US 11,532,093 B2 · First floor height estimation from optical images

Abstract

Systems and methods of automating the generation of an estimate of the first floor elevation of a building, using a Digital Terrain Model (DTM) and an image of the front of the building. First, machine learning (ML) algorithms are used to estimate the height of the first floor of the building above ground from the image. The estimated height is added to a DTM height of the geographical area corresponding to the location of the building to provide an estimate of the FFE. The disclosed embodiments can be applied in estimating the first floor height of any kind of building. For example, single-family homes, townhomes, high rise buildings, stores, malls, shops, movie theaters, or any other building. Additionally, the disclosed embodiments are camera-agnostic and can be applied on any type of image collected by any type of camera.

Claims20

Dependent claims

A computing system for automating estimation of a first floor height of an architectural structure the comprising:

at least one aerial sensor configured to collect one or more terrain characteristics of a geographical area corresponding to a location of the structure;
at least one memory device storing instructions for causing at least one processor to:
receive an image of the structure;
segment the image of the structure to generate a segmented image comprising data relating to a plurality of physical layout elements identified from processing the image of the structure;
determine, using the segmented image of the structure, a first pixel location and a second pixel location corresponding respectively to a first floor position and a grade position of the structure;
and
generate an estimate of a first floor height of the structure, based at least in part on a pixel height differential computed from the first pixel location and the second pixel location, wherein the estimate of the first floor height corresponds to a true, first-floor height of the structure.

4 more elements

A method for automating estimation of a first floor height of an architectural structure comprising:

receiving an image of a structure;
segmenting the image of the structure to generate a segmented image comprising data relating to a plurality of layout elements identified from processing the image of the structure;
determining, using the segmented image of the structure, a first pixel location and a second pixel location corresponding respectively to a first floor position and a grade position of the structure;
and
generating an estimate of a first floor height of the structure, based at least in part on a pixel height differential computed from the first pixel location and the second pixel location, wherein the estimate of the first floor height corresponds to a true, first-floor height of the structure.

2 more elements

A non-transitory computer-readable storage medium storing instructions configured to cause at least one computing device to perform a method for automating estimation of a first floor height of an architectural structure comprising:

receiving an image of a structure;
segmenting the image of the structure to generate a segmented image comprising a plurality of physical layout elements identified from processing the image of the structure;
determining, using the segmented image of the structure, a first pixel location and a second pixel location corresponding respectively to a first floor position and a grade position of the structure;
and
generating an estimate of a first floor height of the structure, based at least in part on a pixel height differential computed from the first pixel location and the second pixel location, wherein the estimate of the first floor height corresponds to a true, first-floor height of the structure.

2 more elements

Dependencies are inferred from the claim text.

First floor height estimation from optical images

Life of this patent

First floor height estimation from optical images

Life of this patent

Abstract

Claims20