Supervised automatic roof modeling
An automated method is disclosed that classifies a first component and a second component of a roof with data points being part of, or extracted from at least one image. The first component and the second component have a ridge, at least one eave parallel to the ridge and a rectangular base. Evidence of a soft constraint and a hard constraint is identified via the data points and such evidence is associated with the first component and the second component. At least one hypothesis model of the roof is generated using relationships between the first component and the second component, the soft constraint and the hard constraint. The hypothesis model is transformed into a three dimensional model, and the three dimensional model is used to generate a roof report of the roof.
1. One or more non-transitory computer readable medium storing a set of computer executable instructions for running on one or more computer systems that when executed cause the one or more computer systems to:
determine a hypothesis model of a roof by:
analyzing data points indicative of a roof to locate evidence of predetermined roof components, soft constraints and hard constraints related to the predetermined roof components, the hard constraints being conditions of variables within an objective function for modeling the roof that are required to be satisfied, the soft constraints being conditions of the objective function of the roof that do not have to be satisfied, but contribute to the objective function of the roof in an optimization of the objective function;
identifying the predetermined roof components using the evidence of the predetermined roof components, the soft constraints and the hard constraints, the predetermined roof components defined as a rectangular base and at least one eave;
for each of the predetermined roof components, determining at least one type of roof configuration, and geometry of the roof component using common-crest component parameters; and
determining an overlapping relationship between the multiple predetermined roof components using the evidence of soft constraints and hard constraints; and
translating the hypothesis model into a three-dimensional model of the roof using an iterative optimization process.
2. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 1 , wherein the common-crest component parameters are sufficient to distinguish hip and gable ends, sheds and pyramids.
3. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 1 , wherein a first hypothesis model and a second hypothesis model of the roof are determined.
4. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 3 , further comprising determining a first score for the first hypothesis model and a second score for the second hypothesis model, the first score and the second score being different.
5. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 4 , further comprising ranking the first score and the second score and providing each score in a report.
6. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 4 , wherein the determination of the score includes comparison of the first hypothesis model and the second hypothesis model with at least one model in a model database.
7. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 1 , further comprising performing at least one iterative optimization process such that the hypothesis model satisfies all of the conditions for the hard constraint and the soft constraint for the at least one component.
8. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 1 , wherein the three-dimensional model of the hypothesis model is configured for importation within third party software without manual manipulation.
9. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 1 , wherein the three-dimensional model of the hypothesis model is configured for importation within CAD software without manual manipulation.
10. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer systems of claim 1 , further comprising generating a roof report having at least one image showing the three-dimensional model and an estimated roof pitch, the three-dimensional model being an optimized component-based model.
11. The one or more non-transitory computer readable medium storing the set of computer executable instructions for running on one or more computer system of claim 1 , further comprising cataloguing at least one of the hypothesis model and the three-dimensional model of the roof within a model database.
12. An automated method performed by at least one processor running computer executable instructions stored on at least one non-transitory computer readable medium, comprising:
generating at least one hypothesis model of the roof by:
classifying a first component and a second component of a roof with data points being part of, or extracted from at least one image, the first component and the second component having a ridge, at least one eave parallel to the ridge and a rectangular base, the ridge of the first component being a first ridge, and the ridge of the second component being a second ridge;
determining type of configuration of the first ridge and the second ridge using common-crest component parameters;
identifying evidence of a soft constraint and a hard constraint via the data points and associating the evidence with the first component and the second component;
determining an overlapping relationship between the first component and the second component;
transforming the hypothesis model into a three-dimensional model using an iterative optimization process;
analyzing the three-dimensional model for edges representing outer boundaries of the roof; and,
generating a roof report of the roof.
13. The automated method of claim 12 , wherein at least some of the data points are part of a point cloud.
14. The automated method of claim 12 , further comprising collecting evidence and constraints via supervisory feedback of one or more of data points, feature derived from the data points, elements of a hypothesis model, elements of an optimized component based model, or elements of an object model.
15. The automated method of claim 14 , further comprising altering at least one hypothesis model via the evidence and constraints provided by the supervisory feedback.
16. The automated method of claim 12 , wherein at least one of the hypothesis model is classified and stored in a model database.
17. The automated method of claim 12 , wherein generating at least one hypothesis model of the roof comprises generating a first hypothesis model and a second hypothesis model of the roof, and further comprising determining a first score for the first hypothesis model and a second score for the second hypothesis model by at least comparing the first hypothesis model and the second hypothesis model with at least one model stored within a model database, the first score and the second score being different.
18. An automated computerized system, comprising:
a computer system executing image display and analysis software reading:
at least one image depicting an object of interest;
supervisory feedback of at least one hard constraint or at least one soft constraint related to the object of interest in the image; and
wherein the image display and analysis software executed by the computer system determines at least one hypothesis model having multiple components and information indicative of an overlapping relationship between the components, the components formed of a ridge, at least one eave, and a rectangular base, and the image display and analysis software performs at least one iterative optimization process on the hypothesis model such that the hypothesis model satisfies all of the at least one hard constraint or the at least one soft constraint, and wherein the hypothesis model is provided in a form of a list of the components including absolute locations and relative parameters of the components and rendered to form a scene model, the scene model parameterized to absolute components, and wherein the image display and analysis software executed by the computer system forms a three-dimensional model from the scene model, the three dimensional model capable of importation into a CAD program without additional supervisory feedback to the three-dimensional model.