Getting Started

Before you can get started with Agouti, there are several steps to complete. First you create an account, then apply for a new project and finally receive access.

Creating an account

Navigate to https://www.agouti.eu.
Select your preferred language at the top right.
Click Register to create an account.
Fill in the form, choose a sufficiently strong password of at least 8 characters.
Read the GDPR statement and check the box if you agree. Agouti minimises the storage of personal data, and does not use any trackers or third-party cookies.

Your account is now set up and ready to use. However, by design, you will not yet have access to a project. To apply for a project, please follow the steps below.

Applying for a project

Agouti is available to all individuals and organizations that work with camera traps. We facilitate academic, non-commercial and government projects free of charge. Commercial organizations can try out Agouti free of charge, but are expected to make a contribution to the development and hosting costs if they plan to keep using Agouti. Contact us to discuss options. To apply for a project:

From the Agouti homepage or directly from your email write a message to agouti@wur.nl. Please include whether you are working in a commercial or non-profit setting.
You will receive a response with an invitation link to your newly created project. Please use the link and notify agouti@wur.nl when that’s done.
The admins will make you PI on the project. You can now enter the project from your personal dashboard and get started.

Setting up your project

Your project is completely empty at this point. We recommend taking some time to set up some initial settings that will help you to get started.

Project settings

Log in and enter your project. Go to the project setting in the left hand navigation. The following settings are available:

The default title is not very descriptive. Please update it to reflect your project’s subject. For example: ‘Predator densities in Bardiya National Park, Nepal’. This text is shown on the public pages, if your project is set to public.
Optional: set a location for your project. This can be anywhere so doesn’t have to match your actual study area. This location is used to plot your project on a public map, if your project is set to public.
Description: write a few lines about your project. We recommend to include: the aim of the study, the sampling strategy and the location. For example: ‘We study the community composition of predators in Bardiya National Park, Nepal. A permanent grid of 50 cameras, spaced 1 km apart, was installed. Each camera was checked monthly.’ This text is shown on the public pages, if your project is set to public.
Image: here, you can upload an image that is shown on the project’s tile. This image is shown on the public pages, if your project is set to public.
Restricted images: this setting controls whether your project’s assets can be accessed publicly by using their URLs. When checked, image URLs in data exports/published data will not be accessible to external users and require authentication by a user that has access to the project. Sequences that contain one or more frames with humans always require authentication.
Restricted project: this setting controls whether your project’s name, description and location can be shown publicly. When checked, your project will be hidden from the public homepage and portals on agouti.eu.
Restricted locations: This controls the visibility of locations by users in our project. When unchecked, deployment locations can be seen by all project members. When checked, photo processors and volunteers will only see the locations for their own deployments.
Deployments are assigned: when checked, deployments can be assigned to specific users as a task.
Nextcloud uploads enabled: this setting controls whether users can upload deployments from their NextCloud storage. For detailed information on using NextCloud, see NextCloud.
Camera AI viewer: this setting controls whether users can access CamerAI, an AI-powered image processing interface.
Observation.org uploads: this setting controls whether observations from this project can be uploaded to Observation.org.
PI: set this to the person that is responsible for the project.
Contact: set this to the person that is coordinating the project.
Organisation: set this to the organization running the project. Contact agouti@wur.nl if your organisation is not listed.

Sampling design

The optional fields allow you to define your project’s sampling design in terms of camtrap-dp definitions.

Deployment defaults

You can define default values for deployment settings, such as default detection distance, camera height, camera heading, camera tilt, coordinate uncertainty, and bait use. The default values can still be adjusted for individual deployments during deployment creation.

Default UTC offset: this sets a timezone for your project. Set it to prevent having to enter this for each deployment manually. If your location observes daylight saving time you can adjust the timezone to match. For example: if you run your cameras in Western European summertime, you can set the offset to +2. The description for each timezone are indicative only, and not exhaustive.
Detection distance: the maximum distance (in centimetres) at which the camera is expected to reliably detect movement.
Camera height: the height of the camera (in centimetres) above ground level.
Camera heading: the compass direction the camera is facing.
Camera tilt: the angle of the camera relative to the horizontal plane.
Coordinate uncertainty: an estimate of the accuracy of the deployment coordinates, expressed in metres.
Bait: indicates the type of bait or attractant that was used at the deployment location, or whether no bait was used.

Annotation

Annotate timelapses: this setting controls how timelapse images are dealt with. If enabled, timelapse images are included in the deployments, and thus annotatable. If not enabled, timelapse images will be left out of the deployment.
Calibration and tracking: When checked, calibration and tracking options are available during annotation.
Sequence cut-off: this setting controls the division of images into sequences. The time difference between subsequent images is used. If the time difference is equal to or smaller than this cut-off, the images are put in the same sequence. If the time difference is greater than this setting, a new sequence is started. The default setting of 120 seconds works well for most projects. Should you wish to change it, please contact agouti@wur.nl. Changing this setting does not affect existing data and is only applied to uploads made after the change.
Automatic annotation: here, you can select an AI model to automatically process your deployments. After selecting a model, a button ‘Annotate by AI’ will show up for each deployment listed on the Deployments page. For more details, please see the dedicated reference automatic annotation page.
Automatically with new photo series: when enabled, new deployments are first processed by the AI and can be edited after this is completed.
Filter blank images on upload: when enabled, this setting will identify blank images and mark them as such directly after upload.
Filter images containing humans on upload: this setting controls how sequences containing humans are handled. New uploads are automatically scanned for humans. Three options are available:
- Disabled: no action is taken and sequences containing humans remain unchanged.
- Blurring: images in sequences containing humans are blurred.
- Obscuring: images in sequences containing humans are hidden and replaced with a placeholder image.

When either blurring or obscuring is enabled, all sequences in the project that contain a human observation, including sequences from previous uploads, will also be blurred or obscured.

Behaviour

Optional: Here, you can define behaviours for your project. Each behaviour has a name and description. These behaviours are available during annotation. You can select none, one or multiple behaviours for each observation.

Individuals

Optional: To keep track of individual animals, add them here. An individual always has a name, species and description. Added individuals are available during annotation.

Deployment tags

Optional: Deployment tags are used to add descriptive labels or metadata to individual deployments. Unlike deployment groups, they do not necessarily represent a formal grouping structure and are mainly intended to store additional information that may be useful for filtering, searching, or describing deployments.

Habitats

Optional: Create custom habitat tags that can be assigned to your locations.

Deployment groups

Optional: Deployment groups are used to organise deployments into meaningful study units that share a common spatial, temporal, or experimental context. A deployment can belong to one or more groups, such as a survey season, sampling session, study area, camera array, or grid cell. These groupings are typically used for analysis and reporting.

Species

The main taxonomic list we use is the Catalogue of Life. You can add class, order, family, genus, species and subspecies from this main list to your project, making them available for selection during annotation. Including higher taxonomic levels is helpful for hard to distinguish groups of species, such as rodents, or allows you to quickly process species groups that are not of interest to your project. If you find a species is missing from the main list, please contact agouti@wur.nl and mention the link to CoL page.

Inviting your collaborators

In the user management you can see who has access to your project, and assign roles to each user. For a description of roles available please see roles & permissions. Access to projects in Agouti is granted by invitation only. As principal investigator or project coordinator you can invite members by following these steps:

Go to the Users menu and select the Copy invite link button.
Send the link to your collaborator, ask them to request access by using the link and to notify you when they’ve done so.
The request is now shown in your project’s Users menu. You can accept or reject the request. After accepting, the default role is volunteer. Adjust this if necessary by editing the user in the list of members.
When a collaborator leaves your project, remember to revoke their access. Do this by clicking the edit button in the list of users and selecting Revoke access.