Geospatial Imaging System Bare Earth Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current topographical models, such as digital elevation maps (DEMs), lack the ability to effectively classify and display different types of geospatial data, particularly in distinguishing bare earth segments from non-bare earth features, which can lead to inaccurate terrain representations and increased manual processing time.
Innovation Solution
A geospatial imaging system that uses a processor to determine segments within a geospatial dataset based on common geometric characteristics, classifies border points to identify bare earth segments, and employs a support vector machine (SVM) for training and classification, allowing for the display of bare earth data points with varying confidence values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated DEM generation is used, then processing speed is improved, but classification accuracy deteriorates
Solution Approach 1:
The patent segments the geospatial data into distinct classes (bare earth, vegetation, buildings, water) by dividing the data into multiple segments based on geometric characteristics. This segmentation allows the system to maintain automated processing while improving classification accuracy by analyzing specific geometric properties of each segment rather than treating all data uniformly.
Solution Approach 2:
The patent applies local quality analysis by examining the geometric characteristics of each segment individually. The system determines whether segments represent bare earth by analyzing local geometric properties such as slope, aspect, and elevation patterns, rather than applying a single global classification rule. This localized analysis improves accuracy while maintaining automation.
2Measurement precision
If manual classification is used, then classification accuracy is improved, but processing time increases
Solution Approach 1:
The patent implements self-service automation where the system automatically classifies geospatial data without requiring manual intervention. The automated classification algorithm analyzes geometric characteristics and confidently identifies bare earth segments, eliminating the need for manual classification while maintaining high accuracy and reducing processing time.
Solution Approach 2:
The patent changes the classification parameters from manual visual assessment to automated geometric parameter analysis. By using measurable geometric characteristics (slope, aspect, elevation) as classification criteria, the system achieves both high accuracy and automated processing, eliminating the time-consuming manual classification process.
3Device complexity
If simple DEM modeling is used, then device complexity is reduced, but information completeness deteriorates
Solution Approach 1:
The patent segments the terrain model into multiple classes (bare earth, vegetation, buildings, water) rather than creating a single simple DEM. This segmentation preserves important information about different terrain types and features, preventing information loss while maintaining reasonable modeling complexity through systematic classification.
Solution Approach 2:
The patent maintains local quality variations in the terrain model by preserving distinct geometric characteristics of different segments. Instead of smoothing all variations into a uniform DEM, the system retains local geometric properties that distinguish bare earth from other features, thereby preserving information completeness without excessive complexity.
Data Source
AI summary
A geospatial imaging system may include a geospatial data storage device configured to store a geospatial dataset including geospatial data points. A processor may cooperate with the geospatial data storage device to determine segments within the geospatial dataset, with each segment including neighboring geospatial data points within the geospatial dataset sharing a common geometric characteristic from among different geometric characteristics. The processor may further determine border geospatial data points of adjacent segments, compare the border geospatial data points of the adjacent segments to determine bare earth segments having respective heights below those of the border geospatial data points of adjacent segments, and classify geospatial data points within each bare earth segment as bare earth geospatial data points.