shijan
-
Posts
139 -
Joined
-
Last visited
Content Type
Profiles
Forums
Blogs
Events
Downloads
Gallery
Store
Everything posted by shijan
-
Once again - my examples are not 100% sharp because i don't have a proper film holder yet to share some decent frames. Currently working on quality film holder project that may be useful for many people in future. Digital noise in digital sensors - in shadows. Film negative - Inverted. Digital noise in film negative became visible - in lights. All my current examples are with NEGATIVES. Diffused light - softer film grain look, but also near invisible dust and scratches. Collimated light (usually used in film scanners) - sharper film grain look, but all dust and scratches became extremely visible. Defocused or soft lens - Softer film grain, but same amount of digital noise. 26MP is not a enough resolution to capture sharpest possible film grain details. My examples only about 2 stops underexposed from native camera ISO 160.
-
Add fake noise to that image again and it will look subjectively not blurred :) It is a known - visual effect more noise add fake feeling of more detailed image. I just do some tests here and try to amplify some effects. Hope somewhere i'll be able to get more focused and grainy camera scan. Seems you confused with everything in this world. I really tied to explain every basic thing to you here.
-
Here is better underexposed example crop. Not a sharpest example because i still don't have stable setup. But it is enough to see that digital noise may be very easy messed up with film grain :)
-
Yep, probably "dabyter" is not a correct world for single or triple row linear scanners, but anyway they do some sensor-to-image pixels processing and usually have pretty hard aliasing. And yep, brightest parts of the negative will show most digital noise because negatives are inverted. And the most digital noise in blue channel, because orange color cast in negatives.
-
And some Anti Aliasing tests. If turn Anti Aliasing test OFF image looks very close to source form scanner. Huge amount of rainbow patterns and dots over film grain structure and aliasing artifacts around contrast white dust and scratches. This probably means that scanners use very basic internal debayering and don't do any anti aliasing reduction. Too many anti aliasing filtration also looks not good. Film negatives are very sensitive to small changes in source image, so even small amount of extra filtration may decrease global saturation in final processed image. So after some tests i can suggest: Anti Aliasing: 2 (instead of default 3) No Luma/Chroma noise reduction. No sharpening. And of course it is all depends of camera and sensor technology. Sensors with OLPF filters need less AA filtration. Normal Bayer sensors need more filtration that Fuji X-Trans sensors.
-
And here are some interesting resolution and artifacts tests. In the past i was lucky (actually not lucky) to use MINOLTA DiMAGE SCAN 5400II. It is a great scanner, but it had a global problem with non uniform backlight stripes, described here Backlight stripes in MINOLTA DiMAGE Scan 5400 II or what? (too bad uploaded photos are gone over time) After many attempts to modify LEDs or diffuse that defect i gave up with it. I decide to take off the lens form that scanner and build a camera scan system. Same time i had access to smaller MINOLTA DiMAGE SCAN Elite II model, so i was able to take some shots from it for compare. For this test i build input ICC profiles for camera and both scanners based on Kodak IT 8.7 Scanner Calibration Target from Affordable IT 8.7 (ISO 12641) Scanner Color Calibration Targets Profile Type: Single gamma + Matrix Scan from camera debayered in Iridient Developer with Anti Aliasing setting: 2. Custom contrast camera curve removed. No noise reduction, no sharpening, no any other adjustments applied to RAW file. 5 frames stacked in Median mode. Poor quality consumer furniture LED panel used as backlight. Scans from scanners are in linear gamma with disabled color management. All images processed with my workflow described earlier: Transform from Camera input ICC profile to ProPhotoRGB with L* gamma ICC profile -> Invert -> Apply RGB AutoLevels -> Recover back clipped data from RGB AutoLevels -> Contrast. No custom White point picker. This is rather complicated frame i use specially for tests. Is is very scratched and it have a lot of extreme saturated colors.
-
Temporal noise reduction don't blur any details, until you stack 100-1000 of images. From my observation Median produce slightly higher micro contrast than Mean. But the difference is near invisible. Increased illumination level won't help a lot. I inspect IT7.8 test slide frame and i can see that very small amount of digital noise always started somewhere from the middle grey patches. Sure it is near invisible but it is always there. To get rid of all noise with illumination you need to increase exposure a lot and so clip huge amount of dynamic range. And this is not only about noise. It is about pixels density and pixels quality. With stacked images from multiple 14 bit "relaxed" bayer pattern-based images you get solid and dense real 16 bit color data. From my personal tests 5 images stack is OK, but some tiny amount of noise in deepest shadows is still there. 10 stacked images look perfect, but take more time and process. Stack more than 10-20 images probably useless for film scans. But for sure is may depend of camera sensor. Lower quality sens0d -> more noise -> need more stacking. Also don't forget that stacking is highly exponental. Dfference between 0 and 5 images always look huge. Difference between 10 and 100 stacked images may be near invisible.
-
Here is also an idea how to emulate scanner multisampling with DSLR. There is a well known very effective lossless temporal noise reduction method. It is based on multiple similar images blended with "Mean", "Median" or some other special blending modes. It is well described in this article Pat David: Noise Removal in Photos with Median Stacks (GIMP/G'MIC & Imagemagick) This stacking option also available in Photoshop, but as usual designed in rather hidden and confused way: File -> Scripts -> Load Files Into Stack -> check "Create Smart Object" Layer -> Smart Objects -> Stack Mode -> Mean This stacking option also available in Affinity Photo, but currently not available PhotoLine. Hope PhotoLine developers will add it in future. Mean — averages pixel content across the stack of images. Good for long exposure simulation and noise reduction. Median — removes pixel content that is not consistent in each image. Suitable for object removal and noise reduction. Outlier — exposes pixel content that differs in each image: great for sequence composites. Maximum — uses the maximum pixel values from each image. Can be used for creative exposure blending where the subject is lighter than the background. Minimum — uses the minimum pixel values from each image. Suitable for exposure blending where the subject is darker than the background. Range — indicates areas that change across the image stack. Good for analyzing what has changed between each image. Mid-Range — uses the middle pixel values from each image. Can be used to increase tonal range if used with bracketed exposures. Total — produces the total value of pixels from each image. Usually results in overexposure, but can be used to lighten very underexposed imagery. Standard Deviation — analytical: measures the distribution of information between the images. Useful for object removal as it clearly indicates areas that will be averaged out with a Median operator. Variance — analytical: as Standard Deviation, indicates how pixel values are spread between images. More intense distributions are shown very clearly. Skewness — analytical: highlights edge detail and indicates the intensity of pixel value distribution. Can be used to determine tonal and spatial differences between images. Kurtosis — analytical: detects the peakedness of an image. A brighter result represents low noise levels and a tonal uniformity (most pixels at dominant gray level). Darker results represent greater noise and less tonal uniformity (more pixels further away from dominant gray level). Entropy — analytical: represents the number of bits required to encode information in the stack. Could be used with stacked video frames (within the same scene or shot). So in short: 1. You just need to quickly shoot 5-10 similar copies of the same film. Continuous Shooting (Burst Mode) will do the trick. Make sure your setup is stable and you don't move camera or film during shooting. 2. Process RAW files to TIFFs as it was described earlier, but don't do invert and don't do other processing yet. 3. Stack images into one single file with "Mean" blending mode and save as single TIFF. 4. Process negative with workflow described earlier in my posts. This will clean up all possible digital noise without touching film grain structure and will make your source file more dense at pixel level. Also this method probably should be way less risky than HDR stacking, because it will not change original tonal relations taken from linear sensor data. And a quick test to proof my theory. Here is crop of inverted and processed film negative. This film negative sample was scanned with camera and was underexposed more than usual to amplify the noise and see the camera limits. This is 400% scaled crop, but i can see the difference at 100% as well.
-
Note that "Gamma" and "Color Space" are separate parts of the image. With tools described earlier you can generate custom working ICC profile with any gamma (tonal curve) you like and with any color space you like. L* gamma is just a real world lightness-based tonal curve. It allow to work in RGB color model and use advantage of same "symmetrical" "physically based" tonal response as Lightness in Lab color model. This is why it is so unique and useful for film negative invert. ICC profile with L* gamma is for image editing only, it should never be used in final exported 8 bit JPEGs.
-
- AutoLevels clip data because they need some amount of highlights and shadows to detect where each color channel starts and ends. - There is a "Clip" setting in AutoLevels. It define the amount of clipped highlights and shadows. - Is it possible to set "Clip" to 0.0 and apply AutoLevels without clipping, but in this situation colors usually will be detected slightly incorrectly, because instead of normal well defined colors, AutoLevels will detect some garbage and artifacts in shadows and extremely bright highlights. This is why "Clip" setting in AutoLevels usually should be set somewhere from 0.01 to 0.1. - AutoLevels is very delicate step in workflow. In simple images with bright colors it usually don't require too many tuning, but with images that have fog, dimmed, sky or unusual lighting you may need to experiment with different "Clip" setting in AutoLevels until you see subjectively pleasant colors. - If you ok with slightly clipped shadows and extreme highlights after AutoLevels, you may keep image as is, don't recover dynamic range and don't add additional contrast. Even with slightly clipped highlights and shadows, dynamic range of of final processed film negative outperforms any digital photo. As for Kodak Ektar 100 negative - just scan it same as any other film and process and described. I never scan films with Epson scaners so i can't help a lot with settings. But based on my V10 reflective scanner, i guess this is how it should work: - Turn off color management and scan as positive film. Even with disabled color management Epson scanners can't scan to Linear gamma, but scan to gamma 2.2 also will work more-less ok. - In Epson Scan app, If it possible adjust Exposure settings manually and make sure that Red channel don't clipped in histogram preview. - Open scanned image in graphic editor and ASSIGN Epson input ICC color profile (for example in Photoshop it is: edit -> assign profile...). Epson input ICC profile usually located in system profiles folder and named something like Perfection V100 ref and Perfection V100 film. Not sure which one will work better here. It could be that Perfection V100 film designed for some specific negative invert tasks in Epson Scan app. So it is a good idea to start with Perfection V100 ref ICC profile. - Now CONVERT image from scanner input ICC profile to ProPhotoRGB L* working profile (for example in Photoshop it is: edit -> convert to profile...) If you don't want to generate L* working ICC profile manually in apps described in first post, you can simply download Elle Stone's Well-Behaved ICC Profiles package here ellelstone/elles_icc_profiles (click to green button named "Code" and select Download ZIP) For some reason Elle Stone use name "LargeRGB" instead of "ProPhotoRGB". Also he use name "labl" instead of L* (or L star) gamma. More info described here Elle Stone's well-behaved ICC profiles and code To avid possible compatibility problems i suggest use V2 ICC profile type but not V4. So in that package you need to look for LargeRGB-elle-V2-labl.icc, or simply download single ICC profile here https://github.com/ellelstone/elles_icc_profiles/raw/master/profiles/LargeRGB-elle-V2-labl.icc Put it in system profiles folder. Next follow the workflow described in first posts of this thread...
-
LOL, you are really strange person. I explained how to build L* gamma ICC profile in first posts and provide links to 3 sources. You didn't read it. You ask to explain about L* gamma. Sorry, no more explanations for you. Do whatever you want and be happy.
-
As for camera exposure during scan - you need to rely only on RGB histograms, but not on single Luma histogram. Also, if camera allow, you need to switch histogram to mode that shows histogram based on uncorrected RAW image. In Fujifilm cameras this option named "Natural Live View" Fuji Natural Live View | Exposing to the Right | Fuji Frame
-
P.S. to everyone: I want to notice again - this thread is not about some abstract workflow and some guesses how to process film negatives. It is about technically correct, well established and polished workflow that provides quick, solid and predictable final results as described in first posts. So please make sure you at least read the topic and understand how it works before post something here. If you want to discuss, confirm or deny something do some compare tests with your own workflow and workflow described at the start of this topic, provide decent info of your workflow and post source RAW test images.
-
Not again please :) First of all your Non mask-corrected version source is clipped in red channel, so you just operate with already broken image from start. Delete it and re-scan without clipping. "experimented both with MakeTiff and applying a Prophoto profile to the RAW import" This is that you never should do, because you just destroy color management when apply ProPhotoRGB (or any other working ICC profile) instead of proper input camera profile. This is main problem of ColorPerfect/MakeTiff developer - he did not provide proper color transformations system form camera input profile to working profile. At the first step image always should be CONVERTED from camera/scanner input profile to working profile. "Is it not obvious which version will be easier to adjust to acceptable colour?" No it is not. And i post extreme examples and link to explanation earlier. All i can suggest - read faq in first post and learn how color management works. Until that i can't see any point to explain you any things further. Debayer RAW file and convert it from Camera input ICC profile to ProPhotoRGB with L* gamma custom-made ICC profile -> Invert -> Apply RGB AutoLevels -> Recover back clipped data from RGB AutoLevels -> Adjust tonal curve and Contrast. Compare test results to your methods. Maybe later you will see your mistakes. I am not interested anymore to discuss with you problems and mistakes that i already fixed for myself 5-7years ago.
-
Few more very basic extreme examples: 8 bit Image processed in sRGB gamma and sRGB color space. AutoLevels Clip was set to 0.05 % As you may see from RGB histogram in Levels tool, relations between channels became very non uniform after invert. Red channel stretched in very strange way. You can see that greenish color cast there from start. Tonal arrangement require large amount of additional manual adjustments to bring back things to normal. All these problems caused by too tiny for such editing sRGB color space and not too "symmetrical" sRGB gamma. Same 8 bit Image converted (transformed) from sRGB to ProPhotoRGB color space and L* gamma. AutoLevels Clip was set to same 0.05 % Due larger color space and due more "symmetrical" L* gamma, there is enough of empty space for adjustments in each channel. Relations between channels stay uniform after invert. Tonal arrangement looks nice and require very small amount of adjustment.
-
The only problem that sunset became yellow-green in your version instead of orange:) Formally the whole concept of "Remove Orange Mask and all colors magically became correct" is incorrect from some point of view. Orange mask is not a some separate evil part of the of image, it is a part of the film negative and it is linked to relations between colors, so is not something that should be somehow removed or just hard clipped to white. It is a symmetrical system that was build for film negative + print paper. Technically In 16 bit it makes no difference for Photoshop or any other app if orange mask "removed" or not. It is all just a adjustment (alignment) of dark and light points in channels. The main problem is how to keep relations between these aligned channels as original as possible. So when some people first attempt to "remove orange mask" with aligning RGB channels by picking WB, and then "fix colors" by AutoLevels, they formally in best case scenario just move same data from one place to another and back. In worst case scenario these additional "movements" generate shifts between colors and as result "corrected" image require more additional corerctioin to adjust colors back to subjective normal look.
-
The main problem with film negatives that they are not accurate by nature. Even same roll many variations. it is impossible to print tech chart for calibration on them natively. So the closest compromise is calibration test chart printed on slide film. It also not clinically perfect because each slide film brand also have some own character. All i done is just use IT8.7 target and provided data for it from Affordable IT 8.7 (ISO 12641) Scanner Color Calibration Targets and build input ICC profile. This works subjectively well with film negatives from scanners as well as from DSLR cameras, this fix problems with low quality light sources and makes final image look subjectively better. By the way, digitaldog, i watched your videos long time ago and learned a lot with them. They really inspire people to digg into color management stuff.
-
There are a lot of apps with some abstract sliders that may invert negative. Some do it well, some don't. The main point of this thread was to show the core of the invert and RGB Levels process, to help to understand hidden technical problems, to explain how tonal arrangement works in L* gamma and how camera calibration affect colors in final image... By the way, Ed_Ingold, here is image converted from your JPEG example with method explained first post :) Jpeg only was converted to ProPhotoRGB and L* gamma for start. I done this just as extreme example. I don't suggest to process film negatives from JPEG sources.
-
Yes, there are a lot of other tools. I collelct some very subjective personal thoughts about alternate apps and tools for negative film processing that i use and don't use: darktable negadoctor standalone app darktable - the photo workflow software Free all-in-one tool. Allow native RAW workflow. Nice debayer quality and nice options for Fujifilm X-Trans sensors. For my personal taste it is too "consumer" and too abstract to use. It works OK with simple images, but don't allow to fine tune films with complicated colors and unusual tonality. darktable allow to use custom working color spaces but looks like operates with film negatives only in sRGB gamma. Controls may feels not too intuitive for the first look, but there are a lot of video tutorials on Youtube and Vimeo. Negative Lab Pro plug-in for Adobe Lightroom Negative Lab Pro Overall it produce same result as darktable negadoctor. Same time it feels like slower and less comfortable to use. It have some amount of useless consumer presets and "looks" variations. ColorPerfect plug-in for Adobe Photoshop ColorPerfect Plugin for Adobe Photoshop | PSE | PL Worst investment ever. Overpriced app that seems like designed to made impression of something "very complicated" and "very important". Worst ever made Interface. Very slow processing. Low quality results. Clipped dynamic range. Hard clipped contrast adjustment. No real color management. Can't read camera input color profile. Communication with developer is possible but goes to nothing. Instead of clean answers to simple straight questions he wrote back giant pseudo technical texts that explains nothing and suggest to process images in sRGB color space. FilmLab standalone app FilmLab I check demo version and it feels like some joke. Very limited controls. Colors are always off and it feels like it attempt to prettify image and add some sort of fake "filmic look" LUT on top. Grain2Pixel plug-in for Adobe Photoshop Grain2Pixel – FIlm Photography Tools I attempt to test Grain2Pixel, but can't run it because it require newest version of Photoshop. Very strange limitation. CNMY plug-in for Adobe Photoshop CNMY film inversion I didn't test this app yet, but hope to check it someday. Negmaster plug-in for Adobe Photoshop NEGMASTER - A brilliant Photoshop plugin for converting negative and slide film I didn't test this app yet, but hope to check it someday. SilverFast NegaFix I didn't test this app yet, but hope to check it someday. From online video examples tutorials user interface looks very outdated, ugly and not logical. Analogue Toolbox for Capture One plug-in for Capture One Analogue Toolbox for Capture One - Curious Photography by Michael Wilmes I didn't test this app yet, but hope to check it someday. SILKYPIX Negative film inversion tool standalone app [Pro10] The Negative film inversion tool | SILKYPIX I didn't test this app yet, but inversion examples on their website looks very bad. Hope to check it someday. VueScan standalone app VueScan Scanner Software for macOS Big Sur, Windows 10, and Linux This is another example how to make simple things in too complicated and non human friendly way. It works, it scans, but i hate the logic of this app and hope i will never to use it. Also it may produce strange pixel patterns artifacts with some film scanners models. Have no idea was this problem fixed or not, but i always recommend to use original scanners drivers and original software if possible. PhotoLine standalone app PhotoLine: Photo Editing, Vector Editor, Design Software Sort of "Secret weapon". Without digging into details, from technical point of view compare to Photoshop it is like intergalactic spaceship compare to basic airplane from 1970 era. PhotoLine was started in early 90's (near same time as Photoshop). It is made by very small team of developers in Germany and not too advertised as most other graphic editors. Developers focused mostly on technical side and don't care too much about interface look and pretty icons. So visually it is just not as pretty looking as Photoshop and i guess this is the main reason why many people reject to use it. Tools and logic in PhotoLine in most cases are superior to Photoshop, but some tools in some places feels slightly limited and require some updates and improvements. I started to use it about year ago and same time start to help on forum with bugfixes and ideas as a volunteer. During this time developers add huge amount of changes, bugfixes and improvements. I also start PhotoLine UI Icons Customization Project to make it look more user friendly PhotoLine UI Icons Customization Project by shijan on DeviantArt Iridient Developer standalone app Iridient Digital Very solid, powerful and simple RAW editor. Great debayer options and sharpen quality. Probably the best one for Fujifilm X-Trans sensors. A lot of other unique options. Quick and easy to use. UI is very basic and not too pretty.
-
digitaldog, It is subjective opinion that image looks better after calibration and i don't have any special test measurements. The difference is easy to see with naked eye, so i really never thinking about inspecting this in more scientific way. Here is link to reference data file for this test chart scan if it helps, i guess it should have all reference L a b color target values there http://www.colorreference.de/targets/E130401.zip Here are also visualizations of ICC profiles generated with this slide film TI8.7 chart compared to sRGB. As you can see original color space of the slide film is way larger than sRGB and AdobeRGB color spaces. Film negative color space arranged is near same way. This is why it is not recommended to invert and process your scans in tiny sRGB or AdobeRGB color spaces. Use ProPhotoRGB color space and keep all original colors.
-
Just look at Yellow patches, they are the most notisable. Also weak greens. Due low quality light source, yellows have strong green tint in image that use original camera input profile. It is partially fixed by Matrix-based ICC profile. And very way better fixed by Lab cLUT ICC profile.
-
Ok, back to technical research. Here are some examples of camera calibration importance. This example was taken with Fujifilm X-Trans sensor and low CRI consumer LED backlight. As you can see proper camera/scanner calibration applied to image may fix a lot of problems caused by poor quality lighting. And as expected, Lab cLUT ICC profiles produce the most accurate colors.
-
File opened in ACR and imported into Photoshop with no changes is a bad idea because ACR always apply additional tonal curve to image and it destroy original tonality of film source. Processing film negatives in sRGB also very bad idea because sRGB color space is way smaller than native film color space and negatives processed like this can't be counted as real digital copies. Mistakes in digital workflow and color management are not the same as variations in film color. But from some point of view it all depends of luck and presence of extreme saturated colors. Sometimes even images processed in sRGB and ACR may look nice. "I still don't fully understand why the red channel has a tendency to become overly saturated by using a filter. It needs further investigation. However, I see use of a mask correction optical filter as a step forward in the workflow." because you process in sRGB, because ACR add additional contrast to source image and because blue filter changes color balance too much.
-
OK, after all that fight against AutoLevels and 32 bit processing it all ended up with PhotoShop - Autocolor :) Your examples only show that there is a some difference between WB and optical blue filters. I wrote the same thing in the first post - WB in RAW is not a good way neutralise the mask, so i have no idea what "BS or graphs" are you talking about. To be honest i have no idea which image from your examples looks better. If compare side by side, image with blue filter just have more cyan tint, so blue sky is cooler, but grass and train are more brownish. In image without optical filter sky looks more transparent, and green grass looks more fresh and greener. A lot of things are unknown from your examples: How did you process RAW file? What light source are you use? What color space and gamma are you use to invert and process images? Are you add blue filter and set camera WB, or set WB and add blue filter after that? Are you use custom-made camera input ICC profile (and if yes, so what type or profile?) or you use basic camera profile bundled in RAW editor? All these little things permanently affect colors. For example Lab_cLUT-based input ICC profile type produce very accurate colors but require more accurate workflow because may clip color space if something set up wrong. Same time Matrix-based input ICC profiles are less accurate in color reproduction but more flexible and simpler to operate. In most cases RAW editors use Matrix-based dual illuminant input profiles, so RAW files are less affected by problems with color shifts during WB adjustments. Also in many situations film negative image that looks less neutralised at the start may produce better colors after editing, because original relations between colors where preserved in that image. Examples from link in my first post scanned with Minolta Dimage Scan. I specially edit these images in ACES color space to illustrate how far colors may shift in largest color spaces: WB adjusted in scanner (orange mask removed, but relations between colors are lost): WB untouched in scanner (orange mask untouched, but relations between colors are preserved):
-
It is same as discuss "other opinion" about flat earth. You post some imaginary tech fake that based on nothing and ask to discuss it as other opinion. And seems you do it permanently in other threads as well and try to confuse people. 1. There is no relation between Bayer pattern that have more green pixels, and debayered to RGB image color. 2. Tools in any app that can process data in 32 bit depth work in same "clipless" way. Davinci Resolve, PhotoLine and many others ... Same goes to audio apps that process music effects in 32 bit depth in same clipless way.