Articles

Oct 17 2021

My “Recipe” – Digital Processing of Bird Photos (from RAW)

Author: DakotaBirder • Discussion: No Comments

This morning I had a nice short excursion to try to grab some photos of migrating sparrows. It was just an hour, but I managed to photograph 6 species, and got some photos in nice early morning light. As I start to process those photos, I thought I’d share my “typical” recipe for processing photos from RAW from my Canon 90D, something I always wanted to blog about.

I don’t do a huge amount of digital processing. I try to use the WYSIWYG (what you see is what you get) mantra, and keep the habitat and bird as I saw it. I’m trying to show the whole flow for a typical photo, and while the steps below may look daunting, processing from RAW to a final TIF for long-term digital archiving is typically just a couple of minutes for me. That involves two software packages: Adobe Photoshop (and plugins), and Topaz DeNoise AI. A very good example of a typical processing flow follows here, for a Harris’s Sparrow I photographed this morning on top of a fence post.

Open and evaluate the photo in Adobe Camera Raw. Here’s the original photograph straight from my 90D, as it appears when I first open it in Adobe Camera Raw. Note that as is often the case when you’re shooting birds, there’s a lot of “empty” space. You often try to “fill the frame” with the bird to get maximum detail, but birds don’t often cooperate! Thankfully with today’s DSLRs, there are plenty of pixels where you can crop some of the frame, and still have sufficient detail for large-size prints. I’ll eventually crop, but for now my main focus is on the overall exposure.

Example - Adobe Camera Raw screen
Original photo, straight from my Canon 90D, and what it looks like when I open in Camera RAW.

Correcting for exposure: The first thing I notice in the photo is that it’s overexposed in the whites on the underside of the sparrow, and a bit too bright for the scene overall. Not an uncommon situation, as in this case, the sparrow popped up on the fence post for just a second, and I quickly grabbed a photo with a lot of time to change camera settings. If you don’t like to deal with the hassle of shooting RAW and shoot in JPG, you’re already in trouble, as that exposure is baked in and you don’t have nearly as much flexibility in saving that detail in the whites. However, in RAW, you can adjust that exposure before converting to your long term archive format (TIF for me).

Below is a zoom of the bird as I correct exposure in Camera Raw. It can often be done simply in the “Basic” tab, with slider bars where you can correct exposure, contrast, and other elements. Here I’ve simply made three quick changes to lower exposure overall and bring out some detail in those whites (with all other settings shown on the right representing default values for this shot):

  • Drop exposure overall for the whole scene by -0.40
  • Change the “highlights” setting to -80. This lowers overall exposure only for the brightest parts of the scene, a very useful feature in this situation when I want lower exposure in the whites.
  • “Dehaze” increased by +10. Dehaze is described as basically compensating for light scatter in the atmosphere. It’s a strong tool that really is wonderful in providing increased clarity to an image when shooting in hazy conditions, but overall it does a nice job in deepening contrast. Here just a touch of an increase results in some darker background tones and darker tones on the bird, giving the image more “pop”.

Note the end result of this basic adjustment on the bottom, with whites on both the fence post and bird showing much more detail, and a less starkly bright exposure overall. There are other tools in Camera RAW you can also use for exposure adjustment and the like, with Curves being the other tool I often use.

Before and after - Changing exposure settings in Adobe Camera RAW
Before and after applying exposure and other corrections prior to RAW conversion

Other settings for RAW conversion: For this photo there’s not much else I want to do before converting from RAW to TIF. Note the “Detail” tab. I set everything to 0, as I don’t want Camera RAW applying sharpening or noise reduction, as I don’t think it does nearly as good a job as Topaz DeNoise AI (coming step). Also note the “Optics” tab. I always make sure I check the box for “remove chromatic aberration”. It’s not going to make much of a difference for this image, but for an image such as a bird in flight against a bright blue sky, it corrects for the “color fringing” that can sometimes occur. This is very lens dependent, and my primary birding lens (Canon 100-400mm II) has a fluorite and other elements that help minimize chromatic aberration. Still, it is sometimes noticeable a the fringe of a very high contrast area as a pixel or two coloring, and this setting helps correct.

Camera RAW provides a specific correction for the lens you’re using, as it does for “profile corrections”, another box I always check. This corrects for distortion and vignetting. Neither are huge issues with the Canon 100-400, but there are other lenses I have where use of this setting has been quite helpful for correcting vignetting (a shadowing around the corners or edges of a photo).

That’s it! White balance overall looks pretty good on this photo, but correcting for color imbalances is about the only other correction I typically handle in the RAW conversion stage. If white balance is off and there’s an odd color cast to a photo, there are ample tools in Camera RAW for correcting, from basic changes in color “temperature” and “tint”, to more specific corrections. Now I’m ready to convert to TIF…

Basic adjustments in Adobe Camera RAW
Settings for both “detail” and “optics” in Adobe Camera Raw

Conversion to TIF: There’s not much to this step. There’s a “down” arrow in the upper right of Camera RAW that opens up the conversion to TIF (or other format). I use TIF for long-term archiving of a photo as it’s a “lossless” form of photo storage, as opposed to JPEG. JPEG provides a smaller file, but at the cost of image quality. To be honest, it’s virtually impossible to visually distinguish between the highest-quality settings JPEGs, and a TIF. You do have peace of mind with a TIF though in that your photo will always be saved at the highest quality, while each new save of a JPEG may cause an additional small loss of information.

Settings are below…not much to it! The only other thing to really consider is colorspace for your output. That’s a topic for another blog post. Take my word for it and just choose SRGB, and go ahead and convert.

The result here…a beautiful, “lossless” TIF file with 6960 x 4640 pixels that’s corrected for exposure, chromatic aberration, and vignetting.

Converting from RAW to TIF in Adobe Camera RAW
Screen in Adobe Camera RAW for conversion to TIF

Noise Removal: This photo was shot in pretty early morning light, and I used ISO 1000 to ensure enough shutter speed to get a sharp photo. Shooting at as low an ISO as possible is always a good idea to reduce noise, but thankfully there are some wonderful tools for correcting noise in an image. For the last year or so I’ve been doing it with Topaz DeNoise AI, simply…awesome…software that I use for both noise correction and sharpening.

In the zoomed out photos above, it’s hard to see much of the noise, but for “pixel peepers” like myself, I’m not fond of the noise you see when viewing an image at full resolution. Remember when converting from Camera RAW, I didn’t do any correction for noise, so my converted TIF right now is showing all the noisy warts from the sensor on my Camera 90D. Next thing I do is load the TIF into Topaz DeNoise AI.

Topaz will both remove noise and sharpen the image. When you load the image in Topaz DeNoise AI, it will select default “remove noise” and “enhance sharpness” settings, based on its own evaluation of your photo. To show you the effects of each in turn, first here are default settings from Topaz DeNoise AI in reducing noise for this photo. Note the background color around the bird and the dramatic improvement in noise (image immediately below). Also note the eye itself, with the textured look in the original image, vs. the smooth, noise-free look in the corrected version.

I flipping love this software, but do realize it’s not everybody’s preference to dramatically reduce noise. Some people prefer some of that original “grain” in a photograph, which I think harkens back to the days of film photography where that grain was part of the image. I say bull-puckey to that! Noise in a digital camera is just that…noise…not a feature of the subject or environment you’re trying to photography. It’s a digital artifact. Reducing that noise provides a much cleaner image, with the smooth even tones that represent the reality around us. Particularly for a photo like this where the background is this wonderful smooth bokeh from the original photo, noise reduction makes for a much more gorgeous background.

The software is great, but beware getting TOO aggressive in trying to remove noise, as you will start to lose detail in areas where you don’t want that detail lost. If you get too aggressive in noise reduction in Topaz DeNoise AI, I’ve also noticed some smudgy edges that can result, as as the border between a bird and a background.

Noise Reduction in Topaz DeNoise AI
Noise reduction example in Topaz DeNoise AI

Sharpening: I also use Topaz DeNoise AI for basic sharpening of the full resolution imaging. Don’t expect miracles with any sharpening software. Garbage in = garbage out, and if you don’t have a good sharp image to begin with, a sharpening algorithm isn’t going to save the shot. However, it can give it that extra “crispness” that can really help bring it over the top.

I used to use Photoshop sharpening tools, but now find the default sharpening in Topaz DeNoise AI to be superior. It helps too at how easy and fast it is, in that I can both remove noise and sharpen an image simultaneously. The first image below shows the software screen itself, and how very simple the basic operation is. If you set it to ‘auto’ it evaluates your image and chooses the ‘correct’ level of noise reduction and sharpening. I always let the software calculate auto settings, and then it shows you a preview of the impact of those settings. Pretty simple…if you want to increase noise reduction from that default setting…simply slide the bar and get an instant preview of the impact. Same for sharpness. You can also play with “recovering original detail”, which “places back” detail that may be lost by the noise reduction, or you can specifically target color noise reduction. In this case, basic noise reduction and sharpness effects are damned good, and I didn’t change the settings. I will often play around with the settings though before saving a new output image.

The image at the bottom shows a before and after, for doing BOTH the noise reduction, and sharpening. For sharpening, it’s subtle, as the original image was quite sharp on it’s own. But you do see some more crispness in the feathering around and in front of the eye and elsewhere in the image.

Topaz DeNoise AI - Example of screen
What the Topaz DeNoise AI software looks like as you’re processing.
Before and after - Sharpening with Topaz DeNoise AI
Before and after processing through Topaz DeNoise AI, showing the improvement in sharpening in the crisper feathering on the bird.

Final step – Cropping to preference – That’s pretty much it for most photos! I do keep it simple. At this stage I still have all the extra space around the bird, so my final step before presentation online, or for a print, is to crop to a pleasing composition. I do typically keep the corrected TIF file though with all the original pixels, in case I want to reframe a photo later with a different presentation. Here’s the final output (compare to the top original photo above). Even with the cropping of all the empty space, thanks to the 32.5 MB sensor on the 90D, the photo has enough remaining pixels to retain a lot of detail, and provide 300 DPI prints up to 10×14 or more.

Note some additional processing steps I sometimes take are below.

Harris's Sparrow - Zonotrichia querula

“Other” processing steps: This example with the Harris’s Sparrow is pretty typical, when I’m dealing with a pretty high-quality photo to begin with. Dependent upon the photo some other steps I sometimes take are:

  • Shadow/highlight adjustment – You can’t ask for much better lighting than the Harris’s Sparrow above, with warm, early morning, and even lighting and a bird that’s posed perpendicular to the light source (behind me as the photographer). That’s obviously often not the case! In harsh lighting situations or other cases where the result is a high-contrast image, sometimes deep shadows are a result, and you’d like to try to bring out some detail in those shadows. There are shadow/highlight tools in Photoshop that I’ll use in those situations. Note dependent upon the photo, trying to bring out detail in the shadows can end up simply bringing out the noise in those shadows. Again, garbage in = garbage out, and if there’s not detail in those shadows to begin with, you’re not going to magically pull it out in post-processing. The best time to try to try to pull out details from the shadows is during RAW conversion, analogous to what I did with this photo in trying to bring out detail in the (overexposed) whites.
  • Selective adjustments – Sometimes you don’t want a specific adjustment to impact the whole image, but just one part. In that case there are many tools in Photoshop that allow you to select specific parts of the image, and apply color adjustments, exposure, sharpness, noise, or other changes to only the selected area.
  • Cloning out unwanted elements – I try to keep my photos as close to the original scene as possible, and that means inclusion of habitat elements if they were in the shot at the time. However, Photoshop has tools that do make it quite easy to “get rid of” unwanted elements in an image. For example, in the final image above, it’s simply the sparrow on a post, with a perfectly “clean” background. What if there were a few blades of grass or an overhanging branch that intersected one corner of the photo (for example)? In that case, it is extremely easy to use the “clone” tool…selecting an area of the background nearby the unwanted feature, and effectively copying and brushing over the unwanted feature. You can often thus end up with that “perfect” background with the unwanted element removed. I try to keep my use of this to a minimum, for those really “special” images where I want that last bit of perfection before printing or selling.

That’s it!! – I’m not selling my typical “recipe” or toolset as the best answer for everyone. A bit part of your own personal workflow is 1) what software you have, 2) what you’re comfortable with, and 3) what effects you’re trying to achieve. The process above “works” for me as I’ve used it for countless thousands of photos, and I can whip through any one photo quite quickly. Your mileage may vary, but I hope the examples above are helpful.

Apr 04 2017

Evolution in the blink of an eye…

Author: DakotaBirder • Discussion: No Comments

Prairie Deer Mouse - Peromyscus maniculatus bairdii

The cool thing about science and nature is that interesting stories are all around us. The tiny Deer Mouse, shown here, has overcome long odds, with the vast majority of its historical habitat gone. However, through some remarkable, fast-track evolutionary adaptation, they’re now able to cope with their new world. Photo by Gregory Smith.

It’s been a busy last week, without any time for birding or photography.  Or blogging, for that matter. I was down in Nebraska for a few days, mixing work and pleasure. The “pleasure” part was my fantasy baseball draft in Omaha Saturday.  Our fantasy league is likely one of the longest running leagues in the country, going back to 1985 during our freshman year in college, when fantasy baseball was still very new.  What’s great about it is that many of the original league members are still participating! It’s great fun, not only the draft itself, but catching up with old college friends.

The “work” part of my Nebraska trip was participation in the 2017 Great Plains Symposium, on the campus of the University of Nebraska-Lincoln. Much like the baseball draft, the symposium too was like stepping back in time, as I reconnected with some of my old college professors who were participating in the symposium. The focus of the symposium was “Flat Places, Deep Identities: Mapping Nebraska and the Great Plains”.  I gave a talk one some of the work I’ve been doing, mapping past, present, and potential future landscapes in the Great Plains.  It was a great symposium, a little different kind of crowd than I’m used to.  Given the work I do, most of the conferences and symposiums I attend deal with the physical sciences. This conference melded mapping, history, socioeconomics, and other social sciences that I’m not exposed to as much.  It was quite fascinating, particularly hearing about the history of Nebraska, using maps to help tell the “story” of change over time.

As part of the symposium “goodies”, participants were given a copy of The New Territory, a quarterly magazine that focuses on Missouri, Arkansas, Kansas, Nebraska, and Oklahoma.  I admit I’d never heard of the publication before. The content fits quite well with the focus of the symposium itself, with many human interest stories about the geography and people of the region. As a physical scientist, one piece caught my eye though. entitled “Evolution in the Cornbelt“, by Conor Gearin. The story focuses on the Prairie Deer Mouse (Peromyscus maniculatus bairdii), a common little fellow from the Great Plains that feeds on the tiny seeds of grasses and weeds in the prairies.

Researchers at Iowa State and Purdue University were curious how a species so adapted to life in the Great Plains has been able to thrive, given that >99% of the original tallgrass prairie in the region has been plowed under, converted to agriculture, urban land, or other man-made land uses. The grass and weed seeds the Prairie Deer Mouse had historically fed on were much more sparsely distributed than they were 200 years ago, yet the species is still quite common.  They started field work to assess the distribution of the nice, including setting up artificial nest boxes that the mice could use for habitation and food storage.  The results astounded the scientists.

Prior to beginning the work, it was assumed that deer mice populations would be the highest in “edge” habitat, areas such as grassy ditches, fencelines, or other “boundary” conditions where remnants of their traditional food sources may still be found.  However, they quickly found that the highest populations of deer mice were often right in the middle of very large corn and soybean fields, far from any traditional food source.  Clearly, Prairie Deer Mice had adapted to an agricultural setting, and were feeding on man-raised grains and pulses. The question was, how could a tiny mouse that was so well adapted to eating tiny grass and weed seeds shift gears and start feeding on corn and soybeans?

The researchers found historical deer mice in historical museums, creatures that had been preserved with taxidermy. Anatomical comparisons with Prairie Deer Mice from today found some stark differences.  The older specimens were well adapted to feeding on tiny seeds, with small mandibles and jaws that didn’t open very far.  The modern specimens had 1) significantly longer lower mandibles, 2) structural changes that allowed their mouths to open wider, and 3) larger upper mandibles. Accompanying the larger mandibles were more robust “hardware” for linking bone to muscle, with beefed up jaw muscles that enabled the tiny mice to feed on much larger food items than they had historically.

In the blink of an eye, geologically speaking, Prairie Deer Mice had shown measurable, obvious evolutionary adaptation in response to their new environment and food sources.  The researchers found high densities of deer mice in the middle of corn and soybean fields.  Some inevitably will succumb to the mechanical tools humans use to turn and manipulate the soil, but with such a rich, dense, bountiful food source, the mice had quickly evolved to fill the new ecological niche and feed on corn and soybean waste.

For a scientist like myself, I’m completely dumbfounded by the sheer ignorance of those who doubt science…who doubt climate change is real…who doubt in evolution.  The actual empirical evidence is overwhelming, conclusive, and “in-your-face”, for those who bother to open their eyes to the world around them. It’s a fascinating story, and the writer (Conor Gearin) did a great job not only summarizing the research, but telling it in a true story-teller’s fashion.  To me, this is exactly the kind of story, and writing style, that could perhaps help to turn the tide against the anti-science wave that seems to be cresting in the U.S. right now. Great story, and The New Territory really looks like a publication that’s worth subscribing to or picking up if you get a chance.

Grand River National Grasslands, Harding County, South Dakota

Expansive grasslands of the Grand River National Grasslands, in Harding County, in far northwestern South Dakota. Grassland habitat like this is greatly reduced in the Great Plains. However, that doesn’t seem to be a problem for one species, the Prairie Deer Mouse, who evidently can do quite well without an actual “prairie”.

Feb 26 2017

Losing the Rainforests…and South Dakota habitat

Author: DakotaBirder • Discussion: No Comments

This week the New York Times had a wonderful (as always), yet sad piece about deforestation in the Amazon rainforest. Approximately 360,000 acres in the region were deforested every year during the 1990s, a number that jumped to 660,000 acres a year during the early 2000s.  A massive push to slow deforestation rates occurred in the 2000s, temporarily slowing the rate of deforestation.  However, in the last few years the deforestation rate has skyrocketed, to over 850,000 acres every year, an area the size of Rhode Island. Every. Year.  Clearly that’s not sustainable.  Even the supposed “success” in the mid-2000s of slowing the rate of deforestation only “slowed” it, it certainly didn’t stop it or reverse the trend.  That’s the world we live in now, where SLOWING the inexorable loss of habitat is considered a major conservation success story, even if those slower rates still would have wiped out most of the rainforest during this century.

We don’t have rainforests in South Dakota.  From a birder’s perspective, we don’t have much bird habitat whatsoever in the eastern part of the state, given the preponderance of corn and soybeans that takes up the vast majority of the landscape.  Still, as a birder, I have reveled in the little reservoir pockets of remaining habitat, small micro-habitats where birds have thrived, despite the massive use of the landscape for agricultural production.  I used to bring my camera with me EVERYWHERE.  Every day when I went to work, my camera came with me.  I would stop at these little pockets of habitat, and take bird photos. Over the years, I’ve gotten some truly wonderful photographs in these small remaining pockets of habitat.

I don’t bring my camera with me to work any more. I don’t bring my camera with me when I run errands. In fact, I don’t do nearly as much birding right around Brandon and Sioux Falls as I used.  Much of the reason is that many of my former little micro-habitat “hotspots” are gone, something that’s just happened in the last several years.  There have been multiple reasons behind it.  The first is simple economics…with demand for corn and soybeans, farmers are cultivating every possible patch of land to maximize production. Fence rows, shelter belts, and other little pockets of habitat are being plowed under to expand planted acreage. There have also been active “safety” programs in the last few years to clear brush and trees from the edges of the roads.  It’s been a truly massive project, with roads all over the state undergoing this kind of “grooming”, removing habitat that is anywhere close to road edges.

We don’t have the rainforest like the Amazon, but habitat loss is having an impact right here in South Dakota.  Here are some small, and some larger, examples of what’s happening with habitat change in South Dakota, and how it’s affecting bird species. Bird photos accompanying each image are some of the actual bird photos I’ve gotten from each location over the years.

Ditch Road - Minnehaha County, South Dakota

Ditch Road, just north of Sioux Falls in Minnehaha County. There’s a stretch of road that runs over 5 miles that has a straight drainage ditch running along side it. In the last year, nearly all of the thick trees and shrubs that were found between the road and the waterway have been removed (the area encircled in red shows the vegetation that used to be there). It’s part of the aforementioned “safety” program to remove things that people could evidently crash into and get hurt.  (I always thought the point of driving was to stay ON the road). With the water and vegetation, it used to be an absolutely wonderful spot for songbirds and even some waterfowl, particularly in migration.  Warblers, Vireos, and Chickadees and Nuthatches, many woodpeckers, and other songbird species were often found here.  Not any more…

Big Sioux Recreation Area near Brandon, South Dakota

Big Sioux Recreation near Brandon, South Dakota – One that’s near and dear to my heart, given that we live on the edge of Brandon across the street from the park (house shown above). One of the best places to bird in the park used to be right amidst the campground areas themselves. The looped road shown above was lined with cedar trees, and thick brush separated many of the camping stalls from each other. The image above shows what it used to look like. The cedars in particular really attracted many birds, including one memorable winter when a hoarde of about 20 Long-eared Owls took up residence in the Campground (see photo above). In the last 2 years, all of the cedars have been removed from the campground area, as have most of the shrubs that separated camping stalls. If you want to play football? Thanks to all the vegetation clearing, it’s now MUCH more open in the campgrounds! If you really LOVE being close to other people while camping, with no pesky vegetation to get between you and the next guy, you’ll love the changes!  If you’re a bird lover? Not so much…

Minnehaha County Wetland

Minnehaha County Wetland — This one has a Google Earth image that actually catches the transformation as it happened. This is a small area in northern Minnehaha County on 253rd Street and near 481st Avenue.  Prior to 2015, the area in red was a mix of wetland, damp grasses, and weedy patches.  It was an absolutely WONDERFUL place to bird, a little patch of wet, weedy habitat that attracted species like Le Conte’s Sparrows, Swamp Sparrows, Bobolinks, Sedge Wrens, Marsh Wrens, and many other birds. In 2015, the owner installed drain tile, shown as the lines that run through the image above.  The drain tile dried out the land so it could be used for cropland, and today, this entire patch is a corn field. Drain tile installation has been RAMPANT in eastern South Dakota in the last few years. In some cases it’s been done to improve conditions on existing cropland.  In many areas though, like this, it’s being installed in areas that are naturally too poor to support crops, and need an artificial drainage system.  It’s hard to have Swamp Sparrows in an area, if you have no “swamp”.

South Dakota Shelterbelt

Shelterbelts – There aren’t a lot of woodlands and forests in eastern South Dakota, but the variety of birdlife in these little oases of trees can be truly astounding.  During migration they are definite bird “traps”, with tired songbirds stopped to rest and feed in these areas before continuing on with their migration. The photos are just examples of what you can find in these shelterbelts, as I haven’t birded this specific location before. However, it’s a great example of what’s been going on in much of South Dakota. This is north of Sioux Falls, near the intersections of Highways 121 and 122. This shelterbelt had been there the entire 23+ years we’ve lived in South Dakota, and it had some very large mature, old trees.  That changed last year, when all of the woody vegetation in the entire area circled in red was removed. This last summer it was all a big open field, planted in corn. There are many places in eastern South Dakota where this is happening, as farmers try to compensate for lowered corn and soy prices in the last few years by planting more and more acres.

Increase in Cropland Cash Rents - 2009 to 2014

What’s behind South Dakota cropland gains? — So what’s driving the agricultural change in South Dakota?  Beyond the little micro-habitat examples given above, there are some very large swaths of grassland being converted to cropland, with some much of the new cropland on land that had never been plowed before. The map above gives you some indication of the economic forces driving cropland gains in the state, and the concomitant loss of vegetated habitats.  Some of the largest recent changes in cash rent values for cropland in recent years are concentrated right in eastern South Dakota. As this article states, South Dakota had the largest increases in overall cropland value from 2004 to 2014, an increase of over 350% in just 10 years as average cropland values rose from $734 an acre to over $3,400 an acre. Prices for the major crop commodities of corn and soy have softened substantially in recent years, but that seems to have driven an intensification of land use in some parts of the state, as farmers try to maximize production by expanding the acreage that they plant.

Grassland conversion in the Great Plains

Grassland Conversion in the Great Plains – In 2013, colleagues/acquaintances published a paper the summarized the recent loss of grassland and wetland in the northern Great Plains. The map above shows the percentage of grasslands in an area that were converted to soybeans or corn between 2006 and 2011. Southern Iowa is a bit misleading, given that this shows “percentage” change, and there wasn’t nearly as much grassland there in 2006 as there was in parts of eastern South Dakota.  It really has been the eastern Dakotas where a huge chunk of cropland gains in the U.S. have occurred in recent years.  From a birder’s perspective…it hasn’t been a happy story.  Click here for the journal paper from Chris Wright and Mike Wimberly.

%d bloggers like this: