Route 15 Bridge Avoids Photoshop

Sunday, April 14th, 2013

Route 15 Bridge, Point of Rocks, MD

Recent posts might give the inaccurate impression that all of my favorite photos have spent long amounts of time in the digital darkroom. While this shot did benefit from the careful selection of 14 different sliders in Adobe Lightroom, along with a minor crop, it’s a single layer, and was not processed in Photoshop (or anything else).

Route 15 Bridge, Point of Rocks, MD

A lot of photographers are blissfully unaware of the degree to which their camera, or the film in their non-digital camera, is making aesthetic decisions on their behalf, and without their cooperation. If you are just using the JPGs spit out by your camera, and haven’t put any significant effort into adjusting the camera’s various processing parameters, then at a minimum, you are letting some Japanese engineer decide how colorful, bright, and contrasty your images are.

The lower image comes as close as possible to showing what the camera saw (Canon EOS 50D, Canon EF-S 17-55 f/2.8 IS USM @37mm, ISO 125, f/5.0, 1/160).  RAW, which confusingly is not an acronym so much as a pretentiously capitalized description, refers to a bitmap dumped directly from the sensor to the memory card, without any processing. Lacking contrast, sharpness, white balance, and the loving touch of a digital darkroom technician, RAW images need to be processed into JPG or TIF before they can be used. The advantage of always shooing in RAW (which is not the same as always shooting in the raw), is that you have the largest possible amount of picture data available to process at your leisure on your PC.

This misty photo of the Rt 15 bridge over the Potomac at Point of Rocks was taken through my Subaru’s window (trick: wait for the wiper to go past) during a record-setting rain in March 2011 (another trick: even if you don’t get out, stop the car—especially if you are the driver).

All Aboard: 20 Layers in Photoshop

Thursday, April 11th, 2013

All Aboard

A few pictures leap right out of the memory card, through your printer, and into the eyes of admiring viewers, looking just like you expected when you took the picture in the first place. Very few.  Most digital images benefit from some time in the digital darkroom, and sometimes it can take several hours of computer work to create a finished image that is worth the price of a sheet of fine art paper.

The intriguingly rusty interior of  Shaker Heights Rapid Transit Car #76, a Pullman-Standard special for that line which served from 1947 to 1975,  motivated me to setup my Canon DSLR on a tripod and spend some time exploring angles and exposures.  For each of several different angles, I took 3 bracketed exposures, recognizing that the tonal range exceeded the capture ability of my camera.  The image below shows the ‘normal’ exposure for that image without any manual attention in Photoshop.

RAW image before postprocessing

My first step was to combine 3 images into a single HDRI image using Photomatix, and then tone map it back into a standard bitmap for editing in Photoshop.  The advantage of HDRI is that it allows you to equalize out the exposure, even on a high contrast scene, such as this shadowy streetcar interior surrounded by the bright sunlight.

High Dynamic Range techniques allow to pull a lot of detail out of the shadow, without blowing out the highlights. It is also an effective way to emphasize texture, so it is well suited to grotty and rusty subjects.  Its also easy to turn a photograph into a cartoon, with distorted and unnaturally vivid colors, and terminally high levels of dirty-looking mid tones.  If I was going to reprocess this image, I think I’d tone down the HDRishness, but printed on Velvet Fine Art paper, I’ve had nice compliments on this image, and it one a first prize at a camera club end of year competition.

All-Aboard-Photoshop-Layers

The image above shows the multiple layers required to pull these surprisingly complex image together. The black & white rectangle visible to the right in most of the layers is a mask, allowing me to apply the effect to selected parts of the image.  I did a lot of local exposure correction to make the steps visible, to ensure that the seat was well lit, to open up some of the details, and to highlight the incredibly filthy window.  I ended up reconstructing the metal strip along the left side of the image to neatly frame it, and I copied the ‘rd’ from the mid-car door, pasting it here to complete ‘Aboard’.  I also pasted in the window and reflection behind it.  Some of the above layers were tonemapped through Photomatix, and some are from a single exposure. Confronted with different textures, I ended up doing different levels of sharpening to different parts of the picture.

Shaker Rapid #71 & #76

I love all the details in the interior, with the handset on the floor, and some kind of control box lying on top of a driver’s seat with a lot of miles under and over it.  The rust and grime, and decay are evidence of the authenticity of use and time. It did feel like I put a lot of Photoshop work into this image, but in comparison to the amount of time Ansel Adams spent in the darkroom on some prints, a couple hours isn’t so bad.  It’s the price you pay when you enjoy found subjects, instead of studio subjects, where light can be much better controlled.

To Tweak or not to Tweak, that is the digital darkroom question

Thursday, August 28th, 2008

For the past year, I’ve enjoyed membership in the Bracknell Camera Club. The speakers, and competition judges, for that matter, seem to be chosen on the basis of how colorful they are. At least from my POV, growing up in America, some of these crusty old analog dudes are…..unusual.

We had a neurosurgeon speak once. He did beautiful B&W photos of symphony performers, and in his oh-so-received accent he explained that he was actually a very simple man who used the simplest gear possible: a Leica with an F/1 lens. I’m never sure if the English actually are polite or not, but nobody snorted out loud.

A more recent speaker (in calendar terms, not age), was a bit more wired, and he actually used our club digital projector, raving over the potential offered by PowerPoint as a slideshow platform. He animated lines across his photos to demonstrate how in spite of his being a completely intuitive photographer, when you looked at his photos, you had to agree that he was instinctively following well-recognized principles of composition. It would be unfair of me to quote him as having said “One doesn’t tweak.” What he actually said was, and you have to imagine an outdated sort of BBC English that is no longer taught (nor received), “Well, of course I use Photoshop, but just for minor exposure and light balance changes. I do not tweak.”

In terms of either the digital or the analog darkroom, just what constitutes a tweak?

I grew up using an old-fashioned darkroom. Thanks to an extremely talented and dedicated teacher, Judi Coolidge, my high school had a serious journalism program, with one of USA’s consistently rated yearbooks and a weekly, multi-page newspaper. You quickly learn that journalistic photography is about coming up with the goods as requested, on time. When you are sent to cover a story, the editor is planning on using your picture, and if you fail, for either technical or aesthetic reasons, then everybody has a problem.

You try your best to expose correctly, and develop the film properly, but if you screw up, you do what you can to make a usable photo. Once you’ve got a negative in hand, there are a lot of decisions to make with exposure, contrast, dodging, and burning. I actually spent a lot of time reprinting other people’s pictures, and sometimes had to use 5-grade high contrast paper to try and make a decent picture out of muddy underexposed images. Was that tweaking? If so, then long live tweaking.

A few years ago, it was all the rage for artsy photographers to file out their enlarger’s negative carrier so that the edges of the film showed, demonstrating that their photo was framed in-camera exactly as printed. What kind of a strange little contest is that? Books and magazines are generally not publishing pictures using the exact aspect ratio of the cameras which took them, so most published pictures are cropped. Does it truly improve the viewer’s aesthetic experience when they know that the photo wasn’t cropped? For that matter, why should the viewer care how many layers I used in Photoshop?

The only thing that counts is what is on the final print. Certainly it helps to optimize the capture and reduce the need for later manipulation, but ultimately, only the photographer knows.

At the top of this posting is a photo I took last year on a Young Life service project to an orphanage in Bulgaria. It was a harsh place, and the kids in the Tran orphanage had a difficult time of it, but for a week, the kids on our service project managed to connect with them.

This picture to the immediate left is closer to the original, but it still has undergone a lot of processing, and it should be pretty obvious that I removed a large brown splotch on the wall. We had just primed the walls, but no amount of primer was capable of fully covering the filth that had been allowed to collect on this wall. From a journalistic point of view, I would probably have left the splotch in, although I’m not certain of that. I don’t think anyone can doubt that as an aesthetic effort, and one that properly portrays the mood of these two girls, the ‘tweaked’ picture at the top of this post, is superior. If you look carefully, you’ll see that the shadow under the redhead’s chin is darker in the ‘original’ than it is above. More tweaking on my part.

The ‘original’ file created by my camera is in camera RAW format, and it cannot be viewed by a human being without being processed in a way that affects the colors, exposure, and contrast. I used Adobe Lightroom to adjusted white balance, exposure and color as my starting point. Then I used CS3 to clone out the splotch on the wall and lighten the shadows, along with a couple other ‘tweaks’ that you wouldn’t notice.

I do feel that much of the power of this picture lies in its authenticity, and that it would be wrong to make substantial manipulations. The redhead actually does have paint all over her fingernails, having ‘helped’ us do the hallways of the orphanage. The other girl was recovering from eye surgery, paid for by a Christian charity. If she had been paying me for a professional portrait, she might have expected that I would open up her eye and align the pupils of both–I really don’t know. I’m comfortable that what I did to the picture was appropriate, and does not misrepresent the lives or spirits of these teens. I tweaked the picture, and I’m proud to admit it.

Lots of photographers like to brag that they capture the picture in camera, and that their pictures don’t need any additional processing. Well, you can still buy transparency film, and as long as you don’t scan it or print it in a darkroom, then you can claim that you didn’t tweak it. Last I heard, a couple guys in Paris had bought up the last of the small-format Polaroid instant film, but even much SX-70 photography is subject to after the fact ‘tweaking.’

The fact is, you can’t reproduce either a film negative or a digital one without making multiple aesthetic decisions about the output appearance. That holds true for journalism and documentary photography as much as it does for purely abstract photography.

The question of what constitutes representational truth and integrity is a deep one, and I’ll be exploring it, and look forward to discussing it. But it isn’t about tweaking. This is a picture. Ceci n’est pas une pipe.

A fake photo on the web? How could that happen?

Wednesday, July 23rd, 2008

It was claimed this month that a picture of an Iranian missile launch was doctored to look more significant. A recent Scientific American article asks Hany Farid (a researcher into digital forensics) how likely that is, and he offers several subtle clues to suggest that this image is not authentic.

Digital Photographs are inherently vulnerable to manipulation. Indeed, manipulation is an integral part of the process of capturing and reproducing an image digitally. Factors that affect the appearance of the final image are set even before a photo is captured, starting with the photographer’s choice of shutter speed, aperture, focal length, position, perspective, timing. etc. All of these are of course relevant to analog photography. Just as different films react to light in different ways, resulting in different images, the camera sensor and associated processing mechanisms output a bitmap that is not identical with the light that fell on the sensor. Before that bitmap can be turned into a visual image, it needs to be further processed, especially if it is a RAW image, which is what most pros and advanced amateurs use.

Although the journalism field has been discussing the issues and problems associated with digital imagery since the 90s, embarrassing photo ‘fakes’ have continued to leak into the major media, which has encouraged the publication of increasingly stringent guidelines. Last year, Reuters shared their Photoshop guidelines on a blog, an interesting example of transparency in the media. In it, photographers are discouraged from doing any manipulation of their photo, including exposure and white balance, and are encouraged to rely on in house experts.

This is an especially sensitive issue with Reuters after a rather obviously manipulated photo caught the attention of the blogging world, and then the mass media (a lengthy and interesting analysis of this and several other faked Reuters pictures from the same period appears on another blog). It is certainly not the case that such incidents are limited to Reuters. A Toledo Blade photographer was let go for pasting a basketball into a game shot, which turned out not to be the first time he had manipulated a picture in a way that was considered inappropriate for the journalistic context.

Photographic integrity comes down to meeting the expectations of the anticipated viewer. If the person who you intend to view your picture knew what you did to it, would they approve? If they saw a before and after, would they consider that you had attempted a deception? Would the deception be for their gain at your loss? Expectations vary widely between media. No reasonable person should expect that the photos in glamor magazines of models look anything like a live human, but in Journalism, even the suggestion of manipulation is unacceptable.

In this month’s case of the faked Iranian missile launch, the New York Times reports that Agence France-Presse picked up the original picture from a political web site, and then it was published on the front page of several American newspapers before France-Presse retracted it. In an age when newspapers are imposing ever-stricter standards on their own people, how responsible is it for a news agency to sell a picture that they copied off the web?

Technorati : , , , , , , , ,