Determining localized weather by media classification

Patent metadata

Field	Value
Publication number	US-11113534-B1
Application number	US-201916403833-A
Country	US
Kind code	B1
Filing date	May 6, 2019
Priority date	May 7, 2018
Publication date	Sep 7, 2021
Grant date	Sep 7, 2021

Abstract

Systems and techniques are described for utilizing video classification capabilities for providing accurate local weather. In some implementations, the techniques include the actions of obtaining images from cameras located at a monitored property. An expected weather forecast and an actual weather condition is obtained for the monitored property. A machine-learning model is trained to classify a current weather condition for the monitored property using the images from the cameras, the expected weather forecast, and the actual weather condition. A weather condition is obtained from the trained machine-learning model that indicates a particular weather condition at the monitored property based on one or more images from a camera and the expected local weather forecast at the monitored property.

First claim

What is claimed is:

1. A computer-implemented method, comprising:
obtaining images from a camera located at a monitored property;
determining that the monitored property is located within a particular geographic region;
obtaining, for each of the images from the camera at the monitored property and based on the particular geographic region, an expected weather forecast for the geographic region at a time the image was captured, and an actual weather condition at the monitored property at the time the image was captured;
generating a training set that includes each of the images labeled with both (i) an indication of the expected weather forecast for the geographic region at the time the image was captured and (ii) an indication of the actual weather condition at the monitored property at the time the image was captured;
training a machine-learning model to classify a current weather condition for the monitored property using the training set that includes the images from the camera, the indications of expected weather forecast for the geographic region at the time the images were captured, and the actual weather condition at the monitored property at the time the images were captured;
obtaining a subsequent image from the camera and a subsequent expected local weather forecast for the monitored property at the time the subsequent image was captured;
providing the subsequent image from the camera and the subsequent expected local weather for the monitored property at the time the subsequent image was captured as inputs to the machine-learning model; and
receiving, in response to providing the subsequent image from the camera and the subsequent expected local weather for the monitored property at the time the subsequent image was captured as inputs to the machine-learning model, a weather condition from the trained machine-learning model that indicates a particular weather condition at the monitored property based on the subsequent image from the camera and the subsequent expected local weather forecast for the monitored property at the time the subsequent image was captured.

2. The computer-implemented method of claim 1 , wherein obtaining the expected weather forecast and the actual weather condition for the monitored property further comprises:
obtaining the expected weather forecast from a third party resource; and
obtaining the actual weather condition from water sensors at the monitored property.

3. The computer-implemented method of claim 1 , further comprising:
determining a device at the monitored property that exposes a portion of the monitored property to the particular weather condition; and
providing an instruction to the device that adjusts a position of the device to reduce an exposure of the portion of the monitored property to the particular weather condition.

4. The computer-implemented method of claim 3 , wherein providing the instruction to the device further comprises providing an instruction to close and lock a front door of the monitored property when the particular weather condition includes rain at the monitored property.

5. The computer-implemented method of claim 1 , further comprising:
providing the current weather condition to a client device owned by a property owner of the monitored property;
receiving a correction to the current weather condition from the client device; and
training the trained machine-learning model to generate the correction to the current weather condition using the correction to the current weather condition, the one or more images from the camera and the expected local weather forecast used to generate the current weather condition.

6. The computer-implemented method of claim 1 , comprising:
providing the trained machine-learning model to each of the cameras at the monitored property; and
receiving an additional weather condition from the trained machine-learning at each of the cameras.

7. A system comprising:
one or more computers; and
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
obtaining images from a camera located at a monitored property;
determining that the monitored property is located within a particular geographic region;
obtaining, for each of the images from the camera at the monitored property and based on the particular geographic region, an expected weather forecast for the geographic region at a time the image was captured, and an actual weather condition at the monitored property at the time the image was captured;
generating a training set that includes each of the images labeled with both (i) an indication of the expected weather forecast for the geographic region at the time the image was captured and (ii) an indication of the actual weather condition at the monitored property at the time the image was captured;
training a machine-learning model to classify a current weather condition for the monitored property using the training set that includes the images from the camera, the indications of expected weather forecast for the geographic region at the time the images were captured, and the actual weather condition at the monitored property at the time the images were captured;
obtaining a subsequent image from the camera and a subsequent expected local weather forecast for the monitored property at the time the subsequent image was captured;
providing the subsequent image from the camera and the subsequent expected local weather for the monitored property at the time the subsequent image was captured as inputs to the machine-learning model; and
receiving, in response to providing the subsequent image from the camera and the subsequent expected local weather for the monitored property at the time the subsequent image was captured as inputs to the machine-learning model, a weather condition from the trained machine-learning model that indicates a particular weather condition at the monitored property based on the subsequent image from the camera and the subsequent expected local weather forecast for the monitored property at the time the subsequent image was captured.

8. The system of claim 7 , wherein obtaining the expected weather forecast and the actual weather condition for the monitored property further comprises:
obtaining the expected weather forecast from a third party resource; and
obtaining the actual weather condition from water sensors at the monitored property.

9. The system of claim 7 , wherein the operations comprise:
determining a device at the monitored property that exposes a portion of the monitored property to the particular weather condition; and
providing an instruction to the device that adjusts a position of the device to reduce an exposure of the portion of the monitored property to the particular weather condition.

10. The system of claim 9 , wherein providing the instruction to the device further comprises providing an instruction to close and lock a front door of the monitored property when the particular weather condition includes rain at the monitored property.

11. The system of claim 7 , wherein the operations comprise:
providing the current weather condition to a client device owned by a property owner of the monitored property;
receiving a correction to the current weather condition from the client device; and
training the trained machine-learning model to generate the correction to the current weather condition using the correction to the current weather condition, the one or more images from the camera and the expected local weather forecast used to generate the current weather condition.

Assignees

• Alarm Com Inc

Inventors

• Conger Dave
• Silverman Christopher

Classifications

• G06V10/774

  Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06V10/803

  of input or preprocessed data · CPC title

• G06V20/52

  Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) · CPC title

• G06V20/41

  Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49) · CPC title

• G01W1/10 Primary

  Devices for predicting weather conditions (computers per se G06; display devices G09) · CPC title