In an era where satellite imagery is critical for Earth observation, IRIS, or Intelligently Reinforced Image Segmentation, emerges as a powerful tool designed for manual image segmentation. This blog will guide you through the installation, usage, and troubleshooting of IRIS, making it user-friendly even for those new to the field.

Features of IRIS

• AI support using gradient boosted decision trees for enhanced image segmentation.
• Multiple configurable views for managing multispectral imagery.
• Quick setup with pip and a single configuration file.
• Platform-independent, compatible with Linux, Windows, and Mac OS.
• Multi-user support, allowing teams to work on datasets collaboratively and merge results.

Installation Instructions

Getting started with IRIS is straightforward. Follow the steps below:

1. Clone the IRIS repository:

git clone git@github.com:ESA-PhiLab/iris.git

2. Navigate into the cloned directory:

cd iris

3. Install the package and its dependencies (preferably within a conda environment, using Python 3.8 or 3.9):

python setup.py install

4. If you wish to make alterations to the IRIS source code, consider installing it as follows to avoid repeated installations:

pip install -e .

Usage of IRIS

Once installed, you can run a demo version of IRIS:

iris demo

After running the demo, create a personalized config file based on democloud-segmentation.json. With your custom config file, you can instantiate your project using:

iris label your-config-file

For optimal experience, it’s recommended to use a keyboard and mouse with a scroll wheel, as control through a trackpad may be limited.

Using Docker for Deployment

If you prefer using Docker, here’s how to deploy IRIS:

1. First, build an image from IRIS’s root directory:

docker build --tag iris .

2. Then, use docker run to launch IRIS:

docker run -p 80:80 -v dataset_path:dataset --rm -it iris label dataset cloud-segmentation.json

Running on GitHub Codespaces

To run IRIS on a GitHub Codespace, fork the repository, select Code, and then open it in Codespace. After that, run:

pip install -e .

Finally, launch the demo using:

iris demo

A pop-up will notify you of the app running on port 5000; click the link to view IRIS in action!

Understanding the Code: An Analogy

Imagine setting up a new kitchen for gourmet cooking. Your ingredients (data) are fresh, but how you prepare them is crucial. Cloning the repository is akin to setting up your kitchen (gathering your tools). Navigating your directory is like organizing your ingredients and utensils. Installing the package is like ensuring that all your pots and pans are clean and ready for use. Running the demo? That’s your first dish attempt, a chance to test your skills. The personalized configuration is your recipe—specific to your chosen cuisine, allowing you to meticulously prepare a unique meal (or in this case, a unique project).

Troubleshooting Your Experience

Should you encounter any issues while using IRIS, consider the following troubleshooting tips:

• Ensure all dependencies are installed; running pip install -r requirements.txt can help.
• If using Docker, double-check your port-forwarding settings and ensure you’re pointing to the correct volume path.
• For keyboard and mouse control issues, switching to a mouse can alleviate difficulties.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.

For more information, visit the official IRIS GitHub page.