The request was for feedback regarding “image file labelling and metadata use” prompts two issues, the file name used for the image and labels on the images. For images that are linked to a database record I encourage people to start the file name of the image with the DwC term catalogNumber, some people refer to it as the barcode, especially when the catalogNumber is embedded in an actual barcode. Whether it is a specimen label with a 2D matrix barcode, and alphanumeric or both, this catalogNumber label should always be with the specimen. I think this same catalogNumber should also put on the image. So, even if someone changed the file name, there is a good chance anyone that ended up with this image could trace it back to the owner. Even if you do decide to put a scientific name on the image (I would not recommend) you can find the current status of information regarding the specimen that was imaged (unless someone crops it out). Below is a example of a good image and appropriate labelling. It has a neutral gray background, a scale bar next the specimen and the file name (minus the extension) off in the lower corner. Following the catalogNumber is an underscore that is followed by information regarding the image. If someone takes 1,000s of images and each image file name starts with catalogNumber then we can do a batch upload into a database and create skeletal records that can be filled in later. If your catalogNumber is unique enough (i.e. a nice alphanumeric that minimally include institution code) then it also satisfies Chris’s USI (Unique specimen identifier) criteria.
Although TDWG suggests that the catalogNumber be unique, unfortunately one of the examples they provide is a simple number. I strongly encourage people to embed the institution and or collection code in the catalogNumber . Ideally that institution code is that same one that is registered http://grbio.org/ . Even if collections use their collection code or collections have different institutional codes (legacy and current) or collections within the same institution use different institution codes, you can probably trace the image back to the specimen owner. GBIF and iDigBio typically do not serve private collection data but they can still register at http://grbio.org/ .
For images that do not reference a specimen, Sharon’s naming convention would be unique enough, but is there a photographer repository so that somebody could trace “Moth_Silverspotted-
Douglasfir-2017-03-10-sjc- P0111” back to Sharon Collman? Cleary that naming convention works for her own tracking, I am only thinking about that same image being copied and transferred several times and ends up on someone’s computer and they want to know where it originated from.
The coverage about metadata by Adrian was great, for example there are several programs that strip Exif data and we should provide better support to prevent people from accidently losing metadata.
I would caution against using taxonomic info in image names. Taxonomic information changes, which may force you to rename or edit image names in the future. Much easier to dot that in a database. Hence, I agree with Chris. A file name that is a numeric or alpha-numeric identifier that corresponds to a specimen and/or image record in a database and storing all pertinent info about the image or specimen in the database allows for easy searching and no need to ever edit image names.
On 3/20/2017 11:01 AM, Christine Johnson wrote:
I have found that if you are trying to add images to a database where there is a database record for that specimen, the best file name to use is the USI (Unique specimen identifier) to that record. Otherwise bulk attaching images to correct records is difficult. But I guess this is only a good naming convention if that is where you intend to access your images (in a database with its database record).
Division of Invertebrate Zoology
Central Park West at 79th Street
New York, NY 10024
I worked out a naming convention that suits me but I'd love to be able to shorten the name files while still capturing the information. Storing the files is basically like a file cabinet.
Starts with the Host for everything but the insects/arthropods or insect if it is an insect or arthropod
Host-pest-location-date (Year-MO-Da)-my initials or photographer-original number given by the camera.
Groundcovers (includes grass)
I'm sure there are wonderful new ways to do this so I'm looking forward to how others name photos and file them.
From: Entomological Collections Network Listserve [[log in to unmask]] on behalf of Felix Sperling [[log in to unmask]]
Sent: Sunday, March 19, 2017 1:52 AM
To: [log in to unmask]
Subject: data on insect photos
A friend who is an accomplished photographer, Adrian Thysse, has just written a blog on the kinds of data that may be added to photos of specimens, with particular reference to insects:
He is especially interested to get feedback on his suggestions for a methodology for image file labelling and metadata use for photographers.
What are other people on the ECN list doing to label your image files?
--Gil Nelson, PhDResearch FacultyiDigBio Steering CommitteeIntegrated Digitized BiocollectionsInstitute for Digital Information and Scientific CommunicationCollege of Communication and InformationCourtesy ProfessorDepartment of Biological SciencesRobert K. Godfrey HerbariumFlorida State University