Abstract

This research aims to develop a web-based application for the identification of mosquito species using image data of adult mosquitoes and larvae. Mosquitoes are known carriers of various diseases and their accurate identification is crucial for effective control and prevention strategies.  The system utilizes a dataset of adult mosquitoes and larvae, consisting of various species and their corresponding features such as morphological characteristics extracted from image datasets. The results demonstrate the effectiveness of machine learning algorithms in accurately identifying mosquito species based on adult mosquitoes and larvae data. The mosquito identification module will be based on machine learning and image processing techniques. The system shows promising potential in providing rapid and reliable mosquito species identification, which can aid in the surveillance and control of mosquito-borne diseases. The proposed system offers a valuable tool for entomologists, public health officials, and researchers working on mosquito control and disease prevention, ultimately contributing to improved strategies for combating mosquito-borne diseases. The adult mosquito identification model achieved a test accuracy of 83.55% and the larvae genera identification model achieved a test accuracy of 87.30%. But the prototype validation, according to the user-supplied images, produced accuracies of 69.57% and 60.47%, highlighting the challenges in real-world deployment. The findings show the potential of machine learning based image analysis to support mosquito surveillance and public-health monitoring tasks. 

Keywords:
Mosquito Identification, Machine Learning, Image Processing, Mosquito Species, Web Based Applications

Download the article (pdf)