Geospatial Data Conflation Tile Disagreement Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Geospatial data conflation tools are inefficient when dealing with dissimilar data sources, requiring extensive manual preprocessing and laborious adjustments, as they only produce acceptable results when sources are substantially similar, and fail to systematically identify and prioritize areas of disagreement.
Innovation Solution
A method that involves obtaining data from multiple sources, determining bounding polygons, dividing them into tiles, extracting features, and calculating disagreement levels to identify and display areas of dissimilarity, allowing for strategic prioritization and automated processing of geospatial data conflation using machine-readable instructions and hardware processors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If geospatial data conflation tools are applied to dissimilar data sources, then automated processing is performed, but the quality of results deteriorates and extensive manual preprocessing is required
Solution Approach 1:
The system performs preliminary actions by automatically identifying and flagging areas of disagreement between data sources before the conflation process begins. This preliminary identification allows users to focus manual efforts only on problematic areas rather than preprocessing entire datasets, thereby maintaining automated processing while ensuring high conflation quality.
Solution Approach 2:
The system introduces an intermediary component that acts as a bridge between dissimilar data sources and the conflation process. This intermediary automatically detects disagreements, quantifies their severity, and prioritizes them, enabling automated processing of dissimilar sources while maintaining quality by guiding subsequent manual or automated resolution efforts.
2Manufacturing precision
If users manually inspect all areas of disagreement between data sources, then thorough reconciliation is achieved, but the time and effort required increase dramatically
Solution Approach 1:
Instead of uniformly inspecting all areas, the system applies local quality by identifying and highlighting only those specific local regions where disagreements exist between data sources. The system quantifies disagreement levels and prioritizes them, allowing users to focus their inspection time and efforts precisely on areas that require reconciliation, thereby achieving thorough reconciliation without examining every area in detail.
Solution Approach 2:
The system performs partial action by identifying and presenting only the most significant areas of disagreement to users rather than requiring inspection of all possible disagreement areas. By prioritizing disagreements based on their impact and severity, the system enables users to achieve satisfactory reconciliation accuracy by addressing only the most critical areas, significantly reducing inspection time while maintaining adequate quality.
3Measurement precision
If users zoom in to inspect particular regions for closer examination, then detailed analysis is possible, but the ability to maintain strategic overview is lost
Solution Approach 1:
The system resolves this contradiction by adding another dimension to the visualization - a hierarchical or multi-scale view that allows users to see both detailed disagreements and the overall strategic picture simultaneously. The system provides an overview map showing the distribution and severity of disagreements across the entire dataset, while enabling drill-down to detailed inspection of specific areas, thereby maintaining strategic overview while allowing detailed analysis when needed.
Data Source
AI summary
A system may be configured to conflate vectorized source data. Some embodiments may: obtain first data from a first source and second data from a second source; determine a first polygon that encloses all features of the first data and a second polygon that encloses all features of the second data; determine a larger polygon that encloses the first and second polygons; divide the larger polygon into first tiles; extract, from each of the first tiles overlaying the first data and from the each tile overlaying the second data, a first set of features and a second set of features, respectively; and identify, based on a computed disagreement level satisfying a set of criteria, each of one or more of the tiles. A set of identified tiles or all of the tiles may then be displayed, including shaded indicators overlaying features of respective portions of the first and second data.