The following instructions are general recommendations for loading images directly into any Symbiota data portal, though specific requirements may differ from portal to portal depending on the project managers. If your portal is housed on an iDigBio server through the ADBC program then see “iDigBio Image Ingestion Utility” below about batch uploading.
Portals are typically used to publish web-ready images (e.g. JPGs). Archival images (e.g. TIF, DNG, RAW, etc) are best managed in a number of other software products specially designed for that purpose. JPGs should be web optimized with minor JPG compression applied. Field images are typically less than 1600 pixels wide and no more than 2M (<1M is ideal). Images should be cropped before loading to best highlight the subject matter. Some photographers prefer to crop images into a square format since they display best within the image thumbnail galleries. Specimen images are often larger (e.g. 4000px wide, 1-6M), but are typically linked to the portals using a batch processing workflow rather than loading via the user interface.
One must have a system login and the proper permission assignments before they will be able to load images through the user interface. If you do not yet have a login name, look in the site menu for a link to create a new account. Once you have an account, the next step is to contact a portal administrator to obtain the necessary permissions and instructions for uploading images. There should be a portal administrator contact link somewhere on the front page of the website.
There are three different categories of images that can be linked to a data portal. Use the descriptions below to determine which category best fits your material.
- Field images linked to taxon name only
- Image of a physical specimen held within an actively curated collection (or associated with Personal Specimen Management)
- Image voucher (field observation supported by an image)
Simple field images without specific locality information (e.g. lat/long coordinates) are linked only to the scientific name of the organism. These images can be viewed on the Species Profile Page along with other general information such as species descriptions, distribution maps, synonyms, common names, etc. Field images are uploaded and managed through the Species Profile Editing interface. Users with Species Profile editing permissions can submit an image by clicking on the editing symbol located in the upper right of any Species Profile page, or through the image submission links available on the sitemap page. Field images with specific locality details (e.g. coordinates) can be loaded as Image Vouchers (see below).
Species Profile Examples:
Images of physical specimens should be linked to their respective specimen record. Specimen images are available both from the Occurrence Details and the Species Profile page that is associated with the most recent determination. If the specimen identification is annotated, the identification of the image will be automatically adjusted. Note that if the image is given a sort order value of greater than 500, that image will be displayed on the Occurrence Details page but not on the Species Profile display. This is typically done for images of poor quality or due to aesthetic issues (e.g. road kills). Specimen images can be added via the user interface by clicking on the Image tab within the Occurrence Editor. See below for screen captures of the Occurrence Editor interface. See the Image Batch Loading page for workflows being used to load images in mass. A field image can be linked to a specimen record, but the image must be of the specific individual that was collected. Field images of a vouchered specimen are considered high valued images since species identifications can be verified or annotated years afterwards by close inspection of the physical specimen.
Occurrence Details Examples:
Occurrence Editor Screen Captures:
Main page of editor
Image tab of editor
Field images with specific locality information can adequately serve as vouchers for field observations of many birds, mammals, and easy-to-identify plants. In some instances, they may actually be preferable, for instance if the organism is rare or the observer does not have the appropriate collecting permit. These images are similar to georeferenced specimen records in that they are searchable occurrence records that place a dot on the species distribution maps. There are, however, some taxonomic groups that typically require microscopic or chemical analysis for accurate identification (e.g. lichens, bryophytes, fungi). For such organisms, field images will have limited value. For this reason, some data portals might not allow the submission of observations without a physical specimen.
It is particularly important that images intended as vouchers for an observation be of high quality and clearly display diagnostic morphology of the organism.
As is the case with images of physical specimen, Image Vouchers are displayed on both the Occurrence Details and the Species Profile pages. Image-supported observation can be uploaded using the Observation Submission page. See links below for a screen capture of this page. Up to three images can be submitted for a single observation. Data fields required for an image to qualify as a voucher include: observer name, observation date, country, state, locality description, latitude, longitude, and at least one image. Contact your data portal administrator for information on whether observations are allowed and the necessary protocols for submitting an observation. Some portals require the review of a tutorial before one is authorized to submit observation.
For some portals that import research-grade images and occurrences from iNaturalist you should be able to submit directly through iNaturalist and they will show up in the portal as an image voucher once they are designated research grade and harvested by the portal.
iDigBio Image Ingestion Utility + Symbiota Quick Start Guide
For Use with up-to-date Symbiota installations
- Ensure that you have the latest version of the iDigBio Image Ingestion tool and Symbiota – it may be found here: https://github.com/iDigBio/idigbio-media-appliance/releases
- Installation guide for the utility: https://github.com/iDigBio/idigbio-media-appliance/wiki/Installing
- It is highly recommended that you also have the latest version of Symbiota installed. Contact your systems administrator if you are unsure of the version of Symbiota your data portal is running.
- In order to run the utility, you will need a UUID and API key from iDigBio. This is like a username and password. If you are a Symbiota data collection manager and do not have either of these things, you may request them from the iDigBio Technical Team (firstname.lastname@example.org).
- Launch the image ingestion utility. Log in with your UUID and API key.
PC users should use a browser other than Internet Explorer. In other words, use Chrome or Firefox.
For Windows users to launch the iDigBio Media Appliance once the install is finished, go to the start menu and type: idigbio_media_appliance.
You can also create a shortcut to the program by creating a new shortcut and entering that command in the “command to run” box. This should open a command prompt, and then shortly afterwards your default browser should open a web page pointed at the iDigBio Media Appliance.
To close the appliance once you are done using it (and all uploads have finished). Close the browser tab with the appliance in it, then click the X box in the upper right hand corner of the appliance command prompt window.
- From the main page, you will see options for generating a CSV. This CSV is used to index the images you wish to load into iDigBio (and later your Symbiota portal).
- You will see three fields: Upload Path, GUID Syntax, and GUID Prefix.
- Upload Path: The directory where your images are stored. Clicking within this field should bring up a window allowing you to point-and-click to where the directory is located on your computer. (Note: the window may pop up in the background, minimize your browser to check) If your directory has nested subdirectories, be sure to tick the “Also Search Files in Subdirectories” box.
- GUID Syntax: For linking images to records within Symbiota data portals, you’ll need to choose the [GUID Prefix][File Name] option. Your file name MUST start with the catalog number of the corresponding record in Symbiota. After the catalog number follow with an underscore and use whatever information you want to describe the image. For example, a record’s catalog number is WFBM0000008, and the image filename is “WFBM0000008_dorsal.jpg”
- GUID Prefix: If your institution code should already a part of your image filenames, you may leave this field blank. Otherwise, enter your institution code and/or collection code as instructed by the data portal manager.
- Click “Generate CSV and Upload Files” to begin the upload to iDigBio. Depending on the amount of images you are uploading, this may take a long time.
- When the upload procedure is complete, from the Upload History tab, click the green “Download Media as CSV” button. You will use this file to link your uploaded images to Symbiota.
- Log in to your Symbiota portal and navigate to the collection whose images you had loaded into iDigBio.
- From the collections administration menu, select “Processing Toolbox”, and then the “Image Loading” tab.
- If you don’t see an option for “iDigBio CSV Upload” like in the above screenshot, it will need to be set up by the portal manager. This only needs to be done once:
- Select “iDigBio CSV Upload” from the Upload Type drop-down menu.
- Enter a regular expression that will correctly capture the catalog number from the image file name. Remember to enclose the portion of the expression representing the catalog number in parentheses. Regular expressions can be difficult to set up http://www.regexr.com is a useful resource in setting one up. If problems persist, don’t hesitate to ask Cat Chapman (email@example.com) for help in setting up the regular expression.
- You will now need to choose the CSV file that you downloaded from the Upload History page in the image loader.
- Click the Process Output File
- If you have followed the steps correctly and encountered no errors, Symbiota will extract the catalog number from the filename; check to see if a record with that catalog number exists within the Symbiota data portal; if a record does exist with that catalog number, the image will be linked to the record; if a record does not exist with the that catalog number, an empty record with only the image and catalog number will be created.