| CPC G06V 20/176 (2022.01) [G06T 7/0002 (2013.01); G06V 10/245 (2022.01); G06V 10/26 (2022.01); G06V 10/32 (2022.01); G06V 10/82 (2022.01); G06T 2200/24 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30184 (2013.01)] | 19 Claims |

|
1. A method implemented by a roof analysis system and comprising:
preprocessing overhead images in obtained imagery data based on a request comprising a geographic location received from a user device, wherein each of the overhead images depicts a building at a different historical point in time identified in the obtained imagery data;
cropping the overhead images to a dilated and expanded representation of the footprint or the property boundary to preprocess the overhead images:
applying an artificial intelligence model to one or more input data structures into which the overhead images are converted, wherein the artificial intelligence model is trained to extract relationships between features from the input data structures indicating one or more changes in a roof of the building and generate an output;
analyzing one or more patterns in the output to determine one or more likely instances of change of the roof;
selecting a most likely one of the one or more likely instances of change of the roof that reflects the most recent change in the building's roof indicating that the building's roof was replaced; and
outputting an estimated roof age for the roof to the user device in response to the request, wherein the estimated roof age is generated based on temporal data corresponding to one or more overhead images associated with the most likely one of the one or more likely instances of change.
|