napari-pixel-correction

Plugin to correct manually pixel wrongly predicted on image by annotation

This napari plugin was generated with Cookiecutter using @napari's cookiecutter-napari-plugin template.

This plugin allows you to manually correct the images of the apple tree flowers by annotation. Below, a piece of an image shows the predicted pixels (in brown). A pixel in brown is assigned to the flower class. We can see that the brown colour does not necessarily cover a flower in this image.

Installation

You can install napari-pixel-correction via pip:

pip install napari-pixel-correction

To install latest development version :

pip install git+https://github.com/hereariim/napari-pixel-correction.git

How does it work

First, you need a compressed file (in .zip format) were you have all your images. For a compressed file named as input.zip, the compressed file should be built like :

.
└── input.zip
    └── repository
        ├── image
        │   ├── im_1.JPG
        │   ├── im_2.JPG  
        │   ├── im_3.JPG
        │   ...
        │   └── im_n.JPG
        │
        └── mask
            ├── im_1_mask.JPG
            ├── im_2_mask.JPG
            ├── im_3_mask.JPG
            ...
            └── im_n_mask.JPG

In repository, each image folder should have two elements : image in RGB and the segmented mask in binary image (where no-flower class is 0 and flower class is 255)

Contributing

Contributions are very welcome. Tests can be run with tox, please ensure the coverage at least stays the same before you submit a pull request.

License

Distributed under the terms of the BSD-3 license, "napari-pixel-correction" is free and open source software

Issues

If you encounter any problems, please file an issue along with a detailed description.