Automated Identification of Mosquitoes using acoustic signals

Mosquitoes are considered to be among the biggest disease spreading flying insects causing worldwide health hazards.  In Pakistan, Dengue fever is an infectious viral disease transmitted through the bite of infected Aedes aegypti and Aedes albopictus female mosquitoes. Ma

Project Title

Automated Identification of Mosquitoes using acoustic signals

Project Area of Specialization

Artificial Intelligence

Project Summary

Mosquitoes are considered to be among the biggest disease spreading flying insects causing worldwide health hazards.  In Pakistan, Dengue fever is an infectious viral disease transmitted through the bite of infected Aedes aegypti and Aedes albopictus female mosquitoes. Manual identification of dengue mosquito is hard and dangerous leading to severe health risks. In this project, we are proposing an automated system that is able to identify aedes species that is the dengue causing mosquitoes species among Anopheles and Culex species using the acoustic recordings of their wingbeat frequency. We will demonstrate the proposed approach on a standard dataset and compare it to human annotations to determine accuracy.

Project Objectives

• Identify mosquitoes species using their sounds
• Help setup acoustic traps for dangerous mosquitoes
• Identify dengue and malaria causing mosquitoes

Project Implementation Method

We will implement machine learning techniques that takes mosquitoes songs audio files as input and use these sounds for training. Then we will calculate distance (euclidean/cosine similarity/mahalanobis) to classify unknown mosquitoe species. We will test the results using on a dataset that we will collect using microphones attached to a raspberry pi and solar panel to avoid human intervention. The sounds will be annotated by specialists from entomolog department.

Benefits of the Project

• Identify mosquitoes species using their sounds
• Help setup acoustic traps for dangerous mosquitoes
• Identify dengue and malaria causing mosquitoes

Technical Details of Final Deliverable

The final software would be delivered as an android app that can be used by anyone from anywhere to identify mosquitoes around them. The app will take mosquitoe sound recorded from mobile device as input and classify it based on the trained dataset.

Final Deliverable of the Project

Software System

Type of Industry

IT , Health

Technologies

Artificial Intelligence(AI), Others

Sustainable Development Goals

Good Health and Well-Being for People, Clean Water and Sanitation

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com