Artsorakel

Artsorakel is a species identification app that uses artificial intelligence to identify wild species from photos. Developed by NBIC, the Norwegian Biodiversity Information Centren (Artsdatabanken).

The app is available for both Android and iOS.

Features

• Identify species from photos using AI
• Works offline for viewing history
• Supports multiple languages
• No account required
• Free to use

Prerequisites

Before building, ensure you have the following installed:

• Python 3 - for the sync script
• Java - required by vd-tool for SVG to Android VectorDrawable conversion
• vd-tool - npm install -g vd-tool
• librsvg - brew install librsvg (macOS)
• ImageMagick - brew install imagemagick (macOS)
• jq - brew install jq (macOS)

For Android development:

• Android Studio
• Android SDK

For iOS development:

• Xcode
• macOS

Setup

1. Clone the repository

   git clone https://github.com/Artsdatabanken/artsorakel.git
   cd artsorakel

2. Configure API credentials

   Copy the secrets template and add your bearer tokens:

   cp shared/config/secrets.json.template shared/config/secrets.json

   Edit shared/config/secrets.json and replace the placeholder values with your actual bearer tokens:

   {
     "api": {
       "bearerToken": {
         "android": "YOUR_ANDROID_BEARER_TOKEN_HERE",
         "ios": "YOUR_IOS_BEARER_TOKEN_HERE"
       }
     }
   }

3. Run the sync script

   The sync script downloads fonts, generates platform-specific resources, and syncs shared assets:

   python3 sync_resources.py

   This will:

   • Download fonts from Fontsource
   • Generate localized strings for both platforms
   • Convert SVG icons to Android VectorDrawables
   • Generate iOS color assets and app icons
   • Sync design system themes and colors

Building

Android

1. Open the Android folder in Android Studio
2. Sync Gradle
3. Run on device/emulator

iOS

1. Open iOS/Artsorakel/Artsorakel.xcodeproj in Xcode
2. Select your development team for signing
3. Run on device/simulator

Project Structure

artsorakel/
├── Android/                 # Android app (Kotlin)
├── iOS/                     # iOS app (Swift/SwiftUI)
├── shared/                  # Shared resources
│   ├── config/              # App configuration and secrets
│   ├── content/             # HTML content (about, FAQ, etc.)
│   ├── designsystem/        # Design tokens and themes
│   ├── fonts/               # Downloaded fonts (auto-generated)
│   ├── images/              # Logo and image assets
│   ├── vectors/             # SVG icons
│   └── strings.csv          # Localization strings
├── sync_resources.py        # Resource sync script
├── LICENSE                  # MIT License
├── README.md                # This file
├── CONTRIBUTING.md          # Contribution guidelines
└── CODE_OF_CONDUCT.md       # Community code of conduct

How It Works

The app sends photos to NBIC's AI middleware, which augments and returns species predictions based on a model trained and hosted by Naturalis Biodiversity Center, based on images contributed by citizen scientists on Artsobservasjoner.no and similar platforms. The model can only identify wild species that have been reported with images - it does not recognize humans, domestic animals, or garden plants. Based on the approximate location of the provided picture (if shared by the user), or the country of their IP address, the Norwegian, Swedish or European modal is selected.

Images and user information are not stored, nor made available to NBIC or third parties.

Contributing

Contributions are welcome! Please read our Contributing Guidelines before submitting issues or pull requests. All participants are expected to follow the Code of Conduct.

Support

Questions and feedback can be sent to support@artsobservasjoner.no.

License

This project is licensed under the MIT License - see the LICENSE file for details.