Status: Completed
Start date: 1 March 2018
Completion date: 30 May 2019
Project code: CT-4566
Species: Wild dogs
Wild dogs are a major problem for livestock producers across Australia. Often, wild dog control has been reactive and expensive, with landholders and contractors effectively forced to ‘chase’ dogs after livestock have been maimed and killed. In chronic cases this can go on for weeks, months or even years, taking a heavy toll on enterprises, families and communities. Tracking studies show that wild dogs may be present on farms for days or weeks before losses occur.
Wild Dog Alert is a system that firmly places livestock producers and other land managers on the front foot to manage wild dogs. The prototype identifies wild dogs through a series of sophisticated facial recognition algorithms.
By combining automated recognition of camera trap images with real-time messaging, the Wild Dog Alert system notifies producers that wild dogs have been detected at a camera trap on their farm before attacks occur to enable producers to act early. This gives farmers a ‘first strike’ capability in their fight against wild dogs, so they can be proactive and put in place immediate and targeted management strategies to avoid stock losses.
This project:
The project team developed and field tested four outputs that can provide land managers with accurate real-time evidence that wild dogs are present on their land. The alert system uses images gathered via remote camera traps and transmitted to the cloud for processing. The team developed the algorithms that, once trained, identify wild dogs from these images. Once detected, the system then sends an alert to the land manager advising the presence of the wild dogs and the location at which they were detected.
Several automated detection systems were developed to allow users to tailor the system to their needs and the availability, or lack of, cellular mobile services across their property.
Development and testing of these products is complete, and the core project is complete. The project has entered the commercialisation phase with the State of New South Wales acting through the Department of Primary Industries within the Department of Regional NSW being nominated as the Designated Partner to take the product forward for commercialisation. The Centre and the project’s Commercialisation Governance Committee will monitor the progress of that process.
Falzon G, Lawson C, Cheung K-W, Vernes K, Ballard G, Fleming P, Milne H, Mather-Zardain AT and Meek PD (2019) ClassifyMe: A Field-Scouting Software for the Identification of Wildlife in Camera Trap Images. Animals, 10(1). https://doi.org/10.3390/ani10010058
Meek PD, Ballard G, Falzon G, Williamson J, Milne H, Farrell R, Stover J, Mather-Zardain AT, Bishop JC, Cheung K-W, Lawson CK, Munezero AM, Schneider D, Johnston BE, Kiani E, Shahinfar S, Sadgrove EJ and Fleming PJS (2020) Camera Trapping Technology and Advances: into the New Millennium. Australian Zoologist 40(3), 392-403. https://doi.org/10.7882/AZ.2019.035
