Thrips are serious pest insects in greenhouses. Traditional trapping systems, such as sticky traps, water traps and funnel traps are based on trial and error, without fundamental knowledge of host seeking behavior. Measuring 3D flight of host seeking thrips will provide the information required to develop more efficient traps.
In various fields of research there is growing demand for quantifying 3D flight behavior of insects. Monitoring flight behavior is critically important for developing efficient traps for pest insects such as blood-feeding mosquitos and phytophagous thrips.
Because individual insects behave in a highly variable manner, research on host seeking behavior requires extensive measurements and large data sets. In this project we aim to meet these demands by developing automated, real-time tracking of flight paths. Whereas separate components, such as control of high-speed cameras, object detection and path tracking are readily available, constructing an integrated setup for instantaneous tracking of multiple flying insects is still a major challenge.
We will tackle this problem for monitoring thrips in host-seeking experiments. Thrips are guided to their host plant by a combination of chemical and visual cues. The experiments aim to optimize these stimuli for attracting thrips to their target, which can then be used to develop efficient traps for pest control.
The experiments will be carried out in a climate-controlled wind tunnel (see figures). Thrips will be released downstream of one or two targets. Targets can differ in colour, intensity and contrast with the background, by means of LED lights adjusted to the visual sensitivity of thrips. By monitoring flight paths we will be able to identify and quantify approaches, landings and searching behavior on the target. Each target will be equipped with two high-speed cameras allowing for full three-dimensional reconstruction of travel paths.
|Examiner:||Prof. Dr. Ir. Johan L. van Leeuwen|
|Rob van Tol|
|Contact:||M.J.M. Lankheet / F.T. Muijres|
|Begin date:||15/10/2017 (variable)|
|Credits:||24 ECTS (minimal), 36 (preferred)|
|Requirements:||Good programming skills in MatLab, Python or C++, familiarity with Machine vision desirable|