Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

An Atmospheric Correction Integrated LULC Segmentation Model for High-Resolution Satellite Imagery

Soham Mukherjee, Yash Dixit, Naman Srivastava, Joel D Joy, Rohan Olikara, Koesha Sinha, Swarup E, Rakshit Ramesh

TL;DR

Look-up-table-based radiative transfer simulations are employed to estimate the atmospheric path reflectance and transmittance for atmospherically correcting high-resolution CARTOSAT-3 Multispectral (MX) imagery for several Indian cities, demonstrating stability in multi-class LULC segmentation accuracy.

Abstract

The integration of fine-scale multispectral imagery with deep learning models has revolutionized land use and land cover (LULC) classification. However, the atmospheric effects present in Top-of-Atmosphere sensor measured Digital Number values must be corrected to retrieve accurate Bottom-of-Atmosphere surface reflectance for reliable analysis. This study employs look-up-table-based radiative transfer simulations to estimate the atmospheric path reflectance and transmittance for atmospherically correcting high-resolution CARTOSAT-3 Multispectral (MX) imagery for several Indian cities. The corrected surface reflectance data were subsequently used in supervised and semi-supervised segmentation models, demonstrating stability in multi-class (buildings, roads, trees and water bodies) LULC segmentation accuracy, particularly in scenarios with sparsely labelled data.

An Atmospheric Correction Integrated LULC Segmentation Model for High-Resolution Satellite Imagery

TL;DR

Look-up-table-based radiative transfer simulations are employed to estimate the atmospheric path reflectance and transmittance for atmospherically correcting high-resolution CARTOSAT-3 Multispectral (MX) imagery for several Indian cities, demonstrating stability in multi-class LULC segmentation accuracy.

Abstract

The integration of fine-scale multispectral imagery with deep learning models has revolutionized land use and land cover (LULC) classification. However, the atmospheric effects present in Top-of-Atmosphere sensor measured Digital Number values must be corrected to retrieve accurate Bottom-of-Atmosphere surface reflectance for reliable analysis. This study employs look-up-table-based radiative transfer simulations to estimate the atmospheric path reflectance and transmittance for atmospherically correcting high-resolution CARTOSAT-3 Multispectral (MX) imagery for several Indian cities. The corrected surface reflectance data were subsequently used in supervised and semi-supervised segmentation models, demonstrating stability in multi-class (buildings, roads, trees and water bodies) LULC segmentation accuracy, particularly in scenarios with sparsely labelled data.
Paper Structure (11 sections, 2 equations, 3 figures, 1 table)

This paper contains 11 sections, 2 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Materials and Methods
  3. Materials
  4. Raster Data
  5. Vector Data
  6. Methods
  7. Atmospheric Correction
  8. LULC Model Training
  9. LULC Model Evaluation
  10. Results and Discussion
  11. Conclusion

Figures (3)

  • Figure 1: Workflow for LULC segmentation
  • Figure 2: (a) TOA reflectance for the subset MX image acquired over Hyderabad, (b) AC obtained BOA reflectance for the same extent as panel (a),(c) CPS model predictions, (d) inter-comparison of TOA and BOA reflectance for four LULC classes, (e) LUT obtained mean Path reflectance for the LULC classes. The shaded regions show the 1$\sigma$ uncertainty for each class.
  • Figure :