hyperspectral-1d-cnn
Science Score: 67.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 7 DOI reference(s) in README -
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (10.5%) to scientific vocabulary
Repository
Basic Info
Statistics
- Stars: 1
- Watchers: 1
- Forks: 1
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
CNN Soil Texture Classification
1-dimensional convolutional neural networks (CNN) for the classification of soil texture based on hyperspectral data.
Description
We present 1-dimensional (1D) convolutional neural networks (CNN) for the classification of soil texture based on hyperspectral data. The following CNN models are included:
LucasCNNLucasResNetLucasCoordConvHuEtAl: 1D CNN by Hu et al. (2015), DOI: 10.1155/2015/258619LiuEtAl: 1D CNN by Liu et al. (2018), DOI: 10.3390/s18093169
These 1D CNNs are optimized for the soil texture classification based on the hyperspectral data of the Land Use/Cover Area Frame Survey (LUCAS) topsoil dataset. It is available here. For more information have a look in our publication (see below).
Requirements
- see Dockerfile
- download
coord.pyfrom titu1994/keras-coordconv based on arXiv:1807.03247
Setup
```bash git clone https://github.com/felixriese/CNN-SoilTextureClassification.git
cd CNN-SoilTextureClassification/
wget https://raw.githubusercontent.com/titu1994/keras-coordconv/c045e3f1ff7dabd4060f515e4b900263eddf1723/coord.py . ```
Usage
You can import the Keras models like that:
```python import cnn_models as cnn
model = cnn.getKerasModel("LucasCNN") model.compile(...)
```
Example code is given in the lucas_classification.py. You can use it like that:
```python from lucasclassification import lucasclassification
score = lucasclassification( data=[Xtrain, Xval, ytrain, yval], modelname="LucasCNN", batchsize=32, epochs=200, randomstate=42)
print(score) ```
Owner
- Name: Fijishi
- Login: Fijishi-Enterprises
- Kind: organization
- Email: opensource@fijishi.com
- Location: United Kingdom
- Website: https://fijishi.com/fijishi-sdgs/
- Repositories: 1
- Profile: https://github.com/Fijishi-Enterprises
Open source projects and SDGs initiatives from Fijishi
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this software, please cite both the article from preferred-citation and the software itself."
authors:
- family-names: Riese
given-names: Felix M.
orcid: https://orcid.org/0000-0003-0596-9585
title: "CNN Soil Texture Classification"
version: 1.1
doi: "10.5281/zenodo.2540718"
date-released: 2020-06-09
repository-code: https://github.com/felixriese/CNN-SoilTextureClassification
license: MIT
preferred-citation:
authors:
- family-names: Riese
given-names: Felix M.
- family-names: Keller
given-names: Sina
title: "Soil Texture Classification with 1D Convolutional Neural Networks based on Hyperspectral Data"
type: article
year: 2019
doi: "10.5194/isprs-annals-IV-2-W5-615-2019"
journal: "ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences"
volume: IV-2/W5
url: https://www.mdpi.com/2072-4292/12/1/7
pages: "615-621"
GitHub Events
Total
Last Year
Dependencies
- actions/checkout v2 composite
- actions/setup-python v2 composite
- codecov/codecov-action v1 composite
- tensorflow/tensorflow 2.0.0-gpu-py3 build
- codecov *
- matplotlib *
- pydot *
- pytest >=6.0.0
- pytest-cov *
- requests *
- scikit-learn *
- seaborn *
- tensorflow >=2.5.0