High speed counting of black soldier flies
The Challenge
The world is generating mixed solid waste (MSW) in volumes far exceeding its capacity to recycle and manage effectively. In Australia, the total volume of MSW generated in 2018 was 67 million tonnes, of which 45.6 million tonnes was buried into landfill resulting in significant environmental pollution. The usage of black solider fly larvae to process MSW provides an environmentally friendly option. However, optimising the breeding process of the flies is yet to be fully developed and is still an on-going active area of research.
In particular, measuring the effectiveness of breeding methods are based on best practice and there is no accurate way to measure the fly reproduction behavior. The number and size of the flies in the breeding process make it difficult for conventional cameras and sensors to capture reliable data.
The Solution
The project utilised the world-leading neuromorphic vision capabilities at Western Sydney University to help industry partner, ARC Ento Tech, improve operational efficiencies in breeding flies. Researchers created a system based on a neuromorphic camera: technology which uses sensors to collect only useful data, and which has been inspired by biology to accurately observe fast-moving objects, including flies. This system measured the number of black soldier flies moving from the hatching and breeding room to another area where grown flies would mate.
The project was completed in collaboration with the mechatronic engineering capability of researchers from Macquarie University, who delivered the hardware which allowed the sensor to operate inside the breeding cages. The FlyCount system was developed with a custom spike detection algorithm for real-time, accurate fly counting, which ensures 95 percent accuracy in dynamic lighting. By integrating neuromorphic sensing technologies to an IoT enabled system, the system demonstrated the effectiveness and accuracy of the event-based counting system for tracking and analysing the movements of black soldier flies remotely.
The results of the project were published in IEEE Sensors.
Read more about the project in this news article.
Industry Partner
University Partners
