zhixuhao/unet
A U-Net convolutional neural network for biomedical image segmentation implemented with Keras functional API.
Not currently ranked — collecting fresh signals.
star history
This repository implements U-Net, a fully convolutional neural network architecture, for semantic image segmentation tasks. The model takes 512x512 input images and outputs pixel-level segmentation masks using a sigmoid activation function. It uses Keras functional API for flexible architecture experimentation and includes data augmentation via Keras ImageDataGenerator to compensate for limited training data. The implementation targets biomedical image segmentation, originally from the ISBI challenge.
Frequently asked
- What is zhixuhao/unet?
- A U-Net convolutional neural network for biomedical image segmentation implemented with Keras functional API.
- Is unet open source?
- Yes — zhixuhao/unet is open source, released under the MIT license.
- What language is unet written in?
- zhixuhao/unet is primarily written in Jupyter Notebook.
- How popular is unet?
- zhixuhao/unet has 4.9k stars on GitHub.
- Where can I find unet?
- zhixuhao/unet is on GitHub at https://github.com/zhixuhao/unet.