Roof report generation
A computer-implemented method for generating roof reports. In one embodiment, a user begins by entering an address into an application and clicking a “request” button. A 3D model of the structure at the address is prepared. The application may then proceed with manual or automatic roof extraction. For manual roof extraction, an analyst may outline roof facets to create a roof model. For automatic roof extraction, the application may use machine learning or another technique to automatically extract and then simplify roof facets. In either case, the application may then present the roof model to a reviewer for inspection. The reviewer may then accept or reject the roof model. If the reviewer rejects the roof model, an analyst may correct the roof model and resuming it for review. Once the roof model passes a review, the application may then proceed with converting the roof model into a roof report.
1. A computer-implemented method of identifying roof features and generating a roof report, the method comprising:
generating a digital 3D model of a structure from imagery of the structure;
generating, via an electronic interface that receives user input, a roof model from the digital 3D model of the structure;
displaying the digital 3D model of the structure and the roof model in the electronic interface; and
generating, using computer processing elements an electronically viewable roof report based on the roof model.
2. The method of claim 1 , further comprising outlining roof facets of a roof of the structure based on user input received via the electronic interface.
3. The method of claim 2 , further comprising automatically adjusting facet vertex elevations based on the digital 3D model of the structure, or
manually adjusting, based on the user input received via the electronic interface, the facet vertex elevations.
4. The method of claim 2 , wherein facet vertices snap to at least one of a surface, other facet vertices, and facet edges of the digital 3D model of the structure.
5. The method of claim 2 , further comprising indicating, via the electronic interface, whether a facet is substantially planar.
6. The method of claim 2 , further comprising identifying, via the electronic interface, one or more right-angle corners of the roof model.
7. The method of claim 2 , further comprising drawing guidelines based on user input receives by the electronic interface.
8. The method of claim 7 , wherein facet vertices snap to at least one of the guidelines and guide line intersections.
9. The method of claim 1 , wherein the roof report comprises at least one of a roof model rendering, data regarding one or more edge lengths, data regarding one or more edge types, data regarding one or more facet areas, data regarding one or more facet pitches, data regarding roof area, data regarding a predominant pitch, overhead imagery, oblique imagery, and an estimate for replacing a roof of the structure.
10. The method of claim 1 , further comprising displaying the roof model in the electronic interface for user review.
11. The method of claim 1 , wherein the roof model is displayed as overlaid on the digital 3D model of the structure in the electronic interface.
12. A computer-implemented method of identifying roof features and generating a roof report, the method comprising:
generating a digital 3D model of a structure from imagery of the structure;
using computer processing components to automatically extract roof features from the 3D model of the structure to create a roof model;
displaying the 3D model of the structure and the roof model in an electronic interface; and
using computer processing components to generate an electronically viewable roof report based on the roof model.
13. The method of claim 12 , further comprising automatically generating the roof model; and
manually editing the roof model via the electronic interface.
14. The method of claim 13 , further comprising adding, removing, or modifying, facets, edges, and vertices of the roof model based on user input via the electronic interface.
15. The method of claim 12 , wherein the roof report comprises at least one of a roof model rendering, data regarding one or more edge lengths, data regarding one or more edge types, data regarding one or more facet areas, data regarding one or more facet pitches, data regarding roof area, data regarding a predominant pitch, overhead imagery, oblique imagery, and an estimate for replacing a roof of the structure.
16. The method of claim 12 , wherein using computer processing components to automatically extract the roof features from the 3D model of the structure to create the roof model comprises applying a convolutional neural network to the 3D model of the structure.
17. The method of claim 16 , wherein a training set of the neural network is automatically augmented by randomizing at least one of the translation, scale, and rotation of training models.
18. The method of claim 16 , further comprising training the neural network using stochastic gradient descent with cross-entropy loss to optimize network weights.
19. The method of claim 12 , wherein the roof model is displayed as overlaid on the digital 3D model of the structure in the electronic interface.
20. The method of claim 12 , further comprising displaying inset images of the structure in the electronic interface.