Les Systèmes Mobiles de Cartographie basés LIDAR permettent d’obtenir des cartes 3D de l’environnement, qui sont géo-référencées grâce à d’autres capteurs embarqués sur le véhicule : GPS, centrale inertielle, ou encore odomètre sont de tels capteurs qui permettent de localiser le véhicule mobile pendant la campagne d’acquisition. Toutefois, ces cartes manquent de précisions et un affinage des cartes est essentiel dans de nombreux cas d’applications où une précision fine est requise sur les cartes 3D, comme pour des applications de classifications par exemple.Lors de la création de cartes 3D géoréférencées,les données sont tout d’abord acquises par le capteur LIDAR et référencées dans le repère cartésien du laser à l’aide d’un calibrage intrinsèque du capteur d’acquisition. Ensuite, un calibrage extrinsèque du capteur permet de caractériser la transformation entre le capteur et le véhicule, et permet de référencer les données dans le repère « body», lié au véhicule d’acquisition. Enfin, avec la trajectoire du véhicule obtenue en fusionnant les données issues des GPS, centrale inertielle et odomètre, il est possible de géoréférencer les données lasers.Nous proposons dans cette thèse d’affiner les relevés laser issus d’acquisitions effectuées à l’aide d’un véhicule mobile de cartographie, en optimisant plusieurs paramètres différents qui entrent en compte dans le géoréférencement des données. Nous nous sommes intéressés à l’affinement des nuages de points par optimisation des paramètres decalibrage extrinsèque dans un premier temps, puis par optimisation des paramètres de calibrage intrinsèque, et enfin par optimisation des paramètres de translations liés à la trajectoire du véhicule mobile. / LIDAR based Mobile Mapping Systems allowto get 3D maps of the environment, which are globally referenced with the help of others sensors embedded on the vehicle: GPS,Inertial Measurement Unit, or odometer are such sensors which allow localizing the vehicle during the acquisition process. However, these maps lack of precision, and are finement of the maps is necessary for manywork where a good precision is needed on the 3D maps, like classification applications for example.When creating the globally referenced 3D maps, the data are firstly acquired by the LIDAR sensor and referenced in the Cartesian reference frame of the sensor withan intrinsic calibration of the sensor. Then, anextrinsic calibration gives the transformation between the sensor and the vehicle, and gives data referenced in the « body »reference frame, linked to the vehicle. Finally, with the fusion of the data coming from the GPS, the Inertial Measurement Unit and theodometer, the laser data can be globally referenced.In this thesis, we propose to refine the point clouds coming from acquisitions done with a mobile mapping system, by optimizing some parameters which are used in the georeferencing process of the data. Firstly, we were interested in the refinement of point clouds by optimizing the extrinsic calibration parameters, and then we were interested in the refinement of point clouds by optimizing the intrinsic calibration parameters; finally by optimizing the translation parameters of the mobile vehicle trajectory.

L'intérêt toujours grandissant pour les données cartographiques fiables, notamment en milieu urbain, a motivé le développement de systèmes de cartographie mobiles terrestres. Ces systèmes sont conçus pour l'acquisition de données de très haute précision, telles que des nuages de points LiDAR 3D et des images optiques. La multitude de données, ainsi que leur diversité, rendent complexe le traitement des données issues de ce type de systèmes. Cette thèse se place dans le contexte du traitement de l'image appliqué au nuages de points LiDAR 3D issus de ce type de système.Premièrement, nous nous intéressons à des images issues de la projection de nuages de points LiDAR dans des grilles de pixels 2D régulières. Ces projections créent généralement des images éparses, dans lesquelles l'information de certains pixels n'est pas connue. Nous proposons alors différentes méthodes pour des applications telles que la génération d'orthoimages haute résolution, l'imagerie RGB-D et l'estimation de la visibilité des points d'un nuage.De plus, nous proposons d'exploiter la topologie d'acquisition des capteurs LiDAR pour produire des images de faible résolution: les range-images. Ces images offrent une représentation efficace et canonique du nuage de points, tout en étant directement accessibles à partir du nuage de points. Nous montrons comment ces images peuvent être utilisées pour simplifier, voire améliorer, des méthodes pour le recalage multi-modal, la segmentation, la désoccultation et la détection 3D. / The ever growing demand for reliable mapping data, especially in urban environments, has motivated the development of "close-range" Mobile Mapping Systems (MMS). These systems acquire high precision data, and in particular 3D LiDAR point clouds and optical images. The large amount of data, along with their diversity, make MMS data processing a very complex task. This thesis lies in the context of 2D image processing applied to 3D LiDAR point clouds acquired with MMS.First, we focus on the projection of the LiDAR point clouds onto 2D pixel grids to create images. Such projections are often sparse because some pixels do not carry any information. We use these projections for different applications such as high resolution orthoimage generation, RGB-D imaging and visibility estimation in point clouds.Moreover, we exploit the topology of LiDAR sensors in order to create low resolution images, named range-images. These images offer an efficient and canonical representation of the point cloud, while being directly accessible from the point cloud. We show how range-images can be used to simplify, and sometimes outperform, methods for multi-modal registration, segmentation, desocclusion and 3D detection.

<div>LiDAR units onboard airborne and terrestrial platforms have been established as a proven technology for the acquisition of dense point clouds for a wide range of applications, such as digital building model generation, transportation corridor monitoring, precision agriculture, and infrastructure monitoring. Furthermore, integrating such systems with one or more cameras would allow forward and backward projection between imagery and LiDAR data, thus facilitating several high-level data processing activities such as reliable feature extraction and colorization of point clouds. However, the attainment of the full 3D point positioning potential of such systems is contingent on an accurate calibration of the mobile mapping unit as a whole. </div><div> </div><div> This research aims at proposing a calibration procedure for terrestrial multi-unit LiDAR systems to directly estimate the mounting parameters relating several spinning multi-beam laser scanners to the onboard GNSS/INS unit in order to derive point clouds with high positional accuracy. To ensure the accuracy of the estimated mounting parameters, an optimal configuration of target primitives and drive-runs is determined by analyzing the potential impact of bias in mounting parameters of a LiDAR unit on the resultant point cloud for different orientations of target primitives and different drive-run scenarios. This impact is also verified experimentally by simulating a bias in each mounting parameter separately. Next, the optimal configuration is used within an experimental setup to evaluate the performance of the proposed calibration procedure. Then, this proposed multi-unit LiDAR system calibration strategy is extended for multi-LiDAR multi-camera systems in order to allow a simultaneous estimation of the mounting parameters relating the different laser scanners as well as cameras to the onboard GNSS/INS unit. Such a calibration improves the registration accuracy of point clouds derived from LiDAR data and imagery, along with their accuracy with respect to the ground truth. Finally, in order to qualitatively evaluate the calibration results for a generic mobile mapping system and allow the visualization of point clouds, imagery data, and their registration quality, an interface denoted as Image-LiDAR Interactive Visualization Environment (I-LIVE) is developed. Apart from its visualization functions (such as 3D point cloud manipulation and image display/navigation), I-LIVE mainly serves as a tool for the quality control of GNSS/INS-derived trajectory and LiDAR-camera system calibration. </div><div> </div><div> The proposed multi-sensor system calibration procedures are experimentally evaluated by calibrating several mobile mapping platforms with varying number of LiDAR units and cameras. For all cases, the system calibration is seen to attain accuracies better than the ones expected based on the specifications of the involved hardware components, i.e., the LiDAR units, cameras, and GNSS/INS units.</div>

Submitted by Mariana Dias Chaves null (mariana.unesp@hotmail.com) on 2018-02-16T19:43:53Z No. of bitstreams: 1 Batistao, MDC-TeseDr.pdf: 6348711 bytes, checksum: 0f1b9c7f3392530f6d2f279ee0b58768 (MD5) / Approved for entry into archive by Claudia Adriana Spindola null (claudia@fct.unesp.br) on 2018-02-19T11:31:34Z (GMT) No. of bitstreams: 1 batistao_mdc_dr_prud.pdf: 6348711 bytes, checksum: 0f1b9c7f3392530f6d2f279ee0b58768 (MD5) / Made available in DSpace on 2018-02-19T11:31:34Z (GMT). No. of bitstreams: 1 batistao_mdc_dr_prud.pdf: 6348711 bytes, checksum: 0f1b9c7f3392530f6d2f279ee0b58768 (MD5) Previous issue date: 2018-02-02 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Essa pesquisa apresenta um estudo acerca dos fatores contribuintes de acidentes rodoviários com o objetivo de fornecer evidências para analisar o comportamento dos fatores contribuintes envolvidos nesses acidentes, mais especificamente nos trechos críticos. Desejase identificar a relação dos fatores com o grau de severidade de um acidente (danos materiais, sem vítimas fatais e com vítimas fatais) e o impacto de cada classe de fator na ocorrência de um acidente. A intensão é embasar uma análise geoespacial levando em consideração técnicas estatísticas e cartográficas e contribuir para melhorar a qualidade das informações sobre segurança viária no país e seu atual cenário crítico. A estrutura metodológica da pesquisa consiste em três etapas principais: (I) Identificação e determinação de segmentos de trechos críticos, (II) Mapeamento dos fatores contribuintes “via” para o acidente e (III) Investigação e estudo dos fatores contribuintes. Quatro trechos de rodovias do oeste do estado de São Paulo foram escolhidos como área de estudo. Na etapa I propôs-se um método de interpolação espacial de escolha de segmentos de trechos críticos levando a premissa existência da dependência geográfica dos acidentes em consideração. No total, foram identificados oito segmentos de trechos críticos na área de estudo. A etapa II concentrou-se no mapeamento dos fatores contribuintes desses segmentos de trechos críticos. Essa etapa trouxe o caráter tecnológico à pesquisa por fazer uso da integração de geotecnologias e a contribuição das Ciências Cartográficas para os estudos de segurança viária, por gerar informação a partir do mapeamento da localização dos fatores contribuintes. Das quatro classes de fatores (humano, ambiente, veículo e via) as características da via foram escolhidas para serem mapeadas, tendo-se deparado com a ausência de qualquer dado dessa classe de fatores tanto no banco de dados dos acidentes como no boletim de ocorrências. A relação com as outras três classes de fatores foi tratada na etapa III da pesquisa, cujos resultados proporcionaram montar o ranking dos seis fatores contribuintes da via mais frequentes nos segmentos de trechos críticos. Adicionalmente, foram construídos três modelos de regressão logística ordinal para investigar o impacto de cada uma das quatro classes de fatores no grau de severidade do acidente (três graus de severidade). Para isso, o grau foi tratado como variável dependente dos modelos. Quatro variáveis independentes (fatores contribuintes) foram consideradas significativas e escolhidas para compor os modelos: consumo de drogas (da classe de fator contribuinte humano), estado dos pneus (da classe de fator veículo), vegetação (da classe de fator via) e sinalização (da classe de fator via). Por fim, os modelos puderam ser analisados a partir da razão de chances (odds ratio) para complementar as informações e sintetizar os resultados como contribuições da pesquisa. / This research presents a study about the contributing factors of road accidents in order to provide evidences to analyse the behaviour of contributing factors involved in these accidents more specifically in critical sections. The intention is to identify the relationship between those factors and the severity degree of an accident (material damage, no fatalities and fatalities) and the impact of each factor class on an accident occurrence. The aim is to base on geospatial analysis taking into account statistical and cartographic techniques and contribute to improve the quality of the road safety information in the country which has a current critical scene. The methodological structure of this thesis consists of following three main steps: (I) Identification and determination of critical sections segments, (II) mapping “road” contributing factors for each accident and (III) Investigation and study of the contributing factors. Four sections of highways in the west of São Paulo state were chosen as the study area. In Step I, proposed a spatial interpolation method to choose critical sections segments premising the existence of geographical dependence of the considered accidents. In entire, eight critical sections segments were identified in the study area. Step II focused on mapping the contributing factors of these segments. This step brought the technological character to this research by making use of geotechnologies integration and the contribution of Cartographic Sciences to road safety by generating information of the contributing factors location from mapping. Of the four factors classes (human, environment, vehicle and road), the road characteristics were chosen to be mapped, since no data from this factor class was found in both the accident database and the occurrence report. The relation with the other three factors classes was the subject of step III, which results provided a ranking of the six most frequent contributing factors in critical sections segments. In addition, three ordinal logistic regression models were constructed to investigate the impact of each of the four factors classes on the accident severity degree (three severity degrees). For this, the severity degree was considered as the models dependent variable. Four significant independent variables (contributing factors) were chosen to compose the following models: drug consumption (from the human contributing factor class), tire condition (vehicle factor class), vegetation (road factor class) and signaling (road factor class). Lastly, the models could be analysed by the odds ratio method to complement the information and synthesize the results as research contributions.

<div>Leaf Area Index (LAI) is an important variable for both for characterizing plant canopy and as an input to many crop models. It is a dimensionless quantity broadly defined as the total one-sided leaf area per unit ground area, and is estimated over agriculture row crops by both direct and indirect methods. Direct methods, which involve destructive sampling, are laborious and time-consuming, while indirect methods such as remote sensing-based approaches have multiple sources of uncertainty. LiDAR (Light Detection and Ranging) remotely sensed data acquired from manned aircraft and UAVs’ have been investigated to estimate LAI based on physical/geometric features such as canopy gap fraction. High-resolution point cloud data acquired with a laser scanner from any platform, including terrestrial laser scanning and mobile mapping systems, contain random noise and outliers. Therefore, outlier detection in LiDAR data is often useful prior to analysis. Applications in agriculture are particularly challenging, as there is typically no prior knowledge of the statistical distribution of points, description of plant complexity, and local point densities, which are crop dependent. This dissertation first explores the effectiveness of using LiDAR data to estimate LAI for row crop plants at multiple times during the growing season from both a wheeled vehicle and an Unmanned Aerial Vehicle (UAV). Linear and nonlinear regression models are investigated for prediction utilizing statistical and plant structure-based features extracted from the LiDAR point cloud data and ground reference obtained from an in-field plant canopy analyzer and leaf area derived from destructive sampling. LAI estimates obtained from support vector regression (SVR) models with a radial basis function (RBF) kernel developed using the wheel-based LiDAR system and UAVs are promising, based on the value of the coefficient of determination (R2) and root mean squared error (RMSE) of the residuals. </div><div>This dissertation also investigates approaches to minimize the impact of outliers on discrete return LiDAR acquired over crops, and specifically for sorghum and maize breeding experiments, by an unmanned aerial vehicle (UAV) and a wheel-based ground platform. Two methods are explored to detect and remove the outliers from the plant datasets. The first is based on surface fitting to noisy point cloud data based on normal and curvature estimation in a local neighborhood. The second utilizes the deep learning framework PointCleanNet. Both methods are applied to individual plants and field-based datasets. To evaluate the method, an F-score and LAI are calculated both before and after outlier removal for both scenarios. Results indicate that the deep learning method for outlier detection is more robust to changes in point densities, level of noise, and shapes. Also, the predicted LAI was improved for the wheel-based vehicle data based on the R2 value and RMSE of residuals. </div><div>The quality of the extracted features depends on the point density and laser penetration of the canopy. Extracting appropriate features is a critical step to have accurate prediction models. Deep learning frameworks are increasingly being used in remote sensing applications. In the last objective of this study, a feature extraction approach is investigated for encoding LiDAR data acquired by UAV platforms multiple times during the growing season over sorghum and maize plant breeding experiments. LAI estimates obtained with these inputs are used to develop support vector regression (SVR) models using plant canopy analyzer data as the ground reference. Results are compared to models based on estimates from physically-based features and evaluated in terms of the coefficient determination (R2). The effects of experimental conditions, including flying height, sensor characteristics, and crop type, are also investigated relative to the estimates of LAI.</div><div><br></div>

<p>Emerging mobile mapping systems include a wide range of platforms, for instance, manned aircraft, unmanned aerial vehicles (UAV), terrestrial systems like trucks, tractors, robots, and backpacks, that can carry multiple sensors including LiDAR scanners, cameras, and georeferencing units. Such systems can maneuver in the field to quickly collect high-resolution data, capturing detailed information over an area of interest. With the increased volume and distinct characteristics of the data collected, practical quality control procedures that assess the agreement within/among datasets acquired by various sensors/systems at different times are crucial for accurate, robust interpretation. Moreover, the ability to derive semantic information from acquired data is the key to leveraging the complementary information captured by mobile mapping systems for diverse applications. This dissertation addresses these challenges for different systems (airborne and terrestrial), environments (urban and rural), and applications (agriculture, archaeology, hydraulics/hydrology, and transportation).</p> <p>In this dissertation, quality control procedures that utilize features automatically identified and extracted from acquired data are developed to evaluate the relative accuracy between multiple datasets. The proposed procedures do not rely on manually deployed ground control points or targets and can handle challenging environments such as coastal areas or agricultural fields. Moreover, considering the varying characteristics of acquired data, this dissertation improves several data processing/analysis techniques essential for meeting the needs of various applications. An existing ground filtering algorithm is modified to deal with variation in point density; digital surface model (DSM) smoothing and seamline control techniques are proposed for improving the orthophoto quality in agricultural fields. Finally, this dissertation derives semantic information for diverse applications, including 1) shoreline retreat quantification, 2) automated row/alley detection for plant phenotyping, 3) enhancement of orthophoto quality for tassel/panicle detection, and 4) point cloud semantic segmentation for mapping transportation corridors. The proposed approaches are tested using multiple datasets from UAV and wheel-based mobile mapping systems. Experimental results verify that the proposed approaches can effectively assess the data quality and provide reliable interpretation. This dissertation highlights the potential of modern mobile mapping systems to map challenging environments for a variety of applications.</p>

Malgré les récentes avancées des Systèmes de Cartographie Mobile, la reconstruction automatique d’objets à partir des données acquises est encore un point crucial. Dans cette thèse, nous nous intéresserons en particulier à la reconstruction tridimensionnelle du marquage au sol à partir d’images acquises sur le réseau routier par une base stéréoscopique horizontale d’un système de cartographie mobile, dans un contexte urbain dense. Une nouvelle approche s’appuyant sur la connaissance de la géométrie 3D des marquages au sol est présentée, conduisant à une précision de reconstruction 3D centimétrique avec un faible niveau de généralisation. Deux objets de la signalisation routière horizontale sont étudiés : les passages piétons et les lignes blanches discontinues. La stratégie générale est composée de trois grandes étapes. La première d’entre elles permet d’obtenir des chaînes de contours 3D. Les contours sont extraits dans les images gauche et droite. Ensuite, un algorithme reposant sur une optimisation par programmation dynamique est mis en oeuvre pour apparier les points de contours des deux images. Un post-traitement permet un appariement sub-pixellique, et, les chaînes de contours 3D sont finalement obtenues par une triangulation photogrammétrique classique. La seconde étape fait intervenir les spécifications géométriques des marquages au sol pour réaliser un filtrage des chaînes de contours 3D. Elle permet de déterminer des candidats pour les objets du marquage au sol. La dernière étape peut être vue comme une validation permettant de rejeter ou d’accepter ces hypothèses. Les candidats retenus sont alors reconstruits finement. Pour chaque bande d’un passage piéton ou d’une ligne discontinue, le modèle est un quasi-parallélogramme. Une contrainte de planéité est imposée aux sommets de chaque bande, ce qui n’est pas le cas pour l’ensemble des bandes formant un marquage au sol particulier. La méthode est évaluée sur un ensemble de 150 paires d’images acquises en centre ville dans des conditions normales de trafic. Les résultats montrent la validité de notre stratégie en terme de robustesse, de complétude et de précision géométrique. La méthode est robuste et permet de gérer les occultations partielles ainsi que les marquages usés ou abîmés. Le taux de détection atteint 90% et la précision de reconstruction 3D est de l’ordre de 2 à 4 cm. Finalement, une application de la reconstruction des marquages au sol est présentée : le géoréférencement du système d’acquisition. La majorité des systèmes de cartographie mobile utilisent des capteurs de géoréférencement direct comme un couplage GPS/INS pour leur localisation. Cependant, en milieu urbain dense, les masques et les multi-trajets corrompent les mesures et conduisent à une précision d’environ 50 cm. Afin d’améliorer la qualité de localisation, nous cherchons à apparier les images terrestres avec des images aériennes calibrées de la même zone. Les marquages au sol sont alors utilisés comme objets d’appariement. La validité de la méthode est démontrée sur un exemple de passage piéton / Despite advances in ground-based Mobile Mapping System (MMS), automatic feature reconstruction seems far from being reached. In this thesis, we focus on 3D roadmark reconstruction from images acquired by road looking cameras of a MMS stereo-rig in dense urban context. A new approach is presented, that uses 3D geometric knowledge of roadmarks and provides a centimetric 3D accuracy with a low level of generalisation. Two classes of roadmarks are studied: zebra-crossing and dashed-lines. The general strategy consists in three main steps. The first step provides 3D linked-edges. Edges are extracted in the left and right images. Then a matching algorithm that is based on dynamic programming optimisation matches the edges between the two images. A sub-pixel matching is computed by post processing and 3D linked-edges are provided by classical photogrammetric triangulation. The second step uses the known specification of roadmarks to perform a signature based filtering of 3D linked-edges. This step provides hypothetical candidates for roadmark objects. The last step can be seen as a validation step that rejects or accepts the candidates. The validated candidates are finely reconstructed. The adopted model consists of a quasi parallelogram for each strip of zebra-crossing or dashed-line. Each strip is constrained to be flat but the roadmark as a whole is not planar. The method is evaluated on a set of 150 stereo-pairs acquired in a real urban area under normal traffic conditions. The results show the validity of the approach in terms of robustness, completeness and geometric accuracy. The method is robust and deals properly with partially occluded roadmarks as well as damaged or eroded ones. The detection rate reaches 90% and the 3D accuracy is about 2-4 cm. Finally an application of reconstructed roadmarks is presented. They are used in georeferencing of the system. Most of the MMSs use direct georeferencing devices such as GPS/INS for their localisation. However in urban areas masks and multi-path errors corrupt the measurements and provide only 50 cm accuracy. In order to improve the localisation quality, we aim at matching ground-based images with calibrated aerial images of the same area. For this purpose roadmarks are used as matching objects. The validity of this method is demonstrated on a zebra-crossing example