Mono Imagery Building Height Calculation via Edge Detection and Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring building heights, such as using mono 2D aerial imagery or stereo pairs, are impractical for large-scale applications due to the lack of 3D information and high costs associated with stereo photogrammetry and LiDAR scanning.
Innovation Solution
A method utilizing mono imagery that identifies building footprints and rooftops through edge detection and machine learning, calculates horizontal offsets, and applies RPC algorithms to determine building heights based on camera angles and pixel offsets, enabling cost-effective height calculation using satellite, plane, or aerial vehicle imagery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stereo pairs of 2D imagery are used to measure building heights, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent introduces an intermediary approach by using a single 2D image combined with a digital terrain model (DTM) and shadow analysis to derive building heights. Instead of requiring stereo pairs, the method uses the DTM as an intermediary reference surface and shadow characteristics as mediators to calculate height, thereby resolving the contradiction between measurement precision and device complexity
Solution Approach 2:
The patent replaces the mechanical/physical requirement of stereo image pairs with a computational approach using 2D image processing, DTM data, and shadow analysis algorithms. This substitution of physical measurement requirements with computational methods reduces device complexity while maintaining measurement capability
2Measurement precision
If LiDAR scanning is used to measure building heights, then measurement precision is improved, but cost increases
Solution Approach 1:
The patent employs inexpensive 2D aerial imagery and readily available digital terrain models as substitutes for expensive LiDAR scanning. These cheaper data sources, while individually less informative, are combined through computational methods to achieve reliable height measurements, effectively replacing costly measurement tools with affordable alternatives
Solution Approach 2:
The patent uses shadow characteristics and digital terrain models as intermediaries to bridge the gap between inexpensive 2D imagery and accurate 3D height measurement. This intermediary approach allows the system to achieve LiDAR-level precision without the associated costs by using available 2D data in creative ways
3Loss of energy
If mono 2D aerial imagery is used, then cost is reduced, but measurement precision deteriorates due to lack of 3D information
Solution Approach 1:
The patent applies dimensionality change by incorporating temporal and contextual dimensions into the 2D image analysis. By adding the time component (shadow position relative to sun angle) and using DTM data for the third spatial dimension, the system extracts 3D height information from 2D imagery, effectively adding dimensions to overcome the limitation of mono imagery
Solution Approach 2:
The patent changes parameters by utilizing shadow length, sun angle, and DTM elevation data in combination with 2D image coordinates. By transforming the problem from direct 3D measurement to parameter-based calculation using multiple variables, the system achieves accurate height measurement from 2D imagery through mathematical relationships
Data Source
AI summary
A technique is directed to methods and systems for calculating building heights from mono imagery. In some implementations, a building height calculation system performs orthorectification of an image of buildings against a digital terrain model to remove effects of terrain distortion from the image. The building height calculation system can execute an edge detection algorithm on the image to identify the edges of the building in the image. The edges can provide a rooftop vector of the building. The building height calculation system can execute, using image data at input, a machine learning algorithm to determine the footprint vector of the building in the image. The building height is calculated based on a camera angle, a distance from the camera to the building, and a pixel offset from the footprint vector to the rooftop vector.