Convolutional Neural Networks to Estimate Dry Matter Yield in a Guineagrass Breeding Program Using UAV Remote Sensing.

Forage dry matter is the main source of nutrients in the diet of ruminant animals. Thus, this trait is evaluated in most forage breeding programs with the objective of increasing the yield. Novel solutions combining unmanned aerial vehicles (UAVs) and computer vision are crucial to increase the efficiency of forage breeding programs, to support high-throughput phenotyping (HTP), aiming to estimate parameters correlated to important traits. The main goal of this study was to propose a convolutional neural network (CNN) approach using UAV-RGB imagery to estimate dry matter yield traits in a guineagrass breeding program. For this, an experiment composed of 330 plots of full-sib families and checks conducted at Embrapa Beef Cattle, Brazil, was used. The image dataset was composed of images obtained with an RGB sensor embedded in a Phantom 4 PRO. The traits leaf dry matter yield (LDMY) and total dry matter yield (TDMY) were obtained by conventional agronomic methodology and considered as the ground-truth data. Different CNN architectures were analyzed, such as AlexNet, ResNeXt50, DarkNet53, and two networks proposed recently for related tasks named MaCNN and LF-CNN. Pretrained AlexNet and ResNeXt50 architectures were also studied. Ten-fold cross-validation was used for training and testing the model. Estimates of DMY traits by each CNN architecture were considered as new HTP traits to compare with real traits. Pearson correlation coefficient r between real and HTP traits ranged from 0.62 to 0.79 for LDMY and from 0.60 to 0.76 for TDMY; root square mean error (RSME) ranged from 286.24 to 366.93 kg·ha-1 for LDMY and from 413.07 to 506.56 kg·ha-1 for TDMY. All the CNNs generated heritable HTP traits, except LF-CNN for LDMY and AlexNet for TDMY. Genetic correlations between real and HTP traits were high but varied according to the CNN architecture. HTP trait from ResNeXt50 pretrained achieved the best results for indirect selection regardless of the dry matter trait. This demonstrates that CNNs with remote sensing data are highly promising for HTP for dry matter yield traits in forage breeding programs.

Mapping Utility Poles in Aerial Orthoimages Using ATSS Deep Learning Method.

Mapping utility poles using side-view images acquired with car-mounted cameras is a time-consuming task, mainly in larger areas due to the need for street-by-street surveying. Aerial images cover larger areas and can be feasible alternatives although the detection and mapping of the utility poles in urban environments using top-view images is challenging. Thus, we propose the use of Adaptive Training Sample Selection (ATSS) for detecting utility poles in urban areas since it is a novel method and has not yet investigated in remote sensing applications. Here, we compared ATSS with Faster Region-based Convolutional Neural Networks (Faster R-CNN) and Focal Loss for Dense Object Detection (RetinaNet ), currently used in remote sensing applications, to assess the performance of the proposed methodology. We used 99,473 patches of 256 × 256 pixels with ground sample distance (GSD) of 10 cm. The patches were divided into training, validation and test datasets in approximate proportions of 60%, 20% and 20%, respectively. As the utility pole labels are point coordinates and the object detection methods require a bounding box, we assessed the influence of the bounding box size on the ATSS method by varying the dimensions from 30×30 to 70×70 pixels. For the proposal task, our findings show that ATSS is, on average, 5% more accurate than Faster R-CNN and RetinaNet. For a bounding box size of 40×40, we achieved Average Precision with intersection over union of 50% (AP50) of 0.913 for ATSS, 0.875 for Faster R-CNN and 0.874 for RetinaNet. Regarding the influence of the bounding box size on ATSS, our results indicate that the AP50 is about 6.5% higher for 60×60 compared to 30×30. For AP75, this margin reaches 23.1% in favor of the 60×60 bounding box size. In terms of computational costs, all the methods tested remain at the same level, with an average processing time around of 0.048 s per patch. Our findings show that ATSS outperforms other methodologies and is suitable for developing operation tools that can automatically detect and map utility poles.

Storm-Drain and Manhole Detection Using the RetinaNet Method.

As key-components of the urban-drainage system, storm-drains and manholes are essential to the hydrological modeling of urban basins. Accurately mapping of these objects can help to improve the storm-drain systems for the prevention and mitigation of urban floods. Novel Deep Learning (DL) methods have been proposed to aid the mapping of these urban features. The main aim of this paper is to evaluate the state-of-the-art object detection method RetinaNet to identify storm-drain and manhole in urban areas in street-level RGB images. The experimental assessment was performed using 297 mobile mapping images captured in 2019 in the streets in six regions in Campo Grande city, located in Mato Grosso do Sul state, Brazil. Two configurations of training, validation, and test images were considered. ResNet-50 and ResNet-101 were adopted in the experimental assessment as the two distinct feature extractor networks (i.e., backbones) for the RetinaNet method. The results were compared with the Faster R-CNN method. The results showed a higher detection accuracy when using RetinaNet with ResNet-50. In conclusion, the assessed DL method is adequate to detect storm-drain and manhole from mobile mapping RGB images, outperforming the Faster R-CNN method. The labeled dataset used in this study is available for future research.

Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery.

Monitoring biomass of forages in experimental plots and livestock farms is a time-consuming, expensive, and biased task. Thus, non-destructive, accurate, precise, and quick phenotyping strategies for biomass yield are needed. To promote high-throughput phenotyping in forages, we propose and evaluate the use of deep learning-based methods and UAV (Unmanned Aerial Vehicle)-based RGB images to estimate the value of biomass yield by different genotypes of the forage grass species Panicum maximum Jacq. Experiments were conducted in the Brazilian Cerrado with 110 genotypes with three replications, totaling 330 plots. Two regression models based on Convolutional Neural Networks (CNNs) named AlexNet and ResNet18 were evaluated, and compared to VGGNet-adopted in previous work in the same thematic for other grass species. The predictions returned by the models reached a correlation of 0.88 and a mean absolute error of 12.98% using AlexNet considering pre-training and data augmentation. This proposal may contribute to forage biomass estimation in breeding populations and livestock areas, as well as to reduce the labor in the field.

A deep learning approach based on graphs to detect plantation lines.

Identifying plantation lines in aerial images of agricultural landscapes is re-quired for many automatic farming processes. Deep learning-based networks are among the most prominent methods to learn such patterns and extract this type of information from diverse imagery conditions. However, even state-of-the-art methods may stumble in complex plantation patterns. Here, we propose a deep learning approach based on graphs to detect plantation lines in UAV-based RGB imagery, presenting a challenging scenario containing spaced plants. The first module of our method extracts a feature map throughout the backbone, which consists of the initial layers of the VGG16. This feature map is used as an input to the Knowledge Estimation Module (KEM), organized in three concatenated branches for detecting 1) the plant positions, 2) the plantation lines, and 3) the displacement vectors between the plants. A graph modeling is applied considering each plant position on the image as vertices, and edges are formed between two vertices (i.e. plants). Finally, the edge is classified as pertaining to a certain plantation line based on three probabilities (higher than 0.5): i) in visual features obtained from the backbone; ii) a chance that the edge pixels belong to a line, from the KEM step; and iii) an alignment of the displacement vectors with the edge, also from the KEM step. Experiments were conducted initially in corn plantations with different growth stages and patterns with aerial RGB imagery to present the advantages of adopting each module. We assessed the generalization capability in the other two cultures (orange and eucalyptus) datasets. The proposed method was compared against state-of-the-art deep learning methods and achieved superior performance with a significant margin considering all three datasets. This approach is useful in extracting lines with spaced plantation patterns and could be implemented in scenarios where plantation gaps occur, generating lines with few-to-no interruptions.

Mauritia flexuosa palm trees airborne mapping with deep convolutional neural network.

Accurately mapping individual tree species in densely forested environments is crucial to forest inventory. When considering only RGB images, this is a challenging task for many automatic photogrammetry processes. The main reason for that is the spectral similarity between species in RGB scenes, which can be a hindrance for most automatic methods. This paper presents a deep learning-based approach to detect an important multi-use species of palm trees (Mauritia flexuosa; i.e., Buriti) on aerial RGB imagery. In South-America, this palm tree is essential for many indigenous and local communities because of its characteristics. The species is also a valuable indicator of water resources, which comes as a benefit for mapping its location. The method is based on a Convolutional Neural Network (CNN) to identify and geolocate singular tree species in a high-complexity forest environment. The results returned a mean absolute error (MAE) of 0.75 trees and an F1-measure of 86.9%. These results are better than Faster R-CNN and RetinaNet methods considering equal experiment conditions. In conclusion, the method presented is efficient to deal with a high-density forest scenario and can accurately map the location of single species like the M. flexuosa palm tree and may be useful for future frameworks.

Climatic seasonality and water quality in watersheds: a study case in Limoeiro River watershed in the western region of São Paulo State, Brazil.

Applying the climatological water balance (WB) concept to describe the relationship between climatic seasonality and surface water quality according to different forms of land use and land cover (LULC) is an important issue, but little explored in the literature. In this paper, we evaluate the influence of WB on surface water quality and its impacts when interacting with LULC. We monitored 11 sampling points during the four seasons of the year, from which we estimate WQI (water quality index) and TSI (trophic state index). We found an effect of the seasonality factor on both WQI values (F(3,30) = 12.472; p < 0.01) and in TSI values (F(3,30) = 6.967; p < 0.01). We noticed that LULC interferes in the way that the water balance influences the WQI and TSI values since in sampling points closest to higher urban density, with little or no riparian protection, the correlation between water balance and water quality was lower. In the stations that had the lowest water surplus and deficit, there was positive linearity between water balance and WQI. However, in the seasons when the surplus and water deficit recorded were extreme, there was no linearity. We conclude that water deficiency impairs the quality of surface water. In the extreme surplus water season, the homogeneity of WQI samples was lower, suggesting a higher interaction between rainwater and LULC. This study contributes to design management strategies of water resources, considering the climatic seasonality for optimization.

Tuberculosis space-temporal distribution from 2011 to 2016 in the municipality of Maputo, Mozambique / Distribución espacio-temporal de la tuberculosis de 2011 al 2016 en el municipio de Maputo, Mozambique

Abstract: Objective: Carry out a spatial-temporal characterization of the incidence of tuberculosis (TB) in Maputo, Mozambique. Method: a descriptive ecological study of tuberculosis cases reported in an information system. The annual mean incidence rate and the number of TB notification cases in the municipality of Maputo from 2011 to 2016 were analyzed. Descriptive statistics were used with calculations of measures of central tendency (mean) and an application of the Poisson linear regression model. Trimester notifications were stratified by district, clinical form, and age group. The quarterly average temperature of the evaluated area was added as a covariate in the model seasonal. Results: 34,623 TB cases were notified from 2011 to 2016, with a trimester average of 1,443 cases. The average annual incidence was higher in the Kampfumo district, with 909.8 per 100 thousand inhabitants (95% CI 854.1 - 968.2); almost twice as much as the incidence of the municipality of Maputo, 527.8 (95% CI 514, 3-541.6), and the country of Mozambique, 551 (95% CI 356 - 787). The clinical diagnosis of the tested cases was higher concerning the bacteriological diagnosis; 44%, and 35%, respectively. Conclusion: Maputo had similar incidence rates to the country of Mozambique, however, there was a heterogeneity rate by district and a reduction in the number of TB cases in both the general population (not co-infected with HIV) and those over 15 years old, being higher in the first trimester.

Resumen: Objetivo: realizar una caracterización espacio-temporal de la incidencia de tuberculosis (TB) en Maputo, Mozambique. Método: estudio ecológico descriptivo de casos de tuberculosis reportados en un sistema de información. Se analizó la tasa de incidencia media anual y el número de casos de notificación de TB en el municipio de Maputo entre 2011 y 2016. Se utilizó estadística descriptiva para calcular las medidas de tendencia central (media) y la aplicación del modelo de regresión lineal de Poisson Las notificaciones trimestrales se estratificaron por distrito, forma clínica y grupo de edad. Resultados: se notificaron 34,623 casos de TB entre 2011 y 2016, con un promedio trimestral de 1,443 casos. La incidencia anual promedio fue mayor en el distrito de Kampfumo, 909.8 por cada 100 mil habitantes (IC 95% 854.1 - 968.2), casi el doble que la incidencia del municipio de Maputo, 527.8 (IC 95% 514 , 3-541.6), y el país de Mozambique, 551 (95% CI 356 - 787). El diagnóstico clínico de los casos fue mayor en relación al diagnóstico bacteriológico, 44% y 35%, respectivamente. Conclusión: Maputo tuvo tasas de incidencia similares a las del país, sin embargo, hubo una heterogeneidad en las tasas por distrito y una reducción en el número de casos de TB en la población general (no coinfectados con VIH) y en los mayores de 15 años, siendo mayores en el primer trimestre.