Development of a Free Open-Source Hybrid Segmentation Plug-In to Extract Agricultural Field Boundaries in a Heterogeneous Land System

DOI

10.2991/978-94-6463-252-1_31How to use a DOI?

Keywords

Hybrid segmentation; Sentinel-2A; Cartosat-2E

Abstract

Accurate representation of agricultural field boundaries is a prerequisite to perform crop classification, evaluating sustainable strategies, and implementing agricultural policies. However, this is an elusive task in the Indian agro-economic setting due to land fragmentation, and small and irregular farm boundaries. In spite of significant advances in the development of efficient segmentation algorithms, their application is limited due to a lack of implementation tool/software, resulting in a wide gap between research and practice. An open-source tool is developed to automatically extract the agricultural field parcels using Red channel (R), Green channel (G), Blue channel (B), and Near Infrared Region (NIR) spectral channels. The tool will work based on Object-based Image Analysis (OBIA) methodology, i.e., the Sobel edge extraction operator will obtain the edges from each band and the edges were added from RGB, and NIR bands to obtain the final edge map. Watershed Segmentation is used to form the regions from the edges. The correctness of the segmentation is computed by using Quantitative completeness (QC) measure. The effectiveness of the plugin was explained on the marginal land system in India using Sentinel-2A and Cartosat-2E imagery. While Sentinel-2A has resulted in under-segmentation (QC = 0.84) with an overall accuracy (OA) of 37.05%, Cartosat-2E has resulted in over-segmentation (QC = 1.49) with an OA of 76.67%. Field boundary extraction with Cartosat-2E imagery performed well in extracting the field parcels relative to ground truth samples.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Sravani Duvvuri
AU  - B. V. N. P. Kambhammettu
AU  - S. S. S. V. Gopala Raju
PY  - 2023
DA  - 2023/11/09
TI  - Development of a Free Open-Source Hybrid Segmentation Plug-In to Extract Agricultural Field Boundaries in a Heterogeneous Land System
BT  - Proceedings of the Second International Conference on Emerging Trends in Engineering (ICETE 2023)
PB  - Atlantis Press
SP  - 282
EP  - 290
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-252-1_31
DO  - 10.2991/978-94-6463-252-1_31
ID  - Duvvuri2023
ER  -