Accuracy Assessment of Land Use and Land Cover Classification: A Case Study in the São Luís Intermediate Region with MapBiomas

This repository contains the source code and procedures used in the article: "Avaliação da Acurácia na Classificação de Uso e Cobertura da Terra: Um Estudo de Caso na Região Intermediária de São Luís com o MapBiomas".

The study provides a localized accuracy assessment of MapBiomas data, building upon the original source code developed for the Collection 8 accuracy analysis. This repository is a fork of the original project, specifically adapted for the São Luís intermediate region.

• Original Repository: MapBiomas Accuracy - Collection 8
• Full Article: Read it here (DOI: 10.33360/geonordeste.v36i.21680)

Authors

• Thomas Victor de Sousa Malheiros Rocha

  • M.Sc. in Environmental Science and Technology
  • Federal University of Maranhão (UFMA)
  • thomasvictor990@gmail.com

• Sérgio Souza Costa

  • Ph.D. in Applied Computing
  • Associate Professor at the Federal University of Maranhão (UFMA)
  • sergio.costa@ufma.br

• Luís Fernando Cirqueira da Silva Correia

  • M.Sc. in Environmental Science and Technology
  • Federal University of Tocantins (UFT)
  • luis.correia@discente.ufma.br

• Denilson da Silva Bezerra

  • Ph.D. in Earth System Science
  • Assistant Professor at the Federal University of Maranhão (UFMA)
  • denilson.bezerra@ufma.br

Getting Started

Follow the steps below to reproduce the analysis:

1. Clone the repository

git clone [https://github.com/LambdaGeo/acuracia_slz.git](https://github.com/LambdaGeo/acuracia_slz.git)
cd acuracia_slz

2. Create a virtual environment
It is highly recommended to use a virtual environment to isolate project dependencies.
python -m venv venv
source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

3. Install dependencies
The necessary libraries are listed in the requirements.txt file.
pip install -r requirements.txt

4. Run the Notebook
The main analysis is performed within a Jupyter Notebook. Ensure your virtual environment is active and launch Jupyter:
jupyter notebook 2_acuracia_sa23z.ipynb