Mosquito larva classification method based on convolutional neural networks

In Mexico a great number of diseases spread by the mosquitos Aedes has been reported. There are some regions on the country that this number is alarming. The spread of this disease becomes a public health problem and the government is worried about this situation and applied some methods for reducing the infection rate. One of principal methods relies on the localization of the mosquito's larvae and then fumigates them. The localization of Aedes larvae is accomplished through state programs which take a considerable time, making them not efficient enough. In this paper we propose a novel method based on convolutional neural networks, where a dataset of larva is used in training in order that the machine learns two types of mosquitos, genus Aedes and "others" genera. The digital images of larva are processed using a set of machine learning algorithms and as a result, the classification task is done. The proposed method would make the larva identification process more efficient, automatic and faster than the conventional methods, and thus the infection rates would be decrease. The results show a good performance on Aedes larva identification, proving that the system can be applied in the real world.