Processing system having a machine learning engine for providing a surface dimension output

Patent metadata

Field	Value
Publication number	US-12299719-B2
Application number	US-202217893995-A
Country	US
Kind code	B2
Filing date	Aug 23, 2022
Priority date	May 4, 2018
Publication date	May 13, 2025
Grant date	May 13, 2025

Abstract

Systems and apparatuses for generating object dimension outputs and predicted object outputs are provided. The system may collect an image from a mobile device. The system may analyze the image to determine whether it contains one or more standardized reference objects. Based on analysis of the image and the one or more standardized reference objects, the system may determine an object dimension output. The system may also determine a predicted object output that includes additional objects predicted to be in a room corresponding to the image. Using object dimension outputs and the predicted object output, the system may determine an estimated repair cost.

First claim

What is claimed is:

1. A computing platform, comprising:
at least one processor;
a communication interface commutatively coupled to the at least one processor; and
memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
receive at least one image;
execute an image analysis operation causing an image analysis and device control system to generate an object dimension output by at least:
determining a plurality of bounding boxes comprising the at least one image, wherein at least some of the plurality of bounding boxes have dimensions that match predetermined dimensions for a neural network;
reducing image quality of the plurality of bounding boxes;
transposing the plurality of bounding boxes on top of a black image that comprises the predetermined dimensions for the neural network; and
determining a pixel dimension for each bounding box of the plurality of bounding boxes;

causing an object prediction control platform to:
determine source data corresponding to the at least one image and a user, and
determine a predicted object output by inputting the source data into one or more machine learning models to output the predicted object output, and wherein determining the predicted object output comprises:
determining, based on a room type corresponding to the at least one image, objects predicted to be in a room,
identifying a correlation between each of the objects predicted to be in the room and the source data, and
in response to determining that a particular correlation exceeds a predetermined threshold, adding the corresponding objects predicted to be in the room to the predicted object output.

2. The computing platform of claim 1 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to:
determine a reference object in the at least one image; determine pixel dimensions of the reference object; and
determine, using predetermined actual dimensions of the reference object and the pixel dimensions of the reference object, an actual to pixel ratio for the at least one image.

3. The computing platform of claim 2 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to:
determine an object boundary corresponding to an object in the at least one image;

determine pixel dimensions corresponding to the object; and
determine, using the pixel dimensions corresponding to the object and the actual to pixel ratio for the at least one image, actual dimensions corresponding to the object.

4. The computing platform of claim 2 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to determine, using the actual to pixel ratio for the at least one image, actual surface dimensions of a surface in the at least one image.

5. The computing platform of claim 4 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to determine a material corresponding to the surface in the at least one image.

6. The computing platform of claim 4 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to determine a cause of damage to the surface in the at least one image.

7. The computing platform of claim 1 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to determine an estimated repair cost corresponding to damage shown in the at least one image by:
generating one or more commands directing an object replacement and advisor platform to determine the estimated repair cost;
sending, along with the one or more commands directing the object replacement and advisor platform to determine the estimated repair cost and to the object replacement and advisor platform, the predicted object output; and
receiving, in response to the one or more commands directing the object replacement and advisor platform to determine the estimated repair cost, the estimated repair cost.

8. The computing platform of claim 7 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to:
generate one or more commands directing the object replacement and advisor platform to determine a claim advisor output;
send, to the object replacement and advisor platform, the one or more commands directing the object replacement and advisor platform to determine the claim advisor output; and
receive, in response to the one or more commands directing the object replacement and advisor platform to determine the claim advisor output, the claim advisor output.

9. The computing platform of claim 8 , wherein the one or more commands directing the object replacement and advisor platform to determine the estimated repair cost further direct the object replacement and advisor platform to cause the objects included in the predicted object output to be added to an online shopping cart corresponding to the user.

10. The computing platform of claim 1 , wherein the memory stores additional computer-readable instructions that, when executed by the at least one processor, further cause the computing platform to generate, based on the at least one image, a room indication output comprising an indication of the room type.

11. The computing platform of claim 7 , wherein the one or more commands further comprises receiving third party source data, the third party source data comprising information that corresponds to the room type.

12. The computing platform of claim 11 , wherein the one or more machine learning models are associated with one or more machine learning datasets, the one or more machine learning datasets comprising a plurality of images corresponding to at least one of (1) one or more damage types and (2) one or more material types, and a combination of circumstances indicated by the third party source data.

13. A method comprising: receiving at least one image;
executing an image analysis operation, the image analysis operation comprising causing an image analysis and device control system to generate an object dimension output by at least: determining a plurality of bounding boxes comprising the at least one image, wherein at
least some of the plurality of the bounding boxes have dimensions that match predetermined dimensions for a neural network;
reducing image quality of the plurality of bounding boxes;
transposing the plurality of bounding boxes on top of a black image that comprises the predetermined dimensions for the neural network; and
determining a pixel dimension for each bounding box of the plurality of bounding boxes; causing an object prediction control platform to:
determine source data corresponding to the at least one image and a user, and determine a predicted object output by inputting the source data into one or more machine learning models to output the predicted object output, and wherein determining the predicted object output comprises:
determining, based on a room type corresponding to the at least one image, objects predicted to be in a room,
identifying a correlation between each of the objects predicted to be in the room and the source data,
in response to determining that a particular

Assignees

• Allstate Insurance Co

Inventors

• Zahn David M
• Daniels Andrew
• Patel Pinal
• Gilkison David L
• Genc Steven

Classifications

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/0895

  Weakly supervised learning, e.g. semi-supervised or self-supervised learning · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N20/00

  Machine learning · CPC title

• G06T7/62

  of area, perimeter, diameter or volume · CPC title