https://github.com/alejandromllo/laguerre-gauss-preprocessing

Laguerre-Gauss Preprocessing: Line Profiles as Image Features
https://github.com/alejandromllo/laguerre-gauss-preprocessing

aerial-imagery artificial-intelligence image-classification image-processing knn-classifier machine-learning mlp-classifier preprocessing

Last synced: 2 months ago
JSON representation

Laguerre-Gauss Preprocessing: Line Profiles as Image Features

• Host: GitHub
• URL: https://github.com/alejandromllo/laguerre-gauss-preprocessing
• Owner: AlejandroMllo
• Created: 2019-12-13T16:38:11.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-07-12T00:54:25.000Z (over 4 years ago)
• Last Synced: 2023-08-23T03:27:30.989Z (over 1 year ago)
• Topics: aerial-imagery, artificial-intelligence, image-classification, image-processing, knn-classifier, machine-learning, mlp-classifier, preprocessing
• Language: Python
• Size: 43.1 MB
• Stars: 1
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Laguerre-Gauss Preprocessing: Line Profiles as Images Features

Code implementation of the results presented on *Laguerre-Gauss Preprocessing: Line Profiles

as Images Features for Aerial Images Classification*.

https://arxiv.org/abs/1912.06729

## Usage

Using Python (v3.7.3) run the files MLP.py, kNN.py and CNN.py to test the results on the trained

models using a _subsample_ of the test set of aerial images.

### Subsample of test set

The folder @sample_test_data@ contains a randomly chosen subsample of the test set for aerial

images classification. The images are 64x64 RGB classified as 0 (nothing of interest: populations,

rivers, forest, etc.) or 1 (something of interest: heavy machinery, boats, deforestation, etc.).