Object-oriented multi-feature ensemble classification of mangrove species using UAV imagery

Mangrove species classification is particularly important for coastal wetland protection and global warming mitigation. However, it is challenging to identify species-level differences in practical applications. In the paper, we propose an object-oriented multi-feature ensemble classifier for the fine classification of mangrove species. First, mangrove images are segmented into objects by the multi-resolution segmentation method. Second, multiple features of each object are extracted by feature calculation methods (Gray Level Co-occurrence Matrix, Standard Deviation, Mean, etc.), and appropriate features are selected for species classification by weight estimation. Finally, the selected features are fed to an ensemble classifier to generate the final mangrove species classification results. Experiments performed on in-situ unmanned aerial vehicle (UAV) images collected in Yingzai, Guangdong Province demonstrate that the proposed multi-feature ensemble classifier achieves superior classification results to its single classifier counterparts.

• Publication year

  2023

• Venue

  International Conference on Geographic Information and Remote Sensing Technology

• Publication date

  2023-02-10

• Fields of study

  Computer Science, Engineering, Environmental Science

• Identifiers
  DOI 10.1117/12.2667351
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Combining UAV-based hyperspectral and LiDAR data for mangrove species classification using the rotation forest algorithm

  2021cited by this paper
• On the Ability of PolSAR Measurements to Discriminate Among Mangrove Species

  2020cited by this paper
• Object-Oriented Mangrove Species Classification Using Hyperspectral Data and 3-D Siamese Residual Network

  2020cited by this paper
• Combining GF-2 and RapidEye satellite data for mapping mangrove species using ensemble machine-learning methods

  2019cited by this paper
• NDVI Based Change Detection in Sundarban Mangrove Forest Using Remote Sensing Data

  2019cited by this paper
• Object-Based Mangrove Species Classification Using Unmanned Aerial Vehicle Hyperspectral Images and Digital Surface Models

  2018cited by this paper
• Landsat-Based Estimation of Mangrove Forest Loss and Restoration in Guangxi Province, China, Influenced by Human and Natural Factors

  2015cited by this paper
• Mangrove Mapping and Change Detection in Ca Mau Peninsula, Vietnam, Using Landsat Data and Object-Based Image Analysis

  2015cited by this paper
• Annual review of marine science

  2013cited by this paper
• Uncertain future of New England salt marshes

  2011cited by this paper
• Centuries of human-driven change in salt marsh ecosystems.

  2009cited by this paper
• Evaluation of Morphological Texture Features for Mangrove Forest Mapping and Species Discrimination Using Multispectral IKONOS Imagery

  2009cited by this paper
• Mangrove forests: Resilience, protection from tsunamis, and responses to global climate change

  2008cited by this paper
• Present state and future of the world's mangrove forests

  2002cited by this paper
• The value of salt marsh as a sea defence

  1995cited by this paper

• No citing papers are available for this paper.