Canopy Identification via Shadow Offset Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying canopy structures from overhead images are inefficient and require manual inspection or resource-intensive data collection, making it difficult to accurately categorize buildings in geographic databases.
Innovation Solution
A computer-implemented method that calculates offsets between the start and end points of shadows cast by buildings and their vertices in overhead images, using these offsets to determine if a building footprint represents a canopy by comparing them to a threshold, thereby automating the identification process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection of street level images is used to differentiate canopies from other buildings, then identification accuracy can be maintained, but productivity is significantly reduced and time consumption increases
Solution Approach 1:
The patent replaces manual visual inspection with an automated computer-implemented method that uses overhead images and shadow analysis. The system automatically calculates offsets between shadow vertices and building footprint vertices, compares these offsets to thresholds, and determines canopy status without human intervention, thereby maintaining accuracy while dramatically improving productivity
Solution Approach 2:
The patent uses overhead images (aerial views) as a copy or alternative representation of the building and its shadow, replacing the need to manually inspect street level images. This copied perspective enables automated measurement of shadow offsets and building footprint characteristics, facilitating rapid automated classification
2Reliability
If resource-intensive data collection methods are used to identify canopy structures, then identification reliability can be improved, but use of energy and time consumption increase significantly
Solution Approach 1:
The patent uses readily available overhead images and existing building footprint data as substitutes for resource-intensive field surveys or specialized data collection campaigns. By working with existing aerial imagery and geographic database records, the system achieves reliable canopy identification without requiring additional time-consuming data gathering efforts
Solution Approach 2:
The patent employs overhead images that serve multiple purposes: they provide both the building footprint information and the shadow cast by the building. This multi-functional use of a single data source enables simultaneous extraction of multiple geometric parameters needed for canopy identification, reducing the need for separate specialized data collection efforts
Data Source
AI summary
Methods, apparatus and computer program for identifying canopy structures are provided. In one aspect, a method includes receiving a first calculated offset between a start point of a shadow cast by a building and a proximate first vertex of a building footprint of the building, and receiving a second calculated offset between an end point of the shadow cast by the building and a proximate second vertex of the building footprint, the shadow identified from an overhead image of the building, and wherein the building footprint comprises a shadowed side and a non-shadowed side, each running between the start point and the end point of the shadow on a different respective side of the building. The method also includes comparing the offsets to an offset threshold, and in response to both the first and second received offsets exceeding the offset threshold, determining that the building footprint represents a canopy.