Presentation (Project) Title

Identification of Vector Mosquitoes: A Field Implementation of Novel Hardware and AI Algorithm

Mentor Information

Sriram Chellappan (College of Engineering)

Presentation Format

Event

Abstract

Mosquito-borne diseases continue to be one of the leading causes of death globally. They arise from vector mosquitoes which can transmit various diseases such as malaria, yellow fever, dengue, etc. Vector-borne diseases account for 17% of all infectious diseases and affect millions of people around the world causing 700,000 deaths annually according to the WHO. Consequently, real-time identification of these mosquitoes is essential for pest control agencies which have the goal of reducing and eventually eradicating these pernicious diseases. In an effort to reduce the prevalence of vector mosquitoes, an Artificial Intelligence (AI) system has been developed to effectively identify them, so that targeted mosquito- control can be applied to eliminate these species as they are detected. In order to create an environment for the AI system, modifications were made to the CDC Miniature Light Trap, which is used throughout the world. Added features include a digital imaging chamber, heated CO2 attractant, and a control system to capture the images required for the AI system. Mosquitoes are attracted to the heated resting surface inside of the imaging chamber, where pictures are taken as the presence of mosquitoes is detected. Currently, mosquitoes are efficiently attracted and caught in the trap. However, there are issues with the camera focus, resulting in blurry images of mosquitoes inside the imaging chamber. This results in the AI being unable to process these images, and effectively identify these mosquitoes. The focus for future work is to correct this issue in order to reliably capture clear images.

Streaming Media

This document is currently not available here.

Share

COinS
 

Identification of Vector Mosquitoes: A Field Implementation of Novel Hardware and AI Algorithm

Mosquito-borne diseases continue to be one of the leading causes of death globally. They arise from vector mosquitoes which can transmit various diseases such as malaria, yellow fever, dengue, etc. Vector-borne diseases account for 17% of all infectious diseases and affect millions of people around the world causing 700,000 deaths annually according to the WHO. Consequently, real-time identification of these mosquitoes is essential for pest control agencies which have the goal of reducing and eventually eradicating these pernicious diseases. In an effort to reduce the prevalence of vector mosquitoes, an Artificial Intelligence (AI) system has been developed to effectively identify them, so that targeted mosquito- control can be applied to eliminate these species as they are detected. In order to create an environment for the AI system, modifications were made to the CDC Miniature Light Trap, which is used throughout the world. Added features include a digital imaging chamber, heated CO2 attractant, and a control system to capture the images required for the AI system. Mosquitoes are attracted to the heated resting surface inside of the imaging chamber, where pictures are taken as the presence of mosquitoes is detected. Currently, mosquitoes are efficiently attracted and caught in the trap. However, there are issues with the camera focus, resulting in blurry images of mosquitoes inside the imaging chamber. This results in the AI being unable to process these images, and effectively identify these mosquitoes. The focus for future work is to correct this issue in order to reliably capture clear images.