Building Extraction from High-Resolution Remote Sensing Images by Adaptive Morphological Attribute Profile under...

Area of Science:

• Remote Sensing
• Computer Vision
• Geographic Information Systems

Background:

• Accurate building extraction from high-resolution remote sensing (HRRS) images is crucial for urban planning and management.
• Existing methods often struggle with diverse urban scenes and complex object characteristics.

Purpose of the Study:

• To propose a novel adaptive morphological attribute profile under object boundary constraint (AMAP-OBC) method for automatic building extraction.
• To enhance the accuracy and robustness of building extraction in HRRS images.

Main Methods:

• Developed an adaptive morphological attribute profile under object boundary constraint (AMAP-OBC).
• Implemented a preprocessing step to screen non-building objects using rule-based candidate object extraction.
• Utilized adaptive scale parameter extraction and object boundary constraint strategies for initial building set generation.
• Employed a further identification strategy with adaptive threshold combination for final building extraction.

Main Results:

• The proposed AMAP-OBC method demonstrated outstanding performance in automatic building extraction.
• The method showed effectiveness across multiple groups of HRRS images from different sensors.
• Successful extraction of diverse geographic objects in urban scenes was achieved.

Conclusions:

• The AMAP-OBC method offers a robust and accurate solution for automatic building extraction from HRRS imagery.
• The integration of adaptive strategies and boundary constraints improves extraction accuracy in complex urban environments.
• This approach holds significant potential for applications in urban remote sensing and geographic information systems.