Photo Notes
Updated 10/4/2014 / superseded by revision on 10/25/2014

My mechanical setup for flying photos is pretty simple.  I carry a small compact camera in my top pocket and take it out when I can.  I don't have a leash, so I only get it out on glides when I can let go of the control lines.  It is a awkward to take photos to the rear.  Since I only shoot on glide, I mostly take photos of where I'm going rather than where I've been.  For rear shots I hold the camera out and guess at the composition, then deal with cropping later.

My flying camera is a Cannon S100, which was state of the art for compact point and shoot cameras when I bought it in the spring of 2012.  It has a large sensor for a small camera, but that isn't necessary for flight photos because our lighting is usually strong.  It records photos in both JPG and CR2, which is the Cannon Raw format.  I convert the CR2 files to DNG, which is Adobe's Raw file format.

I set the camera on Aperture Priority (Av) and let the shutter speed float.  I crank the ISO all the way down to about 100 on sunny days, and sometimes 200 on cloudy dark days.  The ISO relates to the scale of light amplification.  Higher numbers provide more amplification at the sensor level.  High amplification also introduces noise, so using a lower setting will yield a finer grained image, but going below about 200 on my camera doesn't seem to make a noticeable difference.  If it's real bright, I sometimes reduce the exposure by 1/3 of a stop.  The exposure isn't real critical when shooting in Raw because you can adjust it in editing, but the editing will only correct so far, so it's nice to be somewhat close to what you need.  I usually shoot at wide angle because I'm mostly capturing scenery, but will sometimes zoom in when trying to capture a specific object like another glider.  I'm usually flying alone so you won't see many other gliders in my photos.  When the light is bright I'll choke down on the aperture a bit to reduce the light, which also provides more depth of field, which doesn't matter in the flying shots because we are focused at infinity.  The lenses will open up to 2.0 when zoomed fully out, so I might use 4.0 to minimize running out of room on the top end of the shutter speed (higher F stop numbers equate to a smaller aperture opening).  I like the shutter speed to be around 1000 or more to avoid blur from shaky hands, but it only goes to 2000 so there is only one F Stop to play with.  Perhaps I should consider shooting in Shutter Priority rather than Aperture Priority?  A review of the meta data from my 10/4/2014 photo set shows the camera hitting the maximum shutter speed on most of the shots.  If I shoot in Shutter Priority, I think the risk is underexposure, compared to shooting in Aperture Priority, where the risk is over exposure?   I could fiddle with the settings on each shot, but I'm in point and shoot mode.  I configure the settings before launch and sometimes evaluate during editing.

The editing workflow consist of an initial viewing pass to select photos, then another pass to thin the list again and give them names that force a sort order and contain some information to facilitate quick ID from a text list without having to view a thumbnail.  I then put a batch suffix on the original jpg files and save them for reference.  I don't directly use the original jpg files, but they do contain useful meta data like the time stamp and camera settings which is easy to access by right clicking in Windows Explorer.

Next comes the editing in Adobe Camera Raw.  The advantage of Raw photos is they contain much more luminance information than a JPG file which discards the unused luminance detail.  Most lighting scenarios have some parts of the frame that will be overexposed and some areas will be under exposed.  I use Adobe's Raw editor to make the luminance adjustments and add a little pop with contrast, saturation, and temperature applied with gradient filters and brushes.  I also adjust the rotation if it isn't too far off, and if there is a definitive horizon, I correct the lens distortion.  Raw editing is non destructive.  It doesn't change any pixels, it simple creates a text based instruction set on how to render the image, so you can tweak away, stepping back and forth if you're fickle about the results.

One of the issues with the editing for screen presentation is that everyone is viewing on displays with different color profiles, so there is no way to control exactly how a photo will appear.  For photo editing I use a 1920x1080 monitor with the color adjusted to closely equate with the results I get out of my photo printer, but most monitors are oversaturated a bit to provide more pop, so my renderings may appear oversaturated on some monitors.  I also tend to overdue the temperature toward the blue side near the horizon because the horizon details are typically washed out.  Locally reducing the highlights and exposure and adding a little blue seems to bring out the outline edge details in the washed out horizon profiles.  Realizing that the renderings may appear oversaturated and blue on many monitors, I try to restrain the temptation to add more pop.

After Camera Raw editing, the photos are "rendered" or opened in Photoshop.  Photoshop has some tools that are better than Camera Raw, like the cropping tools, and more rotation range, but rotation is destructive (pixels are square), so it's best to only do it once.  Some images like Google Earth screen captures go directly to Photoshop without passing through Camera Raw, so I'll use the Photoshop adjustment layers for non-destructive luminance adjustments for those images.  Photoshop also has better tools for various local touchup task, but they aren't usually needed for scenery shots.  If I need to stitch photos together, which is sometimes necessary with Google Earth screen captures, Photoshop has good tools for those task.

When I'm done with the Photoshop editing, I set the DPI to 200 and reduce the size to a maximum of either 1920 horizontal or 1080 vertical, which equates to 1080 HD TV.  I used  to make the photos smaller, and sometimes bigger, but storage is cheaper than it was 10 years ago, and most people have access to larger monitors.  I rarely post bigger photos because almost none of them will be printed.  Most readers will just flip through without wanting to zoom in.  For what we are doing, you can use Google Earth for zooming and further study.  If I capture a cherished quality shot that I might want to print, I'll keep a full size Photoshop version.  DPI doesn't' usually matter for a screen presentation, but by making all the photos the same DPI there is consistency in the text point sizing.  A 1920x1080 photo at 200 DPI will fit on letter size photo paper without needing to tweak the DPI settings.  I don't try to optimize the photos for small phone screens.  The communication intent is better realized on larger monitors or TVs.  If someone just wants the latest news, they can scan the headlines with their smart phone, but to fully engage in the presentation, you need a bigger screen.

When the Photoshop files are done, I open them in Illustrator for the markups because Illustrator has much better drawing and markup tooling than Photoshop.  To nail the registration, I initially "open" the Photoshop file in Illustrator, which forces the correct size, then delete the image and "place" it back into the file.  I then force registration (alignment) by setting the upper left corner of the placed object to 0-0.  By placing the image, the photo isn't part of the file.  The Illustrator file only the references the Photoshop file, which results in smaller Illustrator file sizes, and I can still make adjustments to the referenced image in Photoshop if necessary.  One disadvantage to "placing" the image object compared to "embedding" is that you have to maintain referential integrity, or re-define the path if the link is broken.  After placing and forcing registration, I lock the image layer and make a new layer (or more if needed) above the image layer for the markups.  I use the markups to communicate, but you can't put a lot of text on a photo, so the flight article narration is still functional.  I do all the markups before making the jpg files and review the set for communication intent.

When I'm done with the markups, I switch off the image layer in Illustrator file and "place" the Illustrator markup layer (or layers) back into the Photoshop file.  It is important to turn of the image layer in Illustrator because you only want to import the markups, not the image.  You don't want more than one copy of the image in the Photoshop file.  It should be noted that unlike going the other direction, when "placing the Illustrator markup layers into Photoshop, they are embedded as a new Photoshop layer rather than externally linked, so if you want to make changes to the markup you need to delete the placed layer in Photoshop and place the revised markup again.

I then make the JPG files from Photoshop.  I mostly use an upper medium compression setting of  7 for flying photos.  The Photoshop scale goes to 12, with 7 being the highest medium setting, and 8 being the lowest high setting.  I sometimes use higher quality settings, but the quality difference between 7 and 8 is minimal.  If I need to print, I don't print from the JPG files, I use the Photoshop file.

When I have all the JPG files, I import them into my home made web template.  I usually present them in 3 variations, thumbnails, half size, and full size.  Most web browsers will fit a photo to the available screen size if opened as a jpg, which the template will do when someone clicks on a thumbnail, but when the photo is part of a web page, it will not resize.  I recommend that viewers download the entire photo set for viewing with a viewer of their choice.

A final note on copying and sharing photos.  Raw images are too large for email, so cloud drives are the best way to give large file sets to someone.  If a photo is displayed as a simple HTML image, then saving the photo will save the full size underlying image, but some programs generate the photos on the fly, like Google Drive, so the underlying image won't be the full size image.  In that case, do not simply right click and save the photo.  You want to download it.  On my home made simple photo display template, the photos are displayed in 3 sizes; Thumbnails, Half Size, and Full Size.  The thumbnails are separate files, but the Half Size and Full Size photos are actually the same underlying image.  As noted above, if a photo is part of a web page, its screen size is fixed and will run off the screen if too large for the monitor.  If a jpg image is opened in a browser as an individual file not wrapped in an HTML container (as with clicking on the thumbnail), the web browser will fit the image to the available screen size, but the underlying image is still the full size photo, so right clicking and saving will save the full size image.  With my home made template, you can save all the photos by opening either the small size or HD size web page containing all the photos, and then saving the webpage (complete), which will make a folder and put all the full size photos in that folder.  Right clicking to save any of the photos regardless of size will save the full size image (except the thumbnails, which are actually small separate files).

If you want higher resolution versions of any of the shots, let me know and I'll give them to you in whatever state of edit you want, or all the edit states.