Large Scale Detailed Mapping of Dengue Vector Breeding Sites Using Street View Images
|Authors:||Schöning, Johannes||Supervisor:||Technical Report University of Bremen, Germany||Other Authors:||Haddawy, Peter
Targeted environmental and ecosystem management remain crucial in control of dengue. But providing detailed environmental information on a large scale to effectively target dengue control efforts remains a challenge. In this paper we present the design and implementation of a pipeline to detect potential dengue vector breeding sites from geotagged images to create highly detailed container density maps at unprecedented scale. We implement the approach using Google Street View images which have the advantage of broad coverage and of being somewhat historical so that the data can be aligned with other types of data for analysis. Containers comprising eight of the most common breeding sites are detected in the images using convolutional neural network transfer learning. Over a test set of images the object recognition algorithm has an accuracy of 0.91 in terms of F-score. Container density counts are generated and displayed on a decision support dashboard. Extensive analyses of the approach is carried out over three provinces in Thailand. Results show that the container density counts agree well with manual container counts, with larval survey data, and with dengue case data. To delineate conditions under which the density counts are indicative of risk, a number of factors affecting agreement with larval survey and dengue case data are analyzed. We conclude that creation of container density maps from geotagged images is a promising approach to providing detailed risk maps at large scale. Ultimately, we intended to include our newly proposed index in the identification of dengue high-risk areas in Thailand.
|Keywords:||AI, Machine Learning, Deep Learning, Dengue Vector, Dengue Breeding Sites, Google Street View Images, Street View Images, Dengue, Mapping, Thailand||Issue Date:||2019||Pages:||39||Type:||Bericht, Report||URN:||urn:nbn:de:gbv:46-00107025-12||Institution:||Universität Bremen||Faculty:||FB3 Mathematik/Informatik|
|Appears in Collections:||Forschungsdokumente|
checked on Sep 23, 2020
checked on Sep 23, 2020
Items in Media are protected by copyright, with all rights reserved, unless otherwise indicated.