Geospatial Dataset on Deforestation and Urban Sprawl in Dhaka, Bangladesh: A Resource for Environmental Analysis

Published: 8 November 2024| Version 2 | DOI: 10.17632/hst78yczmy.2
Contributors:
Md Fahad Khan, Md Rakibul Islam, Ahmed Imtiaz,

Description

Google Earth Pro facilitated the acquisition of satellite imagery to monitor deforestation in Dhaka, Bangladesh. Multiple years of images were systematically captured from specific locations, allowing comprehensive analysis of tree cover reduction. The imagery displays diverse aspect ratios based on satellite perspectives and possesses high resolution, suitable for remote sensing. Each site provided 5 to 35 images annually, accumulating data over a ten-year period. The dataset classifies images into three primary categories: tree cover, deforested regions, and masked images. Organized by year, it comprises both raw and annotated images, each paired with a JSON file containing annotations and segmentation masks. This organization enhances accessibility and temporal analysis. Furthermore, the dataset is conducive to machine learning initiatives, particularly in training models for object detection and segmentation to evaluate environmental alterations.

Files

Institutions

American International University Bangladesh

Categories

Computer Vision, Environmental Science, Remote Sensing, Deforestation, Deep Learning

Licence