Predicting wildfires on the basis of biophysical indicators and spatiotemporal properties using a convolutional neural network

Patent metadata

Field	Value
Publication number	US-10990874-B2
Application number	US-201715601704-A
Country	US
Kind code	B2
Filing date	May 22, 2017
Priority date	May 22, 2017
Publication date	Apr 27, 2021
Grant date	Apr 27, 2021

Abstract

Systems, software, and computer implemented methods can be used to predict wildfires based on biophysical and spatiotemporal data. A method includes receiving a request for a wildfire prediction for at least one geographical area. At least one biophysical indicator is identified. Each biophysical indicator provides biophysical data for the at least one geographical area. The at least one biophysical indicator is provided to a convolutional neural network (CNN). The CNN is trained using ground truth data that includes historical information about wildfires for at least one ground truth geographical area. The CNN is used to generate at least one prediction for wildfire risk for the at least one geographical area. The at least one prediction is provided responsive to the request.

First claim

What is claimed is:

1. A computer-implemented method, the method comprising:
receiving a request for a wildfire prediction for at least one geographical area;
retrieving image data for the at least one geographical area, wherein the image data includes overhead images of the at least one geographical area for at least one time point;
generating, from the image data, at least one biophysical indicator, each biophysical indicator providing image-based biophysical data for the at least one geographical area;
providing the at least one biophysical indicator to a convolutional neural network (CNN), the CNN trained using ground truth data that includes historical information about wildfires for at least one ground truth geographical area;
using the CNN to generate at least one prediction for wildfire risk for the at least one geographical area;
evaluating the at least one prediction to generate at least one corresponding prediction evaluation by comparing the at least one prediction to the ground truth data for the at least one geographical area using categorical cross-entropy as an objective function; and
providing the at least one prediction and the at least one corresponding prediction evaluation responsive to the request.

2. The method of claim 1 , wherein the at least one geographical area is different than the at least one ground truth geographical area.

3. The method of claim 1 , wherein the at least one biophysical indicator includes at least one of a vegetation index, a dry matter index, a leaf area index, and a fraction of absorbed photosynthetically active radiation index.

4. The method of claim 1 , wherein the CNN includes an input layer, at least one rectified convolutional layer, at least one fully connected layer, and an output layer.

5. The method of claim 1 , wherein the at least one prediction comprises a first output neuron and a second output neuron, the first output neuron and the second output neuron indicating a probability of a wildfire and a probability of no wildfire for the geographic area for an upcoming time period, respectively.

6. The method of claim 1 , wherein providing the at least one prediction comprises presenting prediction information on a map that displays the at least one geographic area.

7. A system, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor, the memory storing instructions which, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
receiving a request for a wildfire prediction for at least one geographical area;
retrieving image data for the at least one geographical area, wherein the image data includes overhead images of the at least one geographical area for at least one time point;
generating, from the image data, at least one biophysical indicator, each biophysical indicator providing image-based biophysical data for the at least one geographical area;
providing the at least one biophysical indicator to a convolutional neural network (CNN), the CNN trained using ground truth data that includes historical information about wildfires for at least one ground truth geographical area;
using the CNN to generate at least one prediction for wildfire risk for the at least one geographical area;
evaluating the at least one prediction to generate at least one corresponding prediction evaluation by comparing the at least one prediction to the ground truth data for the at least one geographical area using categorical cross-entropy as an objective function; and
providing the at least one prediction and the at least one corresponding prediction evaluation responsive to the request.

8. The system of claim 7 , wherein the at least one geographical area is different than the at least one ground truth geographical area.

9. The system of claim 7 , wherein the at least one biophysical indicator includes at least one of a vegetation index, a dry matter index, a leaf area index, and a fraction of absorbed photosynthetically active radiation index.

10. The system of claim 7 , wherein the CNN includes an input layer, at least one rectified convolutional layer, at least one fully connected layer, and an output layer.

11. The system of claim 7 , wherein the at least one prediction comprises a first output neuron and a second output neuron, the first output neuron and the second output neuron indicating a probability of a wildfire and a probability of no wildfire for the geographic area for an upcoming time period, respectively.

12. The system of claim 7 , wherein providing the at least one prediction comprises presenting prediction information on a map that displays the at least one geographic area.

13. One or more computer-readable media storing instructions which, when executed by at least one processor, cause the at least one processor to perform operations comprising:
receiving a request for a wildfire prediction for at least one geographical area;
retrieving image data for the at least one geographical area, wherein the image data includes overhead images of the at least one geographical area for at least one time point;
generating, from the image data, at least one biophysical indicator, each biophysical indicator providing image-based biophysical data for the at least one geographical area;
providing the at least one biophysical indicator to a convolutional neural network (CNN), the CNN trained using ground truth data that includes historical information about wildfires for at least one ground truth geographical area;
using the CNN to generate at least one prediction for wildfire risk for the at least one geographical area;
evaluating the at least one prediction to generate at least one corresponding prediction evaluation by comparing the at least one prediction to the ground truth data for the at least one geographical area using categorical cross-entropy as an objective function; and
providing the at least one prediction and the at least one corresponding prediction evaluation responsive to the request.

14. The computer-readable media of claim 13 , wherein the at least one geographical area is different than the at least one ground truth geographical area.

15. The computer-readable media of claim 13 , wherein the at least one biophysical indicator includes at least one of a vegetation index, a dry matter index, a leaf area index, and a fraction of absorbed photosynthetically active radiation index.

16. The computer-readable media of claim 13 , wherein the CNN includes an input layer, at least one rectified convolutional layer, at least one fully connected layer, and an output layer.

17. The computer-readable media of claim 13 , wherein the at least one prediction comprises a first output neuron and a second output neuron, the first output neuron and the second output neuron indicating a probability of a wildfire and a probability of no wildfire for the geographic area for an upcoming time period, respectively.

18. The computer-readable media of claim 13 , wherein providing the at least one prediction comprises presenting prediction information on a map that displays the at least one geographic area.

Assignees

• Sap Se

Inventors

• Tschernezki Vadim
• Blum Oliver
• Gildhoff Hinnerk
• Wyss Michèle
• Deiseroth Bjoern
• Svojanovsky Wenzel

Classifications

• G06N3/084 Primary

  Backpropagation, e.g. using gradient descent · CPC title

• G06N3/048

  Activation functions · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06N3/09 Primary

  Supervised learning · CPC title