Stereoscopic Point Cloud Roof Modeling for Accurate Cost Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for processing aerial images to generate 3D models of structures are inadequate in accurately depicting elevation, detecting internal line segments, and segmenting models for cost-accurate cost estimation, resulting in incomplete or inaccurate 3D models.
Innovation Solution
A system that processes stereoscopic image pairs to compute disparity values, generate 3D point clouds, and fuse them to create a final point cloud model, allowing for the delineation of structural features and generation of three-dimensional wireframe models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional software systems process aerial images to generate 3D models, then processing can be performed, but the models are inaccurate and incomplete in depicting elevation, internal line segments, and structural features
Solution Approach 1:
The patent segments the 3D modeling process into distinct stages: generating initial point clouds from aerial images, identifying structural features within those point clouds, and creating refined 3D models based on identified features. This segmentation allows each stage to be optimized independently, improving both elevation accuracy and model completeness.
Solution Approach 2:
The patent introduces point clouds as an intermediary representation between aerial images and final 3D models. By first converting images to point clouds and then analyzing structural features within the point cloud data, the system achieves more accurate elevation depiction and complete structural representation than direct image-to-model conversion.
2Manufacturing precision
If conventional systems generate 3D models from aerial images, then models can be created, but they lack accurate internal line segments and feature delineation
Solution Approach 1:
The patent performs preliminary identification of structural features within point clouds before generating the final 3D model. By pre-identifying walls, roofs, doors, and windows in the point cloud data, the system establishes accurate feature boundaries and line segments that guide subsequent model generation, ensuring precise feature delineation.
Solution Approach 2:
The patent transitions from 2D aerial images to 3D point clouds, adding a dimensional perspective that reveals internal line segments and structural features invisible in 2D projections. This dimensional transformation enables accurate depiction of three-dimensional structural relationships and feature boundaries.
3Measurement precision
If detailed 3D models with complete structural features are generated, then accurate cost estimation is possible, but processing time and computational resources increase
Solution Approach 1:
The patent segments the modeling process to focus computational resources on identifying and modeling only the structural features relevant to cost estimation (walls, roofs, doors, windows) rather than processing every detail of the entire scene. This selective segmentation maintains cost estimation accuracy while reducing overall processing time.
Solution Approach 2:
The patent applies partial action by generating point clouds and identifying features only in regions containing structures of interest, rather than processing entire aerial imagery datasets. This approach achieves sufficient model detail for cost estimation purposes without the computational burden of exhaustive processing.
Data Source
AI summary
A system for modeling a roof structure comprising an aerial imagery database and a processor in communication with the aerial imagery database. The aerial imagery database stores a plurality of stereoscopic image pairs and the processor selects at least one stereoscopic image pair among the plurality of stereoscopic image pairs and related metadata from the aerial imagery database based on a geospatial region of interest. The processor identifies a target image and a reference image from the at least one stereoscopic pair and calculates a disparity value for each pixel of the identified target image to generate a disparity map. The processor generates a three dimensional point cloud based on the disparity map, the identified target image and the identified reference image. The processor optionally generates a texture map indicative of a three-dimensional representation of the roof structure based on the generated three dimensional point cloud.