Kauê de Moraes Vestena

Kauê de Moraes Vestena

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Kauê de Moraes Vestena has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Impact of ZTD Estimation and Tropospheric Gradients on Kinematic PPP Accuracy in the Amazon and Southern Regions of Brazil

    by Lais Thuany Cardoso Theodoro, Kaue de Moraes Vestena, Tiago Lima Rodrigues, Paulo Sergio De Oliveira Junior

    The study investigated the influence of tropospheric gradients on the accuracy of Precise Point Positioning (PPP), using data from Brazilian Network for Continuous Monitoring of the GNSS Systems (RBMC) in the Amazon and southern regions of Brazil. The research was conducted due to the need to better understand how tropospheric gradients affect the accuracy of kinematic PPP. Understanding these effects is crucial to improving the accuracy of Global Navigation Satellite System (GNSS) positioning, especially in regions with variable atmospheric conditions. The study used RBMC data and tropospheric delay modeling based on Numerical Weather Prediction (PNT) data. The results showed significant improvements in the planimetric accuracy of kinematic PPP with the inclusion of tropospheric gradients. The improvements were observed on the order of millimeters, with increases of up to 3% in horizontal accuracy and 7% in vertical accuracy. The analysis also revealed that estimating tropospheric gradients was more significant in the vertical component and regions with higher humidity tropospheric scenarios. It was concluded that including tropospheric gradients is crucial to improving the accuracy of kinematic PPP, especially in the Amazon region, where tropospheric scenarios are more humid. Additionally, estimating tropospheric gradients was considered as or even more significant in planimetric accuracy than using a Mapping Function based on PNT data. This research highlights the importance of considering tropospheric gradients in GNSS positioning and provides valuable insights for future studies and practical applications in the field of geodesy and satellite navigation.

  • Open Access

    Investigating the Performance of Open-Vocabulary Classification Algorithms for Pathway and Surface Material Detection in Urban Environments

    by Kauê de Moraes Vestena, Silvana Phillipi Camboim, Maria Antonia Brovelli, Daniel Rodrigues dos Santos

    Mapping pavement types, especially in sidewalks, is essential for urban planning and mobility studies. Identifying pavement materials is a key factor in assessing mobility, such as walkability and wheelchair usability. However, satellite imagery in this scenario is limited, and in situ mapping can be costly. A promising solution is to extract such geospatial features from street-level imagery. This study explores using open-vocabulary classification algorithms to segment and identify pavement types and surface materials in this scenario. Our approach uses large language models (LLMs) to improve the accuracy of classifying different pavement types. The methodology involves two experiments: the first uses free prompting with random street-view images, employing Grounding Dino and SAM algorithms to assess performance across categories. The second experiment evaluates standardized pavement classification using the Deep Pavements dataset and a fine-tuned CLIP algorithm optimized for detecting OSM-compliant pavement categories. The study presents open resources, such as the Deep Pavements dataset and a fine-tuned CLIP-based model, demonstrating a significant improvement in the true positive rate (TPR) from 56.04% to 93.5%. Our findings highlight both the potential and limitations of current open-vocabulary algorithms and emphasize the importance of diverse training datasets. This study advances urban feature mapping by offering a more intuitive and accurate approach to geospatial data extraction, enhancing urban accessibility and mobility mapping.


  • Open Access

    Investigating the Performance of Open Vocabulary Classification Algorithms for Pathway and Surface Material Detection in Urban Environments

    by Kauê de Moraes Vestena, Silvana Phillipi Camboim, Maria Antonia Brovelli, Daniel Rodrigues dos Santos

    Mapping pavement types, especially in sidewalks, is essential for urban planning and mobility studies. Identifying pavement materials is a key factor in assessing mobility, such as walkability and wheelchair usability. However, satellite imagery in this scenario is limited, and in-situ mapping can be costly. A promising solution is to extract such geospatial features from street-level imagery. This study explores using open vocabulary classification algorithms to segment and identify pavement types and surface materials in this scenario. Our approach uses Large Language Models (LLMs) to improve the accuracy of classifying different pavement types. The methodology involves two experiments: the first uses free prompting with random street-view images, employing Grounding Dino and SAM algorithms to assess performance across categories. The second experiment evaluates standardized pavement classification using the Deep-Pavements Dataset and a fine-tuned CLIP algorithm optimized for detecting OSM-compliant pavement categories. The study presents open resources, such as the Deep Pavements Dataset and a fine-tuned CLIP-based model, demonstrating a significant improvement in the True Positive Rate (TPR) from 56.04% to 93.5%. Our findings highlight both the potential and limitations of current open vocabulary algorithms and emphasize the importance of diverse training datasets. This study advances urban feature mapping by offering a more intuitive and accurate approach to geospatial data extraction, enhancing urban accessibility and mobility mapping.


  • Open Access

    A AMAZÔNIA NO MAPA: DESAFIO OLÍMPICO NO CONTEXTO DA V OLIMPÍADA BRASILEIRA DE CARTOGRAFIA

    by Angelica Carvalho Di Maio, Marcus Vinícius Alves de Carvalho, Kellen Milene Gomes e Santos, Raquel Dezidério Souto, Kauê De Moraes Vestena

    A V edição da Olimpíada Brasileira de Cartografia (OBRAC 2023), com o tema "A Amazônia no Mapa", focou no mapeamento de recursos naturais e povos indígenas da Amazônia utilizando geotecnologias. O evento destacou a importância da Amazônia através de mapas que exploraram biodiversidade, desmatamento, bacias hidrográficas e questões sociais. A metodologia envolveu provas teóricas e práticas, com atividades de mapeamento participativo e colaborativo, utilizando plataformas como OpenStreetMap (OSM) e ferramentas como uMap, QGIS e Jamboard. O mapeamento participativo engajou estudantes e comunidades na criação de mapas que representassem a cultura e as potencialidades da Amazônia, enquanto o mapeamento colaborativo permitiu a participação remota via internet, utilizando bases cartográficas digitais. Esses métodos visaram empoderar comunidades e tornar visíveis suas questões locais, além de preencher lacunas cartográficas, especialmente em áreas rurais e subnormais. As equipes produziram mapas que refletiram as dinâmicas regionais, destacando aspectos positivos e desafios da região amazônica. Durante a etapa final presencial, as equipes visitaram instituições como o Instituto Nacional de Pesquisas Espaciais e o CEMADEN. A equipe vencedora do Instituto Federal do Pará desenvolverá um projeto de iniciação científica júnior sobre a Floresta de Carajás em 2024.


  • Open Access

    Automatic detection of urban infrastructure elements from terrestrial images using deep learning

    by Jaime Carlos Macuácua, Jorge António Silva Centeno, Fernando Alves Barros Firmino, Jorgiana Kamila Teixeira Do Crato, Kauê de Moraes Vestena, Caisse Amisse

    Urban infrastructure element detection is important for the domain of public management in large urban centres. The diversity of objects in the urban environment makes object detection and classification a challenging task, requiring fast and accurate methods. Advances in deep learning methods have driven improvement in detection techniques (processing, speed, accuracy) that do not rely on manually crafted models, but, instead, use learning approaches with corresponding large training sets to detect and classify objects in images. We applied an object detection model to identify and classify four urban infrastructure elements in the Mappilary dataset. We use YOLOv5, one of the top-performing object detection models, a recent release of the YOLO family, pre-trained on the COCO dataset but fine-tuned on Mappilary dataset. Experimental results from the dataset show that YOLOv5 can make qualitative predictions, for example, the power grid pole class presented the mean Average Precision (mAP) of 78% and the crosswalk class showed mAP around 79%. A lower degree of certainty was verified in the detection of public lighting (mAP=64%) and accessibility (mAP=61%) classes due to the low resolution of certain objects. However, the proposed method showed the capability of automatically detection and location of urban infrastructure elements in real-time, which could contribute to improve decision-making.


  • Open Access

    Evaluation of precise point positioning and post-processing kinematic methods for tide measurement in hydrographic surveys

    by Felipe Rodrigues Santana, Claudia Pereira Krueger, Érica Santos Matos Baluta, Tulio Alves Santana, Kaue de Moraes Vestena

    The use of the ellipsoidal heights of a vessel for measuring the tide allows the diminution of uncertainties of sounding reduction in hydrographic surveys. In this article, the Precise Point Positioning (PPP) and Post-Processing Kinematic (PPK) methods were evaluated by comparing them with data from a tide gauge station while the boat remained moored nearby. When traversing sounding lines, the PPK solution was used as a reference to validate the PPP solution. In all, 12 different periods of surveys were analyzed, distributed throughout the year 2021, in the Guanabara and Ilha Grande Bays, in Rio de Janeiro. The results showed that in comparison with the tide station, when the vessel was moored, the PPK was able to fulfill the strictest criteria (Exclusive Order) of the IHO (International Hydrographic Organization) in 100% of the surveys, while the PPP in 50%. Regarding the use of the PPK as a reference, with the boat both moored and sailing, the PPP met these criteria in 67% for Exclusive Order, 25% for Order 1A and 8% for the 1B. Reducing the vertical uncertainties of hydrographic surveys in shallow waters is useful for all marine and coastal vertical positioning applications, such as integrating terrestrial and marine vertical references.


  • Open Access

    Automatic foliar spot detection from low-cost RGB digital images using a hybrid approach of convolutional neural network and random forest classifier

    by Jaime Carlos Macuácua, Jorge António Silva Centeno, Caisse Amisse, Mário Ernesto Jijón-Palma, Kauê de Moraes Vestena

    Tomatoes are widely cultivated, both by family farmers and corporate producers. During the tomato growth cycle, several diseases can affect the plant. The identification of these diseases through short-range images is significant, and computer vision techniques are commonly used to identify diseases in plant leaves. In this paper, a hybrid model that combines a convolutional neural network (CNN) and a Random Forest (RF) decision tree is used for foliar spot detection in tomato leaves. High-level features learned and extracted from CNN are used as input for the RF classifier. To evaluate the proposed model’s performance for plant disease identification, a case study of 2480 low-cost digital RGB images collected in actual field conditions, under different intensities of light exposure, were used, including healthy tomato leaves and leaves with visible symptoms of powdery mildew fungus, which attacks the tomato leaf. The results were compared with six conventional machine learning classifiers: Logistic Regression (LR), Linear Discriminant Analysis (LDA), K- Nearest Neighbors (KNN), Naive Bayes (NB), Support Vector Machine (SVM) and Random Forest (RF). The results show that the proposed model outperformed conventional classifiers, reaching an accuracy of 98%. The results highlight the importance of fusing models to improve the detection plant´s diseases.


  • Open Access

    OSM Sidewalkreator: A QGIS plugin for an automated drawing of sidewalk networks for OpenStreetMap

    by Kauê de Moraes Vestena, Silvana Philippi Camboim, Daniel Rodrigues dos Santos

    Sidewalks are a relevant part of the living space in urban environments but are still rarely mapped. In recent years, the mapping of sidewalks has grown in importance among the OSM and academic communities as a matter of concern for many UN SDGs. To cover this gap, we propose a GitHub-hosted, fully open-source QGIS Plugin entitled "OSM SidewalKreator" to automatically draw the geometries of sidewalks for OSM crossings and curb crossing interfaces. The plugin workflow encompasses the steps of input area selection; data fetching, data cleaning, sidewalk geometries generation; crossings and kerbs generation; optional sidewalk splitting; and data exporting. Furthermore, the tool gives the user the capacity to have control over the process. Our tests revealed that the proposed method embodied by the plugin surpasses the manual process in many contexts, highlighting completeness, topological and thematic accuracies. We conclude that deepening, improving, and increasing the amount of open sidewalk mapping, mainly in the widely available OSM, can be a valuable asset to improve the development of accessibility and mobility worldwide. Highlights: SIdewalks and other footpaths are generally rarely mapped despite their huge importance OSM SIdewalKreator QGIS PLugin is a product of the research, being a valuable support tool A deep analysis compared sidewalks produced with two methods, endorsing the method’s strengths


  • Open Access

    Collaboratively Updating National Cartography: An Automated Tool for Integrating OpenStreetMap Data into Topographic Mapping in Brazil

    by Leonardo Scharth Loureiro Silva, Silvana Philippi Camboim, Kauê de Moraes Vestena

    The availability and maintenance of accurate and up and implementing public policies, particularly in infrastructure, health, education, security, environmental governance, and social promotion. Also, it is essential to support the achievement of Sustainable Development Goals (SDGs). This need is intensified in developing countries due to several particularly challenging circumstances, where resources for official cartography may be scarce. For example, in Brazil, the country lacks comprehensive coverage of reference mapping at larger scales, and much of the currently available national mapping was produced decades ago. this problem, the authors propose investigating alternative which can be a low-cost alternative for updating national cartographic reference mapping. They focus on using data from the OpenStreetMap platform, which is rich in the amount of information available correspondences with the Brazilian official conceptual model. However, several operational issues integrate this data by official data producers effectively. For example, descriptions of utilizing such data (Sarretta and Minghini, 2021; Brovelli and Zamboni, 2018; Mahabir et al occasionally, these are more focused on theoretical aspects and do not address operational aspects effective for integration by official data producers. Thus, this study brings an innovative application to the geospatial community no similar approaches have been found in the literature. methodological questions. To this end, the authors have developed an automated integrates collaborative mapping data from OpenStreetMap into a reference database at a scale of 1:2 This development implements a methodological flow to automate the selection, analysis and import of collaborative data into the reference database in a QGIS plugin to accelerate and simplify the integration. The choice of a free platform already used by the national mapping agency is part of the strategy to integrate this activity into the daily tasks of cartographic updates. The first stage of development involved automating processes with tools written in Python. This phase comprised the implementation of information import flows using QGIS Graphical Modeler and the Overpass API to download data from OpenStreetMap. The imports were carried out for object classes of the official cartographic base of the State of Rio de Janeiro at the scale incorporated in a PostgreSQL/PostGIS database. eventual homologous in the reference database. The geometry of collabo no match. Similarly, names and attributes defined as "of interest" are added only when the reference information is is also worth mentioning that the flow preserves the OpenStreetMap import rec originates from the collaborative platform and keeps its element code (osm id). This link ensures that the entire information history is traceable and grants recognition of its authorship. This concern also ensures com OpenStreetMap's data licensing (OpenStreetMap which facilitated information integration by incorporating geometries and element attributes. Figure 1 shows a diagram of the workflow of the plugin. This automated methodology is of significant value, as it reduces the human interaction of repetitive, tedious and operator-sensitive activities, such as semantic compatibilization. Moreover, it can be reproduced and improved for having bee


  • Open Access

    Mapping Invisible and Inaccessible Areas of Brazilian Cities to Reduce Inequalities

    by Elias Nasr Naim Elias, Everton Bortolini, Jaqueline Alves Pisetta, Kauê de Moraes Vestena, Maurielle Felix da Silva, Nathan Damas, Silvana Philippi Camboim

    This account of the experiences of the first YouthMappers chapter in Brazil aims to present the work done by Mapeadores Livres UFPR focusing on the themes of accessibility (SDG 10 Reduced Inequalities) and favela mapping (SDG 11 Sustainable Cities and Communities), bringing as a transversal discussion of the quality of the geospatial data obtained collaboratively. Mapeadores Livres UFPR proposes to be a formative space in collaborative mapping, and the theme of favela mapping reinforces our character as an extension and research group and as a teaching group with the inclusion of this topic in the course developed. YouthMappers functions as an outreach project at the Universidade Federal do Paraná and is an interface between the universe of social organizations and their humanitarian projects and that of universities and their research topics.