The purpose of this research is to design, deploy and validate artificial intelligence (AI) algorithms operating on drone videos to enable a real time methodology for optimizing predictive mapping unknown, geographic locations (henceforth, geolocation) of potential, seasonal, Anopheles (gambiae, and funestus) larval habitats in an agro-village, epi-entomological, intervention site (Akonyibedo village) in Gulu District, Northern Uganda. Formulae are developed for classifying the drone swath, capture point, land cover in Akonyibedo agro-village. An AI algorithm is designed for constructing a smartphone application (app) in order to enable automatic detection of potential larval habitats from drone videos. The aim of this work is to enable scaling up to larger intervention sites (e.g., district level, sub-county) and then throughout entire Uganda. We demonstrate how capture point, stratifiable, drone swath coverage in Akonyibedo village can be accomplished employing temporal series of re-centered, real time, imaged, Anopheline, larval habitat, seasonal, map projections. We also define a remote methodology for detecting unknown, georeferenceable, capture point geolocations of potential, seasonal, breeding sites employing multispectral, wavelength, signature, reflux emissivities in a drone spectral library. Our results show that high-resolution drone imagery when processed employing state of the art AI algorithms can discriminate a profile of water bodies where Anopheles mosquitoes are most likely to breed (overall ground truth accuracy of 100%). Live, high definition, Anopheline larval habitat signature maps can be generated in real-time drone AI app on a smartphone or Apple device while the image is being captured or larvicidal application is taking place.

We design a framework based on Mask Region-based Convolutional Neural Network to automatically detect and separately extract anatomical components of mosquitoes-thorax, wings, abdomen and legs from images. Our training dataset consisted of 1500 smartphone images of nine mosquito species trapped in Florida. In the proposed technique, the first step is to detect anatomical components within a mosquito image. Then, we localize and classify the extracted anatomical components, while simultaneously adding a branch in the neural network architecture to segment pixels containing only the anatomical components. Evaluation results are favorable. To evaluate generality, we test our architecture trained only with mosquito images on bumblebee images. We again reveal favorable results, particularly in extracting wings. Our techniques in this paper have practical applications in public health, taxonomy and citizen-science efforts.

No abstract

Recent advances in neural machine translation (NMT) have pushed the quality of machine translation systems to the point where they are becoming widely adopted for building competitive systems. However, there is still a large number of languages that are yet to reap the benefits of NMT. In this paper, we provide the first large-scale case study of the practical application of MT in the Turkic language family in order to realize the gains of NMT for Turkic languages under high-resource to extremely low-resource scenarios. In addition to presenting an extensive analysis that identifies the bottlenecks towards building competitive systems to ameliorate data scarcity, our study has several key contributions, including, i) a large parallel corpus covering 22 Turkic languages consisting of common public datasets in combination with new datasets of approximately 2 million parallel sentences, ii) bilingual baselines for 26 language pairs, iii) novel high-quality test sets in three different translation domains and iv) human evaluation scores. All of our data, software and models are publicly available. 1

Recent advances in neural machine translation (NMT) have pushed the quality of machine translation systems to the point where they are becoming widely adopted to build competitive systems. However, there is still a large number of languages that are yet to reap the benefits of NMT. In this paper, we provide the first large-scale case study of the practical application of MT in the Turkic language family in order to realize the gains of NMT for Turkic languages under high-resource to extremely lowresource scenarios. In addition to presenting an extensive analysis that identifies the bottlenecks towards building competitive systems to ameliorate data scarcity, our study has several key contributions, including, i) a large parallel corpus covering 22 Turkic languages consisting of common public datasets in combination with new datasets of approximately 2 million parallel sentences, ii) bilingual baselines for 26 language pairs, iii) novel high-quality test sets in three different translation domains and iv) human evaluation scores. All models, scripts, and data will be released to the public. 1