Contagem de ovos do Aedes Aegypti em palhetas de ovitrampas baseada em Deep Learning
Abstract
Aedes aegypti is the vector of one of the most difficult public health problems to be tackled in the tropical world: the spread of the dengue epidemic. As there is no vaccine or specific treatment, and the eradication of its vector has become practically impossible, the best way to avoid the disease is to control the Aedes aegypti mosquito. With that, this work presents a computational methodology for the segmentation of eggs in ovitrap straws that aims to help specialists in this counting. For the development of this work we performed tests with the convolutional neural networks developed for segmentation: U-Net, Segnet and a pre-trained network. Then, we performed a post-processing step based on mathematical morphology. The results achieved were promising, and the UNet network showed the best performance, with an accuracy of 98.65% in egg segmentation and mean square error of 4.25% in counting.
References
Bandong, S. and Joelianto, E. (2019). Counting of aedes aegypti eggs using image processing with grid search parameter optimization. In 2019 International Conference on Sustainable Engineering and Creative Computing (ICSECC), pages 293–298.
Barros, V. C., Pacheco, A. C. L., Feitosa, L. M., Figueredo, J. S., Batista, F., Lima, I. P., and Barbosa, O. A. A. (2019). Produtos naturais no combate ao mosquito Aedes aegypti. Campinas: Atomo.
Feitosa, L. N. (2015). Sistema de contagem automática de ovos do aedes aegypti a partir de processamento de imagens das palhetas de ovitrampas. Monografia, Instituto Federal de Educação, Ciência e Tecnologia do Rio Grande do Norte.
Gaburro, J., Duchemin, J.-B., Paradkar, P., Nahavandi, S., and Bhatti, A. (2016). Assessment of icount software, a precise and fast egg counting tool for the mosquito vector aedes aegypti. Parasites Vectors, 9:1–9.
Garcia, P. S. C., Martins, R., Coelho, G. L. L. M., and Cámara-Chávez, G. (2019). Acquisition of digital images and identification of aedes aegypti mosquito eggs using classification and deep learning. In 2019 32nd SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), pages 47–53.
Glasser, M. and Gomes, C. (2000). Infestação do estado de são paulo por aedes aegypti e aedes albopictus. Revista de Saúde Pública, 34:570–577.
LONG, J.; SHELHAMER, E. D. (2015). Fully convolutional networks for semantic segmentation. IEEE. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 28:3174 – 3179.
MENDES, C. C. T.; FR ´EMONT, V. W. D. F. (2017). Exploiting fully convolutional neural networks for fast road detected). IEEE. Robotics and Automation (ICRA),, 28.
Rocha, C., Bizerra, A., Coutinho, D., Silva, L., de Deus, M., and Souto, P. (2019). Contagem automática de ovos do Aedes Aegypti em palhetas de ovitrampas: um sistema para aquisição e processamento de imagens, pages 67–74.
Santana, C. et al. (2019). A solution for counting aedes aegypti and aedes albopictus eggs in paddles from ovitraps using deep learning. (17):1987–1994.
Silva, M. A. G. N. M. d., Rodrigues, M. A. A. B., and Araujo, R. E. d. (2012). System for acquisition and processing of ovitraps images to ght dengue. Brazilian journal of biomedical engineering, 28:364 – 374.
Simonyan, K. and Zisserman, A. (2014). Very deep convolutional networks for largeIEEE. Proceedings of the IEEE Conference on Computer scale image recognition. Vision and Pattern Recognition, 28.
Sousa, J. A. and Paiva, A. C. (2014). Contagem automática de ovos de mosquito da dengue em imagem de ovitrampa. In 2014 V Jornada de Informática do Maranhão (JIM), pages 1–4.
Yussof, W., Man, M., Hitam, M., abdul hamid, a. a., Awalludin, E., and Wan Abu Bakar, W. A. (2018). Wavelet-based auto-counting tool of aedes eggs. pages 56–59.
