Skip to content

A semantic segmentation model which processes the given x-ray image of the chest into a segmented map where the area occupied by the lungs is particularly segmented.The model uses the popular U-NET architecture to process the input image and has been trained using the tensorflow's keras api where dice-loss is used as the loss function and dice-c…

Notifications You must be signed in to change notification settings

asiskumarroy/Semantic-segmentation

Repository files navigation

Semantic-segmentation

A semantic segmentation model which processes the given x-ray image of the chest into a segmented map where the area occupied by the lungs is particularly segmented.The model uses the popular U-NET architecture to process the input image and has been trained using the tensorflow's keras api where dice-loss is used as the loss function and dice-coefficient as the performance metric.

About the Dataset:

The dataset provided is actually a part of the original dataset on which the model was trained.The entire dataset is available in kaggle.

About the example:

The predicted output mask for the given input as a test image is an example of how the model works to segment the lungs area after been trained for 57 epochs with a batch size of 8.Further training wasn't performed as the model had reached its minima for which there wasn't any further improve in its loss.

About

A semantic segmentation model which processes the given x-ray image of the chest into a segmented map where the area occupied by the lungs is particularly segmented.The model uses the popular U-NET architecture to process the input image and has been trained using the tensorflow's keras api where dice-loss is used as the loss function and dice-c…

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published