A5076941451- Remote Sensing in Agriculture
- Remote Sensing and LiDAR Applications
- Forest ecology and management
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Land Use and Ecosystem Services
- Remote-Sensing Image Classification
- Geochemistry and Geologic Mapping
- Fire effects on ecosystems
- Geography and Environmental Studies
- Spectroscopy and Chemometric Analyses
- Forest Ecology and Biodiversity Studies
- 3D Surveying and Cultural Heritage
- Conservation, Biodiversity, and Resource Management
- Tree Root and Stability Studies
- Smart Agriculture and AI
- Flood Risk Assessment and Management
- Soil erosion and sediment transport
- Leaf Properties and Growth Measurement
- Environmental and biological studies
- Landslides and related hazards
- Groundwater and Watershed Analysis
- Soil Management and Crop Yield
- Soil Geostatistics and Mapping
- Species Distribution and Climate Change
- Hydrology and Watershed Management Studies
Universidade do Estado de Santa Catarina
2016-2025
National Institute for Space Research
2006-2016
TU Bergakademie Freiberg
2009-2013
Detection and classification of tree species from remote sensing data were performed using mainly multispectral hyperspectral images Light And Ranging (LiDAR) data. Despite the comparatively lower cost higher spatial resolution, few studies focused on captured by Red-Green-Blue (RGB) sensors. Besides, recent years have witnessed an impressive progress deep learning methods for object detection. Motivated this scenario, we proposed evaluated usage Convolutional Neural Network (CNN)-based...
The classification of tree species can significantly benefit from high spatial and spectral information acquired by unmanned aerial vehicles (UAVs) associated with advanced methods. This study investigated the following topics concerning 16 in two subtropical forest fragments Southern Brazil: i) potential integration UAV-borne hyperspectral images 3D derived their photogrammetric point cloud (PPC); ii) performance machine learning methods (support vector – SVM random RF) when employing...
The traditional method of measuring nitrogen content in plants is a time-consuming and labor-intensive task. Spectral vegetation indices extracted from unmanned aerial vehicle (UAV) images machine learning algorithms have been proved effective assisting nutritional analysis plants. Still, this has not considered the combination spectral to predict tree-canopy structures. This paper proposes new framework infer citrus-tree at canopy-level using processed with random forest algorithm. A total...
Studies designed to discriminate different successional forest stages play a strategic role in management, policy and environmental conservation tropical environments. The discrimination of is still challenge due the spectral similarity among concerned classes. Considering this, objective this paper was investigate performance Sentinel-2 Landsat-8 data for discriminating patch located subtropical portion Atlantic Rain Forest Southern Brazil with aid two machine learning algorithms relying on...
This study proposes and evaluates five deep fully convolutional networks (FCNs) for the semantic segmentation of a single tree species: SegNet, U-Net, FC-DenseNet, two DeepLabv3+ variants. The performance FCN designs is evaluated experimentally in terms classification accuracy computational load. We also verify benefits connected conditional random fields (CRFs) as post-processing step to improve maps. analysis conducted on set images captured by an RGB camera aboard UAV flying over urban...
This paper presents a framework based on machine learning algorithms to predict nutrient content in leaf hyperspectral measurements. is the first approach evaluate macro- and micronutrient with both reflectance/first-derivative data. For this, citrus-leaves collected at Valencia-orange orchard were used. Their spectral data was measured Fieldspec ASD FieldSpec® HandHeld 2 spectroradiometer surface reflectance first-derivative spectra from range of 380 1020 nm (640 bands) evaluated. A total...
The use of remote sensing data for tree species classification in tropical forests is still a challenging task, due to their high floristic and spectral diversity. In this sense, novel sensors on board unmanned aerial vehicle (UAV) platforms are rapidly evolving technology that provides new possibilities mapping. Besides the acquisition spatial resolution images, UAV-hyperspectral cameras operating frame format enable produce 3D hyperspectral point clouds. This study investigated...
In recent years, many agriculture-related problems have been evaluated with the integration of artificial intelligence techniques and remote sensing systems. Specifically, in fruit detection problems, several works were developed using Deep Learning (DL) methods applied images acquired different acquisition levels. However, increasing use anti-hail plastic net cover commercial orchards highlights importance terrestrial Apples are one most highly-challenging fruits to be detected images,...
Urban forests are an important part of any city, given that they provide several environmental benefits, such as improving urban drainage, climate regulation, public health, biodiversity, and others. However, tree detection in cities is challenging, the irregular shape, size, occlusion, complexity areas. With advance technologies, deep learning segmentation mapping methods can map accurately. We applied a region-based CNN object instance algorithm for semantic canopies environments based on...
Iraq, a country in the Middle East, has suffered severe drought events past two decades due to significant decrease annual precipitation. Water storage by building dams can mitigate impacts and assure water supply. This study was designed identify suitable sites build new within Al-Khabur River Basin (KhRB). Both fuzzy analytic hierarchy process (AHP) weighted sum method (WSM) were used compared select dam sites. A total of 14 layers as input dataset (i.e., lithology, tectonic zones,...
Fire in Brazilian Pantanal represents a serious threat to biodiversity. The National Institute of Spatial Research (INPE) has program named Queimadas, which estimated from January 2020 October 2020, burned area approximately 40,606 km2. This also provides daily data active fire (fires spots) methodology that uses MODIS (Aqua and Terra) sensor as reference satellites, presents limitations mainly when dealing with small fires. Remote sensing researches on dynamics have contributed wildfire...
Improving management practices in industrial forest plantations may increase production efficiencies, thereby reducing pressures on native tropical forests for meeting global pulp needs. This study aims to predict stem volume (V) of fast-growing Eucalyptus hybrid clones located southeast Brazil using field plot and airborne Light Detection Ranging (LiDAR) data. Forest inventory attributes LiDAR-derived metrics were calculated at 108 sample plots. The best LiDAR-based predictors V identified...
Punjab, Pakistan is famous for rice production in all over the world, but economic indicators are low toward contribution regional economy. Climatic and physical factors responsible yield degradation. Suitable land cultivation can be mapped keeping view these climatic factors. In this research, season was calculated using Moderate Resolution Imaging Spectro-radiometer (MODIS) time series datasets complete year 2014. Landsat 8 thermal were obtained temperature based growth variability maps...
The Brazilian territory contains approximately 160 million hectares of pastures, and it is necessary to develop techniques automate their management increase production. This technical note has two objectives: First, estimate the canopy height using unmanned aerial vehicle (UAV) photogrammetry; second, propose an equation for estimation biomass savanna (Cerrado) pastures based on UAV height. Four experimental units Panicum maximum cv. BRS Tamani were evaluated. Herbage mass sampling,...
Modeling the hyperspectral response of vegetables is important for estimating water stress through a noninvasive approach. This article evaluates water-stress induced lettuce (Lactuca sativa L.) using artificial neural networks (ANN). We evenly split 36 pots into three groups: control, stress, and bacteria. Hyperspectral was measured four times, during 14 days induction, with an ASD Fieldspec HandHeld spectroradiometer (325–1075 nm). Both reflectance absorbance measurements were calculated....
Developing the capacity to monitor species diversity worldwide is of great importance in halting biodiversity loss. To this end, remote sensing plays a unique role. In study, we evaluate potential Global Ecosystem Dynamics Investigation (GEDI) data, combined with conventional satellite optical imagery and climate reanalysis predict situ alpha (Species richness, Simpson index, Shannon index) among tree species. Data from Sentinel-2 imagery, ERA-5 SRTM-DEM simulated GEDI data were selected for...
Os Modelos Digitais de Superfície são utilizados em diversas aplicações, sendo essenciais para modelagem hidrológica. Descrevemos a precisão altimétrica diferentes MDSs na área urbana Campo Grande, Mato Grosso do Sul, Brasil. Como conjunto dados referência, usamos dois conjuntos distintos com coordenadas GNSS 3D. modelos testados foram Tandem-X, ALOS AW3D30, SRTMc, TOPODATA, SRTM v.3 e Aster GDEM v.2. Estimamos as discrepâncias verticais relação ao usando seguintes métricas: discrepância...
Abstract In the absence of regional/local allometric models known accuracy, pantropical (PMs) are often employed for predicting aboveground biomass (AGB) trees growing in (sub)tropical forests. Using accurate a given population is crucial to increase accuracy and reduce uncertainty estimates mean AGB per unit area. This study evaluated effects local (LMs) PMs on large-area (Mg ha$^{-1}$) Brazilian subtropical evergreen rainforest. addition due sampling variability forest inventory dataset,...
Summary Oil seeps pose significant environmental hazards to both terrestrial and aquatic ecosystems. Traditional mapping techniques encounter logistical political challenges, particularly in complex regions, such as Kirkuk, an area rich oil gas fields. These fields contribute the proliferation of through natural processes industrial activities, underscoring need for efficient detection methods. This study introduces a novel hybrid algorithm, SAM-DT, which combines spectral angle (SAM) with...