There are three different pages which give an overview of
your source's data.
Browse through pages of images displayed as thumbnails.
Below the grid, click the arrow buttons to go to the next or
The search box at the top lets you choose a subset of images. Change the search fields, then click Search to apply the filters. The "Image name contains" field accepts punctuation, so you can use search queries like dsc_ .jpg or dupe-name.
You can mouse-over an image thumbnail to see its filename and
annotation status. The thumbnail's border color also indicates the
image's annotation status.
The box below the page navigator lets you run an action on the
current set of images. Click the help button in that box for more details.
Requires source Edit or Admin permissions.
Show images' metadata as a grid of text fields.
You can get a smaller set of images with the Search form,
just like in Browse Images.
Type in the text fields to edit the metadata, then click the Save Edits button below to save your edits. If there are any errors, the errors will display below the metadata grid and nothing will be saved yet. Note that saving might take a while if there are many images in the metadata grid.
You can edit multiple images' metadata at the same time by using
the checkboxes along the left side. If you check multiple images, and
then edit the metadata of one of those checked images, those edits
will be applied to all of the images you've checked. You can also use
the top checkbox to check or un-check all of the images.
The "Height (cm)" field is the number of centimeters of substrate the image covers, from the top of the image to the bottom. This was formerly required by the automatic annotation system, but now it's just an ordinary metadata field that you can optionally fill in for your information.
Similar to Browse Images, except that each thumbnail is an image cut-out
or "patch" which represents a point location within that image.
You can use the Search form to only display patches of a particular Label.
Mouse-over an image patch to see which image and which point the patch is
Select an action with the left dropdown box. Then use the right dropdown box to specify whether to apply the action to all images in the search, or just the images on this page. Click "Go" to run the action.
The available actions are:
Enter Annotation Tool
(Requires source Edit or Admin permissions) This takes you to the annotation tool for the first image in the specified image set. Then, the annotation tool's Prev/Next buttons will only navigate through the specified image set.
Export (download) data about the specified image set from CoralNet to your computer. There are several kinds of export data available, selectable with a dropdown box.
Each export type uses CSV format unless explained otherwise. A CSV (comma-separated values) file can be opened in a spreadsheet program like Excel.
The available types of export data are:
Metadata: This contains anything you would see in the "Edit Metadata" page: image date, auxiliary metadata, and so on.
Annotations, CSV, Simple: Image name, Pixel row, Pixel column, and Label for each annotation point in the image set. TIP: If you are only interested in Confirmed annotations, then first do an image search with "Annotation status" set to "Confirmed".
Annotations, CSV, Full: The fields of "Annotations, CSV, Simple" plus the annotator, annotation date, and the machine annotator's top suggestions for the point.
Annotations, CPCe: (Requires source Edit permissions) Export .cpc files which can be opened in Coral Point Count. There will be one .cpc file per image, and the .cpc files will be contained in a .zip archive file. If you're not familiar with opening .zip files, see this WikiHow article. The .cpc file format contains several pieces of information besides point positions and labels; here's how we handle each one:
Notes codes: Coral Point Count has a Notes column in the list of points, and the Notes codes you enter in this column are saved in the .cpc file. This info is available based on .cpc files you've previously uploaded. If you have uploaded a .cpc file for a particular CoralNet image, that .cpc file's Notes codes will be retained, and those Notes codes will be used when you request a .cpc export for that image. If you have never uploaded a .cpc file for this CoralNet image, the Notes codes will be blank.
Header data: When you create a .cpc file in Coral Point Count, at some point it shows a window with 28 header fields such as Project name, Site name, Latitude, Easting, etc. Similarly to the Notes codes, this info is available based on .cpc files you've previously uploaded. (You might notice that these header fields have some overlap with CoralNet's metadata fields, but CoralNet does not currently make any connection between these fields on import/export.)
Computer environment info: This includes the location of the .txt code file that CPCe uses to match label codes to label names/groups, and the location of the image file that was opened in CPCe. Unlike Notes codes and Header data, this environment info is necessary to make a .cpc work in Coral Point Count. Here's what CoralNet does to make sure this info is filled in:
CoralNet first looks for this info in previously uploaded .cpc files.
If one or more images do not have a previous .cpc, then this action form will show form fields where you can fill in the environment values. These values will be automatically filled in with information from other images' uploaded .cpc files, if available. If no .cpc has been uploaded to this source before, then the fields will be blank and you'll have to fill them in.
If you have to fill in the environment info yourself, the code file location should look something like C:\CPCe_41_inst\coral_codes.txt, and the image folder location should look something like C:\Image_data\Panama_corals (you may have a slash at the end or not). Note that the image folder is a folder location, not a file location; the actual image file might be at C:\Image_data\Panama_corals\DSC_0048.JPG.
Image Covers: Coverage statistics for each image in the set; for example, this image consists of 5% label A, 10% label B, etc. based on the annotations. Also included is the image's "annotation area", which is the area of the image where the random points are scattered.
Note that the exported file may take a while to generate, particularly for sources with many images and points. After you click "Go", be sure to stay on this page until your browser starts downloading the file (or prompts you to choose a download location). If you are exporting data for thousands of images, you might have to wait for a few minutes.
(Requires source Edit or Admin permissions) Permanently delete the images in the specified image set, along with their annotations. When you click "Go", a confirmation dialog will pop up. In the dialog, type "delete" (without the quotes) and hit Enter to proceed with deletion.