Classification Of Aedes Mosquito Larva Using Convolutional Neural Networks And Extreme Learning Machine

The spread of diseases transmitted by Aedes mosquitoes, such as Dengue fever, Chikungunya (CHIKV), and Zika, poses a severe health problem in many countries. Aedes mosquitoes can be found in tropical and subtropical areas. One method to control the diseases is by controlling the population of Aedes mosquitoes, and one way to control the population of Aedes mosquito is by exterminating Aedes mosquito larvae. Identifying mosquito larvae, however, demands a considerable amount of effort, time, and specialized knowledge about the morphology and anatomy of mosquito larvae to accurately classify the various types. Therefore, a system is needed to classify Aedes and non-Aedes mosquito larvae. In this study, classification was carried out using larvae images taken using a digital microscope. The methods used to classify mosquito larvae images are Convolutional Neural Networks (CNN) to extract features from the larvae image and Extreme Learning Machine (ELM) to classify the larvae image. The result obtained from evaluating the CNN-ELM model is an accuracy of 98%, f1-score of 99% for Aedes larvae, and fl-score of 96% for non-Aedes larvae.