Scale error correction in a multi-dimensional model

• US 9,478,031 B2
• Filed: 01/30/2015
• Issued: 10/25/2016
• Est. Priority Date: 01/31/2014
• Status: Active Grant

• Current Assignee: Hover Inc. (Tucows Incorporated)
• Original Assignee: Hover Inc. (Tucows Incorporated)
• Inventors: Bhatawadekar, Vineet, Sun, Shaohui, Pavlidis, Ioannis, Altman, Adam J.
• Primary Examiner(s): Crawford, Jacinta M
• Assistant Examiner(s): Cofino, Jonathan M
• Application Number:
  US14/610,850
• Publication Number:
  US 20150221128A1
• Time in Patent Office: 634 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06T 11/60 Editing figures and text; Combining figures or text
• CPC Class Codes: G06T 17/05 Geographic models
• CPC Class Codes: G06T 19/20 Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
• CPC Class Codes: G06T 2200/04 involving 3D image data
• CPC Class Codes: G06T 2207/30184 Infrastructure
• CPC Class Codes: G06T 2210/04 Architectural design, interior design
• CPC Class Codes: G06T 2219/2016 Rotation, translation, scaling
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/13 Edge detection

Abstract

A system and method is provided for identifying error and rescaling and constructing or reconstructing a multi-dimensional (e.g., 3D) building model using street-level imagery. Street-level imagery is used to identify architectural elements that have known architectural standard dimensions. Dimensional measurements of architectural elements in the multi-dimensional building model (poorly scaled) are compared with known architectural standard dimensions to rescale and construct/reconstruct an accurate multi-dimensional building model.

Claims

1. A method of resealing a multi-dimensional building model comprising:

retrieving an image used during construction of the multi-dimensional building model;

identifying a plurality of boundaries for an architectural element located within the image and included in the multi-dimensional building model;

defining the plurality of boundaries for the architectural element;

measuring a distance between two or more of the plurality of boundaries of the architectural element;

comparing the measured distance against one or more known dimensions, within a threshold of error, of at least one similar architectural element;

determining an error in scaling of the image based on the comparing being performed against a selection of a plurality of sizes of the at least one similar architectural element and selecting a size that is dimensionally closest to the architectural element and within the threshold of error; and

resealing the building model based on a difference between the measured distance and the one or more known dimensions of the similar architectural element of the selected size to correct for the determined error.

2. The method of claim 1, further comprising reconstructing the resealed multi-dimensional building model.

3. The method of claim 1, wherein the plurality of boundaries are defined automatically or manually.

4. The method of claim 3, wherein the plurality of boundaries are defined by a plurality of geo-referenced pixels.

5. The method of claim 1, wherein the architectural element includes any of:

siding row(s), door(s), brick(s), or window(s) having known architectural standard dimensions.

6. The method of claim 1, wherein the retrieved image includes two or more architectural elements.

7. The method of claim 1, wherein the dimensions include any of:

width, height, width and height, area, or a ratio of width-to-height or height-to-width.

8. The method of claim 1, wherein the retrieved images are rectified and correlated to vertices of geometric planes or polygons from the multi-dimensional model.

9. A system for resealing a multi-dimensional building model comprising:

an image database operable to retrieve an image associated with the multi-dimensional building model; and

an image processor communicatively coupled to the image database and operable to;

identify a plurality of boundaries for an architectural element located within the image;

define the plurality of boundaries for the architectural element;

measure a distance between two or more of the plurality of boundaries of the architectural element;

compare the measured distance between the two or more of the plurality of boundaries to a plurality of known similar architectural element'"'"'s standard dimensions, within an error threshold;

select, based on the comparing, actual dimensions of a known similar architectural element that is dimensionally closest to the architectural element;

determine an error in scaling of the image based on comparing the measured distance to a respective dimension on the known similar architectural element; and

rescale the multi-dimensional building model according to the error; and

a weighted decision engine communicatively coupled to the image database and the image processor predicting which architectural element or combination of architectural elements would produce an accurately resealed building model based on repeating the comparing for a plurality of different ones of the architectural elements and selecting a minimum occurrence of the error in scaling.

10. The system of claim 9, wherein the distance measured between the two or more of the plurality of boundaries of the architectural element is extrapolated based on pixel resolution.

11. The system of claim 9, wherein the resealing is implemented during construction of the multi-dimensional building model.

12. The system of claim 9, wherein the architectural element includes any of:

siding row(s), door(s), brick(s), or window(s) having known architectural standard dimensions.

13. The system of claim 9, wherein the retrieved image includes two or more architectural elements.

14. A method of determining scale for a multi-dimensional building model comprising:

a. retrieving an image for a multi-dimensional building model including two or more architectural elements;

b. identifying a plurality of boundaries for each of the two or more architectural elements within the image;

c. defining the plurality of boundaries for the each of the two or more architectural elements;

d. measuring a distance between two or more of the plurality of boundaries of the one of the two or more architectural elements;

e. comparing the measured distance between the two or more of the plurality of boundaries to a known architectural standard dimension, within an error threshold, to determine a actual measurement for each of the two or more architectural elements;

f. determining an error in scaling of the image based on the determined actual measurement; and

g. communicating the error in scaling of the each of the two or more architectural elements for accumulation in computer memory; and

h. determining, using a weighted decision engine, a final scale error based on a minimum scaling error for one of the two or more architectural elements.

15. The method of claim 14, wherein the weighted decision engine predicts, based on the determined errors in scaling, which architectural element of the two or more architectural elements would produce a most accurate resealed multi-dimensional building model.