Single-Aerial Building Area Estimation Using Predicted Height Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating building square footage are often inaccurate and costly, requiring multiple images or pre-existing models, which can be prohibitive.
Innovation Solution
A method utilizing a single aerial image normalized to a nadir perspective to estimate square footage by generating a Canopy Height Model (CHM) through computer vision, estimating building-related pixel heights, and summing story footprints without the need for expensive Digital Surface Models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple aerial images from different oblique viewpoints are used to estimate building geometries, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces expensive, complex multi-image acquisition systems with a single inexpensive aerial image. By using a trained computer vision model to generate a predicted Canopy Height Model from one image, the system eliminates the need for multiple oblique and orthogonal images, significantly reducing device complexity and data acquisition costs while maintaining estimation accuracy
Solution Approach 2:
The patent substitutes the mechanical/physical approach of capturing multiple images from different viewpoints with an information-processing approach. A trained computer vision model processes a single image to generate height predictions, replacing the need for physical multi-angle imaging systems and complex photogrammetry workflows
2Measurement precision
If Digital Surface Models are used to estimate building height, then measurement precision is improved, but cost increases significantly
Solution Approach 1:
The patent replaces expensive Digital Surface Models with a computationally-generated predicted Canopy Height Model derived from a single aerial image. This substitution dramatically reduces the cost of imaging data while maintaining the ability to estimate building heights accurately through machine learning-based height prediction
Solution Approach 2:
The patent creates a synthetic copy of height information through a predicted Canopy Height Model generated by a trained computer vision model. Instead of purchasing or acquiring expensive DSM data, the system synthesizes height predictions from a single image, providing a cost-effective alternative that replicates the functionality of DSMs
3Measurement precision
If manual square footage measurements are performed, then measurement precision is improved, but productivity decreases due to prohibitive costs
Solution Approach 1:
The patent replaces manual field measurement processes with an automated computer vision system. The trained model automatically processes aerial images to generate predicted height models, estimate building geometries, and calculate square footage, eliminating the need for costly manual measurements while improving both accuracy and cost efficiency
Solution Approach 2:
The system enables self-service measurement capabilities where the computer vision model automatically performs all measurement tasks without human intervention. The model processes images, generates height predictions, identifies building structures, and calculates square footage autonomously, providing accurate measurements at a fraction of manual measurement costs
Data Source
AI summary
Systems, methods, and non-transitory computer-readable media are disclosed herein for estimating square footage of a building from a single aerial image. The method includes receiving an aerial image of a property; generating, from the aerial image using a trained computer vision model, a predicted Canopy Height Model (CHM_prd); estimating, using the CHM_prd: a height of building-related pixels with respect to corresponding ground-related pixels in the CHM_prd, a number of stories associated with the building, and a footprint corresponding to each story of the building; summing square footage of each story to determine total square footage of the building; and outputting an indication of the total square footage of the building.