Detecting Aedes Aegypti Mosquitoes through Audio Classification with Convolutional Neural Networks
Authors:
Marcelo Schreiber Fernandes,
Weverton Cordeiro,
Mariana Recamonde-Mendoza
Abstract:
The incidence of mosquito-borne diseases is significant in under-developed regions, mostly due to the lack of resources to implement aggressive control measurements against mosquito proliferation. A potential strategy to raise community awareness regarding mosquito proliferation is building a live map of mosquito incidences using smartphone apps and crowdsourcing. In this paper, we explore the pos…
▽ More
The incidence of mosquito-borne diseases is significant in under-developed regions, mostly due to the lack of resources to implement aggressive control measurements against mosquito proliferation. A potential strategy to raise community awareness regarding mosquito proliferation is building a live map of mosquito incidences using smartphone apps and crowdsourcing. In this paper, we explore the possibility of identifying Aedes aegypti mosquitoes using machine learning techniques and audio analysis captured from commercially available smartphones. In summary, we downsampled Aedes aegypti wingbeat recordings and used them to train a convolutional neural network (CNN) through supervised learning. As a feature, we used the recording spectrogram to represent the mosquito wingbeat frequency over time visually. We trained and compared three classifiers: a binary, a multiclass, and an ensemble of binary classifiers. In our evaluation, the binary and ensemble models achieved accuracy of 97.65% ($\pm$ 0.55) and 94.56% ($\pm$ 0.77), respectively, whereas the multiclass had an accuracy of 78.12% ($\pm$ 2.09). The best sensitivity was observed in the ensemble approach (96.82% $\pm$ 1.62), followed by the multiclass for the particular case of Aedes aegypti (90.23% $\pm$ 3.83) and the binary (88.49% $\pm$ 6.68). The binary classifier and the multiclass classifier presented the best balance between precision and recall, with F1-measure close to 90%. Although the ensemble classifier achieved the lowest precision, thus impairing its F1-measure (79.95% $\pm$ 2.13), it was the most powerful classifier to detect Aedes aegypti in our dataset.
△ Less
Submitted 18 August, 2020;
originally announced August 2020.
An Aiding Tool for Building Design Generation, Thermal Assessment and Optimization -- EnergyPlus Interaction Overview
Authors:
Marco S. Fernandes,
E. Rodrigues,
Adélio R. Gaspar,
Álvaro Gomes
Abstract:
A building design aiding tool for space allocation and thermal performance optimization is being developed to help practitioners during the building space planning phase, predicting how it will behave regarding energy consumption and thermal comfort. The tool evaluates, ranks, and optimizes generated floor plans according to thermal performance criteria, using the dynamic simulation program Energy…
▽ More
A building design aiding tool for space allocation and thermal performance optimization is being developed to help practitioners during the building space planning phase, predicting how it will behave regarding energy consumption and thermal comfort. The tool evaluates, ranks, and optimizes generated floor plans according to thermal performance criteria, using the dynamic simulation program EnergyPlus. The tool is currently able to use a wide variety of EnergyPlus objects, allowing for various template and detailed HVAC, DHW, and thermal and electrical energy production systems and components, as well as numerous internal gains types, construction elements and energy saving controls, to be accounted for and simulated in the generated buildings. This paper presents the tool overall concept as well as the main features regarding dynamic simulation. Some performance results are presented for distinct systems to illustrate the use and potential of the tool.
△ Less
Submitted 14 June, 2018;
originally announced June 2018.
