Enhancing Coastal Wetland Mapping with Advanced Remote Sensing Technology

Researchers from North Carolina Agricultural and Technical State University have made significant strides in developing an accurate method for classifying coastal wetland areas using high-resolution multi-spectral imagery and LiDAR remote sensing data. By leveraging the Random Forest algorithm, the team was able to achieve impressive accuracy rates in different scenarios, with the best result reaching 94.1%. This breakthrough has far-reaching implications for practical applications in wetland mapping and ecological research.

Key Takeaways:

• Researchers from North Carolina Agricultural and Technical State University developed a new method for classifying coastal wetland areas using high-resolution multi-spectral imagery and LiDAR remote sensing data.
• The team used the Random Forest algorithm to enhance classification accuracy, achieving overall accuracy rates ranging from 88.2% to 94.1% in different scenarios.
• The study addressed binary classification for wetland and non-wetland classification and a multi-classification for different wetland classes.
• The Random Forest model's performance in different scenarios was evaluated, with Scenario 1 achieving an overall accuracy of 93.9%, Scenario 2 achieving an overall accuracy of 93.5%, Scenario 3 achieving an overall accuracy of 94.1%, and Scenario 4 achieving an overall accuracy of 88.2%.
• The study highlights the potential of advanced remote sensing technology for practical applications in wetland mapping and ecological research.
• The research has significant implications for environmental studies and conservation efforts in coastal regions.

Statistics:

• The study achieved an overall accuracy rate of 94.1% in Scenario 3.
• The Random Forest model's accuracy rates ranged from 88.2% to 94.1% in different scenarios.
• The study integrated multispectral imagery data, LiDAR, and additional sources to enhance classification accuracy.
• The Random Forest algorithm was used to significantly improve the overall accuracy of wetland mapping.

Sources:

• M. Anokye, Department of Built Environment, North Carolina Agricultural and Technical State University, 1601 E Market St, Greensboro, NC, United States
• L. Hashemi-Beni
• "Evaluating Coastal Wetland Mapping Accuracy with High-Resolution Multi-spectral Imagery and LiDAR Remote Sensing Data" (ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2025, X-G-2025():109-116).