# US 2022/0018989 A1 · Forecasting environmental occlusion events

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0018989 A1
- Document ID: US20220018989A1
- Status: Published application
- Kind: A1
- Application number: 16/930,195
- Filed: 2020-07-15
- Published: 2022-01-20
- Earliest claimed: 2020-07-15
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Jitendra Singh; Mukul Tewari; Seema Nagar; Kuntal Dey
- Web page: https://patentgrep.com/patent/US20220018989A1

## CPC

- G01W 1/10: Devices for predicting weather conditions (computers per se G06; display devices G09)
- G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- G06F 18/2321: using statistics or function optimisation, e.g. modelling of probability density functions
- G06F 18/241: relating to the classification model, e.g. parametric or non-parametric approaches
- G06N 20/00: Machine learning
- G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI]
- G06V 20/13: Satellite images
- G01W: METEOROLOGY (radar, sonar, lidar or analogous systems, designed for meteorological use G01S13/95, G01S15/88, G01S17/95)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

In an approach for forecasting environmental occlusion events, a processor receives a spatio-temporal zone of interest. A processor collects data associated with the spatio-temporal zone of interest. A processor builds a machine-learning model using the data. A processor generates an occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and enriched mathematical operators. A processor outputs the occlusion probability map.

## Claims

### Claim 1

A computer-implemented method for forecasting environmental occlusion events, the computer-implemented method comprising:
receiving, by one or more processors, a spatio-temporal zone of interest;
collecting, by the one or more processors, data associated with the spatio-temporal zone of interest;
building, by the one or more processors, a machine-learning regression model using the data;
generating, by the one or more processors, an occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and enriched mathematical operators; and
outputting, by the one or more processors, the occlusion probability map.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein receiving the spatio-temporal zone of interest further comprises:
receiving, by the one or more processors, the spatio-temporal zone of interest from a user computing device in which a user through a user interface selected the spatio temporal zone of interest using a spatial polygon or a bounding box on a geographical map, wherein the spatio-temporal zone of interest consists of a geo-spatial location and temporal period.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the data comprises historical scene classification maps from remote sensing data for the spatio-temporal zone of interest and historical weather and environmental forecasts for the spatio-temporal zone of interest.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the machine-learning regression model is configured to determine an occlusion type, occlusion intensity, and occlusion probability for the spatio-temporal zone of interest.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein building the machine-learning regression model using the data comprises:
creating, by the one or more processors, a spatial segment at a smallest spatial resolution to be supported by the machine-learning regression model;
creating, by the one or more processors, a temporal unit at a smallest time segment unit to be supported by the machine-learning regression model; and
for each spatial segment and each temporal unit, training, by the one or more processors, the machine-learning regression model on the data.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the enriched mathematical operators are configured to aggregate or disaggregate occlusion probabilities produced by the machine-learning regression model based on a spatial segment size and time segment size used by the machine-learning regression model compared to a granularity of a geo-spatial location and a temporal period of the spatio-temporal zone of interest.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein generating the occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and the enriched mathematical operators comprises:
shading, by the one or more processors, the occlusion probability map using dark to light coloring depending on an intensity or time duration of a predicted occlusion; and
clustering, by the one or more processors, different occlusion types and occlusion intensities for the spatio-temporal zone of interest on the occlusion probability map.

### Claim 8

A computer program product for forecasting environmental occlusion events, the computer program product comprising:
one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the program instructions comprising:
program instructions to receive a spatio-temporal zone of interest;
program instructions to collect data associated with the spatio-temporal zone of interest;
program instructions to build a machine-learning regression model using the data;
program instructions to generate an occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and enriched mathematical operators; and
program instructions to output the occlusion probability map.

### Claim 9 (depends on 8)

The computer program product of claim 8, wherein the program instructions to receive the spatio-temporal zone of interest further comprise:
program instructions to receive the spatio-temporal zone of interest from a user computing device in which a user through a user interface selected the spatio temporal zone of interest using a spatial polygon or a bounding box on a geographical map, wherein the spatio-temporal zone of interest consists of a geo-spatial location and temporal period.

### Claim 10 (depends on 8)

The computer program product of claim 8, wherein the data comprises historical scene classification maps from remote sensing data for the spatio-temporal zone of interest and historical weather and environmental forecasts for the spatio-temporal zone of interest.

### Claim 11 (depends on 8)

The computer program product of claim 8, wherein the machine-learning regression model is configured to determine an occlusion type, occlusion intensity, and occlusion probability for the spatio-temporal zone of interest.

### Claim 12 (depends on 8)

The computer program product of claim 8, wherein the program instructions to build the machine-learning regression model using the data comprise:
program instructions to create a spatial segment at a smallest spatial resolution to be supported by the machine-learning regression model;
program instructions to create a temporal unit at a smallest time segment unit to be supported by the machine-learning regression model; and
for each spatial segment and each temporal unit, program instructions to train the machine-learning regression model on the data.

### Claim 13 (depends on 8)

The computer program product of claim 8, wherein the enriched mathematical operators are configured to aggregate or disaggregate occlusion probabilities produced by the machine-learning regression model based on a spatial segment size and time segment size used by the machine-learning regression model compared to a granularity of a geo-spatial location and a temporal period of the spatio-temporal zone of interest.

### Claim 14 (depends on 8)

The computer program product of claim 8, wherein the program instructions to generate the occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and the enriched mathematical operators comprise:
program instructions to shade the occlusion probability map using dark to light coloring depending on an intensity or time duration of a predicted occlusion; and
program instructions to cluster different occlusion types and occlusion intensities for the spatio-temporal zone of interest on the occlusion probability map.

### Claim 15

A computer system for forecasting environmental occlusion events, the computer system comprising:
one or more computer processors;
one or more computer readable storage media;
program instructions stored on the computer readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to receive a spatio-temporal zone of interest;
program instructions to collect data associated with the spatio-temporal zone of interest;
program instructions to build a machine-learning regression model using the data;
program instructions to generate an occlusion probability map for the spatio-temporal zone of interest based on the machine-learning model and enriched mathematical operators; and
program instructions to output the occlusion probability map.

### Claim 16 (depends on 15)

The computer system of claim 15, wherein the program instructions to receive the spatio-temporal zone of interest further comprise:
program instructions to receive the spatio-temporal zone of interest from a user computing device in which a user through a user interface selected the spatio temporal zone of interest using a spatial polygon or a bounding box on a geographical map, wherein the spatio-temporal zone of interest consists of a geo-spatial location and temporal period.

### Claim 17 (depends on 15)

The computer system of claim 15, wherein the data comprises historical scene classification maps from remote sensing data for the spatio-temporal zone of interest and historical weather and environmental forecasts for the spatio-temporal zone of interest.

### Claim 18 (depends on 15)

The computer system of claim 15, wherein the machine-learning regression model is configured to determine an occlusion type, occlusion intensity, and occlusion probability for the spatio-temporal zone of interest.

### Claim 19 (depends on 15)

The computer system of claim 15, wherein the program instructions to build the machine-learning regression model using the data comprise:
program instructions to create a spatial segment at a smallest spatial resolution to be supported by the machine-learning regression model;
program instructions to create a temporal unit at a smallest time segment unit to be supported by the machine-learning regression model; and
for each spatial segment and each temporal unit, program instructions to train the machine-learning regression model on the data.

### Claim 20 (depends on 15)

The computer system of claim 15, wherein the enriched mathematical operators are configured to aggregate or disaggregate occlusion probabilities produced by the machine-learning regression model based on a spatial segment size and time segment size used by the machine-learning regression model compared to a granularity of a geo-spatial location and a temporal period of the spatio-temporal zone of interest.
