Have you ever spotted kiwi birds in NZ bush? Can you imagine how many of them are living in a forest? It is actually quite rare that you come across birds like kiwis because of their declining population and nocturnal lifestyle. Therefore it is a quite challenging task for the rangers and researchers to study the ecology of such birds without actually staying in a forest for weeks and months. The DOC (Department of Conservation) has been attempting to identify bird calls from audio recordings collected in a forest however so far this has been done completely manually (by hiring people and getting them to literally "listen and identify" bird calls) which is not cost effective and the results can also be inaccurate. Recently the DOC called a group of researchers around NZ to develop an automated system to identify bird calls as well as measuring abundance of species. The supervisor is a part of the group in charge of extracting various information from the recording, which will be used by a machine learning based bird call identification system (uses an algorithm similar to what is known as Artificial Intelligence or AI recently) studied by other researchers in the group.
The ultimate goal of the supervisor's team is providing the identification system with the recording of individual bird's call with the information about its location (or the angle with respect to the recorder's position). The project started in 2017 where previous students investigated the performance of existing sound source localisation algorithms using various shape of microphone arrays using recordings collected in a real kauri forest. This year, the project will expand the study and attempt to localise more than one birds included in the same recording. We would also try to record the targeted calls more clearly by removing background noise using digital filtering techniques. Algorithms studies will be tested using recording database collected by the previous students, however some additional field recordings may be needed throughout the project.
Students suitable for this project would be interested in acoustics and digital signal processing as well as ecology of kiwis. At least one of the group members must have passed MECHENG370 after 2015.
Undergraduate
A signal processing algorithm for localising multiple kiwi birds from their calls and recording the calls with minimum amount of noise.
A set of experimental results conducted to verify the performance of the signal processing algorithm using sound recordings collected in a real bush.
This project requires fundamental knowledge in digital signal processing. At least one of the students taking this project must have passed MECHENG370 after 2015. This project is open only to Mechatronics students.
Acoustics Lab (City 422.154, Lab)