Resumo. The estimation of forest and fuel variables by using LiDAR data and CIR high-resolution i... more Resumo. The estimation of forest and fuel variables by using LiDAR data and CIR high-resolution images is the main objective of the project PTDC/AGR-CFL/72380/2006 financed by the Portuguese foundation Fundação para a Ciência e Tecnologia (FCT). To estimate those variables a novel software system needs to be developed and tested. To this end, control data are necessary. In this article it is addressed the planning and acquisition of traditional forest inventory data in pure eucalyptus strata as well as topographic data. The latter is acquired with the classical topographic methods by means of a total station and by using global positioning techniques with the help of high precision GPS receivers.
The assessment and management of flood risks framework impose the mapping of flood hazard in pote... more The assessment and management of flood risks framework impose the mapping of flood hazard in potential flood risks areas. Floods in urban environments may happen due to rainfall extreme events and be exacerbated by saturated or impervious surfaces. Flood risk is greater in urban areas. The urban flood inundation modelling needs high resolution and accurate geospatial data and also measurements of water depth and flood extent to validate the model. The methods to integrate high resolution geospatial data to derive input to the flood models and their impact on the model output are still a research issue. These topics are addressed in this work. The flood inundation model was applied to a small urban catchment within Vila Nova de Gaia city, at the northern part of Portugal, where it has been observed several flood events in a recent past. The area to be modelled comprises a rectangle of 0.8 km × 1.0 km. The topography, topology and land cover of the study site were characterized with the help of a 1 m x 1m resolution LiDAR Digital Surface Model (DSM), high resolution colour infrared images with a pixel of 0.5 m and digital maps at 1:2000 scale with the buildings locations, road network and land use classes. The LISFLOOD-FP, Version 4.5, a raster-based model for the modelling of urban flood inundation developed by Bates & De Roo, (2000) was used in this work. Several improvements to the model were carried out subsequently (Horritt & Bates, 2001; Hunter et al., 2005). The upstream boundary is a hydrograph calculated by the Hydrologic Engineering Center-Hydrologic Modelling System (HEC-HMS) model, version 3.5. In order to determine the peak flood discharge the Soil Conservation Service method is applied as a consistent methodology for ungauged small basins. The design hydrograph was built from the 100-year return period dimensionless hydrograph. The flood inundation models require input data such as terrain morphology and friction coefficients, which have to be derived from raw data by using adequate methods. The Digital Terrain Model (DTM) is produced by filtering non-terrain features, such as vegetation and buildings, from the Digital Surface Model (DSM) followed by the spatial interpolation among the terrain points. A normalized Digital Surface Model (nDSM) gives the non-terrain feature height values and results from subtracting the DTM to the DSM. The vegetation is extracted from the high-resolution images by the Normalized Difference Vegetation Index. The produced vegetation map is intersected with the nDSM to obtain the vegetation height. In this new map, the cells with a height value greater than 2 m are removed to obtain a Digital Surface Flow Model (DSM f1). Other Digital Surface Flow model (DSM f2) is produced which contains the buildings. This is done by inserting the buildings locations, computed from topographic maps into the DTM. The areas demarked as buildings are raised by 12 m. The spatially distributed friction coefficients are assigned to the land use classes extracted from large scale maps. For this study, the values of each friction class were taken from literature (
Uneven illumination is a recurrent problem in image processing. This is essentially due to image ... more Uneven illumination is a recurrent problem in image processing. This is essentially due to image acquisition sensors' malfunction or external interference. In this paper we propose a variational model for nonuniform illumination correction, that incorporates a penalty term that performs the intensity distribution transfer between two pre-defined sub-regions of the input scalar image, one uniformly illuminated and the other nonuniformly illuminated. This term representing the illumination correction is a Wasserstein distance. It corresponds to the optimal permutation minimizing the cost of rearranging the intensity distribution of the nonuniformly illuminated sub-region into the other, the uniformly illuminated. Simultaneously, this variational model also carries out a regularization of the image by means of a total variation penalty term, to reduce noise. The effectiveness of the model is illustrated for some images.
Digital agriculture has evolved significantly over the last few years due to the technological de... more Digital agriculture has evolved significantly over the last few years due to the technological developments in automation and computational intelligence applied to the agricultural sector, including vineyards which are a relevant crop in the Mediterranean region. In this work, a study is presented of semantic segmentation for vine detection in real-world vineyards by exploring state-of-the-art deep segmentation networks and conventional unsupervised methods. Camera data have been collected on vineyards using an Unmanned Aerial System (UAS) equipped with a dual imaging sensor payload, namely a high-definition RGB camera and a five-band multispectral and thermal camera. Extensive experiments using deep-segmentation networks and unsupervised methods have been performed on multimodal datasets representing four distinct vineyards located in the central region of Portugal. The reported results indicate that SegNet, U-Net, and ModSegNet have equivalent overall performance in vine segmentation. The results also show that multimodality slightly improves the performance of vine segmentation, but the NIR spectrum alone generally is sufficient on most of the datasets. Furthermore, results suggest that high-definition RGB images produce equivalent or higher performance than any lower resolution multispectral band combination. Lastly, Deep Learning (DL) networks have higher overall performance than classical methods. The code and dataset are publicly available on https://github.com/Cybonic/DL_vi neyard_segmentation_study.git.
Unmanned Aerial Systems (UAS, aka drones) are being used to map marine macro-litter on the coast.... more Unmanned Aerial Systems (UAS, aka drones) are being used to map marine macro-litter on the coast. Within the UAS4Litter project, the application of UAS has been applied on three sandy beachdune systems on the wave-dominated North Atlantic Portuguese coast. Several technical solutions have been tested in terms of drone mapping performance, manual image screening and marine litter map analysis. The conceptualization and implementation of a multidisciplinary framework allowed to improve and making more efficient the mapping of marine litter items with UAS on coastal environment.
Unmanned aerial vehicles (UAV, aka drones) are being used for mapping macro-litter in the environ... more Unmanned aerial vehicles (UAV, aka drones) are being used for mapping macro-litter in the environment. As drone images require a manual processing task for detecting marine litter, it is of interest to evaluate the accuracy of non-expert citizen science operators (CSO) in performing this task. Students from Italian secondary schools (in this work, the CSO) were invited to identify, mark, and classify stranded litter items on a UAV orthophoto collected on an Italian beach. A specific training program and working tools were developed for the aim. The comparison with the standard in situ visual census survey returned a general underestimation (50%) of items. However, marine litter bulk categorisation was fairly in agreement with the in situ survey, especially for sources classification. The concordance level among CSO ranged between 60% and 91%, depending on the item properties considered (type, material, and colour). As the assessment accuracy was in line with previous works developed...
The European Parliament Directive 2007/60/EC on the assessment and management of flood risks defi... more The European Parliament Directive 2007/60/EC on the assessment and management of flood risks defines that the states shall prepare flood hazard maps at the most appropriate scale, for the areas with potential significant flood risks. Flood modelling requires several types of input data like topographic data and friction coefficient values. These are considered to be the key data sets for flow drainage and flood modelling in urban areas. The Airborne Laser Scanning (ALS) or LiDAR (Light Detection And Ranging) is one technique used to obtain the topographic data of flood areas and cross sections of a river. This data source can be used to produce highresolution Digital Terrain Model (DTM). The integration of DTM generated from laser scanning altimetry into flood modelling is yet a research topic, because the LiDAR data require a considerable pre-processing effort before their use within hydraulic modelling. Furthermore, it is also necessary to determine the optimal spatial resolution ...
The scientific community involved in the UN-REDD program is still reporting large uncertainties a... more The scientific community involved in the UN-REDD program is still reporting large uncertainties about the amount and spatial variability of CO 2 stored in forests. The main limitation has been the lack of field samplings over space and time needed to calibrate and convert remote sensing measurements into aboveground biomass (AGB). As an alternative to costly field inventories, we examine the reliability of state-of-the-art lidar methods to provide direct retrieval of many forest metrics that are commonly collected through field sampling techniques (e.g., tree density, individual tree height, crown cover). AGB is estimated using existing allometric equations that are fed by lidar-derived metrics at either the individual tree-or forest layer-level (for the overstory or underneath layers, respectively). Results over 40 plots of a multilayered forest located in northwest Portugal show that the lidar method provides AGB estimates with a relatively small random error (RMSE = of 17.1%) and bias (of 4.6%). It provides local AGB baselines that meet the requirements in terms of accuracy to calibrate satellite remote sensing measurements (e.g., the upcoming lidar GEDI (Global Ecosystem Dynamics Investigation), and the Synthetic Aperture Radar (SAR) missions NISAR (National Aeronautics and Space Administration and Indian Space Research Organization SAR) and BIOMASS from the European Space Agency, ESA) for AGB mapping purposes. The development of similar techniques over a variety of forest types would be a significant improvement in quantifying CO 2 stocks and changes to comply with the UN-REDD policies.
Maria do Carmo Bezerra & Mona-Lisa Choas artigo 39 integração de modeloS de SimUlação em Sig: apl... more Maria do Carmo Bezerra & Mona-Lisa Choas artigo 39 integração de modeloS de SimUlação em Sig: aplicação ao caSo da drenagem de ágUaS plUviaiS UrBanaS.
Surface water flooding occurs due to extreme rainfall and the inability of the sewer system to dr... more Surface water flooding occurs due to extreme rainfall and the inability of the sewer system to drain all runoff. As a consequence, a considerable volume of water is carried out over the surface through preferential flow paths and can eventually accumulate in natural (or man-made) ponds. This can cause minor material losses but also major incidents with obvious consequences in
This study presents a robust approach for characterization of multi-layered forests using airborn... more This study presents a robust approach for characterization of multi-layered forests using airborne laser scanning (ALS) data. Fuel mapping or biomass estimation requires knowing the diversity and boundaries of the forest patches, as well as their spatial pattern. This includes the thickness of the main vegetation layers, but also the spatial arrangement and size of the individual plants that compose each stratum. In order to decompose the ALS point cloud into genuine 3-D segments corresponding to individual vegetation features, such as shrubs or tree crowns, we apply a statistical approach based on the mean shift algorithm. The segments are progressively assigned to a forest layer: ground vegetation, understory or overstory. Our method relies on a single biophysically meaningful parameter, the kernel bandwidth, which is related to the local forest structure. It is validated on 44 plots of a Portuguese forest, composed mainly of eucalyptus (Eucalyptus globulus Labill.) and maritime pine (Pinus pinaster Ait.) trees. The number of detected trees varies with the dominance position: from 98.6% for the dominant trees to 12.8% for the suppressed trees. Linear regression models explain up to 70% of the variability associated with ground vegetation and understory height.
Plant communities display a vertical structure based on the size and growth pattern of the domina... more Plant communities display a vertical structure based on the size and growth pattern of the dominant species. To a large extent, this pattern, called vertical stratification, depends on the climatic zone. Vertical structure analysis consists in detecting the number of layers and their limits within a forest stand. So far, there is a lack of robust approaches applied to airborne laser scanning (ALS) data that properly segment the different strata of forests having complex structures. In this study, we propose a procedure to characterize vertical forest stratification based on the mean shift (MS) algorithm. The MS is a non-linear filter that searches for local density maxima (modes). It is a non-parametric and unsupervised approach, which only requires a single criterion, the kernel bandwidth. Since the forest point cloud is a multi-modal distribution, the MS is used to find the modes which are supposed to be the barycenters of vegetation features. Once achieved, the modes are grouped together according to height range and the corresponding ALS points are assigned to each vegetation strata. Due to their complex pattern, using a single scale over the whole space is not recommended for the analysis of such environments. On this basis, the modes are computed using a variable kernel bandwidth according to the forest pattern. To depict such a pattern, we propose a new technique that segments the main forest layers at the plot level: overstory, understory, and surface vegetation. The procedure has been carried out on 45 plots of a Portuguese forest mainly composed of eucalyptus (Eucalyptus globulus) and pine (Pinus pinaster) trees that can be strongly populated by understory and surface vegetation.
IEEE Transactions on Geoscience and Remote Sensing, 2015
The canopy density model (CDM), a new product interpolated from airborne laser scanner (ALS) data... more The canopy density model (CDM), a new product interpolated from airborne laser scanner (ALS) data and dedicated to forest structure characterization is presented. It exploits both the multiecho capability of the ALS and a nonparametric density estimation technique called kernel density estimators (KDEs). The CDM is used to delineate the outmost perimeter of vegetation features and to compute forest crown cover (CrCO). Contrary to other works that focus on single-layer forest canopies, CrCo is derived here for each layer, namely, the overstory, the understory, and ground vegetation. The root-mean-square error of prediction determined by using field data acquired over 44 forest stands in a forest in Portugal allows the testing of the reliability of the method: It ranges from 6.21% (overstory) to 13.76% (ground vegetation). In addition, we investigate the ability of the CDM to map the CrCo for individual trees. Finally, two existing methods have been applied to our study site in order to assess improvements, advantages, and drawbacks of our approach.
O perigo de inundação é um componente da avaliação do risco de inundação. Um dos elementos fundam... more O perigo de inundação é um componente da avaliação do risco de inundação. Um dos elementos fundamentais na avaliação do perigo de inundação é a estimativa da extensão da zona inundável. Neste trabalho é estudado o efeito da resolução espacial do Modelo Digital de Superfície de Escoamento (MDSe) na extensão da zona inundável. São criados três MDSe de diferentes resoluções que resultam da integração de dados LiDAR, cartografia a escala grande e ortoimagens CIR (Colour Infra-Red). Os dados LiDAR, devidamente filtrados dos objetos no terreno, são convertidos em três Modelos Digitais de Terreno (MDT) em formato raster com 1 m, 2 m, e 4 m. Da cartografia é extraído o edificado na forma de um raster. A vegetação com altura inferior a 2 m é por sua vez extraída das ortoimagens CIR. Os rasters dos objetos e vegetação são redimensionados para células de 1 m, 2 m e 4 m. Os três MDSe assim obtidos são utilizados na simulação hidráulica, com a aplicação do modelo LISFLOOD-FP e com o objetivo de ...
A extracção de Modelos Digitais de Terreno (MDT) em zonas urbanas a partir de Modelos Digitais de... more A extracção de Modelos Digitais de Terreno (MDT) em zonas urbanas a partir de Modelos Digitais de Superfície (MDS) é ainda um problema com sucesso limitado devido ao facto da operação de filtragem, ao utilizar apenas a informação geométrica contida no MDS, não permitir distinguir eficazmente todos os objectos situados acima da superfície topográfica que define o MDT. Para ultrapassar os problemas propõe-se uma nova metodologia que, pela integração dos dados obtidos por varrimento aéreo com laser (LiDAR) com orto-imagens CIR (Color Infra-Red, ou ainda falsa cor) de alta resolução, permite a extracção de informação precisa e fiável do terreno e dos objectos considerados não pertencentes ao MDT (edifícios e vegetação alta). Os resultados obtidos mostram que em zonas urbanas complexas a metodologia proposta tem um desempenho superior à metodologia convencional
Abstract. The research Project PTDC/AGR-CFL/72380/2006, entitled "Forest and Fuel Variables ... more Abstract. The research Project PTDC/AGR-CFL/72380/2006, entitled "Forest and Fuel Variables Estimation and Digital Terrain Modelling with Airborne Laser Scanning and High Resolution Multi-Spectral Images", financed by the Portuguese foundation Fundação para a Ciência ...
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
In the context of coastal environment, the combined use of Unmanned Aerial Systems (UAS) with Str... more In the context of coastal environment, the combined use of Unmanned Aerial Systems (UAS) with Structure-from-Motion (SfM) photogrammetry has been demonstrated to be a suitable tool to collect accurate geoinformation. This work aims at investigating the effectiveness of UAS survey to accurately map the geometry of rouble mound groins for monitoring the structural integrity. A set of nadiral images was processed through SfM photogrammetry to obtain the Digital Surface Model (DSM) of the groin. The DSM accuracy was evaluated through the comparison of five profiles measured by the traditional GNSS technique. Overall, vertical accuracy returned a root mean square error of 2.6 cm. The results suggest that UAS combined with SfM photogrammetry can improve the current monitoring of coastal engineering structures by providing high accurate geospatial products.
The abundance of litter pollution in the marine environment has been increasing globally. Remote ... more The abundance of litter pollution in the marine environment has been increasing globally. Remote sensing techniques are valuable tools to advance knowledge on litter abundance, distribution and dynamics. Images collected by Unmanned Aerial Vehicles (UAV, aka drones) are highly efficient to map and monitor local beached (BL) and floating (FL) marine litter items. In this work, the operational insights to carry out both BL and FL surveys using UAVs are detailly described. In particular, flight planning and deployment, along with image products processing and analysis, are reported and compared. Furthermore, analogies and differences between UAV-based BL and FL mapping are discussed, with focus on the challenges related to BL and FL item detection and recognition. Given the efficiency of UAV to map BL and FL, this remote sensing technique can replace traditional methods for litter monitoring, further improving the knowledge of marine litter dynamics in the marine environment. This comm...
Unmanned aerial systems (UASs) have recently been proven to be valuable remote sensing tools for ... more Unmanned aerial systems (UASs) have recently been proven to be valuable remote sensing tools for detecting marine macro litter (MML), with the potential of supporting pollution monitoring programs on coasts. Very low altitude images, acquired with a low-cost RGB camera onboard a UAS on a sandy beach, were used to characterize the abundance of stranded macro litter. We developed an object-oriented classification strategy for automatically identifying the marine macro litter items on a UAS-based orthomosaic. A comparison is presented among three automated object-oriented machine learning (OOML) techniques, namely random forest (RF), support vector machine (SVM), and k-nearest neighbor (KNN). Overall, the detection was satisfactory for the three techniques, with mean F-scores of 65% for KNN, 68% for SVM, and 72% for RF. A comparison with manual detection showed that the RF technique was the most accurate OOML macro litter detector, as it returned the best overall detection quality (F-s...
Resumo. The estimation of forest and fuel variables by using LiDAR data and CIR high-resolution i... more Resumo. The estimation of forest and fuel variables by using LiDAR data and CIR high-resolution images is the main objective of the project PTDC/AGR-CFL/72380/2006 financed by the Portuguese foundation Fundação para a Ciência e Tecnologia (FCT). To estimate those variables a novel software system needs to be developed and tested. To this end, control data are necessary. In this article it is addressed the planning and acquisition of traditional forest inventory data in pure eucalyptus strata as well as topographic data. The latter is acquired with the classical topographic methods by means of a total station and by using global positioning techniques with the help of high precision GPS receivers.
The assessment and management of flood risks framework impose the mapping of flood hazard in pote... more The assessment and management of flood risks framework impose the mapping of flood hazard in potential flood risks areas. Floods in urban environments may happen due to rainfall extreme events and be exacerbated by saturated or impervious surfaces. Flood risk is greater in urban areas. The urban flood inundation modelling needs high resolution and accurate geospatial data and also measurements of water depth and flood extent to validate the model. The methods to integrate high resolution geospatial data to derive input to the flood models and their impact on the model output are still a research issue. These topics are addressed in this work. The flood inundation model was applied to a small urban catchment within Vila Nova de Gaia city, at the northern part of Portugal, where it has been observed several flood events in a recent past. The area to be modelled comprises a rectangle of 0.8 km × 1.0 km. The topography, topology and land cover of the study site were characterized with the help of a 1 m x 1m resolution LiDAR Digital Surface Model (DSM), high resolution colour infrared images with a pixel of 0.5 m and digital maps at 1:2000 scale with the buildings locations, road network and land use classes. The LISFLOOD-FP, Version 4.5, a raster-based model for the modelling of urban flood inundation developed by Bates & De Roo, (2000) was used in this work. Several improvements to the model were carried out subsequently (Horritt & Bates, 2001; Hunter et al., 2005). The upstream boundary is a hydrograph calculated by the Hydrologic Engineering Center-Hydrologic Modelling System (HEC-HMS) model, version 3.5. In order to determine the peak flood discharge the Soil Conservation Service method is applied as a consistent methodology for ungauged small basins. The design hydrograph was built from the 100-year return period dimensionless hydrograph. The flood inundation models require input data such as terrain morphology and friction coefficients, which have to be derived from raw data by using adequate methods. The Digital Terrain Model (DTM) is produced by filtering non-terrain features, such as vegetation and buildings, from the Digital Surface Model (DSM) followed by the spatial interpolation among the terrain points. A normalized Digital Surface Model (nDSM) gives the non-terrain feature height values and results from subtracting the DTM to the DSM. The vegetation is extracted from the high-resolution images by the Normalized Difference Vegetation Index. The produced vegetation map is intersected with the nDSM to obtain the vegetation height. In this new map, the cells with a height value greater than 2 m are removed to obtain a Digital Surface Flow Model (DSM f1). Other Digital Surface Flow model (DSM f2) is produced which contains the buildings. This is done by inserting the buildings locations, computed from topographic maps into the DTM. The areas demarked as buildings are raised by 12 m. The spatially distributed friction coefficients are assigned to the land use classes extracted from large scale maps. For this study, the values of each friction class were taken from literature (
Uneven illumination is a recurrent problem in image processing. This is essentially due to image ... more Uneven illumination is a recurrent problem in image processing. This is essentially due to image acquisition sensors' malfunction or external interference. In this paper we propose a variational model for nonuniform illumination correction, that incorporates a penalty term that performs the intensity distribution transfer between two pre-defined sub-regions of the input scalar image, one uniformly illuminated and the other nonuniformly illuminated. This term representing the illumination correction is a Wasserstein distance. It corresponds to the optimal permutation minimizing the cost of rearranging the intensity distribution of the nonuniformly illuminated sub-region into the other, the uniformly illuminated. Simultaneously, this variational model also carries out a regularization of the image by means of a total variation penalty term, to reduce noise. The effectiveness of the model is illustrated for some images.
Digital agriculture has evolved significantly over the last few years due to the technological de... more Digital agriculture has evolved significantly over the last few years due to the technological developments in automation and computational intelligence applied to the agricultural sector, including vineyards which are a relevant crop in the Mediterranean region. In this work, a study is presented of semantic segmentation for vine detection in real-world vineyards by exploring state-of-the-art deep segmentation networks and conventional unsupervised methods. Camera data have been collected on vineyards using an Unmanned Aerial System (UAS) equipped with a dual imaging sensor payload, namely a high-definition RGB camera and a five-band multispectral and thermal camera. Extensive experiments using deep-segmentation networks and unsupervised methods have been performed on multimodal datasets representing four distinct vineyards located in the central region of Portugal. The reported results indicate that SegNet, U-Net, and ModSegNet have equivalent overall performance in vine segmentation. The results also show that multimodality slightly improves the performance of vine segmentation, but the NIR spectrum alone generally is sufficient on most of the datasets. Furthermore, results suggest that high-definition RGB images produce equivalent or higher performance than any lower resolution multispectral band combination. Lastly, Deep Learning (DL) networks have higher overall performance than classical methods. The code and dataset are publicly available on https://github.com/Cybonic/DL_vi neyard_segmentation_study.git.
Unmanned Aerial Systems (UAS, aka drones) are being used to map marine macro-litter on the coast.... more Unmanned Aerial Systems (UAS, aka drones) are being used to map marine macro-litter on the coast. Within the UAS4Litter project, the application of UAS has been applied on three sandy beachdune systems on the wave-dominated North Atlantic Portuguese coast. Several technical solutions have been tested in terms of drone mapping performance, manual image screening and marine litter map analysis. The conceptualization and implementation of a multidisciplinary framework allowed to improve and making more efficient the mapping of marine litter items with UAS on coastal environment.
Unmanned aerial vehicles (UAV, aka drones) are being used for mapping macro-litter in the environ... more Unmanned aerial vehicles (UAV, aka drones) are being used for mapping macro-litter in the environment. As drone images require a manual processing task for detecting marine litter, it is of interest to evaluate the accuracy of non-expert citizen science operators (CSO) in performing this task. Students from Italian secondary schools (in this work, the CSO) were invited to identify, mark, and classify stranded litter items on a UAV orthophoto collected on an Italian beach. A specific training program and working tools were developed for the aim. The comparison with the standard in situ visual census survey returned a general underestimation (50%) of items. However, marine litter bulk categorisation was fairly in agreement with the in situ survey, especially for sources classification. The concordance level among CSO ranged between 60% and 91%, depending on the item properties considered (type, material, and colour). As the assessment accuracy was in line with previous works developed...
The European Parliament Directive 2007/60/EC on the assessment and management of flood risks defi... more The European Parliament Directive 2007/60/EC on the assessment and management of flood risks defines that the states shall prepare flood hazard maps at the most appropriate scale, for the areas with potential significant flood risks. Flood modelling requires several types of input data like topographic data and friction coefficient values. These are considered to be the key data sets for flow drainage and flood modelling in urban areas. The Airborne Laser Scanning (ALS) or LiDAR (Light Detection And Ranging) is one technique used to obtain the topographic data of flood areas and cross sections of a river. This data source can be used to produce highresolution Digital Terrain Model (DTM). The integration of DTM generated from laser scanning altimetry into flood modelling is yet a research topic, because the LiDAR data require a considerable pre-processing effort before their use within hydraulic modelling. Furthermore, it is also necessary to determine the optimal spatial resolution ...
The scientific community involved in the UN-REDD program is still reporting large uncertainties a... more The scientific community involved in the UN-REDD program is still reporting large uncertainties about the amount and spatial variability of CO 2 stored in forests. The main limitation has been the lack of field samplings over space and time needed to calibrate and convert remote sensing measurements into aboveground biomass (AGB). As an alternative to costly field inventories, we examine the reliability of state-of-the-art lidar methods to provide direct retrieval of many forest metrics that are commonly collected through field sampling techniques (e.g., tree density, individual tree height, crown cover). AGB is estimated using existing allometric equations that are fed by lidar-derived metrics at either the individual tree-or forest layer-level (for the overstory or underneath layers, respectively). Results over 40 plots of a multilayered forest located in northwest Portugal show that the lidar method provides AGB estimates with a relatively small random error (RMSE = of 17.1%) and bias (of 4.6%). It provides local AGB baselines that meet the requirements in terms of accuracy to calibrate satellite remote sensing measurements (e.g., the upcoming lidar GEDI (Global Ecosystem Dynamics Investigation), and the Synthetic Aperture Radar (SAR) missions NISAR (National Aeronautics and Space Administration and Indian Space Research Organization SAR) and BIOMASS from the European Space Agency, ESA) for AGB mapping purposes. The development of similar techniques over a variety of forest types would be a significant improvement in quantifying CO 2 stocks and changes to comply with the UN-REDD policies.
Maria do Carmo Bezerra & Mona-Lisa Choas artigo 39 integração de modeloS de SimUlação em Sig: apl... more Maria do Carmo Bezerra & Mona-Lisa Choas artigo 39 integração de modeloS de SimUlação em Sig: aplicação ao caSo da drenagem de ágUaS plUviaiS UrBanaS.
Surface water flooding occurs due to extreme rainfall and the inability of the sewer system to dr... more Surface water flooding occurs due to extreme rainfall and the inability of the sewer system to drain all runoff. As a consequence, a considerable volume of water is carried out over the surface through preferential flow paths and can eventually accumulate in natural (or man-made) ponds. This can cause minor material losses but also major incidents with obvious consequences in
This study presents a robust approach for characterization of multi-layered forests using airborn... more This study presents a robust approach for characterization of multi-layered forests using airborne laser scanning (ALS) data. Fuel mapping or biomass estimation requires knowing the diversity and boundaries of the forest patches, as well as their spatial pattern. This includes the thickness of the main vegetation layers, but also the spatial arrangement and size of the individual plants that compose each stratum. In order to decompose the ALS point cloud into genuine 3-D segments corresponding to individual vegetation features, such as shrubs or tree crowns, we apply a statistical approach based on the mean shift algorithm. The segments are progressively assigned to a forest layer: ground vegetation, understory or overstory. Our method relies on a single biophysically meaningful parameter, the kernel bandwidth, which is related to the local forest structure. It is validated on 44 plots of a Portuguese forest, composed mainly of eucalyptus (Eucalyptus globulus Labill.) and maritime pine (Pinus pinaster Ait.) trees. The number of detected trees varies with the dominance position: from 98.6% for the dominant trees to 12.8% for the suppressed trees. Linear regression models explain up to 70% of the variability associated with ground vegetation and understory height.
Plant communities display a vertical structure based on the size and growth pattern of the domina... more Plant communities display a vertical structure based on the size and growth pattern of the dominant species. To a large extent, this pattern, called vertical stratification, depends on the climatic zone. Vertical structure analysis consists in detecting the number of layers and their limits within a forest stand. So far, there is a lack of robust approaches applied to airborne laser scanning (ALS) data that properly segment the different strata of forests having complex structures. In this study, we propose a procedure to characterize vertical forest stratification based on the mean shift (MS) algorithm. The MS is a non-linear filter that searches for local density maxima (modes). It is a non-parametric and unsupervised approach, which only requires a single criterion, the kernel bandwidth. Since the forest point cloud is a multi-modal distribution, the MS is used to find the modes which are supposed to be the barycenters of vegetation features. Once achieved, the modes are grouped together according to height range and the corresponding ALS points are assigned to each vegetation strata. Due to their complex pattern, using a single scale over the whole space is not recommended for the analysis of such environments. On this basis, the modes are computed using a variable kernel bandwidth according to the forest pattern. To depict such a pattern, we propose a new technique that segments the main forest layers at the plot level: overstory, understory, and surface vegetation. The procedure has been carried out on 45 plots of a Portuguese forest mainly composed of eucalyptus (Eucalyptus globulus) and pine (Pinus pinaster) trees that can be strongly populated by understory and surface vegetation.
IEEE Transactions on Geoscience and Remote Sensing, 2015
The canopy density model (CDM), a new product interpolated from airborne laser scanner (ALS) data... more The canopy density model (CDM), a new product interpolated from airborne laser scanner (ALS) data and dedicated to forest structure characterization is presented. It exploits both the multiecho capability of the ALS and a nonparametric density estimation technique called kernel density estimators (KDEs). The CDM is used to delineate the outmost perimeter of vegetation features and to compute forest crown cover (CrCO). Contrary to other works that focus on single-layer forest canopies, CrCo is derived here for each layer, namely, the overstory, the understory, and ground vegetation. The root-mean-square error of prediction determined by using field data acquired over 44 forest stands in a forest in Portugal allows the testing of the reliability of the method: It ranges from 6.21% (overstory) to 13.76% (ground vegetation). In addition, we investigate the ability of the CDM to map the CrCo for individual trees. Finally, two existing methods have been applied to our study site in order to assess improvements, advantages, and drawbacks of our approach.
O perigo de inundação é um componente da avaliação do risco de inundação. Um dos elementos fundam... more O perigo de inundação é um componente da avaliação do risco de inundação. Um dos elementos fundamentais na avaliação do perigo de inundação é a estimativa da extensão da zona inundável. Neste trabalho é estudado o efeito da resolução espacial do Modelo Digital de Superfície de Escoamento (MDSe) na extensão da zona inundável. São criados três MDSe de diferentes resoluções que resultam da integração de dados LiDAR, cartografia a escala grande e ortoimagens CIR (Colour Infra-Red). Os dados LiDAR, devidamente filtrados dos objetos no terreno, são convertidos em três Modelos Digitais de Terreno (MDT) em formato raster com 1 m, 2 m, e 4 m. Da cartografia é extraído o edificado na forma de um raster. A vegetação com altura inferior a 2 m é por sua vez extraída das ortoimagens CIR. Os rasters dos objetos e vegetação são redimensionados para células de 1 m, 2 m e 4 m. Os três MDSe assim obtidos são utilizados na simulação hidráulica, com a aplicação do modelo LISFLOOD-FP e com o objetivo de ...
A extracção de Modelos Digitais de Terreno (MDT) em zonas urbanas a partir de Modelos Digitais de... more A extracção de Modelos Digitais de Terreno (MDT) em zonas urbanas a partir de Modelos Digitais de Superfície (MDS) é ainda um problema com sucesso limitado devido ao facto da operação de filtragem, ao utilizar apenas a informação geométrica contida no MDS, não permitir distinguir eficazmente todos os objectos situados acima da superfície topográfica que define o MDT. Para ultrapassar os problemas propõe-se uma nova metodologia que, pela integração dos dados obtidos por varrimento aéreo com laser (LiDAR) com orto-imagens CIR (Color Infra-Red, ou ainda falsa cor) de alta resolução, permite a extracção de informação precisa e fiável do terreno e dos objectos considerados não pertencentes ao MDT (edifícios e vegetação alta). Os resultados obtidos mostram que em zonas urbanas complexas a metodologia proposta tem um desempenho superior à metodologia convencional
Abstract. The research Project PTDC/AGR-CFL/72380/2006, entitled "Forest and Fuel Variables ... more Abstract. The research Project PTDC/AGR-CFL/72380/2006, entitled "Forest and Fuel Variables Estimation and Digital Terrain Modelling with Airborne Laser Scanning and High Resolution Multi-Spectral Images", financed by the Portuguese foundation Fundação para a Ciência ...
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
In the context of coastal environment, the combined use of Unmanned Aerial Systems (UAS) with Str... more In the context of coastal environment, the combined use of Unmanned Aerial Systems (UAS) with Structure-from-Motion (SfM) photogrammetry has been demonstrated to be a suitable tool to collect accurate geoinformation. This work aims at investigating the effectiveness of UAS survey to accurately map the geometry of rouble mound groins for monitoring the structural integrity. A set of nadiral images was processed through SfM photogrammetry to obtain the Digital Surface Model (DSM) of the groin. The DSM accuracy was evaluated through the comparison of five profiles measured by the traditional GNSS technique. Overall, vertical accuracy returned a root mean square error of 2.6 cm. The results suggest that UAS combined with SfM photogrammetry can improve the current monitoring of coastal engineering structures by providing high accurate geospatial products.
The abundance of litter pollution in the marine environment has been increasing globally. Remote ... more The abundance of litter pollution in the marine environment has been increasing globally. Remote sensing techniques are valuable tools to advance knowledge on litter abundance, distribution and dynamics. Images collected by Unmanned Aerial Vehicles (UAV, aka drones) are highly efficient to map and monitor local beached (BL) and floating (FL) marine litter items. In this work, the operational insights to carry out both BL and FL surveys using UAVs are detailly described. In particular, flight planning and deployment, along with image products processing and analysis, are reported and compared. Furthermore, analogies and differences between UAV-based BL and FL mapping are discussed, with focus on the challenges related to BL and FL item detection and recognition. Given the efficiency of UAV to map BL and FL, this remote sensing technique can replace traditional methods for litter monitoring, further improving the knowledge of marine litter dynamics in the marine environment. This comm...
Unmanned aerial systems (UASs) have recently been proven to be valuable remote sensing tools for ... more Unmanned aerial systems (UASs) have recently been proven to be valuable remote sensing tools for detecting marine macro litter (MML), with the potential of supporting pollution monitoring programs on coasts. Very low altitude images, acquired with a low-cost RGB camera onboard a UAS on a sandy beach, were used to characterize the abundance of stranded macro litter. We developed an object-oriented classification strategy for automatically identifying the marine macro litter items on a UAS-based orthomosaic. A comparison is presented among three automated object-oriented machine learning (OOML) techniques, namely random forest (RF), support vector machine (SVM), and k-nearest neighbor (KNN). Overall, the detection was satisfactory for the three techniques, with mean F-scores of 65% for KNN, 68% for SVM, and 72% for RF. A comparison with manual detection showed that the RF technique was the most accurate OOML macro litter detector, as it returned the best overall detection quality (F-s...
Uploads
Papers by Gil Gonçalves