Detailed Urban Land Use Land Cover Classification at the Metropolitan Scale Using a Three-Layer Classification...

Area of Science:

• Remote Sensing
• Geospatial Analysis
• Urban Planning

Background:

• Detailed urban Land Use/Land Cover (LULC) information is crucial for effective urban and environmental management.
• Existing remote sensing data (medium or very high resolution) has limitations in extracting detailed urban LULC for large areas due to resolution, cost, or data availability.

Purpose of the Study:

• To develop a novel three-layer classification scheme for deriving detailed urban LULC information.
• To integrate Chinese GF-1 (medium resolution) and GF-2 (very high resolution) satellite imagery for enhanced LULC mapping.
• To overcome the limitations of existing methods for large-scale, detailed urban LULC extraction.

Main Methods:

• Integration of GF-1 and GF-2 satellite imagery with varying spatial resolutions.
• Application of multi-resolution image segmentation and object-based image classification (OBIA).
• Incorporation of geometry, texture, and spectral information using a random forest algorithm.

Main Results:

• Successful extraction of homogeneous LULC types (water, vegetation) using GF-1 imagery.
• Accurate identification of heterogeneous urban LULC types (buildings, roads) using GF-2 imagery.
• Achieved overall accuracies of 0.89 and 0.87 for the second and third level LULC classification maps, respectively.

Conclusions:

• The developed three-layer classification scheme effectively integrates medium and very high-resolution imagery for detailed urban LULC mapping.
• This approach provides a viable solution for accurate LULC information extraction in large urban areas.
• The methodology demonstrates significant potential for improving urban and environmental management through enhanced geospatial data.