Light Detection and Ranging (LIDAR) är en teknik för att samla in data om terräng. Genom att använda dessa data kan man skapa olika terrängmodeller. Denna studie syftar till att undersöka hur olika procentuella reduceringar av ursprungsdata påverkar kvalitén hos genererade höjdmodeller i form av Triangular Irregular Network (TIN). Detta görs genom att med hjälp av statistiska metoder göra jämförelser mellan punkter i den genererade TIN modellen och motsvarande punkter i det ursprungliga LIDAR punktmolnet. Studien visar att, beroende på noggrannhetskrav och topografi, en så liten andel som 5 % av punkterna kan vara tillräckligt, samt att noggrannhetsförbättring vid användning av mer än 50 % av ursprungsdata inte kan motivera den ökade arbetsbelastningen för datahantering.

O trabalho se estruturou em torno de dois estudos. O primeiro avaliou o ajuste de um modelo não linear de efeito misto para descrever o afilamento do tronco de árvores clonais de eucalipto. O modelo utilizado para descrever as variações da altura em função do raio foi o logístico de quatro parâmetros que, por integração permitiu a estimação do volume das árvores. A incorporação de funções de variância no processo de ajuste resultou em redução significativa no valor do Critério de informação de Akaike, mas os resíduos não apresentaram melhorias notáveis. Com a finalidade de compatibilizar precisão e parcimônia, o modelo que considera as variações do afilamento como uma função da altura total e do raio à altura do peito mostrou-se como o mais indicado para a estimativa do volume de árvores por funções de afilamento. O segundo estudo analisou uma nova proposta para inventários florestais em plantios clonais de eucalipto que integra modelagem geoestatística, medições de circunferência das árvores em campo e a tecnologia LiDAR aeroembarcada. As estatísticas propostas mostraram que o modelo geoestatístico com função para média foi estatisticamente superior ao modelo com média constante, com erros reduzidos em até 40%. A altura das árvores que compuseram o grid de predição para aplicação do modelo geoestatístico foi obtida pelo processamento da nuvem de pontos dos dados LiDAR. Obtidos os pares de diâmetro e altura, aplicou-se o modelo de afilamento selecionado no primeiro artigo em que se observaram diferenças médias na predição do volume próximas a 0,7%, e 0,18% para contagem de árvores, ambas com tendências de subestimativas. Diante dos resultados obtidos, o método é considerado como promissor e trabalhos futuros visam gerar um banco de parcelas permanentes que propiciem estudos de crescimento e produção florestal. / This study investigates the use of mixed-effect model and the use of LiDAR based model to estimate volume from eucalyptus forest plantation. At the first part, this study evaluates nonlinear mixed-effects to model stem taper of monoclonal Eucalyptus trees. The relation between radius and height variation was described by the four-parameter logistic model that integration returns stem volume. Embedding variance functions to the estimation process decreased significantly the Akaike\'s Information Criterion but did not improve the residual analysis. The best model to estimate stem volume from taper equations explained the stem taper as a function of the commercial height and the radius at breast height. The second part investigated the volume estimation fusing geostatistic derived from field information and airborne laser scanning data. The model based on geostatistic assumptions was statistically superior to the traditional one, with errors 40% lower. Thus, the geostatistical model was applied over tree heights extracted from the laser cloud. To each combination of diameter and height, the taper equation form the first part of this study was used. The volume and the number of trees were underestimated in 0.7% and 0.18%, respectively. The results look promising, and more permanent plots are necessary to allow studies about growth and yield of forest.

Submitted by William Justo Figueiro (williamjf) on 2015-07-01T23:37:49Z No. of bitstreams: 1 10.pdf: 1839460 bytes, checksum: 72eb68f5839e6788cf3f08b69ca3d26c (MD5) / Made available in DSpace on 2015-07-01T23:37:49Z (GMT). No. of bitstreams: 1 10.pdf: 1839460 bytes, checksum: 72eb68f5839e6788cf3f08b69ca3d26c (MD5) Previous issue date: 2011 / FAPERGS - Fundação de Amparo à Pesquisa do Estado do Rio Grande do Sul / O sistema LIDAR obtém nuvens de pontos georreferenciadas que podem ser convertidas em Modelos Digitais de Afloramentos (MDAs). Os pulsos de laser são gerados e emitidos por um Laser Scanner Terrestre, que atinge a superfície do afloramento em diferentes pontos. Estes objetos refletem o pulso incidente, que volta para o equipamento. Com isso, a distância entre o sensor e o objeto é determinada com base no intervalo de tempo entre a emissão e o retorno do pulso. Para fins de Modelagem Digital de Afloramentos (MDA) o uso dessa técnica é recente e necessita do desenvolvimento de pesquisas. Diante disso, o objetivo desse trabalho foi estabelecer uma sequencia de métodos envolvendo a aquisição e processamento de nuvem de pontos e a visualização e interpretação de superfícies e volumes de um Modelo Digital de Afloramentos (MDA). A Rocha da Pedra Pintada, localizado no Rio Grande do Sul na Cidade de Caçapava do Sul, foi imageado a partir de 7 estações e a nuvem totalizou 17 milhões de pontos, que foi convertida, após processamento, em um MDA. Para a interpretação geológica, utilizou-se a técnica de ortorretificação para sobrepor a nuvem de pontos a fotografias de alta resolução do afloramento, o que agregou qualidade na visualização e interpretação do MDA. Dificuldades nas etapas de tratamento dos dados ocorreram em razão do grande volume de dados, da ausência de recursos de otimização de processamento e da limitação no gerenciamento de banco de dados. Além disso, faz-se necessário desenvolver um aplicativo eficiente de visualização tridimensional com ferramentas especificas de interpretação geológica. / The LIDAR system provides georeferrenced clouds with thousand-to-million of points which can be converted in digital outcrops models. A laser beam is emitted and captured by a laser scanner after reaching an outcrop in many different positions. Based on the travel time between sensor and outcrop it is possible to determine the position of each point with high accuracy. This technique is still a novelty for applied studies in Geology, especially in Digital Outcrop Models (DOMs), being necessary research and development. Thus, the goal of this work was establish a workflow concerning acquisition and processing of point clouds, and visualization and geological interpretation of DOMs. The Pedra Pintada, located in the state Rio Grande do Sul in the city Caçapava do Sul outcrop was imaged from seven different stations and the cloud has 17 million points, converted in a DOM after processing. The geological interpretation was made possible by the orthorectification technique, in which a high resolution photograph overlies the point cloud and the visual quality is obtained. The huge volume of data, the lack of optimized processing resources and the inadequate dataset management became visualization and interpretation of DOMs a difficult task. Furthermore, it is necessary to develop a software with an efficient tridimensional visualization system with specific tools for geological interpretations.

The use of Light Detection and Ranging (LiDAR) information is gaining popularity, however its use has been limited in deciduous forests. This thesis describes two studies using LiDAR data in an Eastern Kentucky deciduous forest. The first study quantifies vertical error of LiDAR derived digital elevation models (DEMs) which describe the forests terrain. The study uses a new method which eliminates Global Positioning System (GPS) error. The study found that slope and slope variability both significantly affect DEM error and should be taken in to account when using LiDAR derived DEMs. The second study uses LiDAR derived forest vegetation and terrain metrics to predict terrestrial Plethodontid salamander abundance across the forest. This study used night time visual encounter surveys coupled with zero-inflation modeling to predict salamander abundance based on environmental covariates. We focused on two salamander species, Plethodon glutinosus and Plethodon kentucki. Our methods produced two different best fit models for the two species. Plethodon glutinosus included vegetation height standard deviation and water flow accumulation covariates, while Plethodon kentucki included only canopy cover as a covariate. These methods are applicable to many different species and can be very useful for focusing management efforts and understanding species distributions across the landscape.

O trabalho se estruturou em torno de dois estudos. O primeiro avaliou o ajuste de um modelo não linear de efeito misto para descrever o afilamento do tronco de árvores clonais de eucalipto. O modelo utilizado para descrever as variações da altura em função do raio foi o logístico de quatro parâmetros que, por integração permitiu a estimação do volume das árvores. A incorporação de funções de variância no processo de ajuste resultou em redução significativa no valor do Critério de informação de Akaike, mas os resíduos não apresentaram melhorias notáveis. Com a finalidade de compatibilizar precisão e parcimônia, o modelo que considera as variações do afilamento como uma função da altura total e do raio à altura do peito mostrou-se como o mais indicado para a estimativa do volume de árvores por funções de afilamento. O segundo estudo analisou uma nova proposta para inventários florestais em plantios clonais de eucalipto que integra modelagem geoestatística, medições de circunferência das árvores em campo e a tecnologia LiDAR aeroembarcada. As estatísticas propostas mostraram que o modelo geoestatístico com função para média foi estatisticamente superior ao modelo com média constante, com erros reduzidos em até 40%. A altura das árvores que compuseram o grid de predição para aplicação do modelo geoestatístico foi obtida pelo processamento da nuvem de pontos dos dados LiDAR. Obtidos os pares de diâmetro e altura, aplicou-se o modelo de afilamento selecionado no primeiro artigo em que se observaram diferenças médias na predição do volume próximas a 0,7%, e 0,18% para contagem de árvores, ambas com tendências de subestimativas. Diante dos resultados obtidos, o método é considerado como promissor e trabalhos futuros visam gerar um banco de parcelas permanentes que propiciem estudos de crescimento e produção florestal. / This study investigates the use of mixed-effect model and the use of LiDAR based model to estimate volume from eucalyptus forest plantation. At the first part, this study evaluates nonlinear mixed-effects to model stem taper of monoclonal Eucalyptus trees. The relation between radius and height variation was described by the four-parameter logistic model that integration returns stem volume. Embedding variance functions to the estimation process decreased significantly the Akaike\'s Information Criterion but did not improve the residual analysis. The best model to estimate stem volume from taper equations explained the stem taper as a function of the commercial height and the radius at breast height. The second part investigated the volume estimation fusing geostatistic derived from field information and airborne laser scanning data. The model based on geostatistic assumptions was statistically superior to the traditional one, with errors 40% lower. Thus, the geostatistical model was applied over tree heights extracted from the laser cloud. To each combination of diameter and height, the taper equation form the first part of this study was used. The volume and the number of trees were underestimated in 0.7% and 0.18%, respectively. The results look promising, and more permanent plots are necessary to allow studies about growth and yield of forest.

As more data sources containing 3-D information are becoming available, an increased interest in 3-D imaging has emerged. Among these is the 3-D reconstruction of buildings and other man-made structures. A necessary preprocessing step is the detection and isolation of individual buildings that subsequently can be reconstructed in 3-D using various methodologies. Applications for both building detection and reconstruction have commercial use for urban planning, network planning for mobile communication (cell phone tower placement), spatial analysis of air pollution and noise nuisances, microclimate investigations, geographical information systems, security services and change detection from areas affected by natural disasters. Building detection and reconstruction are also used in the military for automatic target recognition and in entertainment for virtual tourism. Previously proposed building detection and reconstruction algorithms solely utilized aerial imagery. With the advent of Light Detection and Ranging (LiDAR) systems providing elevation data, current algorithms explore using captured LiDAR data as an additional feasible source of information. Additional sources of information can lead to automating techniques (alleviating their need for manual user intervention) as well as increasing their capabilities and accuracy. Several building detection approaches surveyed in the open literature have fundamental weaknesses that hinder their use; such as requiring multiple data sets from different sensors, mandating certain operations to be carried out manually, and limited functionality to only being able to detect certain types of buildings. In this work, a building detection system is proposed and implemented which strives to overcome the limitations seen in existing techniques. The developed framework is flexible in that it can perform building detection from just LiDAR data (first or last return), or just nadir, color aerial imagery. If data from both LiDAR and aerial imagery are available, then the algorithm will use them both for improved accuracy. Additionally, the proposed approach does not employ severely limiting assumptions thus enabling the end user to apply the approach to a wider variety of different building types. The proposed approach is extensively tested using real data sets and it is also compared with other existing techniques. Experimental results are presented.

No description available.

Submitted by William Justo Figueiro (williamjf) on 2015-07-09T22:32:57Z No. of bitstreams: 1 27b.pdf: 4120563 bytes, checksum: 28666d8a39aa4371e2cad8353a3b6fc2 (MD5) / Made available in DSpace on 2015-07-09T22:32:57Z (GMT). No. of bitstreams: 1 27b.pdf: 4120563 bytes, checksum: 28666d8a39aa4371e2cad8353a3b6fc2 (MD5) Previous issue date: 2012-08-01 / Petrobras - Petróleo Brasileiro S. A. / UNISINOS - Universidade do Vale do Rio dos Sinos / O presente estudo visou o desenvolvimento e aplicação de uma metodologia para a detecção e classificação de diferentes respostas espectrais em nuvens de pontos obtidas a partir de escâner a laser terrestre (Laser Scanner Terrestre) com o intuito de identificar a presença de diferentes rochas em afloramentos e a geração de um Modelo Digital de Afloramento. A ferramenta para a classificação de padrões espectrais, denominada K-Clouds, foi desenvolvida com base em análise de agrupamentos (clusters), que a partir de uma indicação do número de classes fornecido pelo usuário através da análise de um histograma dos dados, realiza a classificação da nuvem de pontos. Os dados classificados podem então ser interpretados por geólogos para uma melhor compreensão e identificação das rochas presentes no afloramento. Além da detecção de diferentes rochas, verificouse que é possível detectar pequenas alterações nas características físico-químicas das mesmas, como aquelas causadas por intemperismo e variação composicional. / The present study aimed to develop and implement a method for detection and classification of spectral signatures in point clouds obtained from Terrestrial Laser Scanner in order to identify the presence of different rocks in outcrops and to generate a Digital Outcrop Model. To achieve this objective, a software based on cluster analysis was created, named K-Clouds. This software was developed through a partnership between UNISINOS and the company V3D. This tool was designed to, beginning with an analysis and interpretation of a histogram from a point cloud of the outcrop and subsequently indication of a number of classes provided by the user, process the intensity return values. This classified information can then be interpreted by geologists, to provide a better understanding and identification from the existing rocks in the outcrop. Beyond the detection of different rocks, this work was able to detect small changes in the physical-chemical characteristics of the rocks, as they were caused by weathering or compositional changes.

<p>Imaging and computer vision systems deployed in real-world environments face the challenge of accommodating a wide range of lighting conditions. However, the cost, the demand for high resolution, and the miniaturization of imaging devices impose physical constraints on sensor design, limiting both the dynamic range and effective aperture size of each pixel. Consequently, conventional CMOS sensors fail to deliver satisfactory capture in high dynamic range scenes or under photon-limited conditions, thereby impacting the performance of downstream vision tasks. In this thesis, we address two key problems: 1) exploring the utilization of spatial multiplexing, specifically spatially varying exposure tiling, to extend sensor dynamic range and optimize scene capture, and 2) developing techniques to enhance the robustness of object detection systems under photon-limited conditions.</p> <p><br></p> <p>In addition to challenges imposed by natural environments, real-world vision systems are susceptible to adversarial attacks in the form of artificially added digital content. Therefore, this thesis presents a comprehensive pipeline for constructing a robust and scalable system to counter such attacks.</p>

This work addresses the deep-learning-based 6 Degree-of-Freedom (DoF) pose estimation utilizing 3D cameras on an iPhone 13 Pro. The task of pose estimation is to estimate the spatial rotation and translation of an object given its 2D or 3D images. During the pose estimation network training process, a common way to expand the training dataset is to generate synthetic images, which requires the 3D mesh of the target object. Although several famous datasets provide the 3D object files, it is still a problem when one wants to generate a customized real-world object. The typical 3D scanners are mainly designed for industrial usage and are usually expensive. We investigated in this project whether the 3D cameras on Apple devices can replace the industrial 3D scanners in the pose estimation pipeline and what might influence the results during scanning. During the data synthesis, we introduced a pose sampling method to equally sample on a sphere. Random transformation and background images from the SUN2012 dataset are applied, and the synthetic image is rendered through Blender. We picked five testing objects with different sizes and surfaces. Each object is scanned both by front TrueDepth camera and rear Light Detection and Ranging (LiDAR) camera with the ‘3d Scanner App’ on iOS. The network we used is based on PVNet, which uses a pixel-wise voting scheme to find 2D keypoints on RGB images and utilizes uncertainty-driven Perspective-n-Point (PnP) to compute the pose. We achieved both quantitative and qualitative results for each instance. i) TrueDepth camera outperforms Light Detection and Ranging (LiDAR) camera in most scenarios, ii) when an object has less reflective surface and high-contrast texture, the advantage of TrueDepth is more obvious. We also picked three baseline objects from Linemod dataset. Although the average accuracy is lower than the original paper, the performance of our baseline instances shows a similar trend to the original paper’s results. In conclusion, we proved that the 3D cameras on iPhone are capable of the pose estimation pipeline. / Detta arbete tar upp den djupinlärningsbaserade 6 Degree-of-Freedom (DoF) poseringsuppskattning med 3D-kameror på en iPhone 13 Pro. Uppgiften med poseuppskattning är att uppskatta den rumsliga rotationen och translationen av ett objekt givet dess 2D- eller 3D-bilder. Ett vanligt sätt att utöka träningsdataup- psättningen under träningsprocessen för positionsuppskattning är att generera syntetiska bilder, vilket kräver 3D-nätet för målobjektet. Även om flera kända datamängder tillhandahåller 3D-objektfilerna, är det fortfarande ett problem när man vill generera ett anpassat verkligt objekt. De typiska 3D-skannrarna är främst designade för industriell användning och är vanligtvis dyra. Vi undersökte i detta projekt om 3D-kamerorna på Apple-enheter kan ersätta de industriella 3D-skannrarna i poseskattningspipelinen och vad som kan påverka resultaten under skanning. Under datasyntesen introducerade vi en posesamplingsmetod för att sampla lika mycket på en sfär. Slumpmässig transformation och bakgrundsbilder från SUN2012-datauppsättningen tillämpas, och den syntetiska bilden renderas genom Blender. Vi valde ut fem testobjekt med olika storlekar och ytor. Varje objekt skannas både av den främre TrueDepth-kameran och den bakre ljusdetektions- och avståndskameran (LiDAR) med "3d-skannerappenpå iOS. Nätverket vi använde är baserat på PVNet, som använder ett pixelvis röstningsschema för att hitta 2D-nyckelpunkter på RGB-bilder och använder osäkerhetsdrivet Perspective-n-Point (PnP) för att beräkna poseringen. Vi uppnådde både kvantitativa och kvalitativa resultat för varje instans. i) TrueDepth-kameran överträffar Light Detection and Ranging-kameran (LiDAR) i de flesta scenarier, ii) när ett objekt har mindre reflekterande yta och högkontraststruktur är fördelen med TrueDepth högre. Vi valde också tre baslinjeobjekt från Linemod dataset. Även om den genomsnittliga noggrannheten är lägre än originalpapperet, visar prestandan för våra baslinjeinstanser en liknande trend som originalpapperets resultat. Sammanfattningsvis bevisade vi att 3D-kamerorna på iPhone är kapabla att göra positionsuppskattning.