This thesis attempts to create an application to determine head rotation angle in video recorded from a security camera. The application consists of three parts: face detection, facial landmarks detection and determination of person's head rotation. The face detection has been implemented using Viola-Jones and HOG algorithms. Facial landmarks detection has been done using algorithm based on active shape model. Two methods to calculate the head rotation angles have been used: the first method works with anthropometric head features. The second method uses Perspective-n-Point algorithm to find the right rotation angles. Finally, all algorithms implemented have been tested and the proper parameters have been determined.

Diploma thesis is about image sensor security. Goal of the thesis was study data integrity gained from the image sensors. Proposed method is about source camera identification from noise characteristics in image sensors. Research was about influence of denoising algorithms applied to digital images, which was acquired from 15 different image sensors. Finally the statistical evaluation had been done from computed results.

The goal of this work is to create an application, which will combine procedural and keyfram animations with subsequent visualization. Composition of this two different animations techniques is used to animate a virtual character. To combine this two techniques one starts with interpolations from keyframe animation and then enchance them by procedural animations to properly fit into the characters surroundings. This procedural part of animation is obtained by using forward and inverse kinematics. Whole application is written in C++, uses GLM math library for computations and OpenGL and GLUT for final visualization.

Metrology is the science of measurement. It is also a prerequisite for maintaining a high quality in all manufacturing processes. In this thesis we will present the demands and solutions for ultra-precision metrology in the manufacturing of lithography masks for the TV-display industry. The extreme challenge that needs to be overcome is a measurement uncertainty of 10 nm on an absolute scale of more that 2 meters in X and Y. Materials such as metal, ceramic composites, quartz or glass are highly affected by the surrounding temperature when tolerances are specified at nanometer levels. Also the fact that the refractive index of air in the interferometers measuring absolute distances is affected by temperature, pressure, humidity and CO2 contents makes the reference measurements really challenging. This goes hand in hand with the ability of how to design a mask writer, a pattern generator with a performance good enough for writing masks for the display industry with sub-micron accuracy over areas of square meters.  As in many other areas in the industry high quality metrology is the key for success in developing high accuracy production tools. The aim of this thesis is therefore to discuss the metrology requirements of mask making for display screens. Defects that cause stripes in the image of a display, the so called “Mura” effect, are extremely difficult to measure as they are caused by spatially systematic errors in the mask writing process in the range of 10-20 nm. These errors may spatially extend in several hundreds of mm and are superposed by random noise with significantly higher amplitude compared to the 10-20 nm.  A novel method for measuring chromium patterns on glass substrates will also be presented in this thesis. This method will be compared to methods based on CCD and CMOS images. Different methods have been implementedin the Micronic MMS1500 large area measuring machine, which is the metrology tool used by the mask industry, for verifying the masks made by the Micronic mask writers. Using alternative methods in the same system has been very efficient for handling different measurement situations. Some of  the discussed methods are also used by the writers for calibration purposes. / QC 20110517

Ce mémoire porte sur la conception, l'assemblage et la calibration d'un système permettant l'acquisition de l'apparence et de la géométrie d'objets. Une première section porte sur la conception du système, c'est-à-dire sur le choix de son design et sa réalisation. Ensuite, les calibrations nécessaires au fonctionnement du système sont présentées. Les calibrations géométriques et radiométriques des caméras sont tout d'abord réalisées. Ensuite, les sources de lumière sont à leurs tours calibrées géométriquement et radiométriquement. Finalement, une application de stéréo photométrique est effectuée. Enfin, nous modélisons la fonction de réflectance de quelques objets.

Designing good digital experiences for children can be difficult; designers have to consider children's cognitive and motor skill limitations, understand their target audience, create something entertaining and educational, comply with national and international jurisdiction, and at the same time appeal to parents. We set out to create a general framework which designers and developers can use as a foundation and testing ground for their digital products in the field of user experience. The methods used during the thesis include interviews, literature studies, user testing, case studies, personas, prototyping, and more. The results created are primarily user experience guidelines packaged in a Theoretical Framework, user testing conclusions, along with suggestions on improving the current Lego Star Wars: Force Builders application, a few in the form of prototypes.

[ES] La descarga de vapor en forma de chorro en piscinas con agua subenfriada es una forma muy eficiente de condensar vapor y es muy utilizada en diversas aplicaciones industriales, como la industria farmacéutica, metalúrgica, química, sistemas de propulsión sumergida y nuclear. Siendo especialmente esencial en los sistemas de seguridad de las centrales nucleares, concretamente en la piscina de supresión de presión de los Reactores de Agua en Ebullición (BWR, Boiling Water Reactor) y en el tanque de almacenamiento de agua de reabastecimiento en la contención de los reactores avanzados de agua a presión. La eficiencia de esta técnica, denominada Condensación por Contacto Directo (DCC, Direct Contact Condensation), radica en la rápida condensación del vapor proporcionando una alta capacidad de transferencia de calor e intercambio de masa. En las últimas décadas, se han llevado a cabo muchos experimentos sobre los chorros sumergidos de gases no condensables y vapor puro en piscinas, que han proporcionado mucha información de interés, aunque continúan los esfuerzos para adquirir información adicional. Para estudiar en detalle el comportamiento de las descargas de chorros turbulentos se han llevado a cabo una serie de experimentos en una instalación que ha sido diseñada para representar diversos entornos de interés crítico para la seguridad en centrales nucleares. Su singularidad radica en la capacidad de poder reproducir fenomenologías similares a situaciones específicas que podrían surgir en una central nuclear, desde sistemas de despresurización hasta roturas con pérdida de refrigerante. La instalación posibilita el estudio con vapor, aire o una mezcla de ambos utilizando geometrías simples (boquilla) o complejas (sparger) en una piscina con agua en reposo. Se ha utilizado una cámara de alta velocidad con Dispositivo de Carga Acoplada (CCD, Charged-Coupled Device) para capturar imágenes de la descarga del chorro en la piscina y así determinar la interfase entre el gas y el líquido, utilizando técnicas de visualización directa. Para llevar a cabo el procesamiento de dichas imágenes, se han aplicado una serie de procedimientos de procesado, filtrado y post-procesado utilizando una subrutina implementada en MATLAB. Tras la adquisición de las imágenes, se ha caracterizado el comportamiento del chorro de vapor y aire, el efecto de la fracción volumétrica de aire sobre la mezcla y el efecto de la variación del tamaño de la boquilla. Además, se ha medido la longitud del momento del chorro y se ha comprobado su estrecha relación con el diámetro de la boquilla, la velocidad del chorro y el porcentaje de la mezcla. Por otro lado, se ha caracterizado el comportamiento del chorro horizontal en la zona de transición. La transición indica el cambio de un chorro dominado exclusivamente por la fuerza de momento a un régimen en el que las fuerzas de flotación prevalecen. Esta prueba se ha realizado utilizando cinco boquillas intercambiables y para cada caudal de vapor se ha determinado el porcentaje de aire en el que ocurre la transición. Otro de los principales objetivos de esta tesis es la modelación utilizando herramientas de códigos de Dinámica de Fluidos Computacional (CFD), específicamente descargas de chorros de aire en agua subenfriada. Para esto se ha empleado el software ANSYS CFX. El comportamiento del flujo multifásico se ha simulado utilizando el modelo euleriano-euleriano no homogéneo; debido a que es más conveniente para estudiar fluidos interpenetrados. Los resultados CFD se han validado con datos experimentales y resultados teóricos. Las cifras de mérito de la longitud de penetración adimensional y la longitud de flotabilidad adimensional muestran una buena concordancia con las mediciones experimentales. Las correlaciones para estas variables se obtuvieron en función de números adimensionales para dar generalidad utilizando sólo condiciones iniciales de contorno. / [CA] La descàrrega de vapor en forma de xorrat en piscines amb aigua subrefredada és una forma molt eficient de condensar vapor i s'utilitza àmpliament en diverses aplicacions industrials, com la indústria farmacèutica, metal·lúrgica, química, sistemes de propulsió submergida i nuclear. És especialment essencial en els sistemes de seguretat de les centrals nuclears, concretament en la piscina de supressió de pressió dels Reactors d'Aigua en Bulliment (BWR) i en el dipòsit d'aigua de reaprovisionament a la contenció dels reactors avançats d'aigua a pressió. L'eficiència d'aquesta tècnica, anomenada Condensació per Contacte Directe (DCC), rau en la ràpida condensació del vapor proporcionant una alta capacitat de transferència de calor i intercanvi de massa. En les últimes dècades, s'han realitzat molts experiments sobre els raigs submergits de gasos no condensables i vapor pur en piscines, que han proporcionat molta informació d'interès, tot i que continuen els esforços per adquirir informació addicional. Per estudiar amb detall el comportament de les descàrregues de raigs turbulents s'han dut a terme una sèrie d'experiments en una instal·lació que ha estat dissenyada per representar diversos entorns d'interès crític per a la seguretat en centrals nuclears. La seua singularitat rau en la capacitat de poder reproduir fenomenologies semblants a situacions específiques que podrien sorgir en una central nuclear, des de sistemes de despressurització fins a ruptures amb pèrdua de refrigerant. La instal·lació possibilita l'estudi amb vapor, aire o una barreja de tots dos utilitzant geometries simples (busseta) o complexes (sparger) en una piscina amb aigua en repòs. A més, és possible combinar aquestes situacions amb diferents condicions de cabal, pressió o temperatura. S'ha utilitzat una càmera de alta velocitat (CCD) per capturar imatges de la descàrrega del raig a la piscina i així determinar la interfase entre el gas i el líquid, emprant tècniques de visualització directa. Per dur a terme el processament d'aquestes imatges, s'han aplicat una sèrie de procediments complexes de processat, filtrat i post-processat mitjançant una subrutina implementada en MATLAB. Després de l'adquisició de les imatges, s'ha caracteritzat el comportament del raig de vapor i aire, l'efecte de la fracció volumètrica d'aire sobre la barreja i l'efecte de la variació de la grandària de la busseta. A més, s'ha mesurat la longitud del moment del raig i s'ha comprovat la seua estreta relació amb el diàmetre de la busseta, la velocitat del raig i el percentatge de la barreja. D'altra banda, s'ha caracteritzat el comportament del raig horitzontal en la zona de transició. La transició indica el canvi d'un raig dominat exclusivament per la força de moment a un règim en què les forces de flotació prevalen. Aquesta prova s'ha realitzat utilitzant cinc bussetes intercanviables i per a cada cabal de vapor s'ha determinat el percentatge d'aire en què ocorre la transició. Un altre dels principals objectius d'aquesta tesi és la modelització utilitzant eines de codis de Dinàmica de Fluids Computacionals (CFD), específicament descàrregues de xorrades d'aire en aigua subrefredada. Per a això s'ha utilitzat el software ANSYS CFX. El comportament del flux multifàsic s'ha simulat utilitzant el model eulerià-eulerià no homogeni, ja que és més convenient per estudiar fluids interpenetrats. Els resultats CFD s'han validat amb dades experimentals i resultats teòrics. Les figures de mèrit de la longitud de penetració adimensional i la longitud de flotabilitat adimensional mostren una bona concordança amb les mesures experimentals. Les correlacions per a aquestes variables s'han obtingut en funció de nombres adimensionals per donar generalitat utilitzant només condicions inicials de contorn. El model numèric CFD desenvolupat té la capacitat de simular el comportament de gasos no condensables descarregats en aigua. / [EN] The discharge of steam in the form of a jet in pools with subcooled water is a highly efficient way to condense steam and is widely used in various industrial applications, such as the pharmaceutical, metallurgical, chemical, submerged propulsion, and nuclear industries. It is particularly essential in the safety systems of nuclear power plants, specifically in the pressure suppression pool of Boiling Water Reactors (BWR) and in the water replenishment tank in the containment of advanced pressurized water reactors. The efficiency of this technique, known as Direct Contact Condensation (DCC), lies in the rapid condensation of steam, providing a high heat transfer and mass exchange capacity. In recent decades, numerous experiments have been conducted on submerged jets of non-condensable gases and pure steam in pools, yielding valuable information, although efforts continue to gather additional data. To study in detail the behavior of turbulent jet discharges, a series of experiments have been carried out in a facility designed to simulate various critical environments for safety in nuclear power plants. Its uniqueness lies in the ability to reproduce phenomenologies like specific situations that could arise in a nuclear power plant, from depressurization systems to ruptures with coolant loss. The facility enables the study with steam, air, or a mixture of both using simple geometries (nozzle) or complex geometries (sparger) in a pool with quiescent water. Additionally, it is possible to combine these situations with different flow, pressure, or temperature conditions. A high-speed camera with Charge-Couple Device (CCD) has been used to capture images of the jet discharge in the pool, aiming to determine the gas-liquid interface through direct visualization techniques. Following the image acquisition, a series of complex processing, filtering, and post-processing procedures have been applied using a subroutine implemented in MATLAB. After obtaining the images, the behavior of the steam and air jet, the effect of the volumetric fraction of air on the mixture, and the impact of varying the nozzle size have been characterized. Additionally, the length of the jet momentum has been measured, establishing its close relationship with the nozzle diameter, jet velocity, and mixture percentage. Furthermore, the behavior of the horizontal jet in the transition zone has been characterized. The transition denotes the shift from a jet dominated exclusively by momentum force to a regime where buoyancy forces prevail. This test has been conducted using five interchangeable nozzles, and for each steam flow rate, the percentage of air at which the transition occurs has been determined. Another main objective of this thesis is modeling using Computational Fluid Dynamics (CFD) tools, specifically jet discharges of air into subcooled water. The ANSYS CFX software has been employed for this purpose. Multiphase flow behavior has been simulated using the non-homogeneous Eulerian-Eulerian model, as it is more convenient for studying interpenetrated fluids. CFD results have been validated with experimental data and theoretical results. The dimensionless penetration length and dimensionless buoyancy length merit figures show good agreement with experimental measurements. Correlations for these variables were obtained as a function of dimensionless numbers to provide generality using only initial boundary conditions. The developed CFD numerical model has the capability to simulate the behavior of non-condensable gases discharged into water. / Córdova Chávez, Y. (2024). Experimentación y modelación de mezcla de vapor y gases no-condensables [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/207266

Die Arbeit beschäftigt sich mit zerstörungsfreien Untersuchungen von Kristalldefekten an kompakten Proben mittels röntgenographischer Kossel- u. Weitwinkel-Mikrobeugung im Rasterelektronenmikroskop. Das REM wurde durch ein neu entwickeltes Aufnahmeverfahren so erweitert, daß die äußerst intensitätsschwachen Kossel-Röntgeninterferenzen mittels Phosphorszintillator und hochauflösendem, extrem empfindlichen CCD-Flächendetektor registriert werden können. Das aufwendige Röntgenfilmverfahren wurde damit abgelöst. Die Aufnahme-Techniken wurden so kombiniert, daß die sich gegenseitig ergänzenden Methoden, v.a. die Kossel- u. Pseudo-Kossel-, aber auch die Rückstreu-Elektronen-Beugung, erstmals mit einem einzigen CCD-System im REM ausführbar sind. Die Aufnahme von Kikuchi-Bändern wurde so weit verbessert, daß diese erstmalig bei vertikaler Inzidenz des Elektronenstrahls auf der Probe beobachtet werden konnten. Durch Einsatz einer fokussierenden Polykapillarlinse in einem Röntgenfluoreszenzspektrometer konnte die Anregung von Kossel-Linien durch Bremsstrahlung und erheblich kürzere Belichtungszeiten sowie deutlich höhere laterale Auflösungen erzielt werden. Für die komplementären Mikrobeugungsmethoden wurde ein einheitliches Programm entwickelt, dessen neue Art der Simulation komplizierter Weitwinkel-Kurven 4. Ordnung die Lokalisierung von Gitterbaufehlern im kompakten Kristall ermöglichte. Entsprechende Simulationen und Verfeinerungen der Kanüle erlaubten in feinkörnigen Polykristallen, wie Bariumtitanat, eine Einzelkornanalyse mit der Weitwinkel-Beugung. Insbesondere wurden markante Erhöhungen der Versetzungsdichte nahe der Korngrenze einzelner Kristallite in FeAl festgestellt. An intermetallischen Fe-Al-Verbindungen wurden in Weitwinkel-Kurven Feinstrukturen gefunden, die sich durch Umweganregungen in Zusammenhang mit Überstrukturen erklären lassen. An zugverformten Ni-Kristallen wurden Kossel- u. Weitwinkel-Linienbreiten in Abhängigkeit des Azimuts im symmetrischen Bragg-Fall ausgewertet u. mit theoretischen Modellen verglichen. Anisotrope Linienverbreiterungen durch die Wirkung von Stufenversetzungen konnten quantitativ nachgewiesen werden. Erste Kossel- u. Weitwinkel-Untersuchungen an zyklisch verformten Ni-Einkristallen zeigten, daß beobachtete perlenkettenförmige Intensitätsanhäufungen an Weitwinkelreflexen die Konfiguration von Versetzungswänden aus Stufenversetzungsdipolen im Kristallvolumen widerspiegeln. Erstmals konnte der Einfluß von Stapelfehlern auf Weitwinkel-Linien, der sich durch linsenförmige Intensitätsaufspaltungen zeigte, experimentell nachgewiesen, theoretisch erklärt und quantitativ bestimmt werden. / The thesis deals with the nondestructive investigation of crystal defects by X-ray Kossel and Pseudo Kossel microdiffraction on compact specimens in the scanning electron microscope. The SEM was extended by means of a newly developed detection method in such a way that X-ray Kossel interferences, which are extremely faint in intensity, can be observed by a phosphor scintillator as well as a high resolution and ultra-sensitive CCD area detector. The demanding X-ray film method was thus replaced. The observation techniques were combined so that the mutually complementary methods, above all the Kossel and Pseudo Kossel, but also the electron backscattered diffraction, are made possible for the first time by using just one CCD system in the SEM. The detection of Kikuchi bands was improved to such a degree that these could be recorded even at vertical incidence of the electron beam on the specimen for the first time. It was shown that the lateral resolution of the Kossel technique under polychromatic X-ray tube excitation could be enhanced considerably and the exposure times strongly reduced by using a polycapillary lens in an X-ray fluorescence spectrometer. For the complementary microdiffraction methods a homogeneous simulation program was developed, whose novel way of simulating the complicated Pseudo Kossel curves of the fourth order enable the lattice defect localization in compact crystals. The corresponding simulations and refinements of the target tube allowed a single grain analysis also in fine-grained polycrystals like barium titanate with Pseudo Kossel X-ray diffraction. Particularly, a marked increase of the dislocation density was ascertained near the grain boundary of individual crystallites in FeAl. At intermetallic Fe-Al compounds Pseudo Kossel curves fine structures were found, which can be explained by umweg (detour) excitations in relation to superstructures. Kossel and Pseudo Kossel line widths were analyzed in dependence on the azimuth and compared with theoretical models at tensile deformed Ni-crystals in the symmetrical Bragg case. Anisotropic line broadenings through the effect of edge dislocations could be proved quantitatively. Conclusions could be drawn from the initial Kossel and Pseudo Kossel investigations of cyclically deformed nickel crystals with respect to the observed pearl-necklace-like intensity thickening at Pseudo Kossel lines reflecting the strong local variations of the dislocation density and, thus, the configuration of dislocation walls of edge dislocation dipoles inside the crystalline volume. For the first time, the effect of stacking faults on Pseudo Kossel reflections appearing by lens-shaped intensity splittings could be proved experimentally, explained theoretically and determined quantitatively.

Die Arbeit beschäftigt sich mit zerstörungsfreien Untersuchungen von Kristalldefekten an kompakten Proben mittels röntgenographischer Kossel- u. Weitwinkel-Mikrobeugung im Rasterelektronenmikroskop. Das REM wurde durch ein neu entwickeltes Aufnahmeverfahren so erweitert, daß die äußerst intensitätsschwachen Kossel-Röntgeninterferenzen mittels Phosphorszintillator und hochauflösendem, extrem empfindlichen CCD-Flächendetektor registriert werden können. Das aufwendige Röntgenfilmverfahren wurde damit abgelöst. Die Aufnahme-Techniken wurden so kombiniert, daß die sich gegenseitig ergänzenden Methoden, v.a. die Kossel- u. Pseudo-Kossel-, aber auch die Rückstreu-Elektronen-Beugung, erstmals mit einem einzigen CCD-System im REM ausführbar sind. Die Aufnahme von Kikuchi-Bändern wurde so weit verbessert, daß diese erstmalig bei vertikaler Inzidenz des Elektronenstrahls auf der Probe beobachtet werden konnten. Durch Einsatz einer fokussierenden Polykapillarlinse in einem Röntgenfluoreszenzspektrometer konnte die Anregung von Kossel-Linien durch Bremsstrahlung und erheblich kürzere Belichtungszeiten sowie deutlich höhere laterale Auflösungen erzielt werden. Für die komplementären Mikrobeugungsmethoden wurde ein einheitliches Programm entwickelt, dessen neue Art der Simulation komplizierter Weitwinkel-Kurven 4. Ordnung die Lokalisierung von Gitterbaufehlern im kompakten Kristall ermöglichte. Entsprechende Simulationen und Verfeinerungen der Kanüle erlaubten in feinkörnigen Polykristallen, wie Bariumtitanat, eine Einzelkornanalyse mit der Weitwinkel-Beugung. Insbesondere wurden markante Erhöhungen der Versetzungsdichte nahe der Korngrenze einzelner Kristallite in FeAl festgestellt. An intermetallischen Fe-Al-Verbindungen wurden in Weitwinkel-Kurven Feinstrukturen gefunden, die sich durch Umweganregungen in Zusammenhang mit Überstrukturen erklären lassen. An zugverformten Ni-Kristallen wurden Kossel- u. Weitwinkel-Linienbreiten in Abhängigkeit des Azimuts im symmetrischen Bragg-Fall ausgewertet u. mit theoretischen Modellen verglichen. Anisotrope Linienverbreiterungen durch die Wirkung von Stufenversetzungen konnten quantitativ nachgewiesen werden. Erste Kossel- u. Weitwinkel-Untersuchungen an zyklisch verformten Ni-Einkristallen zeigten, daß beobachtete perlenkettenförmige Intensitätsanhäufungen an Weitwinkelreflexen die Konfiguration von Versetzungswänden aus Stufenversetzungsdipolen im Kristallvolumen widerspiegeln. Erstmals konnte der Einfluß von Stapelfehlern auf Weitwinkel-Linien, der sich durch linsenförmige Intensitätsaufspaltungen zeigte, experimentell nachgewiesen, theoretisch erklärt und quantitativ bestimmt werden. / The thesis deals with the nondestructive investigation of crystal defects by X-ray Kossel and Pseudo Kossel microdiffraction on compact specimens in the scanning electron microscope. The SEM was extended by means of a newly developed detection method in such a way that X-ray Kossel interferences, which are extremely faint in intensity, can be observed by a phosphor scintillator as well as a high resolution and ultra-sensitive CCD area detector. The demanding X-ray film method was thus replaced. The observation techniques were combined so that the mutually complementary methods, above all the Kossel and Pseudo Kossel, but also the electron backscattered diffraction, are made possible for the first time by using just one CCD system in the SEM. The detection of Kikuchi bands was improved to such a degree that these could be recorded even at vertical incidence of the electron beam on the specimen for the first time. It was shown that the lateral resolution of the Kossel technique under polychromatic X-ray tube excitation could be enhanced considerably and the exposure times strongly reduced by using a polycapillary lens in an X-ray fluorescence spectrometer. For the complementary microdiffraction methods a homogeneous simulation program was developed, whose novel way of simulating the complicated Pseudo Kossel curves of the fourth order enable the lattice defect localization in compact crystals. The corresponding simulations and refinements of the target tube allowed a single grain analysis also in fine-grained polycrystals like barium titanate with Pseudo Kossel X-ray diffraction. Particularly, a marked increase of the dislocation density was ascertained near the grain boundary of individual crystallites in FeAl. At intermetallic Fe-Al compounds Pseudo Kossel curves fine structures were found, which can be explained by umweg (detour) excitations in relation to superstructures. Kossel and Pseudo Kossel line widths were analyzed in dependence on the azimuth and compared with theoretical models at tensile deformed Ni-crystals in the symmetrical Bragg case. Anisotropic line broadenings through the effect of edge dislocations could be proved quantitatively. Conclusions could be drawn from the initial Kossel and Pseudo Kossel investigations of cyclically deformed nickel crystals with respect to the observed pearl-necklace-like intensity thickening at Pseudo Kossel lines reflecting the strong local variations of the dislocation density and, thus, the configuration of dislocation walls of edge dislocation dipoles inside the crystalline volume. For the first time, the effect of stacking faults on Pseudo Kossel reflections appearing by lens-shaped intensity splittings could be proved experimentally, explained theoretically and determined quantitatively.

Text is no longer confined to scanned pages and often appears in camera-based images originating from text on real world objects. Unlike the images from conventional flatbed scanners, which have a controlled acquisition environment, camera-based images pose new challenges such as uneven illumination, blur, poor resolution, perspective distortion and 3D deformations that can severely affect the performance of any optical character recognition (OCR) system. Due to the variations in the imaging condition as well as the target document type, traditional OCR systems, designed for scanned images, cannot be directly applied to camera-captured images and a new level of processing needs to be addressed. In this thesis, we study some of the issues commonly encountered in camera-based image analysis and propose novel methods to overcome them. All the methods make use of color connected components. 1. Connected component descriptor for document image mosaicing Document image analysis often requires mosaicing when it is not possible to capture a large document at a reasonable resolution in a single exposure. Such a document is captured in parts and mosaicing stitches them into a single image. Since connected components (CCs) in a document image can easily be extracted regardless of the image rotation, scale and perspective distortion, we design a robust feature named connected component descriptor that is tailored for mosaicing camera-captured document images. The method involves extraction of a circular measurement region around each CC and its description using the angular radial transform (ART). To ensure geometric consistency during feature matching, the ART coefficients of a CC are augmented with those of its 2 nearest neighbors. Our method addresses two critical issues often encountered in correspondence matching: (i) the stability of features and (ii) robustness against false matches due to multiple instances of many characters in a document image. We illustrate the effectiveness of the proposed method on camera-captured document images exhibiting large variations in viewpoint, illumination and scale. 2. Font and background color independent text binarization The first step in an OCR system, after document acquisition, is binarization, which converts a gray-scale/color image into a two-level image -the foreground text and the background. We propose two methods for binarization of color documents whereby the foreground text is output as black and the background as white regardless of the polarity of foreground-background shades. (a) Hierarchical CC Analysis: The method employs an edge-based connected component approach and automatically determines a threshold for each component. It overcomes several limitations of existing locally-adaptive thresholding techniques. Firstly, it can handle documents with multi-colored texts with different background shades. Secondly, the method is applicable to documents having text of widely varying sizes, usually not handled by local binarization methods. Thirdly, the method automatically computes the threshold for binarization and the logic for inverting the output from the image data and does not require any input parameter. However, the method is sensitive to complex backgrounds since it relies on the edge information to identify CCs. It also uses script-specific characteristics to filter out edge components before binarization and currently works well for Roman script only. (b) Contour-based color clustering (COCOCLUST): To overcome the above limitations, we introduce a novel unsupervised color clustering approach that operates on a ‘small’ representative set of color pixels identified using the contour information. Based on the assumption that every character is of a uniform color, we analyze each color layer individually and identify potential text regions for binarization. Experiments on several complex images having large variations in font, size, color, orientation and script illustrate the robustness of the method. 3. Multi-script and multi-oriented text extraction from scene images Scene text understanding normally involves a pre-processing step of text detection and extraction before subjecting the acquired image for character recognition task. The subsequent recognition task is performed only on the detected text regions so as to mitigate the effect of background complexity. We propose a color-based CC labeling for robust text segmentation from natural scene images. Text CCs are identified using a combination of support vector machine and neural network classifiers trained on a set of low-level features derived from the boundary, stroke and gradient information. We develop a semiautomatic annotation toolkit to generate pixel-accurate groundtruth of 100 scenic images containing text in various layout styles and multiple scripts. The overall precision, recall and f-measure obtained on our dataset are 0.8, 0.86 and 0.83, respectively. The proposed method is also compared with others in the literature using the ICDAR 2003 robust reading competition dataset, which, however, has only horizontal English text. The overall precision, recall and f-measure obtained are 0.63, 0.59 and 0.61 respectively, which is comparable to the best performing methods in the ICDAR 2005 text locating competition. A recent method proposed by Epshtein et al. [1] achieves better results but it cannot handle arbitrarily oriented text. Our method, however, works well for generic scene images having arbitrary text orientations. 4. Alignment of curved text lines Conventional OCR systems perform poorly on document images that contain multi-oriented text lines. We propose a technique that first identifies individual text lines by grouping adjacent CCs based on their proximity and regularity. For each identified text string, a B-spline curve is fitted to the centroids of the constituent characters and normal vectors are computed along the fitted curve. Each character is then individually rotated such that the corresponding normal vector is aligned with the vertical axis. The method has been tested on a data set consisting of 50 images with text laid out in various ways namely along arcs, waves, triangles and a combination of these with linearly skewed text lines. It yields 95.9% recognition accuracy on text strings, where, before alignment, state-of-the-art OCRs fail to recognize any text. The CC-based pre-processing algorithms developed are well-suited for processing camera-captured images. We demonstrate the feasibility of the algorithms on the publicly-available ICDAR 2003 robust reading competition dataset and our own database comprising camera-captured document images that contain multiple scripts and arbitrary text layouts.