Vector Extraction Engine for 3D Road Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current linear feature extraction tools, such as ROADTRACKER, are inadequate for accurately extracting two-dimensional linear features from rectangular city road grids and lack three-dimensional extraction capabilities, and semi-automated extraction does not incrementally grow vectors with multiple mouse clicks.
Innovation Solution
A system comprising a vector server, raster server, digital surface model server, and rendering engine that enables two- and three-dimensional vector extraction with incremental growth during semi-automated extraction, using user inputs for guidance and incorporating digital surface models and high-resolution stereo imagery for accurate smoothing and topology cleaning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated bulk extraction is used for road centerlines, then extraction efficiency is improved, but geometric accuracy for rectangular city road grids deteriorates
Solution Approach 1:
The patent segments the extraction process into automated bulk extraction for general areas and manual/interactive refinement for specific problematic areas like rectangular city road grids. This allows high-speed automated processing to handle most cases while targeted manual intervention corrects geometric inaccuracies in challenging scenarios.
Solution Approach 2:
The patent applies partial automation by using automated extraction algorithms for the majority of road segments while requiring manual correction only for specific cases where geometric accuracy is compromised. This partial application of automated action maintains overall efficiency while ensuring precision where needed.
2Adaptability or versatility
If traditional extraction tools are used, then two-dimensional linear feature extraction is enabled, but three-dimensional extraction capability is lost
Solution Approach 1:
The patent extends traditional two-dimensional linear feature extraction by incorporating the third dimension (elevation/depth) through integration with digital surface models and stereo imagery. This dimensionality change enables extraction of three-dimensional linear features such as bridges, tunnels, and elevated roadways while maintaining the established two-dimensional extraction capabilities.
3Manufacturing precision
If semi-automated extraction requires complete click sequence before starting, then processing accuracy is improved, but user interaction efficiency deteriorates
Solution Approach 1:
The patent implements preliminary action by allowing users to place guide points incrementally and having the system pre-compute and display intermediate extraction results after each click. This preliminary visualization enables users to verify accuracy in real-time and make corrections before finalizing the complete extraction sequence, reducing overall interaction time while maintaining accuracy.
Solution Approach 2:
The patent incorporates continuous feedback by displaying incremental extraction results after each user click in the semi-automated process. This real-time feedback allows users to assess extraction accuracy progressively and adjust their guide point placement accordingly, improving both user efficiency and final extraction quality.
Data Source
AI summary
A system for vector extraction comprising a vector extraction engine stored and operating on a network-connected computing device that loads raster images from a database stored and operating on a network-connected computing device, identifies features in the raster images, and computes a vector based on the features, and methods for feature and vector extraction.