Optimising site selection for wildfire detection cameras

January 9, 2020

Placing wildfire surveillance towers at the right sites improves visibility and reduces costs.

By Andries Heyns
Since March 2017, a tower site-selection optimisation framework for the South African-developed ForestWatch camera-based wildfire detection system (by EnviroVision Solutions, based in Durban) has been under development with researchers from the University of Pretoria and the University of Alabama.

Their research has now been published in The International Journal of Wildland Fire, titled Optimisation of tower site locations for camera-based wildfire detection systems.

The optimisation of tower site locations is required to maximise the visibility cover achieved over valuable areas belonging to subscribed clients – typically from the forestry sector. Reducing the number of towers required to achieve satisfactory cover results in reduced installation costs and reduced future expenses on maintenance and upgrades.

Historically, these systems have been planned by foresters and locals with intimate knowledge of the terrain rather than by computational optimisation tools. When entering vast new territories, however, such knowledge and expertise may not be available to system planners, which is why the optimisation framework – which uses cutting-edge geographical analysis techniques and evolution-inspired system layout optimisation algorithms – is so valuable.

The site-selection framework has matured into a fully-functioning optimisation tool, culminating in its recent implementation in the selection of a four-camera wildfire detection system in the Southern Cape.

The optimisation-determined layouts drastically outperformed the coverage achieved by sites proposed by technical experts with years of experience in site selection, resulting in the selection of a more effective tower layout than initially planned.

Furthermore, the numerous superior layouts were obtained by the framework within a number of days (including data collection, processing, optimisation and analysis) – drastically faster than the weeks of planning that were required by the experts to determine their site locations. ForestWatch systems are in operation in South Africa, Spain, Australia, Canada and the United States, and the framework is earmarked for implementation in other system planning problems.

Andries Heyns – University of Alabama, University of Pretoria
Warren du Plessis – University of Pretoria
Mike Kosch – EnviroVision Solutions
Gavin Hough – EnviroVision Solutions

Publication – www.publish.csiro.au/WF/WF18196
EnviroVision solutions – evsolutions.biz

*First published in SA Forestry magazine, Nov 2019

Related article: Predicting the movement of wildfires and prescribed burns

Notify of
Inline Feedbacks
View all comments

Related Stories

Warning: Attempt to read property "taxonomy" on string in /usr/www/users/saforsqgqs/wp-content/plugins/pjs-pesto/src/classes/StoryBlocks.php on line 172 Warning: Attempt to read property "term_id" on string in /usr/www/users/saforsqgqs/wp-content/plugins/pjs-pesto/src/classes/StoryBlocks.php on line 174
linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram