Semi-automatic modeling of buildings from digital surface models

Semi-automatic building detection and extraction is a topic of growing interest due to its potential application in such areas as cadastral information systems, cartographic revision, and GIS. One of the existing strategies for building extraction is to use a digital surface model (DSM) represented by a cloud of known points on a visible surface, and comprising features such as trees or buildings. Conventional surface modeling using stereo-matching techniques has its drawbacks, the most obvious being the effect of building height on perspective, shadows, and occlusions. The laser scanner, a recently developed technological tool, can collect accurate DSMs with high spatial frequency. This paper presents a methodology for semi-automatic modeling of buildings which combines a region-growing algorithm with line-detection methods applied over the DSM.

Orientação semi-automática de uma sequência de pares de imagens frontais por fototriangulação a partir de fotocoordenadas extraídas pelo SIFT

Pós-graduação em Ciências Cartográficas - FCT

Semi-supervised dimensionality reduction based on partial least squares for visual analysis of high dimensional data

Dimensionality reduction is employed for visual data analysis as a way to obtaining reduced spaces for high dimensional data or to mapping data directly into 2D or 3D spaces. Although techniques have evolved to improve data segregation on reduced or visual spaces, they have limited capabilities for adjusting the results according to user's knowledge. In this paper, we propose a novel approach to handling both dimensionality reduction and visualization of high dimensional data, taking into account user's input. It employs Partial Least Squares (PLS), a statistical tool to perform retrieval of latent spaces focusing on the discriminability of the data. The method employs a training set for building a highly precise model that can then be applied to a much larger data set very effectively. The reduced data set can be exhibited using various existing visualization techniques. The training data is important to code user's knowledge into the loop. However, this work also devises a strategy for calculating PLS reduced spaces when no training data is available. The approach produces increasingly precise visual mappings as the user feeds back his or her knowledge and is capable of working with small and unbalanced training sets.

ABO histo-blood group antigen expression on the graft endothelium long term after ABO-compatible, non-identical heart transplantation

We recently reported a complete change in the endothelial ABO histo-blood group phenotype of a cardiac allograft long term after B to O mismatched transplantation. In the context of the current controversy on graft recolonization with recipient endothelial cells and its importance in the development of immunological unresponsiveness, we monitored the expression of endothelial ABH histo-blood group antigens of 10 ABO-compatible, non-identical cardiac allografts over an observation period of at least 30 months. ABH antigens as well as markers for endothelial cells, erythrocytes and thrombocytes were investigated retrospectively by immunohistochemistry using monoclonal antibodies on sections of formalin-fixed, paraffin-embedded biopsies and were evaluated semi-quantitatively by microscopy. In contrast to our earlier finding of the change in the endothelial ABO histo-blood group phenotype long term after ABO- mismatched transplantation, we could not confirm this change in 10 compatible but non-identical cases.

Multimodal approach in the use of clinical scoring, morphological MRI and biochemical T2-mapping and diffusion-weighted imaging in their ability to assess differences between cartilage repair tissue after microfracture therapy and matrix-associated autologous chondrocyte transplantation: a pilot study

OBJECTIVE: The aim of the present pilot study is to show initial results of a multimodal approach using clinical scoring, morphological magnetic resonance imaging (MRI) and biochemical T2-relaxation and diffusion-weighted imaging (DWI) in their ability to assess differences between cartilage repair tissue after microfracture therapy (MFX) and matrix-associated autologous chondrocyte transplantation (MACT). METHOD: Twenty patients were cross-sectionally evaluated at different post-operative intervals from 12 to 63 months after MFX and 12-59 months after MACT. The two groups were matched by age (MFX: 36.0+/-10.4 years; MACT: 35.1+/-7.7 years) and post-operative interval (MFX: 32.6+/-16.7 months; MACT: 31.7+/-18.3 months). After clinical evaluation using the Lysholm score, 3T-MRI was performed obtaining the MR observation of cartilage repair tissue (MOCART) score as well as T2-mapping and DWI for multi-parametric MRI. Quantitative T2-relaxation was achieved using a multi-echo spin-echo sequence; semi-quantitative diffusion-quotient (signal intensity without diffusion-weighting divided by signal intensity with diffusion weighting) was prepared by a partially balanced, steady-state gradient-echo pulse sequence. RESULTS: No differences in Lysholm (P=0.420) or MOCART (P=0.209) score were observed between MFX and MACT. T2-mapping showed lower T2 values after MFX compared to MACT (P=0.039). DWI distinguished between healthy cartilage and cartilage repair tissue in both procedures (MFX: P=0.001; MACT: P=0.007). Correlations were found between the Lysholm and the MOCART score (Pearson: 0.484; P=0.031), between the Lysholm score and DWI (Pearson:-0.557; P=0.011) and a trend between the Lysholm score and T2 (Person: 0.304; P=0.193). CONCLUSION: Using T2-mapping and DWI, additional information could be gained compared to clinical scoring or morphological MRI. In combination clinical, MR-morphological and MR-biochemical parameters can be seen as a promising multimodal tool in the follow-up of cartilage repair.

Multi-temporal mapping of seagrass cover, species and biomass of the Eastern Banks, Moreton Bay, Australia, with links to shapefiles

The spatial and temporal dynamics of seagrasses have been studied from the leaf to patch (100 m**2) scales. However, landscape scale (> 100 km**2) seagrass population dynamics are unresolved in seagrass ecology. Previous remote sensing approaches have lacked the temporal or spatial resolution, or ecologically appropriate mapping, to fully address this issue. This paper presents a robust, semi-automated object-based image analysis approach for mapping dominant seagrass species, percentage cover and above ground biomass using a time series of field data and coincident high spatial resolution satellite imagery. The study area was a 142 km**2 shallow, clear water seagrass habitat (the Eastern Banks, Moreton Bay, Australia). Nine data sets acquired between 2004 and 2013 were used to create seagrass species and percentage cover maps through the integration of seagrass photo transect field data, and atmospherically and geometrically corrected high spatial resolution satellite image data (WorldView-2, IKONOS and Quickbird-2) using an object based image analysis approach. Biomass maps were derived using empirical models trained with in-situ above ground biomass data per seagrass species. Maps and summary plots identified inter- and intra-annual variation of seagrass species composition, percentage cover level and above ground biomass. The methods provide a rigorous approach for field and image data collection and pre-processing, a semi-automated approach to extract seagrass species and cover maps and assess accuracy, and the subsequent empirical modelling of seagrass biomass. The resultant maps provide a fundamental data set for understanding landscape scale seagrass dynamics in a shallow water environment. Our findings provide proof of concept for the use of time-series analysis of remotely sensed seagrass products for use in seagrass ecology and management.

Semi-supervised subspace clustering and applications to neuroscience

Machine learning techniques are used for extracting valuable knowledge from data. Nowa¬days, these techniques are becoming even more important due to the evolution in data ac¬quisition and storage, which is leading to data with different characteristics that must be exploited. Therefore, advances in data collection must be accompanied with advances in machine learning techniques to solve new challenges that might arise, on both academic and real applications. There are several machine learning techniques depending on both data characteristics and purpose. Unsupervised classification or clustering is one of the most known techniques when data lack of supervision (unlabeled data) and the aim is to discover data groups (clusters) according to their similarity. On the other hand, supervised classification needs data with supervision (labeled data) and its aim is to make predictions about labels of new data. The presence of data labels is a very important characteristic that guides not only the learning task but also other related tasks such as validation. When only some of the available data are labeled whereas the others remain unlabeled (partially labeled data), neither clustering nor supervised classification can be used. This scenario, which is becoming common nowadays because of labeling process ignorance or cost, is tackled with semi-supervised learning techniques. This thesis focuses on the branch of semi-supervised learning closest to clustering, i.e., to discover clusters using available labels as support to guide and improve the clustering process. Another important data characteristic, different from the presence of data labels, is the relevance or not of data features. Data are characterized by features, but it is possible that not all of them are relevant, or equally relevant, for the learning process. A recent clustering tendency, related to data relevance and called subspace clustering, claims that different clusters might be described by different feature subsets. This differs from traditional solutions to data relevance problem, where a single feature subset (usually the complete set of original features) is found and used to perform the clustering process. The proximity of this work to clustering leads to the first goal of this thesis. As commented above, clustering validation is a difficult task due to the absence of data labels. Although there are many indices that can be used to assess the quality of clustering solutions, these validations depend on clustering algorithms and data characteristics. Hence, in the first goal three known clustering algorithms are used to cluster data with outliers and noise, to critically study how some of the most known validation indices behave. The main goal of this work is however to combine semi-supervised clustering with subspace clustering to obtain clustering solutions that can be correctly validated by using either known indices or expert opinions. Two different algorithms are proposed from different points of view to discover clusters characterized by different subspaces. For the first algorithm, available data labels are used for searching for subspaces firstly, before searching for clusters. This algorithm assigns each instance to only one cluster (hard clustering) and is based on mapping known labels to subspaces using supervised classification techniques. Subspaces are then used to find clusters using traditional clustering techniques. The second algorithm uses available data labels to search for subspaces and clusters at the same time in an iterative process. This algorithm assigns each instance to each cluster based on a membership probability (soft clustering) and is based on integrating known labels and the search for subspaces into a model-based clustering approach. The different proposals are tested using different real and synthetic databases, and comparisons to other methods are also included when appropriate. Finally, as an example of real and current application, different machine learning tech¬niques, including one of the proposals of this work (the most sophisticated one) are applied to a task of one of the most challenging biological problems nowadays, the human brain model¬ing. Specifically, expert neuroscientists do not agree with a neuron classification for the brain cortex, which makes impossible not only any modeling attempt but also the day-to-day work without a common way to name neurons. Therefore, machine learning techniques may help to get an accepted solution to this problem, which can be an important milestone for future research in neuroscience. Resumen Las técnicas de aprendizaje automático se usan para extraer información valiosa de datos. Hoy en día, la importancia de estas técnicas está siendo incluso mayor, debido a que la evolución en la adquisición y almacenamiento de datos está llevando a datos con diferentes características que deben ser explotadas. Por lo tanto, los avances en la recolección de datos deben ir ligados a avances en las técnicas de aprendizaje automático para resolver nuevos retos que pueden aparecer, tanto en aplicaciones académicas como reales. Existen varias técnicas de aprendizaje automático dependiendo de las características de los datos y del propósito. La clasificación no supervisada o clustering es una de las técnicas más conocidas cuando los datos carecen de supervisión (datos sin etiqueta), siendo el objetivo descubrir nuevos grupos (agrupaciones) dependiendo de la similitud de los datos. Por otra parte, la clasificación supervisada necesita datos con supervisión (datos etiquetados) y su objetivo es realizar predicciones sobre las etiquetas de nuevos datos. La presencia de las etiquetas es una característica muy importante que guía no solo el aprendizaje sino también otras tareas relacionadas como la validación. Cuando solo algunos de los datos disponibles están etiquetados, mientras que el resto permanece sin etiqueta (datos parcialmente etiquetados), ni el clustering ni la clasificación supervisada se pueden utilizar. Este escenario, que está llegando a ser común hoy en día debido a la ignorancia o el coste del proceso de etiquetado, es abordado utilizando técnicas de aprendizaje semi-supervisadas. Esta tesis trata la rama del aprendizaje semi-supervisado más cercana al clustering, es decir, descubrir agrupaciones utilizando las etiquetas disponibles como apoyo para guiar y mejorar el proceso de clustering. Otra característica importante de los datos, distinta de la presencia de etiquetas, es la relevancia o no de los atributos de los datos. Los datos se caracterizan por atributos, pero es posible que no todos ellos sean relevantes, o igualmente relevantes, para el proceso de aprendizaje. Una tendencia reciente en clustering, relacionada con la relevancia de los datos y llamada clustering en subespacios, afirma que agrupaciones diferentes pueden estar descritas por subconjuntos de atributos diferentes. Esto difiere de las soluciones tradicionales para el problema de la relevancia de los datos, en las que se busca un único subconjunto de atributos (normalmente el conjunto original de atributos) y se utiliza para realizar el proceso de clustering. La cercanía de este trabajo con el clustering lleva al primer objetivo de la tesis. Como se ha comentado previamente, la validación en clustering es una tarea difícil debido a la ausencia de etiquetas. Aunque existen muchos índices que pueden usarse para evaluar la calidad de las soluciones de clustering, estas validaciones dependen de los algoritmos de clustering utilizados y de las características de los datos. Por lo tanto, en el primer objetivo tres conocidos algoritmos se usan para agrupar datos con valores atípicos y ruido para estudiar de forma crítica cómo se comportan algunos de los índices de validación más conocidos. El objetivo principal de este trabajo sin embargo es combinar clustering semi-supervisado con clustering en subespacios para obtener soluciones de clustering que puedan ser validadas de forma correcta utilizando índices conocidos u opiniones expertas. Se proponen dos algoritmos desde dos puntos de vista diferentes para descubrir agrupaciones caracterizadas por diferentes subespacios. Para el primer algoritmo, las etiquetas disponibles se usan para bus¬car en primer lugar los subespacios antes de buscar las agrupaciones. Este algoritmo asigna cada instancia a un único cluster (hard clustering) y se basa en mapear las etiquetas cono-cidas a subespacios utilizando técnicas de clasificación supervisada. El segundo algoritmo utiliza las etiquetas disponibles para buscar de forma simultánea los subespacios y las agru¬paciones en un proceso iterativo. Este algoritmo asigna cada instancia a cada cluster con una probabilidad de pertenencia (soft clustering) y se basa en integrar las etiquetas conocidas y la búsqueda en subespacios dentro de clustering basado en modelos. Las propuestas son probadas utilizando diferentes bases de datos reales y sintéticas, incluyendo comparaciones con otros métodos cuando resulten apropiadas. Finalmente, a modo de ejemplo de una aplicación real y actual, se aplican diferentes técnicas de aprendizaje automático, incluyendo una de las propuestas de este trabajo (la más sofisticada) a una tarea de uno de los problemas biológicos más desafiantes hoy en día, el modelado del cerebro humano. Específicamente, expertos neurocientíficos no se ponen de acuerdo en una clasificación de neuronas para la corteza cerebral, lo que imposibilita no sólo cualquier intento de modelado sino también el trabajo del día a día al no tener una forma estándar de llamar a las neuronas. Por lo tanto, las técnicas de aprendizaje automático pueden ayudar a conseguir una solución aceptada para este problema, lo cual puede ser un importante hito para investigaciones futuras en neurociencia.

3D plane-based egomotion for SLAM on semi-structured environment

Several works deal with 3D data in SLAM problem. Data come from a 3D laser sweeping unit or a stereo camera, both providing a huge amount of data. In this paper, we detail an efficient method to extract planar patches from 3D raw data. Then, we use these patches in an ICP-like method in order to address the SLAM problem. Using ICP with planes is not a trivial task. It needs some adaptation from the original ICP. Some promising results are shown for outdoor environment.

Calmness modulus of linear semi-infinite programs

Our main goal is to compute or estimate the calmness modulus of the argmin mapping of linear semi-infinite optimization problems under canonical perturbations, i.e., perturbations of the objective function together with continuous perturbations of the right-hand side of the constraint system (with respect to an index ranging in a compact Hausdorff space). Specifically, we provide a lower bound on the calmness modulus for semi-infinite programs with unique optimal solution which turns out to be the exact modulus when the problem is finitely constrained. The relationship between the calmness of the argmin mapping and the same property for the (sub)level set mapping (with respect to the objective function), for semi-infinite programs and without requiring the uniqueness of the nominal solution, is explored, too, providing an upper bound on the calmness modulus of the argmin mapping. When confined to finitely constrained problems, we also provide a computable upper bound as it only relies on the nominal data and parameters, not involving elements in a neighborhood. Illustrative examples are provided.

Calmness of the Feasible Set Mapping for Linear Inequality Systems

In this paper we deal with parameterized linear inequality systems in the n-dimensional Euclidean space, whose coefficients depend continuosly on an index ranging in a compact Hausdorff space. The paper is developed in two different parametric settings: the one of only right-hand-side perturbations of the linear system, and that in which both sides of the system can be perturbed. Appealing to the backgrounds on the calmness property, and exploiting the specifics of the current linear structure, we derive different characterizations of the calmness of the feasible set mapping, and provide an operative expresion for the calmness modulus when confined to finite systems. In the paper, the role played by the Abadie constraint qualification in relation to calmness is clarified, and illustrated by different examples. We point out that this approach has the virtue of tackling the calmness property exclusively in terms of the system’s data.

Infrared microspectroscopic mapping of the homogeneity of extruded blends: Application to starch/polyester blends

Blends of starch and a biodegradable polyester, produced by an extrusion process, which included a cross-linker/compatibilizer (maleic anhydride) and an initiator (dicumyl peroxide), were studied by infrared (IR) microspectroscopy using an attenuated total reflectance (ATR) objective. Extruded material, which had a diameter of about 3 mm, was sectioned and embedded in epoxy resin prior to IR analysis. Spectra were collected in a grid pattern across the sectioned face of the sample. Measurement of various band parameters from the spectra allowed IR maps to be constructed containing semi-quantitative information about the distribution of blend components. These maps showed the quality of the blend on a microscopic scale and showed how it varied with different concentrations of compatibilizer and initiator. (c) 2005 Elsevier Ltd. All rights reserved.

Linking riparian vegetation spatial structure in Australian tropical savannas to ecosystem health indicators: semi-variogram analysis of high spatial resolution satellite imagery

Government agencies responsible for riparian environments are assessing the combined utility of field survey and remote sensing for mapping and monitoring indicators of riparian zone health. The objective of this work was to determine if the structural attributes of savanna riparian zones in northern Australia can be detected from commercially available remotely sensed image data. Two QuickBird images and coincident field data covering sections of the Daly River and the South Alligator River - Barramundie Creek in the Northern Territory were used. Semi-variograms were calculated to determine the characteristic spatial scales of riparian zone features, both vegetative and landform. Interpretation of semi-variograms showed that structural dimensions of riparian environments could be detected and estimated from the QuickBird image data. The results also show that selecting the correct spatial resolution and spectral bands is essential to maximize the accuracy of mapping spatial characteristics of savanna riparian features. The distribution of foliage projective cover of riparian vegetation affected spectral reflectance variations in individual spectral bands differently. Pan-sharpened image data enabled small-scale information extraction (< 6 m) on riparian zone structural parameters. The semi-variogram analysis results provide the basis for an inversion approach using high spatial resolution satellite image data to map indicators of savanna riparian zone health.

Developments of the generative topographic mapping

The generative topographic mapping (GTM) model was introduced by Bishop et al. (1998, Neural Comput. 10(1), 215-234) as a probabilistic re- formulation of the self-organizing map (SOM). It offers a number of advantages compared with the standard SOM, and has already been used in a variety of applications. In this paper we report on several extensions of the GTM, including an incremental version of the EM algorithm for estimating the model parameters, the use of local subspace models, extensions to mixed discrete and continuous data, semi-linear models which permit the use of high-dimensional manifolds whilst avoiding computational intractability, Bayesian inference applied to hyper-parameters, and an alternative framework for the GTM based on Gaussian processes. All of these developments directly exploit the probabilistic structure of the GTM, thereby allowing the underlying modelling assumptions to be made explicit. They also highlight the advantages of adopting a consistent probabilistic framework for the formulation of pattern recognition algorithms.

Semi-supervised construction of general visualization hierarchies

We have recently developed a principled approach to interactive non-linear hierarchical visualization [8] based on the Generative Topographic Mapping (GTM). Hierarchical plots are needed when a single visualization plot is not sufficient (e.g. when dealing with large quantities of data). In this paper we extend our system by giving the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in [8], whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of GTMs is used. The latter is particularly useful when the plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a data set of 2300 18-dimensional points and mention extension of our system to accommodate discrete data types.

Semi-supervised learning of hierarchical latent trait models for data visualisation

An interactive hierarchical Generative Topographic Mapping (HGTM) ¸iteH_GTM has been developed to visualise complex data sets. In this paper, we build a more general visualisation system by extending the HGTM visualisation system in 3 directions: bf (1) We generalize HGTM to noise models from the exponential family of distributions. The basic building block is the Latent Trait Model (LTM) developed in ¸iteKaban_pami. bf (2) We give the user a choice of initializing the child plots of the current plot in either em interactive, or em automatic mode. In the interactive mode the user interactively selects ``regions of interest'' as in ¸iteH_GTM, whereas in the automatic mode an unsupervised minimum message length (MML)-driven construction of a mixture of LTMs is employed. bf (3) We derive general formulas for magnification factors in latent trait models. Magnification factors are a useful tool to improve our understanding of the visualisation plots, since they can highlight the boundaries between data clusters. The unsupervised construction is particularly useful when high-level plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. We illustrate our approach on a toy example and apply our system to three more complex real data sets.