drift, but the results were poor for the CV 12mm no matter what the settings ... so I set about to try something a bit different. What has worked better for this lens in particular is this: 1) Take a picture of white cardboard with the Heliar with 486 filter (see first attached image) 2) Invert in Photoshop (Image>Adjustments>Invert) see second attached image 3) Paste as a new layer on top of Photo to be corrected 4) Change blending mode of top layer to "color dodge" See before and after bogus test pictures following the two "mask images." It certainly isn't perfect, but it's a whole lot better than trying PanoTools' radial luminosity correction for this lens somehow ... (longer focal length lenses seem to work better with the PanoTools thing for some reason) See what you think .... DH Welcome, dear visitor! As registered member you'd see an image here… Simply register for free here – We are always happy to welcome new members! Link to post Share on other sites Simply register for free here – We are always happy to welcome new members! ' data-webShareUrl='https://www.l-camera-forum.com/topic/16594-alternative-method-for-correcting-486-cyan-drift/?do=findComment&comment=175769'>More sharing options...
deltoid1 Posted February 18, 2007 Share #2 Posted February 18, 2007 Great idea. I imagine you apply this in a Photoshop action too. Just a quick post to give people some ideas for an alternative way to deal with cyan drift that occurs with wide angle lenses and the 486 filters. This is important, because even when firmware 1.10 comes out, there is still a need to deal with this problem in post-processing when dealing with non-Leica lenses that: 1) have no equivalent Leica focal length - like the CV 12mm f/4.5 Heliar 2) cannot easily be hand coded - like the Zeiss 15mm f/2.8 Biogon, because of a bayonet screw that's in the coding area. I have been playing with PanoTools to correct cyan drift, but the results were poor for the CV 12mm no matter what the settings ... so I set about to try something a bit different. What has worked better for this lens in particular is this: 1) Take a picture of white cardboard with the Heliar with 486 filter (see first attached image) 2) Invert in Photoshop (Image>Adjustments>Invert) see second attached image 3) Paste as a new layer on top of Photo to be corrected 4) Change blending mode of top layer to "color dodge" See before and after bogus test pictures following the two "mask images." It certainly isn't perfect, but it's a whole lot better than trying PanoTools' radial luminosity correction for this lens somehow ... (longer focal length lenses seem to work better with the PanoTools thing for some reason) See what you think .... DH Link to post Share on other sites More sharing options...
dhoelscher Posted February 18, 2007 Author Share #3 Posted February 18, 2007 Sure ... you certainly could set up a Photoshop action for this ... I'm still at the "highly experimental" stage ... when I've done some more testing, I'll post ready to use gradients for all the lenses that I have access to for people to try ... DH Link to post Share on other sites More sharing options...
sean_reid Posted February 18, 2007 Share #4 Posted February 18, 2007 Nice idea....I would just add that the appearance of cyan drift can vary with focus distance and aperture because its blended with vignetting. As such, one might want to photograph the test chart at a distance he or she often works at (to the extent possible) and at a favorite aperture for that lens. For the ambitious, one can buy 4' x 8' sheets of foamcore. Finding the right pattern and intensity for the correction may be the challenge. Cheers, Sean Link to post Share on other sites More sharing options...
deltoid1 Posted February 18, 2007 Share #5 Posted February 18, 2007 Or, you can simply choose to correct for the focusing distance that gives the least vignetting, and simply leave the vignetting (the difference between the least and the most,) in the image for the other focusing distances. Can that difference be really that much? Nice idea....I would just add that the appearance of cyan drift can vary with focus distance and aperture because its blended with vignetting. As such, one might want to photograph the test chart at a distance he or she often works at (to the extent possible) and at a favorite aperture for that lens. For the ambitious, one can buy 4' x 8' sheets of foamcore. Finding the right pattern and intensity for the correction may be the challenge. Cheers, Sean Link to post Share on other sites More sharing options...
sean_reid Posted February 18, 2007 Share #6 Posted February 18, 2007 Can that difference be really that much? It depends on which specific lens one is testing for. Link to post Share on other sites More sharing options...
Guest guy_mancuso Posted February 18, 2007 Share #7 Posted February 18, 2007 Advertisement (gone after registration) Thanks David nice work now if i can just get my damn CV15mm to take a code i would be happy. I just can't seem to get it for the WATE code. Now if someone figures how to code that Zeiss 15mm and get IR filters on it than let me know i may just RUN and buy it Link to post Share on other sites More sharing options...
robertwright Posted February 18, 2007 Share #8 Posted February 18, 2007 An additional tweak to this would be to mask the correction layer and then paint the amount of correction you want in, in this way you would not need to shoot additional versions for differences in aperture and focus distance. You can also attach an adjustment layer to the mask to allow you to tweak the color balance of the mask layer. Getting complicated there, but perhaps less so than having to decide which mask to use based on aperture data that is non-existant in exif and focus distance that is a guess also. Link to post Share on other sites More sharing options...
jonoslack Posted February 18, 2007 Share #9 Posted February 18, 2007 Hi David I love it! Congratulations - I love a bit of good lateral thinking, and this is sterling work. I haven't got any filters yet, so it's a bit academic, but this is an answer I won't forget. many thanks you clever person you! Link to post Share on other sites More sharing options...
MP3 Posted February 19, 2007 Share #10 Posted February 19, 2007 Great Work! David, it will surely benefit many non-Leica lenses users here. Cheers Matthew Link to post Share on other sites More sharing options...
dhoelscher Posted February 19, 2007 Author Share #11 Posted February 19, 2007 Nice idea....I would just add that the appearance of cyan drift can vary with focus distance and aperture because its blended with vignetting. ... Very interesting ... thanks, Sean! This must be why my example above isn't perfectly corrected .. I didn't control for aperture.. Correct me on this if I'm wrong --- 1) luminance vignetting IS aperture dependent 2) cyan drift (by itself) per se isn't aperture dependent, BUT ... In this scenario, the appearance of the cyan vignetting is altered by the luminosity vignetting, so the two are tied ... therefore, the cyan vignetting (net) IS aperture dependent. I'm gonna do some more tests tomorrow and control for aperture and focus distance... I'll post some DNG "templates" for people to play with. DH Link to post Share on other sites More sharing options...
Bob Ross Posted February 19, 2007 Share #12 Posted February 19, 2007 Hi David, My plans to do something like this used the Expo Disk. The only thing that I have done so far was to check the vignetting correction on the 28mm Elmarit-M Asph, my only coded lens. An Expo Disk migh be a convenient way to check for the variations pointed out by Sean, but it might not be easy to do on the 12mm/15mm CVs. Bob Link to post Share on other sites More sharing options...
Jamie Roberts Posted February 19, 2007 Share #13 Posted February 19, 2007 Nice work, David! I must admit, I couldn't get the "invert image" method to quite work for me So here's an alternate version which I've been playing around with! See what folks think... 1) Shoot your reference white wall to see vignette at full size 2) Load into PS 3) Draw a round selection with feather of about 200px (this is a nice starting point) 4) Fill the vignette with RED (255R, 0G, 0B) 5) Save this for each lens 6) Add this as a layer, but don't use Color Burn, use OVERLAY mode. 7) Vary transparency to taste Also a PS action, with one "filter" per lens. BTW--if you're feeling clever, you could *also* add a simple curves layer to the mask that accounts for the vignette luminance difference as well (and actually, you could probably do the whole job with just a curves layer, but I haven't had time to do this yet)... BTW#2--to account for vignetting strength differences at different apertures, what you want do is find the worst case, and make sure your "filter" covers that amount; all you do then is reduce transparency for weaker vignettes (since this is essentially a mask layer in PS, you can easily add or subtract "vignette" with the eraser or brush if you need to as well). BTW#3--when I get a chance, an interesting approach to this would be to make the color and luminance correction independtly in L*A*B, where there is no danger of damaging detail by changing colors. Then you'd be saying, well, who needs coded lenses? See Guy? This is why I wanted your M8 "white wall" shots Link to post Share on other sites More sharing options...
Guest guy_mancuso Posted February 19, 2007 Share #14 Posted February 19, 2007 Yes Jamie I know been busy working LOL. Need to make money first to afford all this crap. ROTFLMAO Link to post Share on other sites More sharing options...
tom0511 Posted February 19, 2007 Share #15 Posted February 19, 2007 It depends on which specific lens one is testing for. Now if the cyan changes with distance and f-stop, how should the Firmware 1.10 correct accurately? F-Stop the M8 might know (I say might because so far I havent seen any f-stop in the Exif and therefore I doubt how reliable the f-stop calculation of the M8 really works), and distance the M8 does not know. Mmmmhh, Also, if the cyan vignetting depends on thelight, what happens when I change the white balance later in the converter? I just hope badly that the cyan vignetting does not vary that much and an correction of the average cyan vignetting of a certain lens will lead to acceptable results. Scary Link to post Share on other sites More sharing options...
dhoelscher Posted February 19, 2007 Author Share #16 Posted February 19, 2007 Hi David,My plans to do something like this used the Expo Disk. ... An Expo Disk migh be a convenient way to check for the variations pointed out by Sean, but it might not be easy to do on the 12mm/15mm CVs. Hey thanks, Bob! Great point, cool idea .... I'm now thinking that either an ExpoDisc (if you have one big enough) or a piece of translucent plexiglas or something like that would be easier to photograph for these corrections than pointing the camera at a sufficiently large enough white background or foam core ... I'll post today's experiments later this afternoon ... DH Link to post Share on other sites More sharing options...
johnll Posted February 19, 2007 Share #17 Posted February 19, 2007 ... either an ExpoDisc (if you have one big enough) or a piece of translucent plexiglas or something like that would be easier to photograph for these corrections than pointing the camera at a sufficiently large enough white background or foam core ... Just move in close. It doesn't have to be in focus. Link to post Share on other sites More sharing options...
dhoelscher Posted February 19, 2007 Author Share #18 Posted February 19, 2007 Just move in close. It doesn't have to be in focus. Thanks, John ... you are quite correct .... DH Link to post Share on other sites More sharing options...
dhoelscher Posted February 19, 2007 Author Share #19 Posted February 19, 2007 Now if the cyan changes with distance and f-stop, how should the Firmware 1.10 correct accurately? F-Stop the M8 might know (I say might because so far I havent seen any f-stop in the Exif and therefore I doubt how reliable the f-stop calculation of the M8 really works), and distance the M8 does not know. Mmmmhh, Also, if the cyan vignetting depends on thelight, what happens when I change the white balance later in the converter? I just hope badly that the cyan vignetting does not vary that much and an correction of the average cyan vignetting of a certain lens will lead to acceptable results. Scary Tom, this is what makes me wonder how the WATE with 486 will work out even with firmware 1.10 .... We might just decide that the WATE, for example, just isn't cut out for 486 filters .... I guess we'll see when firmware 1.10 is released into the wild .... DH Link to post Share on other sites More sharing options...
sean_reid Posted February 19, 2007 Share #20 Posted February 19, 2007 Very interesting ... thanks, Sean! This must be why my example above isn't perfectly corrected .. I didn't control for aperture.. Correct me on this if I'm wrong --- 1) luminance vignetting IS aperture dependent 2) cyan drift (by itself) per se isn't aperture dependent, BUT ... In this scenario, the appearance of the cyan vignetting is altered by the luminosity vignetting, so the two are tied ... therefore, the cyan vignetting (net) IS aperture dependent. I'm gonna do some more tests tomorrow and control for aperture and focus distance... I'll post some DNG "templates" for people to play with. DH To the best of my knowledge so far...Luminance vignetting does vary by aperture and, to a lesser extent, focus distance. Cyan drift (red vignetting), in isolation, does not seem to vary according to aperture but since both kinds of vignetting appear together in the file, one can influence the appearance of the other. Cheers, Sean Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.