Scholarlink

<p>Trees serve manifold ecosystem functions including climate change mitigation, biodiversity conservation, landscape restoration etc. yet are facing threats globally due to human intervention. As a result, effective conservation initiatives require quantifying both present and past extents of the tree cover. Remote sensing technologies coupled with machine learning techniques appear to be effective in mapping and monitoring tree cover for the past few decades and offering advantages over traditional approaches. Despite extensive research on vegetation, mangroves, forest health, and urban forests, a comprehensive review focusing solely on remote sensing’s role in tree cover mapping is lacking. This review aims to fill that gap by providing an overview of the studies that mapped tree cover using remote sensing, and discusses spatial context, satellite sensors, classification approaches utilized for mapping tree cover. Literature search using Google Scholar showed that such studies are prevalent in every continent to major climatic domain. From coarse to high resolution satellite data (e.g., Landsat, Sentinel, MODIS, Worldview, ALOS PALSAR etc.) are used independently or with an integration depending on the purpose, availability, and economical feasibility. While Landsat has gained more popularity due to its historical record and free availability, it faces limitations in identifying small fragments of tree cover. A wide range of tree cover mapping methodologies are available, and can be classified into pixel-based or object-based to supervised or unsupervised classification approaches which include machine learning techniques, such as Support Vector Machine (SVM), Decision Tree (DT), Nearest Neighbour (NN), Maximum Likelihood (ML), Artificial Neural Network (ANN), Ensemble etc. However, challenges exist in mapping tree cover using remote sensing. Future research should focus on improving classification performance by leveraging multi-source, high-resolution, multi-temporal, and multi-sensor data, embracing the evolving capabilities of remote sensing technologies along with artificial intelligence to enhance accuracy and ensure reliability in tree cover mapping.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A comprehensive review of tree cover mapping using satellite sensor data

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月2日

查看全文
  • Md. Shamim Reza Saimun,
  • M. Mahmudur Rahman

摘要

Trees serve manifold ecosystem functions including climate change mitigation, biodiversity conservation, landscape restoration etc. yet are facing threats globally due to human intervention. As a result, effective conservation initiatives require quantifying both present and past extents of the tree cover. Remote sensing technologies coupled with machine learning techniques appear to be effective in mapping and monitoring tree cover for the past few decades and offering advantages over traditional approaches. Despite extensive research on vegetation, mangroves, forest health, and urban forests, a comprehensive review focusing solely on remote sensing’s role in tree cover mapping is lacking. This review aims to fill that gap by providing an overview of the studies that mapped tree cover using remote sensing, and discusses spatial context, satellite sensors, classification approaches utilized for mapping tree cover. Literature search using Google Scholar showed that such studies are prevalent in every continent to major climatic domain. From coarse to high resolution satellite data (e.g., Landsat, Sentinel, MODIS, Worldview, ALOS PALSAR etc.) are used independently or with an integration depending on the purpose, availability, and economical feasibility. While Landsat has gained more popularity due to its historical record and free availability, it faces limitations in identifying small fragments of tree cover. A wide range of tree cover mapping methodologies are available, and can be classified into pixel-based or object-based to supervised or unsupervised classification approaches which include machine learning techniques, such as Support Vector Machine (SVM), Decision Tree (DT), Nearest Neighbour (NN), Maximum Likelihood (ML), Artificial Neural Network (ANN), Ensemble etc. However, challenges exist in mapping tree cover using remote sensing. Future research should focus on improving classification performance by leveraging multi-source, high-resolution, multi-temporal, and multi-sensor data, embracing the evolving capabilities of remote sensing technologies along with artificial intelligence to enhance accuracy and ensure reliability in tree cover mapping.