PyImageLabeling 1.0.8

PyImageLabeling is a tool designed to help create image masks, i.e., labeled images for training machine learning models.

PyImageLabeling

• GitHub Repository: View codesource.
• PyPI: Install via pip.

PyImageLabeling is a powerful tool with a user-friendly interface based on PyQT6 for creating image masks. These labeled images are used in the creation of machine learning models dedicated to computer vision tasks.

Two types of labeling are available:

• Pixel-by-Pixel: allows to use the pixel-level precision (paintbrush, magic pen, contour filling).
• Geometric shapes: allows to use different geometric shapes (polygon, rectangle, ellipse) for labeling.

Overview of PyImageLabeling in the context of an application related to pancreatic cancer.

Overview of PyImageLabeling with the Flood Area Segmentation dataset.

Installation and Run

Note that you need first Python 3 (version 3.12, or later) to be installed. You can do it, for example, from Python.org.

PyPi installation (Windows, Mac and Linux)

Check whether you have the last version of PyPi:

python3 -m pip install -U pip

Install PyImageLabeling:

python3 -m pip install -U PyImageLabeling

To launch PyImageLabeling:

python3 -m PyImageLabeling

Executable ".exe" (Windows)

You can download the Windows executable. Just double-click on the executable file to launch PyImageLabeling.

Github Installation (Windows, Mac and Linux)

Here is an illustration for Linux. We assume that Python 3 is installed, and consequently ‘pip3’ is also installed. In a console, type:

git clone https://github.com/crillab/PyImageLabeling.git

You may need to update the environment variable ‘PYTHONPATH’, by typing for example:

export PYTHONPATH="${PYTHONPATH}:${PWD}/.."

Get the last version of pip:

python3 -m pip install --upgrade pip

Executes the pyproject.toml inside the PyImageLabeling directory and installs dependencies ("numpy", "pyqt6", "opencv-python", "pillow", "matplotlib").

python3 -m pip install -e .

To launch PyImageLabeling:

python3 -m PyImageLabeling

Features

PyImageLabeling considers that you have two working directories (which may be the same):

• The working directory containing your images to be labeled.
• The saving directory containing your labeled images and a JSON file.

The saving directory consists of:

• The labeled PNG files: each label of each image is represented by a labeled PNG file.
• The JSON file contains several information about labels: identifier, name, color and type.

Load/Save and Image Management

Tool	Functionality
Load	Load images and/or labeled images from a working or saving directory
Save	Save current labeled images in a saving directory
Export copy	Create a copy of the saving in another folder
Next	Select the next image
Previous	Select the previous image

Navigation & Controls

Movement	Function
Zoom in	Click to zoom in (up mouse wheel shorcut)
Zoom out	Click to zoom out (down mouse wheel shorcut)
Move	Click and drag to move the current (wheel mouse click shorcut)
Reset	Reset to default view (zoom and position)

Label setting

Tool	Functionality
Add Label	Add a new label (by chosing the name, the type and the color of the new label)
Label Opacity	Adjust the transparency of the labeled layers.

Moreover, for each label, you can set some parameters:

Tool	Functionality
Color	Set the color of the selected label
Visibility	Set the visibilty of the selected label
Setting	Set the parameters of the selected label
Remove	Remove the selected label (warning: delete all saving files containing this label)

Labeling Tools

Tool	Pixel-by-pixel
Contour Filling	Shape detection: click inside a detected shape to fill with the label (tolerance in parameter)
Paintbrush	Paint with the selected label (Paintbrush size in parameter)
Magic Pen	Intelligent fill tool based on color similarity (tolerance, pixel limit and method in paramter)

Tool	Geometric shapes
Ellipse	Precision ellipse selection for label assignment
Rectangle	Precision rectangular selection for label assignment
Polygon	Precision polygon selection for label assignment

Editing Tools

Each editing tools is proper to each selected label.

Tool	Functionality
change labeled area	Change the lebel of an area for another one with a click
Undo	Revert the last action with full history support
Eraser	Brush that erases the selected label (Eraser brush size and "all labels" mode in parameter)
Clear All	Clear all label areas.

License

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

Use Cases

• Computer Vision: Create training datasets for object detection and segmentation
• Medical Imaging: Annotate medical scans and diagnostic images
• Autonomous Vehicles: Label road scenes and traffic elements
• Agriculture: Mark crop areas and plant health indicators
• Quality Control: Identify defects and areas of interest in industrial applications

---