Fire urgency estimator in geosynchronous orbit (FUEGO)

Patent metadata

Field	Value
Publication number	US-9792500-B2
Application number	US-201514927137-A
Country	US
Kind code	B2
Filing date	Oct 29, 2015
Priority date	Apr 30, 2013
Publication date	Oct 17, 2017
Grant date	Oct 17, 2017

Abstract

A fire detector is disclosed that successively images a particular area from geosynchronous Earth orbit satellite to attain very good signal-to-noise ratios against Poisson fluctuations within one second. Differences between such images allow for the automatic detection of small fires greater than 12 square meters. Imaging typically takes place in transparent bands of the infrared spectrum, thereby rendering smoke from the fire and light clouds somewhat transparent. Several algorithms are disclosed that can help reduce false fire alarms, and their efficiencies are shown. Early fire detection and response would be of great value in the United States and other nations, as wild land fires destroy property and lives and contribute around five percent of the US global carbon dioxide contribution. Such apparatus would incorporate modern imaging detectors, software, and algorithms able to detect heat from early and small fires, and yield detection times on a scale of minutes.

First claim

What is claimed is:

1. An apparatus for fire urgency estimation, the apparatus comprising:
(a) an imager configured for capturing images, wherein said imager is configured for deployment within a satellite configured for orbiting Earth;
(b) a processor configured for processing said images; and
(c) programming executable on said processor for detecting fire locations across extensive geographic regions while suppressing false alarms by performing steps comprising:
(i) classifying trends in response to temporal subtraction of images received by said imager over a time of seconds to minutes;
(ii) dithering of images to improve resolution;
(iii) adjusting detection thresholds in a time-sensitive autonomous manner to reduce false alarm rates despite constantly varying conditions; and
(iv) generating calibrated decisions regarding fire urgency in response to steps (i) through (iii).

2. A method of fire urgency estimation for detecting fire locations across extensive geographic regions while suppressing false alarms, the method comprising:
(a) classifying trends in response to temporal subtraction of images received by an imager over a time of seconds to minutes, wherein said images are captured by the imager integrated within a satellite orbiting Earth;
(b) dithering of said images to improve resolution;
(c) adjusting detection thresholds in a time-sensitive autonomous manner to reduce false alarm rates despite varying environmental conditions; and
(d) generating calibrated decisions regarding fire urgency in response to steps (a) through (c).

3. A fire urgency estimator, comprising:
(a) a platform comprising an imager configured to capture a plurality of Earth geographic images;
(b) programming executable on the platform capable of performing steps comprising:
(i) subtracting one Earth geographic image from another Earth geographic image;
(A) wherein the two images are separated by a sampling interval to form a radiance difference image; and
(B) wherein the two images view at least a portion of the same Earth geography;

(ii) detecting candidate fires in the radiance difference image within the same Earth geography;
(iii) excluding from the candidate fires those that fail one or more tests to form a set of non-excluded candidate fires;
(iv) evaluating an importance cost of the non-excluded candidate fires; and
(v) reporting the non-excluded candidate fires above a fire importance cost reporting threshold.

4. The fire urgency estimator of claim 3 , wherein the platform comprises either:
(a) a satellite in Earth orbit; or
(b) a manned aerial vehicle; or
(c) an unmanned aerial vehicle (UAV).

5. The fire urgency estimator of claim 3 , wherein the platform further comprises:
(a) a processor physically disposed proximally to the platform; or
(b) a ground based processor; or
(c) a distributed processor comprising:
(i) a processor physically disposed within or proximal to the platform; and
(ii) a ground based processor in communication with the processor physically disposed proximally to the platform;
(iii) wherein processing takes place in both the platform processor and in the ground based processor; and
(iv) wherein the platform processor and the ground based processor are in at least periodic communication.

6. The fire urgency estimator of claim 3 , wherein the imager comprises either a monospectral or multispectral imager.

7. The fire urgency estimator of claim 6 , wherein the imager comprises an ability to produce dithered geographic images.

8. The fire urgency estimator of claim 3 , wherein said detecting candidate fires comprises:
(a) forming a histogram of the radiance difference image; and
(b) calculating one or more properties from the histogram, wherein the one or more properties are selected from a group of properties consisting of: a mean, a maximum value, a standard deviation (σ), and a full width half maximum (FWHM) value.

9. The fire urgency estimator of claim 8 , wherein the step of detecting candidate fires further comprises detecting one or more fire candidates from the histogram.

10. The fire urgency estimator of claim 3 , further comprising:
(a) a fire urgency receiver;
(b) wherein the receiver is configured to receive from the platform at least one of the reported non-excluded candidate fires above the fire importance cost reporting threshold; and
(c) wherein the receiver is configured to receive from the platform a geographic location corresponding to at least one reported non-excluded candidate fire.

11. A fire urgency estimator, comprising:
(a) a computer processor; and
(b) programming in a non-transitory computer readable medium and executable on the computer processor for performing steps comprising:
(i) capturing a plurality of Earth geographic images by using an imager;
(ii) subtracting one Earth geographic image from another Earth geographic image;
(A) wherein the two images are separated by a sampling interval to form a radiance difference image; and
(B) wherein the two images view at least a portion of the same Earth geography;

(iii) detecting candidate fires in the radiance difference image within the same Earth geography;
(iv) excluding from the candidate fires those that fail one or more tests to form a set of non-excluded candidate fires;
(v) evaluating an importance cost of the non-excluded candidate fires; and
(vi) reporting the non-excluded candidate fires above a fire importance cost reporting threshold.

12. A computer readable non-transitory storage medium comprising software configured to be run by at least one computer processor for performing steps comprising:
(a) providing a radiance difference image by performing steps comprising:
(i) acquiring at least two Earth geographic images;
(ii) subtracting one Earth geographic image from another Earth geographic image to form the subtracted difference image;
(iii) wherein the two images are separated by a sampling interval;
(iv) wherein the two images view at least a portion of the same Earth geography; and
(v) wherein the radiance difference image corresponds the portion of the same Earth geography;

(b) detecting candidate fires in the radiance difference image within the same Earth geography;
(c) excluding from the candidate fires those that fail one or more tests to form a set of non-excluded candidate fires;
(d) evaluating an importance cost of the non-excluded candidate fires; and
(e) reporting the non-excluded candidate fires above a fire importance cost reporting threshold.

13. A fire urgency estimator, comprising:
(a) a platform;
(b) the platform comprising an imager configured to capture a plurality of Earth geographic images
(c) a processor disposed on the platform; and
(d) means for reporting from the Earth geographic images one or more non-excluded candidate fires above a fire importance cost reporting threshold by processing the plurality of Earth geographic images on the processor;
wherein the means for reporting comprises programming executable on the processor capable of performing steps comprising:
(1) providing a radiance difference image by performing steps comprising:
(i) acquiring at least two Earth geographic images;
(ii) subtracting one Earth geographic image from another Earth geographic image to form the subtracted difference image;
(iii) wherein the two images are separated by a sampling interval;
(iv) wherein the two images view at least a portion of the same Earth geography; and
(v) wherein the radiance difference image corresponds to the portion of the same Earth geography;

(2) detecting candidate fires in the radiance difference image withi

Assignees

• Univ California

Inventors

• Pennypacker Carlton R
• Tripp Robert D
• Stephens Scott L
• Kelly Nina M
• Jakubowski Marek K
• Lampton Mike

Classifications

• G08B17/005 Primary

  for forest fires, e.g. detecting fires spread over a large or outdoors area (fire fighting forest fires A62C3/02) · CPC title

• H04N23/11

  for generating image signals from visible and infrared light wavelengths · CPC title

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title

• G06T7/97

  Determining parameters from multiple pictures (depth or shape recovery from multiple images G06T7/55; stereo camera calibration G06T7/85) · CPC title