Αρχεία Ημερολογίου για Ιούνιος 2020

Ιούνιος 17, 2020

My iNaturalist Upload Process

I may be the slowest (read: most delinquent) uploader on iNaturalist. For any given effort, especially for the wonderful bioblitz’s that I enjoy participating in, I’m usually the last to get all my stuff up on iNat. An honest mea culpa for such delays will have to admit to being somewhat lazy, somewhat sleep deprived, somewhat addicted to certain TV shows, and generally just interested in what is right in front of me right now, as opposed to “yesterday”. All that said, my editing process is a bit arduous because of my equipment and my personal preferences for image and upload quality. Ignoring all those earlier reasons for my typical pathway to uploads, I thought it might be of some limited interest to at least document the torture I put myself through when I actually do get around to handling my observations. So here are some details.

Equipment:

My trusted photographic equipment is my now-infamous little Canon PowerShot SX620 HS point-and-hope camera. It has the advantage of portability and affordability. I won’t bemoan all the disadvantages, but for present discussion, the main problem—are you listening Canon?—is its lack of GPS datalogging capability with images. That adds one major time-consuming step to my editing pathway, as listed below.

My Editing Steps:

When I’m in the field, especially on a multi-day trip, I download all images daily from the SD card to my travel laptop but keep the images on the card. A 32-gig SD card is sufficient to last me through a 10-day to 2-week trip, depending on how many plant and moth images I take. (I always carry one or two extra cards just in case.) I will often examine my images on the laptop while traveling but I don’t do much editing (other than to examine what I might have documented) because the laptop is not the final destination of the images. The real work begins when I get home to my desktop computer.

I work on an iMac desktop computer (currently running MacOS Mojave 10.14.6) and I’m still using iPhoto (9.6.1 which is now 5 years out of support).

  1. Download the images. I usually let iPhoto separate all the downloaded images into daily “Events”. I manipulate and change those events later on as I organize and group my photos based on destination, subject, etc.
  2. Geotag ALL images. Before any editing starts, I have to make sure all images have proper geographic location data. I add this manually in iPhoto. I can do this in batches based on separate locations, but if I was moving from place to place in a given field day, this can be very tedious. Recall that I don’t have GPS capabilities on the camera, so I either have to take detailed field notes (which I do) to associate batches of images with a known location, or I have to supplement the image set by photographing a screen image of a GPS app when I’m in the field or photograph some other landmark (e.g. a street sign, an Allsups storefront, etc.) so that I can properly place my images.
  3. Rotate, straighten, and crop images. My many iNat friends know how obsessive I am about this step. From a busy bioblitz day in the field, I may have 300 to 600 images to rotate, straighten, and crop (and discard the many blurry ones). Because of the menu structure in iPhoto, these three steps are usually easy to accomplish for a given image in succession. That said, I’m frequently switching between/among image dimensions (4 x 3, 3 x 4, square) from one image to the next; this adds to the editing time significantly. One image proportion does NOT fit all. I may also try two or three different crops for a given image to see which captures the best detail. In some cases, particularly with plants, I may duplicate an image to allow cropping to mutliple details of the plant which might be important for documentation. This all takes time.
  4. Adjust image quality. For many images, the field settings of the camera are often sufficient and the images don’t need post-processing. However, depending on the subject of the pic, or for images taken at dawn, dusk, or at night, it is often useful to brighten an image, brighten the shadows, and/or heighten the contrast to bring out details. As an added burden, when I shoot moth images, to avoid washing out images from a flash at close range I often shoot moths at -1/3 f-stop exposure. I find that on a white sheet or a light gray wall, this can result in images that are a shade too dark—the alternative, over-exposure, results in lost imagery—so I have to brighten many moth images after the fact. Changing the camera settings in real time when obtaining the photos is just too tedious and risky—“shoot now and post-process later!” IMPORTANT NOTE: I rarely adjust colors, hues, or saturation unless the images were obtained in some type of overly intrusive lighting conditions (certain MV and UV lamps, etc.).

    Now the fun begins:

  5. Add keywords and tags. The primary keywords I use on all photos are taxonomic. I use these extensively within iPhoto for organizing and sorting images later. For plants, I have a few general categories including “Plants_flowers”, “Fungi”, “Lichen”, and a few others. For all animals other than insects, I typically use a class or order such as “Mammal”, “Reptile”, “Amphibian”, “Arachnid”, “Opiliones”, etc. For herps, I’ll also add “Snake”, “Frog”, “Turtle”, etc. For all insects except Lepidoptera, I add the order such as “Coleoptera”, “Hemiptera”, “Diptera”, etc. For Leps, I distinguish “Butterfly” and “Moth”. All moths get the “Moth” keyword as well as a family keyword and in a few cases a subfamily tag; these will look like “Gelechiidae”, “Noctuidae”, “Erebidae”, “Arctiinae”, “Pyraustinae”, etc. I have a standby “Fam Unk” for moths which I can’t place; this allows me to collect all those unknowns into one location if desired. For any batch of pics from a field effort, I also apply other keywords for images of “Habitat”, “People”, etc. On long vacations (other than in Texas), I will also add a keyword for the two-letter state abbreviation (AZ, OK, VA, etc.). All this keywording can be done in batches in iPhoto by selecting the appropriate subset of images and adding a keyword once.
  6. Add file titles. My images come off the camera with the sequential filename “IMG_xxxx” numbered from 0001 to 9999 (and repeating). I have found it most convenient to change the “IMG” to the species or taxon identification. With a recent change on iNaturalist, this offers a huge advantage because this filename is now parsed from the image title and added as an identification for each observation. I use scientific names for all of these except for birds which get the standard 4-letter code (such as NOCA_1234 or MODO_5678). Of course, I title my images with the lowest taxon of which I’m certain, so I end up with images named “Melipotis indomita_1234”, “Gelechiidae_1234”, “Calyptocarpus vialis_5678”, “Malvaceae_5678”, etc. Naturally, at this stage, I am identifying all my images, so adding the proper names to each image can either be quick (for familiar plants and animals) OR the culmination of in depth research which takes me all over the place in references, on the internet, etc. This is really the bulk of my iNatting effort at home. Chasing down identifications can lead me down any number of rabbit holes and to various distractions (deep dives in scientific literature, etc.). It also gives me some good exercise as I rifle through the tons of field guides, floras, and manuals on the desk next to the computer. I’ve often joked that when I depart, my obituary will indicate that I was found lifeless under a fallen stack of floras and field guides in my home office.

    Truly, when I finally place a good name on an image for a species new to me, it is one of the most gratifying moments in this whole process. Perhaps that’s the childhood stamp collector in me. Who knows.

    I should add that step 6 and the research that necessarily accompanies it is done in batches. From a given field effort (location, date), I’ll work through all the plants before moving on to other animals, moths, etc. For the Matador WMA bioblitz, for instance, I’m going through all the images of a given day, taking on just the Coleoptera, then the Diptera, then other insects, etc., etc. Then on to the next day or next destination.

  7. Export images. When I have things all or mostly organized, edited, and identified (to a reasonable degree), I export images out of iPhoto to a separate “Uploads” folder on my computer. This separate exporting step allows me to select a good upload size (large but not full sized) and to keep the work flow organized. As with the identification and file naming steps, I will usually do this exporting in batches (plants, beetles, moths, etc.) because it leads directly into the uploading stage. Since this adds a significant amount of disk storage to create these duplicate images (outside of the iPhoto library), I occasionally offload all final images to an external drive and delete them from my desktop computer. I just checked my external drive: The accumulated storage for all my final edited iNat images to date (approaching 26K observations) is about 58 gigabytes.
  8. Upload images. After exporting a batch of final images to my Uploads folder, I use the batch uploader on iNat to upload my observations. I drag and drop and then begin addiing or double-checking IDs, locations, etc., and adding the all-important observation comments. IMPORTANT: Because the batch uploader occasionally hangs up or I make small mistakes in my upload information, I tend to upload only about 10 to 15 observations at a time. That way, I haven’t lost much work if the upload fails or if I have introduced any errors. Even with the efficiency of the batch uploader (a whole other topic), I still make a point of adding geographic names to my observations since iNat’s locational information comes from Google Maps or some other generic source, not my “official” placenames.
  9. Hit the Upload button, sit back, cross fingers, and enjoy the results. I try to glance through all newly uploaded observations just to catch any name or geography errors that I may have introduced. There’s always something. In my senior years now, I find that my brain and fingers don’t communicate as well as they used to when typing. Verb tenses, homonyms and homophones are my bane; my brain knows better, my fingers don’t.

As part of that final enjoyment of an upload—or even during the process of identifying and labeling images (step 6)—I will often take the time to wander through other observations of the same species to see where else the species has been documented. In particular, if it’s a species I am confident at identifying, I’ll take the time to try to upgrade any suitable images of the same species to Research Grade as appropriate.

And that’s my life in a nutshell.

Posted on Ιούνιος 17, 2020 0446 ΜΜ by gcwarbler gcwarbler | 10σχόλια | Αφήστε ένα σχόλιο