:toc: = paddle (PArticle Detection via Deep LEarning)
== ⚠️ DEPRECATED ⚠️ Unfortunately, I cannot provide support for paddle any longer. Fortunately, there has been a lot of progress in the meantime, with regard to the usability of Mask R-CNN for custom applications, such as image-based particle detection. Therefore, I recommend to use the https://github.com/open-mmlab/mmdetection[mmdetection] framework, which has a large community, a thorough documentation and a huge model zoo. If you plan to train on your own data, https://mmdetection.readthedocs.io/en/latest/user_guides/train.html#train-with-customized-datasets[this] is a good place to get started.
image:https://img.shields.io/badge/DOI-10.1016/j.powtec.2019.10.020-blue.svg[link="https://doi.org/10.1016/j.powtec.2019.10.020"] image:https://img.shields.io/badge/arXiv-1907.05112-b31b1b.svg[link="https://arxiv.org/abs/1907.05112"] image:https://img.shields.io/github/license/maxfrei750/paddle.svg[link="https://github.com/maxfrei750/paddle/blob/master/LICENSE"] image:https://img.shields.io/badge/code%20style-black-000000.svg[link="https://github.com/psf/black"]
This repository is a toolbox for the easy, deep learning-based primary particle size analysis of agglomerated, aggregated, partially sintered or simply occluded particles. It is the successor of the https://github.com/maxfrei750/DeepParticleNet[DeepParticleNet] toolbox, which accompanies the following publication:
https://doi.org/10.1016/j.powtec.2019.10.020[Image-Based Size Determination of Agglomerated and Partially Sintered Particles via Convolutional Neural Networks]
image::assets/example_detection.jpg[Example Detection, 804, 300]
image::assets/example_psd.png[Example PSD Measurement, 400, 265]
The utilized convolutional neural network is based on the Mask R-CNN architecture, developed by https://arxiv.org/abs/1703.06870[He et al.]. It was implemented using https://pytorch.org/[PyTorch], https://github.com/pytorch/vision[torchvision] and https://github.com/PyTorchLightning/pytorch-lightning[PyTorchLightning].
== Installation
. Install https://conda.io/en/latest/miniconda.html[conda] for your operating system.
. Open a command line.
. Clone this repository: git clone https://github.com/maxfrei750/paddle.git
. Change into the folder of the repository: cd paddle
. Create a new conda environment: conda env create --file environment.yaml
. Activate the conda environment: activate paddle
. Install paddle into the environment: pip install --editable .
TIP: Creating the environment with conda can take a while. Using https://github.com/mamba-org/mamba[mamba] can speed up the creation significantly.
== Getting started
. Open a command line.
. Activate the conda environment: activate paddle
. Start JupyterLab: jupyter lab and click one of the links to access the JupyterLab server.
. In JupyterLab, navigate to the paddle/demos folder and choose a demo of your liking.
== Citation If you use this repository for a publication, then please cite it using the following bibtex-entry: