/Crop-Type-Mapping

Machine Learning Algorithms based Crop Type Mapping in North-Western part of Bangladesh using Google Earth Engine (GEE) and Python

Primary LanguageJupyter Notebook

This repository contains the group project of "Big Geodata Analytical Methods and Distributed Computing" Course of TMT+ training organized by The Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente.

The title of the project is "Machine Learning Algorithms based Crop Type Mapping in North-Western part of Bangladesh using Google Earth Engine (GEE) and Python"

Precise crop type maps are crucial for monitoring cropping patterns, sustainable use of available natural resources and estimating harvests. Manual digitization and labeling—the usual method of creating crop type maps— is mostly time consuming, expensive and even prone to human errors. The Machine learning algorithms have been evolved as cost-effective alternatives for classifying crop varieties using satellite imageries in recent times.
The objective of this prject is to deploy the recent advances i.e., the machine learning algorithms to classify 6 crop types of north-western part (Godagari Upazila of Rajshahi district) in Bangladesh from Sentinel-2 imagery. Four machine learning algorithms (Random Forest, Artificial Neural Network, KNN and Support Vector Machine) has been investigated for mapping the crop types accurately using the data of Rabi cropping season (October 2020-March 2021).

The study area shapefile can be accesses on:
https://code.earthengine.google.com/?asset=users%2Fshohelovro%2FCropTypePaper%2FGodagari_Upazila&fbclid=IwAR0yP7K6kvB9vsP6GJ7E8SR3wyFAOA6_ab4frGGrHEGRAAu36D5kFt8qbxk

The composite image can be accessed on:
https://drive.google.com/file/d/1KywwGYj1oVxrzyKH6322IE7olfXImT4D/view?fbclid=IwAR3saSt8WEvGpck_4yk-84mv3dqz-L7XjYZ3n-iuYcvDm7oAQKzF4jokY6E