BD Open LULC Map: High-resolution land use land cover mapping & benchmarking for urban development in Dhaka, Bangladesh

TL;DR

This paper introduces the BD Open LULC Map, a high-resolution, pixel-wise land use land cover dataset for Dhaka, Bangladesh, created to improve urban mapping and benchmarking in data-scarce developing regions.

Contribution

The work provides a large-scale, validated LULC dataset for Dhaka using high-resolution imagery, and benchmarks deep learning segmentation models across multiple satellite sources.

Findings

DeepLab V3+ achieves high accuracy on the dataset.

BOLM enables improved urban land cover classification.

Performance varies between Bing and Sentinel-2A imagery.

Abstract

Land Use Land Cover (LULC) mapping using deep learning significantly enhances the reliability of LULC classification, aiding in understanding geography, socioeconomic conditions, poverty levels, and urban sprawl. However, the scarcity of annotated satellite data, especially in South/East Asian developing countries, poses a major challenge due to limited funding, diverse infrastructures, and dense populations. In this work, we introduce the BD Open LULC Map (BOLM), providing pixel-wise LULC annotations across eleven classes (e.g., Farmland, Water, Forest, Urban Structure, Rural Built-Up) for Dhaka metropolitan city and its surroundings using high-resolution Bing satellite imagery (2.22 m/pixel). BOLM spans 4,392 sq km (891 million pixels), with ground truth validated through a three-stage process involving GIS experts. We benchmark LULC segmentation using DeepLab V3+ across five major classes and compare performance on Bing and Sentinel-2A imagery. BOLM aims to support reliable deep models and domain adaptation tasks, addressing critical LULC dataset gaps in South/East Asia.