Building Shape Classification via Skeleton Line Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional template matching methods for building shape classification are limited by local feature interference, lack of comprehensive templates, and reduced accuracy for irregular buildings, particularly failing to accurately recognize shapes with non-right-angled edges.
Innovation Solution
A high-precision building shape recognition method involving the extraction of skeleton lines through the Douglas-Peucker algorithm, constrained Delaunay triangulation, and feature vector calculation using cosine similarity for template matching, which expands the template library and reduces local feature interference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If traditional template matching methods are used to compare building outlines with templates, then the matching process is simple to implement, but the recognition accuracy is limited and local features cause interference
Solution Approach 1:
The patent extracts the core structural information of buildings by generating skeleton lines that represent the essential shape characteristics, removing local feature interference while preserving global shape information for accurate template matching
Solution Approach 2:
The patent segments the building shape analysis into distinct components: skeleton line extraction, feature point identification, and template matching stages, allowing each component to be optimized independently for both simplicity and accuracy
2Reliability
If template matching is performed using building outlines, then right-angled buildings are recognized well, but irregular buildings with non-right-angled edges have poor recognition效果
Solution Approach 1:
The patent transforms the building shape representation from outline-based to skeleton-line-based, changing the fundamental parameter used for matching and enabling the system to handle both regular and irregular building shapes effectively
Solution Approach 2:
The patent uses asymmetric skeleton line structures that can represent both symmetric (right-angled) and asymmetric (irregular) building shapes, making the template library more versatile without sacrificing reliability for regular buildings
3Measurement precision
If a comprehensive template library is constructed to cover all building shapes, then recognition accuracy improves, but the complexity of the template library increases
Solution Approach 1:
The patent creates a universal skeleton line extraction method that can represent diverse building shapes using a unified approach, allowing a more compact and manageable template library while maintaining comprehensive coverage of building shape variations
Solution Approach 2:
The patent transitions from two-dimensional outline matching to a hierarchical approach using one-dimensional skeleton lines with associated feature points, reducing template library complexity while improving recognition accuracy through multi-dimensional feature analysis
Data Source
AI summary
Disclosed in the present disclosure is a classification method based on skeleton lines for building shapes. The method includes: (1) expanding a template library of buildings by combining building shape classification in architecture and building shape features in real life on the basis of inheriting advantages of a traditional template matching method; (2) avoiding influence of small depressions and protrusions of the buildings on overall shapes by constructing a least-squares template; (3) extracting the skeleton lines of the buildings and calculating feature vectors of the buildings; and (4) calculating similarity between feature vectors of skeleton lines of the buildings and skeleton lines of templates by using cosine similarity, and selecting the template with the highest similarity as a classification result of the building shapes.