Orientador: Antonio Manoel Mansanares / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Fisica Gleb Wataghin / Made available in DSpace on 2018-09-26T13:55:51Z (GMT). No. of bitstreams: 1 Freitas_LauraRamosde_D.pdf: 4490607 bytes, checksum: c7b66729395ef47ae8f2ab3bb212544e (MD5) Previous issue date: 2005 / Resumo: A Microscopia Fototérmica de Reflexão vem sendo utilizada na investigação de dispositivos micro e opto-eletrônicos em operação, devido ao seu caráter não destrutivo e por não requerer contato com a superfície da amostra. Esta técnica se baseia na dependência da refletância da amostra com a temperatura, com o campo elétrico local, bem como com a densidade de portadores livres, que são, por sua vez, afetados por defeitos. Este fato torna esta técnica muito adequada para investigar defeitos em processos de fabricação e envelhecimento destas estruturas. Neste trabalho, apresentamos um estudo experimental e teórico sobre a resposta fototérmica em função do comprimento de onda do feixe de prova, para estruturas micro-eletrônicas estratificadas. As amostras consistiram basicamente de trilhas condutoras de silício policristalino de diversos chips. As medidas de termo-refletância foram realizadas na faixa de comprimento de onda de 450nm até 750nm, sendo que as trilhas foram alimentadas sempre com corrente modulada, com três montagens experimentais diferentes. Um padrão oscilatório é observado na região espectral em que a camada superior é transparente. Essas oscilações são causadas pelas múltiplas reflexões nas interfaces. Utilizando um modelo termo-óptico, mostramos que as constantes (n e k), que dependem do comprimento de onda, assim como suas derivadas com relação à temperatura (dn/dT e dk/dT), influenciam fortemente o sinal de termo-refletância. A espessura óptica das camadas, principalmente determinadas pela parte real dos índices de refração, define o período de oscilação. Por outro lado, a parte imaginária estabelece o comprimento de onda em que as oscilações começam. Abaixo de um certo comprimento de onda, a luz de prova não penetra no material e a refletância da camada superficial domina o sinal / Abstract: Photothermal microscopy has been used as a suitable technique for the investigation of micro- and opto-electronic devices in operating cycle, because of its non-contact and non-destructive character. This technique is based on the dependence of the sample reflectance with temperature and with local electric field, as well as with free carrier density, which are in their turn disturbed by defects. This fact makes this technique very useful for investigating defects in fabrication and aging processes of such structures. In the present work, we report an experimental and theoretical study of the thermoreflectance response as a function of the probe wavelength for layered microelectronics structures. The investigated samples consisted of polycrystalline silicon conducting tracks from various chips. Thermore²ectance measurements were carried out in the wavelength range from 450 to 750 nm with the tracks biased in modulated regime, with three diÿerent experimental setups. An oscillating pattern is observed in the spectral region where the upper layer is transparent. Such oscillations are due to the interference resulting from the multiple reflections at the interfaces. Using a thermo-optical model, we show that the optical constants (n and k) of the materials, which are wavelength dependents, as well as their temperature derivatives (dn/dT and dk/dT), strongly in²uence the thermoreflectance signal. The optical thicknesses of the layers, mainly determined by the real part of the refractive indexes, de½ne the period of oscillation. On the other hand, the imaginary part of the refractive indexes establishes the cutoÿ wavelength of the oscillations. Below this cutoÿ wavelength, the probe light does not penetrate the material, and the upper surface reflectance dominates the signal / Doutorado / Física da Matéria Condensada / Doutor em Ciências

O objetivo deste trabalho foi estimar o volume de madeira de um povoamento jovem de Pinus elliottii, localizado no litoral sudeste do Rio Grande do Sul, com imagens dos sensores LISS-III/ResourceSat-1 e TM/Landsat 5, comparando o desempenho destes para tal. Obtiveram-se imagens de setembro de 2010, mês coincidente com o inventário florestal feito na área de estudo. Os valores de reflectância espectral de superfície foram recuperados das imagens originais. Após o georreferenciamento, dos pixels coincidentes com a localização das unidades amostrais do inventário florestal foram extraídos os valores das reflectâncias nas quatro bandas espectrais equivalentes aos dois sensores, cujas respostas foram comparadas. Além das bandas espectrais foram utilizados os índices de vegetação (IV’s) SR, NDVI, SAVI, MVI e GNDVI. Também, foi proposto o ajuste destes IV’s originais pela idade do povoamento, os quais foram identificados por SR_i, NDVI_i, MVI_i e GNDVI_i. A aplicação do logaritmo nas bandas espectrais melhorou os valores dos coeficientes de correlação linear (r), à exceção do IVP, retornando valores entre 0,69 (IVP) a 0,83 (Verde) para o LISS-III e entre 0,68 (Vermelho) a 0,79 (IVM) para o TM; Com os IV’s o logaritmo melhorou os valores de r somente para os IV’s originais, retornando valores de r entre 0,77 (NDVI) a 0,84 (GNDVI) com o LISS-III e entre 0,73 (NDVI) a 0,82 (MVI) para o TM. Com os IV’s ajustados pela idade do povoamento a logaritimização não se mostrou necessária para melhorar a associação linear, retornando valores de r entre 0,79 (NDVI_i) a 0,82 (MVI_i) com o LISS-III e entre 0,74 (SR_i) a 0,80 (MVI_i) com o TM. Além disso, o ajuste pela idade aumentou o intervalo dinâmico dos IV’s ajustados, e, aparentemente, aumentou a sensibilidade nos povoamentos de maior volume. Diferenças significativas na associação linear entre os dados espectrais do TM e LISS-III com o volume só foram encontradas na banda equivalente do verde. Com dados TM, a equação melhor ajustada explicou 68% da variabilidade do volume; com dados LISS-III a equação explicou 72% da variabilidade. Estas equações geraram dois mapas de volume de madeira, onde as médias das estimativas obtidas com LISS-III estiveram dentro do intervalo de confiança da média do inventário florestal em 70% dos talhões considerados; para o TM a coincidência foi de 65% dos talhões. Conclui-se que os sensores LISS-III e TM apresentam alta similaridade e que a metodologia empregada pode ser utilizada para auxiliar no inventário florestal dos povoamentos jovens de P. elliottii na área de estudo principalmente pelo fato das estimativas obtidas pelas imagens cobrirem todo o talhão, ao passo que a amostragem do inventário florestal contempla menos de 2% da área. / The aim of this work was to estimate the wood volume of a young stand of Pinus elliottii, located on the southeastern coast of the state of Rio Grande do Sul, by imagery from LISS-III/ResourceSat-1 and TM/Landsat 5 sensors, comparing their performance for such. Images were obtained on September 2010, the month coincident with the forest inventory made in the study area. The surface spectral reflectance values were retrieved from the original images. After the georeferencing, the sampling units location from the forest inventory were used to select the pixels to extract the reflectance values on the four spectral bands equivalents for the two sensors, which answers were compared. In addition to the bands were used the Vegetation Indices (VI’s) SR, NDVI, SAVI, MVI and GNDVI. Also proposed was the adjusting of these original VI’s by the stand age, which ones were identified by SR_i, NDVI_i, MVI_i and GNDVI_i. The application of logarithm in the spectral bands improved the r values, with exception to NIR, achieving values between 0.69 (NIR) and 0.83 (Green) for LISS-III and between 0.68 (Red) and 0.79 (SWIR) for TM; With the VI’s, the logarithm improved the r values only for the original VI’s, returning r values from 0.77 (NDVI) to 0.84 (GNDVI) with LISS-III and r values from 0.73 (NDVI) to 0.82 (MVI) for TM. With the VI’s adjusted by stand age the logarithm was not necessary to improve the linear association, returning r values from 0.79 (NDVI_i) to 0.82 (MVI_i) with LISS-III and r values from 0.74 (SR_i) to 0.80 (MVI_i) with TM. Moreover, adjusting by age increased the dynamic range of the VI’s adjusted, and apparently increased the sensitivity in stands with larger volume. Significant differences in the linear association between TM and LISS-III spectral data with volume were just found on the green equivalent band. With TM data, the best fitted model explained 68% of the volume variability; with LISS-III data the model explained 72% of the variability. These models generated two wood volume maps, where the average of the estimates achieved with LISS-III were within the confidence level of the average from the forest inventory on 70% of the compartments considered; for TM the coincidence was on 65% of the compartments. It is conclude that the sensors LISS-III and TM presented high similarity and the methodology applied can be used to aid in forest inventory of young stands of P. elliottii in the study area mainly because the estimates obtained by the images cover the entire compartment, while the forest inventory sampling contemplates less than 2% of the area.

This thesis describes the process of creating an augmented reality mobile application which allows designers, architects and researchers to retrieve accurate colour information picked from the real-world environment. Specifically, the goal is not only to obtain colour coordinates for the area of interest but to find a set of matching colour samples in an extensive database of colour atlases provided with the application. To properly understand how the camera sensor perceives colour under current lighting conditions, it must be calibrated beforehand using a physical colour chart. Based on this calibration, we can estimate the picked colour coordinates in a standardised colour space. The mobility aspect is the main advantage of the resulting application. Instead of carrying multiple colour atlases, the user can estimate colour matching just using a mobile device and a portable colour chart.

Many of the colour images captured by different types of digital camera do not provide quality colour image according to human visual perception. In this study we explore technique for colour correction of the colour images from the Waikato Image Ranger. Colour images were captured using three different illuminants with the Waikato Image Ranger. The colour image formed from the ranger data do not have good quality because illuminants used do not match usual RGB standard illuminants. The spectral power distribution values of the illuminants were measured using spectroradiometer. To calculate tristimulus values the reflectance function of the scene is required. A mechanism of calculating the reflectance functions from the ranger data using genetic algorithm was explained. The reflectance functions are approximated using variable Gaussian basis functions, and fit to the ranger colour triplets by genetic algorithms. From the estimated reflectance functions standard CRT RGB values were calculated. It was found that the genetic algorithm approach was for too slow for practical purposes and produced images with far too much colour variation.

Emergent spatial signals which may be interpreted in the context of fluvial sediment transport processes are detected through the use of reflectance spectroscopy in the sand-sized sediments of the Towamba River, southeastern New South Wales. Reflectance spectroscopy of sufficiently fine spectral resolution represents a technique for mineral composition analysis which is complementary to X-ray diffraction, with advantages in terms of ease of sample preparation and rapidity of measurement. Instrumentation is available allowing high-quality spectrum acquisition in the field and from airborne and satellite-borne instruments. The former allows mineral analyses to more easily be incorporated into sediment tracing studies as an additional variable. The latter offers large scale, repeatable areal coverage of a dynamic system in which sediments are exposed to the sky. The Towamba River drains a catchment of c. 1000 km [square] in extensively altered granitic terrain along the south coast of New South Wales, and carries significant quantities of sand-sized sediment through much of the system. Pervasive but spatially variable chlorite, epidote, and sericite have been described in local and neighbouring terrain. These are spectrally active in the SWIR wavelength region in which the PIMA portable spectrometer operates. The airborne HyMap instrument is sensitive through this range as well as through the visible and near-infrared regions. Conventionally such channel sediments would represent a single class in the context of the broader landscape, and comparatively they represent a domain of restricted variance. In this study of samples of sediment were collected for analysis with the PIMA, the results of which supported the efficacy of such an exercise in a conventional tracing context and supported analysis of HyMap imagery. Although issues related to reduction of HyMap-detected radiance to reflectance prevented effective analysis of the shorter wavelengths sensitive to the presence of ferrous and ferric iron, the consideration of absorption feature depths and the application of a matched filtering operator revealed gross-scale spatial patterns which were interpreted as two populations of sand in the main channel. This interpretation is consistent with bank erosion occurring during two very large magnitude flow events in the 1970s, with minor ongoing perturbation of the sediment signal in the main channel by the contribution of sediment from tributaries. The presence of a definite spatial signal having been established, routes for further investigation are suggested. A noisy signal hypothesised on the basis of imagery may be used to better direct a field sampling program for a conventional sediment tracing study. The signal to noise ratio may be improved for example through calibration of radiance to reflectance and removal of atmospheric interference and improved field sampling schemes, after which more rigorous, quantitative exercises such as geostatistical ???field??? trial or spatial series analysis may be performed. Connections to process through sediment transport models are enabled through the use of GIS.

This research investigates the spectral reflectance characteristics of soil and vegetation using multi-angular and single view hyperspectral data. The question of the thesis is ???How much information can be obtained from multi-angular hyperspectral remote sensing in comparison with single view angle hyperspectral remote sensing of soil and vegetation???? This question is addressed by analysing multi-angular and single view angle hyperspectral remote sensing using data from the field, airborne and space borne hyperspectral sensors. Spectral mapping, spectral indices and Functional Data Analysis (FDA) are used to analyse the data. Spectral mapping has been successfully used to distinguish features of soil and cotton with hyperspectral data. Traditionally, spectral mapping is based on collecting endmembers of pure pixels and using these as training areas for supervised classification. There are, however, limitations in the use of these algorithms when applied to multi-angular images, as the reflectance of a single ground unit will differ at each angle. Classifications using six-class endmembers identified using single angle imagery were assessed using multi-angular Compact High Resolution Imaging Spectrometer (CHRIS) imagery, as well as a set of vegetation indices. The results showed no significant difference between the angles. Low nutrient content in the soil produced lower vegetation index values, and more nutrients increased the index values. This research introduces FDA as an image processing tool for multi-angular hyperspectral imagery of soil and cotton, using basis functions for functional principal component analysis (fPCA) and functional linear modelling. FDA has advantages over conventional statistical analysis because it does not assume the errors in the data are independent and uncorrelated. Investigations showed that B-splines with 20-basis functions was the best fit for multi-angular soil spectra collected using the spectroradiometer and the satellite mounted CHRIS. Cotton spectra collected from greenhouse plants using a spectrodiometer needed 30-basis functions to fit the model, while 20-basis functions were sufficient for cotton spectra extracted from CHRIS. Functional principal component analysis (fPCA) of multi-angular soil spectra show the first fPCA explained a minimum of 92.5% of the variance of field soil spectra for different azimuth and zenith angles and 93.2% from CHRIS for the same target. For cotton, more than 93.6% of greenhouse trial and 70.6% from the CHRIS data were explained by the first fPCA. Conventional analysis of multi-angular hyperspectral data showed significant differences exist between soil spectra acquired at different azimuth and zenith angles. Forward scan direction of zenith angle provides higher spectral reflectance than backward direction. However, most multi-angular hyperspectral data analysed as functional data show no significant difference from nadir, except for small parts of the wavelength of cotton spectra using CHRIS. There is also no significant difference for soil spectra analysed as functional data collected from the field, although there was some difference for soil spectra extracted from CHRIS. Overall, the results indicate that multi-angular hyperspectral data provides only a very small amount of additional information when used for conventional analyses.

Humans recognize optical reflectance properties of surfaces such as metal, plastic, or paper from a single image without knowledge of illumination. We develop a machine vision system to perform similar recognition tasks automatically. Reflectance estimation under unknown, arbitrary illumination proves highly underconstrained due to the variety of potential illumination distributions and surface reflectance properties. We have found that the spatial structure of real-world illumination possesses some of the statistical regularities observed in the natural image statistics literature. A human or computer vision system may be able to exploit this prior information to determine the most likely surface reflectance given an observed image. We develop an algorithm for reflectance classification under unknown real-world illumination, which learns relationships between surface reflectance and certain features (statistics) computed from a single observed image. We also develop an automatic feature selection method.

Under normal viewing conditions, humans find it easy to distinguish between objects made out of different materials such as plastic, metal, or paper. Untextured materials such as these have different surface reflectance properties, including lightness and gloss. With single isolated images and unknown illumination conditions, the task of estimating surface reflectance is highly underconstrained, because many combinations of reflection and illumination are consistent with a given image. In order to work out how humans estimate surface reflectance properties, we asked subjects to match the appearance of isolated spheres taken out of their original contexts. We found that subjects were able to perform the task accurately and reliably without contextual information to specify the illumination. The spheres were rendered under a variety of artificial illuminations, such as a single point light source, and a number of photographically-captured real-world illuminations from both indoor and outdoor scenes. Subjects performed more accurately for stimuli viewed under real-world patterns of illumination than under artificial illuminations, suggesting that subjects use stored assumptions about the regularities of real-world illuminations to solve the ill-posed problem.

This paper describes a machine vision system that classifies reflectance properties of surfaces such as metal, plastic, or paper, under unknown real-world illumination. We demonstrate performance of our algorithm for surfaces of arbitrary geometry. Reflectance estimation under arbitrary omnidirectional illumination proves highly underconstrained. Our reflectance estimation algorithm succeeds by learning relationships between surface reflectance and certain statistics computed from an observed image, which depend on statistical regularities in the spatial structure of real-world illumination. Although the algorithm assumes known geometry, its statistical nature makes it robust to inaccurate geometry estimates.

Humans distinguish materials such as metal, plastic, and paper effortlessly at a glance. Traditional computer vision systems cannot solve this problem at all. Recognizing surface reflectance properties from a single photograph is difficult because the observed image depends heavily on the amount of light incident from every direction. A mirrored sphere, for example, produces a different image in every environment. To make matters worse, two surfaces with different reflectance properties could produce identical images. The mirrored sphere simply reflects its surroundings, so in the right artificial setting, it could mimic the appearance of a matte ping-pong ball. Yet, humans possess an intuitive sense of what materials typically "look like" in the real world. This thesis develops computational algorithms with a similar ability to recognize reflectance properties from photographs under unknown, real-world illumination conditions. Real-world illumination is complex, with light typically incident on a surface from every direction. We find, however, that real-world illumination patterns are not arbitrary. They exhibit highly predictable spatial structure, which we describe largely in the wavelet domain. Although they differ in several respects from the typical photographs, illumination patterns share much of the regularity described in the natural image statistics literature. These properties of real-world illumination lead to predictable image statistics for a surface with given reflectance properties. We construct a system that classifies a surface according to its reflectance from a single photograph under unknown illuminination. Our algorithm learns relationships between surface reflectance and certain statistics computed from the observed image. Like the human visual system, we solve the otherwise underconstrained inverse problem of reflectance estimation by taking advantage of the statistical regularity of illumination. For surfaces with homogeneous reflectance properties and known geometry, our system rivals human performance.