System and method for home Façade features in a geographic region
A method including receiving a first set of one or more street level images of houses into a Machine Learned Model trained with a second set of street level images with one or more exterior features of the houses labeled, identifying the one or more exterior features in the first set of street level images by way of the Machine Learned Model, and quantifying and outputting the counts and/or two-dimensional areas for each of the identified exterior features in the first set of one or more street level images is described. Non-transitory, computer-readable storage media having instructions for executing the method steps by one or more processors as well as computer or computer systems capable of performing the method steps are also described.
1. A method, comprising:
providing a street address and the aerial outline of a house from a mapping application;
converting the street address to a geographic location comprising a latitude and longitude;
retrieving a street level image of the house from the geographic location;
cropping the street level image of the house by:
identifying corners of the house in the street level image by:
determining the angular extent of the house from the nearest point on the street from the aerial outline; and
translating distance and angles from the determining step to pixel values in the street level image; and
calculating portions of the street level image in excess of pixel values representing the house; and
cropping the portions from the street level image, thereby removing at least a portion of neighboring houses from the image; and
storing the cropped images.
2. The method of claim 1 , wherein the street address was retrieved within a geographical area of interest from a storage of street addresses, the geographical area of interest inputted on a graphical user interface.
3. The method of claim 1 , further comprising:
providing the stored cropped images to a Machine Learned Model trained with a training set of street level images of houses with one or more exterior features labeled; and
identifying the home exterior features of interest in the cropped images by way of the Machine Learned Model.
4. The method of claim 3 , further comprising:
processing output of the Machine Learned Model by converting pixels of the identified exterior features into area dimensions; and
quantifying two-dimensional areas of the exterior features in one or more of the cropped images.
5. The method of claim 3 , further comprising quantifying a number of identified exterior features in one or more of the cropped images.
6. A method comprising:
receiving output from a Machine Learned Model trained to identify exterior features from houses in a geographic area of interest, the output comprising one or more exterior features identified;
processing the output by:
converting pixel coordinates in the exterior features to distance measurement units;
determining the area of the exterior features from the distance measurements; and
quantifying the number of exterior features identified.
7. The method of claim 6 , further comprising formatting and standardizing the processed output for input into a geographic information system application.
8. The method of claim 6 , wherein output is text having information comprising:
a unique alphanumeric identifier;
street address;
geographic location;
date of image acquisition; and
image URL.
9. The method of claim 6 , wherein the outputting comprises information representing characteristics of each image in the first set of one or more street level images, wherein the information further comprises counts of individual home exterior features identified in the image.
10. The method of claim 6 , wherein the outputting comprises information representing characteristics of each image in the first set of one or more street level images, wherein the information further comprises calculated areas of individual home exterior features of the image.
11. The method of claim 6 , wherein the Machine Learned Model is trained with a training set of street level images of houses with one or more exterior features labeled.
12. The method of claim 11 , wherein the Machine Learned Model is provided with cropped street level images of houses as input and identifies home exterior features of interest in the cropped images.
13. The method of claim 12 , wherein the street level images are cropped by:
identifying corners of the house in the street level images by:
determining the angular extent of the house from the nearest point on the street from the aerial outline;
translating distance and angles from the determining step to pixel values in the street level image;
calculating portions of the street level images in excess of pixel values representing the house; and
cropping the portions from the street level images, thereby removing at least a portion of neighboring houses from the image.
14. The method of claim 13 , wherein the street level images were retrieved as raw street level images corresponding to addresses with a geographic area of interest, the geographic area of interest received as input with home exterior features of interest on a graphical user interface.
15. The method of claim 14 , wherein the addresses of all houses within the geographic area of interest were retrieved after input of the geographic area of interest on the graphical user interface.