In this example, segmentation was done to identify the area of blow hole in the magnetic tiles using deep learning approach.
The dataset is available from:
https://github.com/abin24/Magnetic-tile-defect-datasets.
And the deep learning model UNet is adapted from:
https://github.com/mateuszbuda/brain-segmentation-pytorch
Original paper of UNet: U-Net: Convolutional Networks for Biomedical Image Segmentation (Ronneberger et al., 2015) https://arxiv.org/abs/1505.04597
The code was developed in Google Colab environment.
View 'defects_detect_blow_hole_dataloader.ipynb' for more information
View the folder 'datasets/defects/img_blow_hole' to get more information on the folder structure
Examples of predicted segmentation output: