Mapping coconut plantation in Western Agro-Climatic zone using object-based classification and machine learning technique

Aug 4, 2026

Upload by Levi

Mapping coconut plantations in Tamil Nadu’s Western Agro-Climatic Zone uses Object-Based Classification (OBC) and machine learning with multi-year satellite observations to address the difficulty of monitoring a geographically dispersed, seasonally varying crop. A ten-year Landsat 7 optical time series (2012–2013 and 2022–2023) was integrated with ground-truth surveys. Support Vector Machine (SVM) and Random Forest (RF) classifiers were tested, with RF delivering higher performance (91.7% in 2012–2013; 90.3% in 2022–2023). Change detection quantified coconut area expansion of 3,270 hectares, with Coimbatore contributing 2,560 hectares, supporting OBC plus RF for accurate mapping.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Abstract

Introduction

PLANT SCIENCE TODAY ISSN 2348-1900 (online) Vol 11(sp4): 01–08 [https://doi.org/10.14719/pst.4861](https://doi.org/10.14719/pst.4861) HORIZON e-Publishing Group RESEARCH ARTICLE Mapping coconut plantation in Western Agro-Climatic zone using object-based classification and machine learning technique Nithya Segar VP1, Ragunath Kaliaperumal1*, Pazhanivelan S¹, Kumaraperumal R¹& Latha Paramanandham2 1 Department of Remote Sensing and GIS, Tamil Nadu Agricultural University, Coimbatore 641003, India 2 Coconut Research Station, Tamil Nadu Agricultural University, Coimbatore 641003, India *Email: [ragunathkp@tnau.ac.in](ragunathkp@tnau.ac.in) OPEN ACCESS ARTICLE HISTORY Received: 29 August 2024 Accepted: 29 September 2024 Available online Version 1.0 : 15 November 2024 Additional information Peer review: Publisher thanks Sectional Editor and the other anonymous reviewers for their contribution to the peer review of this work. Reprints & permissions information is available at [https://horizonepublishing.com/](https://horizonepublishing.com/)[ ](https://horizonepublishing.com/)[journals/index.php/PST/open_access_policy](journals/index.php/PST/open_access_policy) [Publisher](Publisher)’s Note: Horizon e-Publishing Group remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Indexing: Plant Science Today, published by Horizon e-Publishing Group, is covered by Scopus, Web of Science, BIOSIS Previews, Clarivate Analytics, NAAS, UGC Care, etc See [https://horizonepublishing.com/journals/](https://horizonepublishing.com/journals/)[ ](https://horizonepublishing.com/journals/)[index.php/PST/indexing_abstracting](index.php/PST/indexing_abstracting) [Copyright](Copyright:)[:](Copyright:) © [The Author](The Author)([s](s))[. This is](. This is) an openaccess article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited ([https://creativecommons.org/licenses/](https://creativecommons.org/licenses/)[ ](https://creativecommons.org/licenses/)[by/4.0/](by/4.0/)) CITE THIS ARTICLE Nithya VP, Ragunath K, Pazhanivelan S, Kumaraperumal R, Latha P. Mapping coconut plantation in Western Agro-Climatic zone using object-based classification and machine learning technique. Plant Science Today.2024;11()sp4:01-08. [https:/doi.org/10.14719/pst.4861](https:/doi.org/10.14719/pst.4861) Abstract Coconut (Cocos nucifera), a key crop for over 10 million farming families in India, is vital in the agricultural economies of southern states like Tamil Nadu. However, traditional methods of monitoring coconut plantations are challenging due to the crop's geographical dispersion and seasonal variations. This study focuses on mapping coconut plantations in the Western Agro-Climatic Zone of Tamil Nadu using Object-Based Classification (OBC) and machine learning techniques. A ten-year time series of Landsat 7 optical satellite data (2012-2013 and 2022-2023) was employed, combined with ground truth surveys across the region. The study utilized Support Vector Machine (SVM) and Random Forest (RF) classifiers, with RF demonstrating superior accuracy. The RF classifier achieved an accuracy of 91.7% in 2012- 2013 and 90.3% in 2022-2023, outperforming SVM, which hovered around 70%. The research also conducted a change detection analysis, revealing anet increase of 3,270 hectares of coconut plantations over the decade, with the Coimbatore district contributing the most significant growth of 2,560 hectares. This study underscores the effectiveness of integrating OBC and machine learning, mainly RF, for accurate and efficient mapping of coconut plantations using Landsat satellite data. Keywords area mapping; change detection; coconut; Landsat 7; machine learning; objectbased classification Introduction Coconut (Cocos nucifera), a perennia

FAQs

Why is coconut plantation mapping challenging in the Western Agro-Climatic zone?

Traditional monitoring is difficult because coconut plantations are geographically dispersed and affected by seasonal variations, while data collection across hierarchical levels can be complex and delayed.

What data and methods were used to map coconut plantations?

The study used a ten-year Landsat 7 optical satellite time series (2012–2013 and 2022–2023) combined with ground-truth surveys, applying Object-Based Classification (OBC) and machine learning classifiers.

Which classifier performed better, and what were its accuracies?

Random Forest (RF) outperformed Support Vector Machine (SVM). RF achieved 91.7% accuracy for 2012–2013 and 90.3% for 2022–2023, while SVM performed around 70%.

Aug 4, 2026

Upload by Levi