Skip to main content

ORIGINAL RESEARCH article

Front. Earth Sci.
Sec. Geoinformatics
Volume 12 - 2024 | doi: 10.3389/feart.2024.1487160
This article is part of the Research Topic Emerging Trends and Advancements of Geoinformatics Applications in Earth and Environmental Systems View all 5 articles

Automating band selection of AVIRIS data for indices using parallel processing

Provisionally accepted
Veerendra Satya Sylesh Peddinti Veerendra Satya Sylesh Peddinti 1Venkata Ravibabu Mandla Venkata Ravibabu Mandla 2*Shashi Mesapam Shashi Mesapam 3Suresh Kancharla Suresh Kancharla 4
  • 1 National Institute of Technology Warangal, Warangal, Telangana, India
  • 2 Other
  • 3 National Institute of Technology, Warangal, India
  • 4 Indian Council of Agricultural Research – IIOPR, Pedavegi, India

The final, formatted version of the article will be published soon.

    AVIRIS data comprises multiple bands, while Sentinel-2 data has a single band spanning a specific wavelength range. Traditionally, only single-band combinations have been used to compute indices with hyperspectral data. The process of selecting AVIRIS bands involves several steps: calculating spectral indices from all possible AVIRIS band combinations, computing the Root Mean Squared Error (RMSE) between the AVIRIS indices and Sentinel-2 index, transforming to reduce RMSE skewness, and choosing bands that deviate less than the [mean standard deviation] for each Land Use Land Cover (LULC) category. In current research, the entire process is automated and employs parallel processing using Python and its libraries. This facilitates the selection of AVIRIS bands for various indices like the Normalized Difference Vegetation Index (NDVI), the Normalized Difference Water Index (NDWI), the Normalized Difference Red Edge (NDRE), the Green Normalized Difference Vegetation Index (GNDVI), and the Normalized Difference Built-up Index (NDBI). The performance of the bands chosen is validated using the proposed methodology against all bands within the spectral range, and single bands. Additionally, the reduction in execution time achieved through parallel processing using a high-performance computational platform was evaluated.

    Keywords: AVIRIS, Automation, Band selection, Hyperspectral data, indices, parallel processing

    Received: 27 Aug 2024; Accepted: 27 Dec 2024.

    Copyright: © 2024 Peddinti, Mandla, Mesapam and Kancharla. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.