In the last decade, volunteers have been contributing massively to what we know nowadays as Volunteered Geographic Information (VGI). Through the research that has been conducted recently, it has become clear that this huge amount of... more
In the last decade, volunteers have been contributing massively to what we know nowadays as Volunteered Geographic Information (VGI). Through the research that has been conducted recently, it has become clear that this huge amount of information might hide interesting and rich geographical information. The Open-StreetMap (OSM) project is one of the most well-known and studied VGI initia-tives. It has been studied to identify its potential for different applications. In the field of Land Use/Cover, an earlier study by the authors explored the use of OSM for Land Use/Cover (LULC) validation. Using the CORINE (COoRdination of IN-formation on the Environment) Land Cover (CLC) database as the Land Use ref-erence data, they analyzed the OSM coverage and classification accuracy, find-ing an interesting global accuracy value of 76.7% for level 1 land classes, for the study area of continental Portugal, despite a very small coverage value of ap-proximately 3.27%. In this chapter we review the existing literature on using OSM data for LULC database production and move this research forwards by explor-ing the suitability of the OSM Points of Interest dataset. We conclude that OSM can give very interesting contributions and that the OSM Points of Interest dataset is suitable for those classified as CLC class 1 which represents artificial surfaces.
Research Interests:
Volunteered Geographic Information has become exponentially available over the Web in the last years. This availability can hide a vast geographical richness and provides us with both a great opportunity to explore new ways to use it and... more
Volunteered Geographic Information has become exponentially available over the Web in the last years. This availability can hide a vast geographical richness and provides us with both a great opportunity to explore new ways to use it and also big challenges related with its unstructured nature. In this paper we conduct a preliminary analysis of the adequacy of photos from the Flickr initiative in order to use them as a source of field data in the quality control of the Land Use/Cover databases production. We evaluate its temporal and spatial distributions over Continental Portugal and also its distribution over Land Use/Cover classes using as a reference the European Corine Land Cover database. We conclude that this source is very valuable but needs to be combined with other sources due to some issues related with its spatial distribution.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Flood events are becoming more severe, causing significant problems to human communities, including physical, psychological, and material damage. For both flood forecasting in emergency response situations and flood mapping,... more
Flood events are becoming more severe, causing significant problems to human communities, including physical, psychological, and material damage. For both flood forecasting in emergency response situations and flood mapping, georeferencing and data curation are paramount in the context of prevention or preparedness. Hence, data display, data management, and articulation with numerical simulation results must occur on GIS platforms. Our research is motivated by recent advances in Web and GIS technologies, social sensing and high-performance computing, and an envisaged wider availability of remote sensing data. This paper presents and discusses an innovative Web GIS platform named "RiverCure Portal" or "RCP" for short. This platform combines observations and hydrodynamic modelling tools to support various stages of the flood risk management cycle, including operational response, emergency preparedness, and risk assessment. RCP is a multi-organisation, multi-context...
Research Interests:
Research Interests:
Research Interests:
Soil classification is a method of encoding the most relevant information about a given soil, namely its composition and characteristics, in a single class, to be used in areas like agriculture and forestry. In this paper, we evaluate how... more
Soil classification is a method of encoding the most relevant information about a given soil, namely its composition and characteristics, in a single class, to be used in areas like agriculture and forestry. In this paper, we evaluate how confidently we can predict soil classes, following the World Reference Base classification system, based on the physical and chemical characteristics of its layers. The Random Forests classifier was used with data consisting of 6 760 soil profiles composed by 19 464 horizons, collected in Mexico. Four methods of modelling the data were tested (i.e., standard depths, n first layers, thickness, and area weighted thickness). We also fine-tuned the best parameters for the classifier and for a k-NN imputation algorithm, used for addressing problems of missing data. Under-represented classes showed significantly worse results, by being repeatedly predicted as one of the majority classes. The best method to model the data was found to be the n first layers approach, with missing values being imputed with k-NN (\(k=1\)). The results present a Kappa value from 0.36 to 0.48 and were in line with the state of the art methods, which mostly use remote sensing data.
Research Interests:
This article describes a novel approach for toponym resolution with deep neural networks. The proposed approach does not involve matching references in the text against entries in a gazetteer, instead directly predicting geo-spatial... more
This article describes a novel approach for toponym resolution with deep neural networks. The proposed approach does not involve matching references in the text against entries in a gazetteer, instead directly predicting geo-spatial coordinates. Multiple inputs are considered in the neural network architecture (e.g., the surrounding words are considered in combination with the toponym to disambiguate), using pre-trained contextual word embeddings (i.e., ELMo or BERT) as well as bi-directional Long Short-Term Memory units, which are both regularly used for modeling textual data. The intermediate representations are then used to predict a probability distribution over possible geo-spatial regions, and finally to predict the coordinates for the input toponym. The proposed model was tested on three datasets used on previous toponym resolution studies, specifically the (i) War of the Rebellion, (ii) Local–Global Lexicon, and (iii) SpatialML corpora. Moreover, we evaluated the effect of u...
Research Interests:
The Open Data movement, aimed at stimulating economic growth, making social changes, encouraging citizen participation and promoting new forms of government accountability, has been growing worldwide during the last decade. Following this... more
The Open Data movement, aimed at stimulating economic growth, making social changes, encouraging citizen participation and promoting new forms of government accountability, has been growing worldwide during the last decade. Following this trend, Open Data initiatives are progressively being used by governments particularly to increase the transparency of public resource investments and their outcomes. Information and Communication Technology tools combined with data analysis techniques can be used to make these Open Data understandable to anyone interested, machine-readable, reusable, and a source for improved decision support, innovation, accountability and efficiency. This paper analyzes some Open Data initiatives within the Brazilian Government in order to verify if they meet the requirements of the Open Data definition. The results indicated that those requirements were all almost fulfilled, but there are still challenges to be overcome, such as greater update frequency, better license-free specification, and effective availability of all data which can be open to citizens. These results can help governments to identify the gaps that need to be addressed and make Open Data initiatives more effective.
Research Interests:
The semantic segmentation of high-resolution aerial images concerns the task of determining, for each pixel, the most likely class label from a finite set of possible labels (e.g., discriminating pixels referring to roads, buildings, or... more
The semantic segmentation of high-resolution aerial images concerns the task of determining, for each pixel, the most likely class label from a finite set of possible labels (e.g., discriminating pixels referring to roads, buildings, or vegetation, in high-resolution images depicting urban areas). Following recent work in the area related to the use of fully-convolutional neural networks for semantic segmentation, we evaluated the performance of an adapted version of the W-Net architecture, which has achieved very good results on other types of image segmentation tasks. Through experiments with two distinct datasets frequently used in previous studies in the area, we show that the proposed W-Net architecture is quite effective in this task, outperforming a baseline corresponding to the U-Net model, and also some of the other recently proposed approaches.
Research Interests:
Toponym resolution refers to the disambiguation of place names and other references to places present in textual documents, resolving them to unambiguous geographical identifiers (e.g., geographic coordinates of latitude and longitude).... more
Toponym resolution refers to the disambiguation of place names and other references to places present in textual documents, resolving them to unambiguous geographical identifiers (e.g., geographic coordinates of latitude and longitude). One of the major challenges in this task is that, usually, place names are highly ambiguous (e.g., there are several locations on the surface of the Earth that share the same name). In this paper, we propose to address the task through a recurrent neural network architecture with multiple inputs and outputs, specifically leveraging pre-trained contextual embeddings (ELMo) and bi-directional Long Short-Term Memory (LSTM) units, both commonly used for textual data modeling. The proposed model was tested on two datasets that were previously used to evaluate toponym resolution systems, namely the War of the Rebellion and the Local-Global Lexicon corpora. The obtained results outperform state-of-the-art results, confirming the superiority of the proposed method over other previous approaches.
Research Interests:
OSM2LULC is a software package developed to automatically convert OpenStreetMap (OSM) data into Land Use Land Cover (LULC) maps using Free and Open Source Software for Geospatial (FOSS4G) tools. It needs to be highly efficient given the... more
OSM2LULC is a software package developed to automatically convert OpenStreetMap (OSM) data into Land Use Land Cover (LULC) maps using Free and Open Source Software for Geospatial (FOSS4G) tools. It needs to be highly efficient given the increasing detail of OSM data and the need to apply it to large extent regions. In this article, a comparison between the implementation of OSM2LULC in different available GIS platforms is made using both vector and raster data structures, which resulted in different versions. A description of the differences of each version is made and, to assess their performance, they were applied to four different study areas with different characteristics, in terms of available OSM data and area size. The performance of each version was evaluated taking into account: the overall processing time required to obtain LULC maps; and differences in the results obtained when different data structures (vector and raster) were used. Results showed that the adoption of a ...
Research Interests:
Research Interests:
In the last decade, volunteers have been contributing massively to what we know nowadays as Volunteered Geographic Information (VGI). Through the research that has been conducted recently, it has become clear that this huge amount of... more
In the last decade, volunteers have been contributing massively to what we know nowadays as Volunteered Geographic Information (VGI). Through the research that has been conducted recently, it has become clear that this huge amount of information might hide interesting and rich geographical information. The OpenStreetMap (OSM) project is one of the most well-known and studied VGI initiatives. It has been studied to identify its potential for different applications. In the field of Land Use/Cover, an earlier study by the authors explored the use of OSM for Land Use/Cover (LULC) validation. Using the COoRdination of INformation on the Environment (CORINE) Land Cover (CLC) database as the Land Use reference data, they analyzed the OSM coverage and classification accuracy, finding an interesting global accuracy value of 76.7 % for level 1 land classes, for the study area of continental Portugal, despite a very small coverage value of approximately 3.27 %. In this chapter we review the existing literature on using OSM data for LULC database production and move this research forwards by exploring the suitability of the OSM Points of Interest dataset. We conclude that OSM can give very interesting contributions and that the OSM Points of Interest dataset is suitable for those classified as CLC class 1 which represents artificial surfaces.
Research Interests:
Research Interests:
Urban and economic development is affected by the value of land. Therefore, land value appraisal methods are important to determine the significance of land segments. A significant number of these models do not apply geospatial analysis... more
Urban and economic development is affected by the value of land. Therefore, land value appraisal methods are important to determine the significance of land segments. A significant number of these models do not apply geospatial analysis in their estimations. This paper presents a land valuation model based on spatial multi criteria decision analysis (SMCA) to assess land values in Abu Dhabi, UAE. Through this study, criteria maps were generated and combined to produce a final map ranking land segments based on their relative value.
Research Interests:
Research Interests:
Research Interests:
Land Cover mapping plays a very important role on a vast number of research fields, such as land use monitoring and modeling, monitoring of tropical deforestation, climate changes, among others. Its production is based on imagery... more
Land Cover mapping plays a very important role on a vast number of research fields, such as land use monitoring and modeling, monitoring of tropical deforestation, climate changes, among others. Its production is based on imagery interpretation and classification made by highly trained and skilled people. One of the most important phases is the assessment phase, where groups of people goes to the field, which makes it a very expensive one. We believe that it is possible to use Volunteered Geographic Information to help in the assessment phase of the Land Cover mapping production, making it less expensive. Thus, a robust data model/system is needed for a proper management of this type of data.
Research Interests:
Research Interests:
Research Interests:
This study demonstrated the potentiality of the use of remote sensing and geospatial techniques for urban climate analysis with a case study of Abu Dhabi city area. A first analysis assessed the variation of Land Surface Temperature (LST)... more
This study demonstrated the potentiality of the use of remote sensing and geospatial techniques for urban climate analysis with a case study of Abu Dhabi city area. A first analysis assessed the variation of Land Surface Temperature (LST) at different spatial scales its relationship with land cover, Normalized Difference Vegetation Index (NDVI), and percentage of Impervious Surface Area (ISA) has also been investigated for the period between 2000 and 2010 using data from multiple satellites and ground measurements. In the second analysis, the data were assimilated in a Surface Energy Balance (SEB) model to assess urban energy fluxes in Abu Dhabi metropolitan area during the winter and the summer seasons. The maps produced from LST and SEB projects will be disseminated through the WebGIS Portal of the Research Center for Renewable Energy Mapping and Assessment.
Research Interests:
We present a novel deep learning approach for spatio-temporal forecasting with remote sensing data, extending a previous model named Spatio-Temporal Convolutional Sequence to Sequence Network (STConvS2S) in several directions. Experiments... more
We present a novel deep learning approach for spatio-temporal forecasting with remote sensing data, extending a previous model named Spatio-Temporal Convolutional Sequence to Sequence Network (STConvS2S) in several directions. Experiments using datasets from previous studies show that the proposed approaches outperform the original STConvS2S and other baseline models on tasks related to predicting future time-steps. In tests related to predicting a missing time-step, some of the proposed extensions also lead to improvements over the original STConvS2S architecture, although simpler models seem to be beneficial in this scenario.
Research Interests:
Research Interests:
The characterization of the of harmful algal blooms (HABs), commonly called red-tide, in terms of location, time of occurrence, and concentration is required to manage the desalination plants over the Arabian Gulf. Several field campaigns... more
The characterization of the of harmful algal blooms (HABs), commonly called red-tide, in terms of location, time of occurrence, and concentration is required to manage the desalination plants over the Arabian Gulf. Several field campaigns have been undertaken over the Arabian Gulf by different organizations such as Regional Organization for the Protection of the Marine Environment (ROPME) and Environment Agency of Abu Dhabi (EAD) to collect the water temperature and chlorophyll samples. In this paper, a geospatial analysis is considered in order to map the distribution of chlorophyll in the Arabian Gulf using the kriging estimators. These maps will allow the identification of the areas mostly affected by the red-tide, periods of occurrences and their concentrations. The cross comparison between chlorophyll and dust concentration obtained from MODIS satellite and some coastal AERONET stations will be considered to clarify the dependency between them since dust contain nutrients which...
Research Interests:
Volunteered Geographic Information has been increasing exponentially over the last years, capturing the attention of the scientific community. Researchers have been very active exploring a vast amount of initiatives and trying to develop... more
Volunteered Geographic Information has been increasing exponentially over the last years, capturing the attention of the scientific community. Researchers have been very active exploring a vast amount of initiatives and trying to develop methodologies and possible real applications for this new source of geographic information. Land Use/Cover production is one of the areas where this type of geographic information might be very useful. In this paper we evaluate if geo-referenced and publicly available photos from the Flickr initiative can be used as a source of geographic information to help Land Use/Cover classification. Using the Corine Land Cover nomenclature, we compare the classification obtained for selected photo locations, against the classification obtained from high resolution satellite imagery for the same locations. We conclude that this source cannot be used alone for the purpose of Land Use/Cover classification but we also believe that it might contain helpful informat...
Research Interests:
Research Interests:
O presente artigo visa rever o conceito de Geomarketing e relaciona-lo no contexto empresarial e na sua aplicacao pratica. O Geomarketing reflete-se num conjunto de tecnicas e ferramentas que permitem analisar geograficamente a realidade... more
O presente artigo visa rever o conceito de Geomarketing e relaciona-lo no contexto empresarial e na sua aplicacao pratica. O Geomarketing reflete-se num conjunto de tecnicas e ferramentas que permitem analisar geograficamente a realidade socioeconomica de uma organizacao. Estas tecnicas e ferramentas estao disponiveis atraves de Sistemas de Informacao Geografica (SIG) que permitem realizar analises envolvendo informacao com carater geografico, juntamente com outro tipo de informacao relacionada com as organizacoes envolvidas. Exemplos de aplicacao como o posicionamento em relacao a concorrencia, otimizacao de rotas, distribuicao de servicos e o conhecimento dos habitos do consumidor, sao fatores padrao no funcionamento aplicavel deste conceito. Tornando-se vital conhecer a aplicabilidade e reconhecer os beneficios do geomarketing, pode contribuir de forma significativa para o sucesso das organizacoes. Devido a escassez de literatura disponivel, continua a ser pertinente o desenvolvi...
Research Interests:
Assessing the data quality of Volunteered Geographic Information (VGI) is important for determining the fitness-for-use of VGI for different applications. This paper provides guidelines for good practice that may assist in quality... more
Assessing the data quality of Volunteered Geographic Information (VGI) is important for determining the fitness-for-use of VGI for different applications. This paper provides guidelines for good practice that may assist in quality assessment, in particular recommendations on additional data that could be used and procedures that could be implemented that will facilitate the assessment of VGI quality. Finally, the role of protocols is discussed in terms of how they might improve data quality assessment.
The severity and impacts of forest fires have increased in the last years in several parts of the world, where devastating fires occur now almost every year. As these types of events are likely to increase due to climate changes, it is... more
The severity and impacts of forest fires have increased in the last years in several parts of the world, where devastating fires occur now almost every year. As these types of events are likely to increase due to climate changes, it is important to develop tools to assist authorities in the early identification and geolocation of ignitions so that they can be tackled as fast as possible. Several types of systems are currently being used, and others are under development, to automatically detect fires, based on, for example, thermal cameras and observation points; even so, the more systems available to detect these events and identify their location at an early stage, the better. This fact motivated the FireLoc project, currently under implementation, which aims to develop a system that will enable citizens to provide georeferenced data allowing the detection and geolocation of spotted fires in real time. The FireLoc system includes a dedicated app that enables citizens to report a s...
A critical goal, is that organizations and citizens can easily access the geographic information required for good governance. However, despite the costly efforts of governments to create and implement Spatial Data Infrastructures (SDIs),... more
A critical goal, is that organizations and citizens can easily access the geographic information required for good governance. However, despite the costly efforts of governments to create and implement Spatial Data Infrastructures (SDIs), this goal is far from being achieved. This is partly due to the lack of usability of the geoportals through which the geographic information is accessed. In this position paper, we present IDEAIS, a research network composed of multiple Ibero-American partners to address this usability issue through the use of Intelligent Systems, in particular Smart Voice Assistants, to efficiently recover and access geographic information.
In this work, a prototype GIS-based platform to integrate Volunteered Geographic Information from various sources with other spatial data is presented, aiming at assisting civil protection authorities in emergency response situations. The... more
In this work, a prototype GIS-based platform to integrate Volunteered Geographic Information from various sources with other spatial data is presented, aiming at assisting civil protection authorities in emergency response situations. The platform is now in the implementation phase, and this paper covers some aspects about its development and preliminary results to demonstrate the potentialities of the approach proposed.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Geographic information has been traditionally produced by mapping agencies and corporations, using highly skilled professionals as well as expensive precision equipment and procedures, in a very costly approach. The production of land use... more
Geographic information has been traditionally produced by mapping agencies and corporations, using highly skilled professionals as well as expensive precision equipment and procedures, in a very costly approach. The production of land use and land cover databases is just one example of such traditional approaches. At the same time, the amount of Geographic Information created and shared by citizens through the web has been increasing exponentially during the last decade as a result of the emergence and popularization of technologies such as the Web 2.0, cloud computing, global positioning systems (GPS), smart phones, among others. This vast amount of free geographic data might have valuable information to extract. Combining data from several initiatives might further increase the value of such data. We propose a conceptual model to integrate data from suitable user generated spatial content initiatives. A prototype to demonstrate the ability of the model to perform such integration, based on two identified use cases, was also developed.
Research Interests:
Research Interests:
Research Interests:
With the Internet growth, the production and amount of available geographic data also increased. There are, at the moment, many servers around the world providing spatial information to several users, and many of them are accessing this... more
With the Internet growth, the production and amount of available geographic data also increased. There are, at the moment, many servers around the world providing spatial information to several users, and many of them are accessing this information through different Geographic Information Systems (GIS) applications. Therefore, it becomes necessary to develop standards in order to solve the interoperability problem inherent to this heterogeneous and distributed data. This work aims at showing the significance of standards, particularly the Web Feature Service - Transaction (WFS-T) standard in the Geographic Information (GI) domain. A prototype that provides GI according with this standard is also implemented, based on PostgreSQL/PostGIS, GeoServer, uDig, Apache and MapFish.