Platform, systems, and methods for identifying characteristics and conditions of property features through imagery analysis
In an illustrative embodiment, methods and systems for automatically categorizing a condition of a property characteristic may include obtaining aerial imagery of a geographic region including the property, identifying features of the aerial imagery corresponding to the property characteristic, analyzing the features to determine a property characteristic classification, and analyzing a region of the aerial imagery including the property characteristic to determine a condition classification.
1. A computer-based method comprising:
receiving, at a computer-based system,
an aerial image of a geographic region including a property, and
an overlay shape map that follows and identifies a boundary line for the property;
identifying, based on a version of the aerial image, a location of at least one rooftop within the boundary line; and
applying two or more of a plurality of image classifiers to data corresponding to at least a portion of the aerial image, wherein each image classifier is configured to classify one or more attributes present in aerial imagery, the plurality of image classifiers comprising
a roof material classifier configured to apply a deep fully convolutional network to identifying a material of the at least one rooftop, and
a roof condition classifier configured to apply a deep fully convolutional network to identifying a condition of the at least one rooftop.
2. The computer-based method of claim 1 , wherein the plurality of image classifiers comprises an inground pool classifier configured to apply a deep fully convolutional network to identify whether an inground pool is present in the aerial image.
3. The computer-based method of claim 1 , wherein the plurality of image classifiers comprises a natural feature classifier configured to apply a deep fully convolutional network to determine a distance between trees, brush, and/or grasses in the aerial image and the at least one rooftop.
4. The computer-based method of claim 1 , further comprising classifying, with a roof type classifier, a roof type for the at least one rooftop.
5. The computer-based method of claim 1 , further comprising, prior to identifying the location of the at least one rooftop, resizing, geometrically correcting, scaling, and/or cropping the aerial image to form the version of the aerial image.
6. The computer-based method of claim 1 , wherein the plurality of image classifiers comprises a skylight feature classifier configured and trained to apply a deep fully convolutional network to identify whether one or more skylights are present on rooftops in the aerial imagery.
7. The computer-based method of claim 1 , wherein identifying the boundary line comprises generating a composite image from the overlay shape map and the aerial image.
8. A computer-based system comprising:
a network interface connection for receiving a plurality of aerial images from a data source over a network, wherein each image has a first visual indicator on the aerial image that follows and identifies a boundary line for a respective property; and
a plurality of image classifiers, wherein each image classifier is configured to classify one or more attributes in the aerial images of properties, the plurality of image classifiers comprising
a rooftop type classifier configured and trained to apply a deep fully convolutional network to identify a rooftop type of rooftops present in aerial imagery,
a roof condition classifier configured and trained to apply a deep fully convolutional network to identify a condition of rooftops present in aerial imagery, and
an additional classifier configured and trained to identify a presence of an additional property feature in aerial imagery.
9. A computer-based method comprising:
receiving, at a computer-based system,
an aerial image of a geographic region including a property, and
an overlay shape map that follows and identifies a boundary line for the property;
identifying one or more rooftops within the boundary line, based on the aerial image;
using a rooftop deep fully convolutional network configured and trained to identify at least one of
i) a given type of a plurality of rooftop types in aerial imagery, or
ii) a given material of a plurality of roofing materials in aerial imagery,
to identify, based on the aerial image, one or more of
a) a type of at least one rooftop of the one or more rooftops, or
b) a material of the at least one rooftop;
using a rooftop condition deep fully convolutional network configured and trained to identify a rooftop condition in aerial imagery to identify a condition of the at least one rooftop; and
using an additional deep fully convolutional network configured and trained to identify a presence of an additional property feature in aerial imagery to identify, based on the aerial image, the presence of the additional property feature within the boundary line of the property.
10. The computer-based method of claim 9 , wherein the additional property feature is an inground pool.
11. The computer-based method of claim 9 , wherein:
the additional property feature is a natural property feature; and
the additional deep fully convolutional network is configured and trained to determine a distance between trees, brush, and/or grasses in the aerial imagery and a given rooftop in the aerial imagery.
12. The computer-based method of claim 9 , further comprising, prior to identifying the one or more rooftops, resizing, geometrically correcting, scaling, and/or cropping the aerial image to form an input for identifying rooftops within the boundary line.
13. The computer-based system of claim 9 , wherein the additional property feature is a skylight.