Thesis (M.S. in Applied Physics)--Naval Postgraduate School, June 2005. / Thesis Advisor(s): Richard M. Harkins, Gamani Karunasiri. Includes bibliographical references (p. 77-78). Also available online.

<p> Studies to date involving the network properties of drainage systems have been theoretical in nature; and the environmental implications of these network characteristics have not been exploited to the extent that would appear warranted. This situation exists due to the lack of an accurate data source. Many studies have recognized this. inadequacy of the conventional data sources to meet the necessary requirements of efficiency (in data production and handling), accuracy, consistency and uniformity. </p> <p> The present study demonstrates that high altitude, small scale colour infrared photography is capable of providing drainage network data that fulfill all these basic requirements. Data derived from the three drainage basins, mapped from a variety of data sources, demonstrate three important points. The level of detail obtained from the small scale colour infrared photography far exceeds that available from more traditional data sources. Secondly, these network data are statistica+ly consistent with the traditional data sources. Thirdly, the basin characteristics derived from the high altitude data source show a marked association with the known surficial environments and an expected variation from one surficial environment to another. </p> / Thesis / Master of Arts (MA)

Imagens térmicas, ou infravermelhas, contém somente informações funcionais do corpo e nenhuma informação anatômica. A falta de um referencial anatômico pode dificultar a inspeção visual da imagem em diagnósticos, motivo pelo qual imagens médicas funcionais são combinadas com imagens anatômicas, como no caso da fusão de imagens anatômicas de ressonância ou ultrassom com imagens funcionais de tomografia computadorizada e PET. Considerando isso, este trabalho apresenta um método para registro e fusão de imagens térmicas com imagens anatômicas, a fim de acrescentar referências anatômicas a imagens térmicas e facilitar a sua analise em diagnósticos. Devido as diferenças morfológicas entre as imagens anatômicas e funcionais a deformação de ao menos uma delas e necessária. Para isso foi adotado o método de transformação não rígida por interpolação utilizando B-splines com marcadores manualmente posicionados. Não há relatos da aplicação desta abordagem no registro e fusão de imagens térmicas em imagens de atlas anatômicos, apesar desta ser utilizada em outras áreas. Foi realizada uma serie de testes com imagens térmicas e anatômicas, essas últimas provindas de atlas anatômicos e fotografias de luz visível, de diferentes partes do corpo humano. O método apresentado e capaz de incluir informações de imagens anatômicas em imagens térmicas pelo registro e fusão destas, apresentando erros de posicionamento da ordem de 5 mm. / Thermal images, or infrared images, contain only functional information about the body, without anatomical information. The lack of an anatomical reference can compromise visual inspection. For this reason, functional medical images are combined with anatomical images, as in the case of the fusion of magnetic resonance or ultrasound anatomical images and PET functional images. Considering this, the present work presents a method for registration and fusion of thermal and anatomical images, with the purpose of providing anatomical references to thermal images and facilitate its analysis in diagnoses. Due to the morphological differences between the anatomical and functional images, it is necessary to transform at least one of them. For this purpose a method of non-rigid transformation with B-splines and manual landmarks was adopted. There are no reports of the application of this technique to thermal images and images from anatomical atlas, despite being used in other fields. Experiments were performed with thermal and anatomical images, the latter from atlas and visible light photographs, from different parts of the human body. The proposed method is capable of including information from anatomical images to thermal images through their registration and fusion, with errors in the order of 5 mm.

Imagens térmicas, ou infravermelhas, contém somente informações funcionais do corpo e nenhuma informação anatômica. A falta de um referencial anatômico pode dificultar a inspeção visual da imagem em diagnósticos, motivo pelo qual imagens médicas funcionais são combinadas com imagens anatômicas, como no caso da fusão de imagens anatômicas de ressonância ou ultrassom com imagens funcionais de tomografia computadorizada e PET. Considerando isso, este trabalho apresenta um método para registro e fusão de imagens térmicas com imagens anatômicas, a fim de acrescentar referências anatômicas a imagens térmicas e facilitar a sua analise em diagnósticos. Devido as diferenças morfológicas entre as imagens anatômicas e funcionais a deformação de ao menos uma delas e necessária. Para isso foi adotado o método de transformação não rígida por interpolação utilizando B-splines com marcadores manualmente posicionados. Não há relatos da aplicação desta abordagem no registro e fusão de imagens térmicas em imagens de atlas anatômicos, apesar desta ser utilizada em outras áreas. Foi realizada uma serie de testes com imagens térmicas e anatômicas, essas últimas provindas de atlas anatômicos e fotografias de luz visível, de diferentes partes do corpo humano. O método apresentado e capaz de incluir informações de imagens anatômicas em imagens térmicas pelo registro e fusão destas, apresentando erros de posicionamento da ordem de 5 mm. / Thermal images, or infrared images, contain only functional information about the body, without anatomical information. The lack of an anatomical reference can compromise visual inspection. For this reason, functional medical images are combined with anatomical images, as in the case of the fusion of magnetic resonance or ultrasound anatomical images and PET functional images. Considering this, the present work presents a method for registration and fusion of thermal and anatomical images, with the purpose of providing anatomical references to thermal images and facilitate its analysis in diagnoses. Due to the morphological differences between the anatomical and functional images, it is necessary to transform at least one of them. For this purpose a method of non-rigid transformation with B-splines and manual landmarks was adopted. There are no reports of the application of this technique to thermal images and images from anatomical atlas, despite being used in other fields. Experiments were performed with thermal and anatomical images, the latter from atlas and visible light photographs, from different parts of the human body. The proposed method is capable of including information from anatomical images to thermal images through their registration and fusion, with errors in the order of 5 mm.

The Cone Calorimeter has been used widely for various purposes as a bench - scale apparatus. Originally the retainer frame (edge frame) was designed to reduce unrepresentative edge burning of specimens. In general, the frame has been used in most Cone tests without enough understanding of its effect. It is very important to have one - dimensional (1D) conditions in order to estimate thermal properties of materials. It has been implicitly assumed that the heat conduction in the Cone Calorimeter is 1D using the current specimen preparation. However, the assumption has not been corroborated explicitly to date. The first objective of this study was to evaluate the heat transfer behavior of a Cone specimen by examining its three - dimensional (3D) heat conduction. It is essential to understand the role of wall lining materials when they are exposed to a fire from an ignition source. Full - scale test methods permit an assessment of the performance of a wall lining material. Fire growth models have been developed due to the costly expense associated with full - scale testing. The models require heat flux maps from the ignition burner flame as input data. Work to date was impeded by a lack of detailed spatial characterization of the heat flux maps due to the use of limited instrumentation. To increase the power of fire modeling, accurate and detailed heat flux maps from the ignition burner are essential. High level spatial resolution for surface temperature can be provided from an infrared camera. The second objective of this study was to develop a heat flux mapping procedure for a room test burner flame to a wall configuration with surface temperature information taken from an infrared camera. A prototype experiment is performed using the ISO 9705 test burner to demonstrate the developed heat flux mapping procedure. The results of the experiment allow the heat flux and spatial resolutions of the method to be determined and compared to the methods currently available.

We focused on a non-cooling room temperature microbolometer infrared imaging array device which includes a sensing layer of p-type a-Si:H component layers doped with boron. Boron incorporation and bonding configuration were investigated for a-Si:H films grown by plasma enhanced chemical deposition (PECVD) at varying substrate temperatures, hydrogen dilution of the silane precursor, and dopant to silane ratio using multiple internal reflection infrared spectroscopy (MIR-IR). This study was then confirmed from collaborators via Raman spectroscopy. MIR-IR analyses reveal an interesting counter-balance relationship between boron-doping and hydrogen-dilution growth parameters in PECVD-grown a-Si:H. Specifically, an increase in the hydrogen dilution ratio (H2/SiH4) or substrate temperature was found to increase organization of the silicon lattice in the amorphous films. It resulted in the decrease of the most stable SiH bonding configuration and thus decrease the organization of the film. The new chemical bonding information of a-Si:H thin film was correlated with the various boron doping mechanisms proposed by theoretical calculations. The study revealed the corrosion morphology progression on aluminum alloy (Al, 0.5% Cu) under acidic chloride solution. This is due to defects and a higher copper content at the grain boundary. Direct galvanic current measurement, linear sweep voltammetry (LSV), and Tafel plots are used to measure corrosion current and potential. Hydrogen gas evolution was also observed (for the first time) in Cu/Al bimetallic interface in areas of active corrosion. Mechanistic insight that leads to effective prevention of aluminum bond pad corrosion is explored and discussed. (Chapter 4) Aluminum bond pad corrosion activity and mechanistic insight at a Cu/Al bimetallic interface typically used in microelectronic packages for automotive applications were investigated by means of optical and scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX) and electrochemistry. Screening of corrosion variables (temperature, moisture, chloride ion concentration, pH) have been investigated to find their effect on corrosion rate and to better understand the Al/Cu bimetallic corrosion mechanism. The study revealed the corrosion morphology progression on aluminum alloy (Al, 0.5% Cu) under acidic chloride solution. The corrosion starts as surface roughening which evolves into a dendrite structure and later continues to grow into a mud-crack type corrosion. SEM showed the early stage of corrosion with dendritic formation usually occurs at the grain boundary. This is due to defects and a higher copper content at the grain boundary. The impact of copper bimetallic contact on aluminum corrosion was explored by sputtering copper microdots on aluminum substrate. Copper micropattern screening revealed that the corrosion is activated on the Al/Cu interface area and driven by the large potential difference; it was also seen to proceed at much higher rates than those observed with bare aluminum. Direct galvanic current measurement, linear sweep voltammetry (LSV), and Tafel plots are used to measure corrosion current and potential. Hydrogen gas evolution was also observed (for the first time) in Cu/Al bimetallic interface in areas of active corrosion. Mechanistic insight that leads to effective prevention of aluminum bond pad corrosion is explored and discussed. Micropattern corrosion screening identified hydrogen evolution and bimetallic interface as the root cause of Al pad corrosion that leads to Cu ball lift-off, a fatal defect, in Cu wire bonded device. Complete corrosion inhibition can be achieved by strategically disabling the mutually coupled cathodic and anodic reaction cycles.