MDPI - Publisher of Open Access Journals

15 pages, 4123 KiB

Open AccessLetter

Preliminary Results from a Wildfire Detection System Using Deep Learning on Remote Camera Images

by Kinshuk Govil, Morgan L. Welch, J. Timothy Ball and Carlton R. Pennypacker

Remote Sens. 2020, 12(1), 166; https://doi.org/10.3390/rs12010166 - 2 Jan 2020

Cited by 92 | Viewed by 14557

Pioneering networks of cameras that can search for wildland fire signatures have been in development for some years (High Performance Wireless Research & Education Network—HPWREN cameras and the ALERT Wildfire camera). While these cameras have proven their worth in monitoring fires reported by other means, we have developed a functioning prototype system that can detect smoke from fires usually within 15 min of ignition, while averaging less than one false positive per day per camera. This smoke detection system relies on machine learning-based image recognition software and a cloud-based work-flow capable of scanning hundreds of cameras every minute. The system is operating around the clock in Southern California and has already detected some fires earlier than the current best methods—people calling emergency agencies or satellite detection from the Geostationary Operational Environmental Satellite (GOES) satellites. This system is already better than some commercial systems and there are still many unexplored methods to further improve accuracy. Ground-based cameras are not going to be able to detect every wildfire, and so we are building a system that combines the best of terrestrial camera-based detection with the best approaches to satellite-based detection. Full article

(This article belongs to the Special Issue Fire Remote Sensing: Capabilities, Innovations, Opportunities and Challenges)

► Show Figures

Graphical abstract

20 pages, 3148 KiB

Open AccessArticle

FUEGO — Fire Urgency Estimator in Geosynchronous Orbit — A Proposed Early-Warning Fire Detection System

by Carlton R. Pennypacker, Marek K. Jakubowski, Maggi Kelly, Michael Lampton, Christopher Schmidt, Scott Stephens and Robert Tripp

Remote Sens. 2013, 5(10), 5173-5192; https://doi.org/10.3390/rs5105173 - 17 Oct 2013

Cited by 18 | Viewed by 24986

Abstract

Current and planned wildfire detection systems are impressive but lack both sensitivity and rapid response times. A small telescope with modern detectors and significant computing capacity in geosynchronous orbit can detect small (12 m²) fires on the surface of the earth, cover most of the western United States (under conditions of moderately clear skies) every few minutes or so, and attain very good signal-to-noise ratio against Poisson fluctuations in a second. Hence, these favorable statistical significances have initiated a study of how such a satellite could operate and reject the large number of expected systematic false alarms from a number of sources. Here we present both studies of the backgrounds in Geostationary Operational Environmental Satellites (GOES) 15 data and studies that probe the sensitivity of a fire detection satellite in geosynchronous orbit. We suggest a number of algorithms that can help reduce false alarms, and show efficacy on a few. Early detection and response would be of true value in the United States and other nations, as wildland fires continue to severely stress resource managers, policy makers, and the public, particularly in the western US. Here, we propose the framework for a geosynchronous satellite with modern imaging detectors, software, and algorithms able to detect heat from early and small fires, and yield minute-scale detection times. Full article

► Show Figures

Search Results (2)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI

Saved Queries

Search Filter Reset All

Years

Feature Papers

Subjects

Journals

Article Types

Countries / Regions

Search Results (2)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI