Computer-aided diagnosis of carotid atherosclerosis based on ultrasound image statistics, laws' texture and neural networks

Quantitative characterisation of carotid atherosclerosis and classification into symptomatic or asymptomatic is crucial in planning optimal treatment of atheromatous plaque. The computer-aided diagnosis (CAD) system described in this paper can analyse ultrasound (US) images of carotid artery and classify them into symptomatic or asymptomatic based on their echogenicity characteristics. The CAD system consists of three modules: a) the feature extraction module, where first-order statistical (FOS) features and Laws' texture energy can be estimated, b) the dimensionality reduction module, where the number of features can be reduced using analysis of variance (ANOVA), and c) the classifier module consisting of a neural network (NN) trained by a novel hybrid method based on genetic algorithms (GAs) along with the back propagation algorithm. The hybrid method is able to select the most robust features, to adjust automatically the NN architecture and to optimise the classification performance. The performance is measured by the accuracy, sensitivity, specificity and the area under the receiver-operating characteristic (ROC) curve. The CAD design and development is based on images from 54 symptomatic and 54 asymptomatic plaques. This study demonstrates the ability of a CAD system based on US image analysis and a hybrid trained NN to identify atheromatous plaques at high risk of stroke.

Low reliability of perceptual priming: consequences for the interpretation of functional dissociations between explicit and implicit memory

In this study, three experiments are presented that investigate the reliability of memory measures. In Experiment 1, the well-known dissociation between explicit (recall, recognition) and implicit memory (picture clarification) as a function of age in a sample of 335 persons aged between 65 and 95 was replicated. Test-retest reliability was significantly lower in implicit than in explicit measures. In Experiment 2, parallel-test reliabilities in a student sample confirmed the finding of Experiment 1. In Experiment 3, the reliability of cued recall and word stem completion was investigated. There were significant priming effects and a dissociation between explicit and implicit memory as a function of levels of processing. However, the reliability of implicit memory measures was again substantially lower than in explicit tests in all test conditions. As a consequence, differential reliabilities of direct and indirect memory tests should be considered as a possible determinant of dissociations between explicit and implicit memory as a function of experimental or quasi-experimental manipulations.

The role of brand marketing in snowboarding: A comparative study of Switzerland and New Zealand

Introduction: Brands play an essential role in the organizational structure of snowboarding by sponsoring athletes, arranging events, contributing to product development and developing long-term partnerships with other key actors. However, the specialities of their role in scene sports, such as creating identities, networking and brand marketing strategies, have not been extensively researched. This study aims to provide an analysis of the function of brands within the snowboarding subculture by comparing how the sport is organized in Switzerland and New Zealand. Sociological theories of subcultures (Hitzler & Niederbacher, 2010) and social networks (Stegbauer, 2008) are used to defi ne the structures of the sport, whereas marketing and branding theories (Adjouri & Stastny, 2006) help to understand the role of the brands. Snowboarding will be defi ned as an alternative sports subculture based on characteristics such as aesthetics, adventure and new resources of performance (Schwier, 2006). Such a defi nition also begs for a novel form of analyzing its organization. Unlike more conventional structures, the organization of snowboarding allows a variety of actors to get involved in leading the sport. By portraying and encouraging differentiated identities and lifestyles, athletes provide a space for other actors to fi nd their place within the sport (Wheaton, 2005). According to Stegbauers network theory, individual actors are able to obtain high positions and defi ne their identity depending on their ties to actors and networks within the subculture (Stegbauer, 2008). For example, social capital, contacts within the sport and insider knowledge on subculture-related information enable actors to get closer to the core (Hitzler & Niederbacher, 2010). Actors who do not have close networks and allies within the subculture are less likely to engage successfully in the culture, whether as an individual or as a commercial actor (Thorpe, 2011). This study focuses on the organizational structure of snowboarding by comparing the development of the sport in Switzerland and New Zealand. An analysis of snowboarding in two nations with diverse cultures and economic systems allows a further defi nition of the structural organization of the sport and explains how brands play an important role in the sport. Methods: The structural organization of the sport will be analyzed through an ethnographic approach, using participant observation at various leading events in Switzerland (Freestyle.ch, European Open) and New Zealand (World Heli Challenge, New Zealand Open, New Zealand Winter Games). The data is analyzed using grounded theory (Glaser & Strauss 1967) and gives an overview of the actors that are playing an important role in the local development of snowboarding. Participant observation was also used as a tool to get inside the sport culture and opened up the possibility to make over 40 semi-structured qualitative expert interviews with international core actors from 11 countries. Obtaining access to one actor as a partner early on helped to get inside the local sport culture. The ‘snowball effect’ allowed the researcher to acquire access, build trust and conduct interviews with experts within the core scene. All the interviewed actors have a direct infl uence on the sport in one or both countries, which permit a cross-analysis. The data of the interviews was evaluated through content analysis (Mayring 2010). The two methods together provided suffi cient data to analyze the organizational structure and discuss the role of brand marketing within snowboarding. Results: An actors mapping by means of a center-periphery framework has identifi ed fi ve main core groups: athletes, media representatives, brand-marketing managers, resort managers and event organizers. In both countries the same grouping of actors were found. Despite possessing different and frequently multiple roles and responsibilities, core actors appear to have a strong common identifi cation as ‘snowboarders’, are considered to be part of the organizational elite of the sport and tend to advocate similar goals. The author has found that brands in Switzerland tend to have a larger impact on the broader snowboarding culture due to a number of factors discussed below. Due to a larger amount of snowboarders and stronger economic power in Europe, snowboarders are making attempts to differentiate themselves from other winter sports, while competing with each other to develop niche markets. In New Zealand, on the other hand, the smaller market enables more cooperation and mutual respect within snowboarders. Further they are more closely linked to other winter sports and are satisfi ed with being lumped together. In both countries, brands have taken up the role of supporting young athletes, organizing competitions and feeding media with subculture-related content. Brands build their image and identity through the collaboration with particular athletes who can represent the values of the brand. Local and global communities with similar lifestyles and interests are being built around brands that share a common vision of the sport. The dominance of brands in snowboarding has enabled them with the power to organize and rule the sport through its fan base and supporters. Brands were defi ned by interviewees as independent institutions led by insiders who know the codes and symbols of the sport and were given trust and credibility. The brands identify themselves as the engines of the sport by providing the equipment, opportunities for athletes to get exposure, allowing media to get exclusive information on activities, events and sport-related stories. Differences between the two countries are more related to the economic system. While Switzerland is well integrated in the broader European market, New Zealand’s geographical isolation and close proximity to Australia tends to limit its market. Further, due to different cultural lifestyles, access to resorts and seasonal restrictions, to name a few, the amount of people practicing winter sports in New Zealand is much smaller than in Switzerland. However, this also presents numerous advantages. For example, the short southern hemisphere winter season in New Zealand enables them to attract international sports athletes, brands and representatives in a period when Europe and North America is in summer. Further, the unique snow conditions in New Zealand and majestic landscape is popular for attracting world renowned photo- and cinematographers. Another advantage is the less populated network as it provides the opportunity for individuals to gain easier access to the core of the sport, obtain diverse positions and form a unique identity and market. In Switzerland, on the other hand, the snowboarding network is dense with few positions available for the taking. Homegrown brands with core recognition are found in both countries. It was found that the Swiss brands tend to have a larger impact on the market, whereas in New Zealand, the sport is more dependent on import products by foreign brands. Further, athletes, events and resorts in New Zealand are often dependent on large brand sponsorships from abroad such as from brand headquarters in the Unites States. Thus, due to its location in the centre of Europe, Swiss brands can take advantage of brands which are closer in proximity and culture to sponsor athletes and events. In terms of media coverage, winter sports in New Zealand tend to have a minor coverage and tradition in local mass media, which leads to less exposure, recognition and investment into the sport. This is also related to how snowboarding is more integrated into other winter sports in New Zealand. Another difference is the accessibility of the ski resort by the population. While in Switzerland the resorts are mostly being visited by day-travelers, ‘weekend warriors’ and holiday makers, the location of the resorts in New Zealand make it diffi cult to visit for one day. This is in part due to the fact that Swiss ski resorts and villages are usually the same location and are accessible through public transportation, while the ski resorts in New Zealand have been built separately from the villages. Further, the villages have not been built to accommodate to high tourist arrivals. Thus, accommodation and food facilities are limited and there is a lack of public transportation to the resorts. Discussion: The fi ndings show that networks and social relations combined with specifi c knowledge on scene-related attributes are crucial in obtaining opportunities within the sport. Partnerships as well as competition between these different actors are necessary for core acceptance, peer credibility and successful commercial interests. Brands need to maintain effective marketing strategies and identities which incorporate subcultural forms of behavior and communication. In order to sustain credibility from its fans, athletes and other snowboarding actors, brands need to maintain their insider status through social networks and commercial branding strategies. The interaction between all actors is a reciprocated process, where social capital, networks and identities are being shared. While the overall structure of snowboard subcultures in Europe and New Zealand are similar, there are some distinct characteristics which make each one unique. References Adjouri, N. & Stastny, P. (2006). Sport-Branding: Mit Sport-Sponsoring zum Markenerfolg. Wiesbaden: Gabler. Glaser, B. & Strauss, K. (1967). The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine. Hebdige, D. (2009). Subculture; The meaning of style. New York: Routledge. Hitzler, R. & Niederbacher, A. (2010). Leben in Szenen: Formen juveniler Vergemeinschaftung heute. Wiesbaden: Verlag für Sozialwissenschaften. Mayring, P. (2010). Qualitative Inhaltsanalyse: Grundlagen und Techniken. Weinheim: Beltz. Schwier, J. (2006). Repräsentationen des Trendsports. Jugendliche Bewegungskulturen, Medien und Marketing. In: Gugutzer, R. (Hrsg.). body turn. Perspektiven der Soziologie des Körpers und des Sports. Bielefeld: transcript (S. 321-340). Stegbauer, C. (2008). Netzwerkanalyse und Netzwerktheorie. Ein neues Paradigma in den Sozialwissenschaften. Wiesbaden: VS Verlag für Sozialwissenschaften. Thorpe, H. (2011). Snowboarding bodies in theory and practice. Basingstoke: Palgrave Macmillan. Wheaton, B. (2005). Understanding lifestyle sports; consumption, identity and difference. New York: Routledge.

Diagnostic confidence and image quality of CT pulmonary angiography at 100 kVp in overweight and obese patients.

AIM To compare image quality and diagnostic confidence of 100 kVp CT pulmonary angiography (CTPA) in patients with body weights (BWs) below and above 100kg. MATERIALS AND METHODS The present retrospective study comprised 216 patients (BWs of 75-99kg, 114 patients; 100-125kg, 88 patients; >125kg, 14 patients), who received 100 kVp CTPA to exclude pulmonary embolism. The attenuation was measured and the contrast-to-noise ratio (CNR) was calculated in the pulmonary trunk. Size-specific dose estimates (SSDEs) were evaluated. Three blinded radiologists rated subjective image quality and diagnostic confidence. Results between the BW groups and between three body mass index (BMI) groups (BMI <25kg/m(2), BMI = 25-29.9kg/m(2), and BMI ≥30kg/m(2), i.e., normal weight, overweight, and obese patients) were compared using the Kruskal-Wallis test. RESULTS Vessel attenuation was higher and SDDE was lower in the 75-99kg group than at higher BWs (p-values between <0.001 and 0.03), with no difference between the 100-125 and >125kg groups (p = 0.892 and 1). Subjective image quality and diagnostic confidence were not different among the BW groups (p = 0.225 and 1). CNR was lower (p < 0.006) in obese patients than in normal weight or overweight subjects. Diagnostic confidence was not different in the BMI groups (p = 0.105). CONCLUSION CTPA at 100 kVp tube voltage can be used in patients weighing up to 125kg with no significant deterioration of subjective image quality and confidence. The applicability of 100 kVp in the 125-150kg BW range needs further testing in larger collectives.

Prospective randomised comparison of diagnostic confidence and image quality with normal-dose and low-dose CT pulmonary angiography at various body weights.

OBJECTIVES To find a threshold body weight (BW) below 100 kg above which computed tomography pulmonary angiography (CTPA) using reduced radiation and a reduced contrast material (CM) dose provides significantly impaired quality and diagnostic confidence compared with standard-dose CTPA. METHODS In this prospectively randomised study of 501 patients with suspected pulmonary embolism and BW <100 kg, 246 were allocated into the low-dose group (80 kVp, 75 ml CM) and 255 into the normal-dose group (100 kVp, 100 ml CM). Contrast-to-noise ratio (CNR) in the pulmonary trunk was calculated. Two blinded chest radiologists independently evaluated subjective image quality and diagnostic confidence. Data were compared between the normal-dose and low-dose groups in five BW subgroups. RESULTS Vessel attenuation did not differ between the normal-dose and low-dose groups within each BW subgroup (P = 1.0). The CNR was higher with the normal-dose compared with the low-dose protocol (P < 0.006) in all BW subgroups except for the 90-99 kg subgroup (P = 0.812). Subjective image quality and diagnostic confidence did not differ between CT protocols in all subgroups (P between 0.960 and 1.0). CONCLUSIONS Subjective image quality and diagnostic confidence with 80 kVp CTPA is not different from normal-dose protocol in any BW group up to 100 kg. KEY POINTS • 80 kVp CTPA is safe in patients weighing <100 kg • Reduced radiation and iodine dose still provide high vessel attenuation • Image quality and diagnostic confidence with low-dose CTPA is good • Diagnostic confidence does not deteriorate in obese patients weighing <100 kg.

Two-Dimensional Crystallisation of Membrane Proteins and Structural Assessment

Two-dimensional (2D) crystallisation of Membrane proteins reconstitutes them into their native environment, the lipid bilayer. Electron crystallography allows the structural analysis of these regular protein–lipid arrays up to atomic resolution. The crystal quality depends on the protein purity, ist stability and on the crystallisation conditions. The basics of 2D crystallisation and different recent advances are reviewed and electron crystallography approaches summarised. Progress in 2D crystallisation, sample preparation, image detectors and automation of the data acquisition and processing pipeline makes 2D electron crystallography particularly attractive for the structural analysis of membrane proteins that are too small for single-particle analyses and too unstable to form three-dimensional (3D) crystals.

Medical image computing in diagnosis and intervention of spinal diseases

Spinal image analysis and computer assisted intervention have emerged as new and independent research areas, due to the importance of treatment of spinal diseases, increasing availability of spinal imaging, and advances in analytics and navigation tools. Among others, multiple modality spinal image analysis and spinal navigation tools have emerged as two keys in this new area. We believe that further focused research in these two areas will lead to a much more efficient and accelerated research path, avoiding detours that exist in other applications, such as in brain and heart.

Automated Tools for Layer Recognition (BMPix) and Counting (PEAK)

We present tools for rapid and quantitative detection of sediment lamination. The BMPix tool extracts color and gray-scale curves from images at pixel resolution. The PEAK tool uses the gray-scale curve and performs, for the first time, fully automated counting of laminae based on three methods. The maximum count algorithm counts every bright peak of a couplet of two laminae (annual resolution) in a smoothed curve. The zero-crossing algorithm counts every positive and negative halfway-passage of the curve through a wide moving average, separating the record into bright and dark intervals (seasonal resolution). The same is true for the frequency truncation method, which uses Fourier transformation to decompose the curve into its frequency components before counting positive and negative passages. We applied the new methods successfully to tree rings, to well-dated and already manually counted marine varves from Saanich Inlet, and to marine laminae from the Antarctic continental margin. In combination with AMS14C dating, we found convincing evidence that laminations in Weddell Sea sites represent varves, deposited continuously over several millennia during the last glacial maximum. The new tools offer several advantages over previous methods. The counting procedures are based on a moving average generated from gray-scale curves instead of manual counting. Hence, results are highly objective and rely on reproducible mathematical criteria. Also, the PEAK tool measures the thickness of each year or season. Since all information required is displayed graphically, interactive optimization of the counting algorithms can be achieved quickly and conveniently.

Vision-based vehicle detection and tracking with a mobile camera using a statistical framework

En esta tesis se aborda la detección y el seguimiento automático de vehículos mediante técnicas de visión artificial con una cámara monocular embarcada. Este problema ha suscitado un gran interés por parte de la industria automovilística y de la comunidad científica ya que supone el primer paso en aras de la ayuda a la conducción, la prevención de accidentes y, en última instancia, la conducción automática. A pesar de que se le ha dedicado mucho esfuerzo en los últimos años, de momento no se ha encontrado ninguna solución completamente satisfactoria y por lo tanto continúa siendo un tema de investigación abierto. Los principales problemas que plantean la detección y seguimiento mediante visión artificial son la gran variabilidad entre vehículos, un fondo que cambia dinámicamente debido al movimiento de la cámara, y la necesidad de operar en tiempo real. En este contexto, esta tesis propone un marco unificado para la detección y seguimiento de vehículos que afronta los problemas descritos mediante un enfoque estadístico. El marco se compone de tres grandes bloques, i.e., generación de hipótesis, verificación de hipótesis, y seguimiento de vehículos, que se llevan a cabo de manera secuencial. No obstante, se potencia el intercambio de información entre los diferentes bloques con objeto de obtener el máximo grado posible de adaptación a cambios en el entorno y de reducir el coste computacional. Para abordar la primera tarea de generación de hipótesis, se proponen dos métodos complementarios basados respectivamente en el análisis de la apariencia y la geometría de la escena. Para ello resulta especialmente interesante el uso de un dominio transformado en el que se elimina la perspectiva de la imagen original, puesto que este dominio permite una búsqueda rápida dentro de la imagen y por tanto una generación eficiente de hipótesis de localización de los vehículos. Los candidatos finales se obtienen por medio de un marco colaborativo entre el dominio original y el dominio transformado. Para la verificación de hipótesis se adopta un método de aprendizaje supervisado. Así, se evalúan algunos de los métodos de extracción de características más populares y se proponen nuevos descriptores con arreglo al conocimiento de la apariencia de los vehículos. Para evaluar la efectividad en la tarea de clasificación de estos descriptores, y dado que no existen bases de datos públicas que se adapten al problema descrito, se ha generado una nueva base de datos sobre la que se han realizado pruebas masivas. Finalmente, se presenta una metodología para la fusión de los diferentes clasificadores y se plantea una discusión sobre las combinaciones que ofrecen los mejores resultados. El núcleo del marco propuesto está constituido por un método Bayesiano de seguimiento basado en filtros de partículas. Se plantean contribuciones en los tres elementos fundamentales de estos filtros: el algoritmo de inferencia, el modelo dinámico y el modelo de observación. En concreto, se propone el uso de un método de muestreo basado en MCMC que evita el elevado coste computacional de los filtros de partículas tradicionales y por consiguiente permite que el modelado conjunto de múltiples vehículos sea computacionalmente viable. Por otra parte, el dominio transformado mencionado anteriormente permite la definición de un modelo dinámico de velocidad constante ya que se preserva el movimiento suave de los vehículos en autopistas. Por último, se propone un modelo de observación que integra diferentes características. En particular, además de la apariencia de los vehículos, el modelo tiene en cuenta también toda la información recibida de los bloques de procesamiento previos. El método propuesto se ejecuta en tiempo real en un ordenador de propósito general y da unos resultados sobresalientes en comparación con los métodos tradicionales. ABSTRACT This thesis addresses on-road vehicle detection and tracking with a monocular vision system. This problem has attracted the attention of the automotive industry and the research community as it is the first step for driver assistance and collision avoidance systems and for eventual autonomous driving. Although many effort has been devoted to address it in recent years, no satisfactory solution has yet been devised and thus it is an active research issue. The main challenges for vision-based vehicle detection and tracking are the high variability among vehicles, the dynamically changing background due to camera motion and the real-time processing requirement. In this thesis, a unified approach using statistical methods is presented for vehicle detection and tracking that tackles these issues. The approach is divided into three primary tasks, i.e., vehicle hypothesis generation, hypothesis verification, and vehicle tracking, which are performed sequentially. Nevertheless, the exchange of information between processing blocks is fostered so that the maximum degree of adaptation to changes in the environment can be achieved and the computational cost is alleviated. Two complementary strategies are proposed to address the first task, i.e., hypothesis generation, based respectively on appearance and geometry analysis. To this end, the use of a rectified domain in which the perspective is removed from the original image is especially interesting, as it allows for fast image scanning and coarse hypothesis generation. The final vehicle candidates are produced using a collaborative framework between the original and the rectified domains. A supervised classification strategy is adopted for the verification of the hypothesized vehicle locations. In particular, state-of-the-art methods for feature extraction are evaluated and new descriptors are proposed by exploiting the knowledge on vehicle appearance. Due to the lack of appropriate public databases, a new database is generated and the classification performance of the descriptors is extensively tested on it. Finally, a methodology for the fusion of the different classifiers is presented and the best combinations are discussed. The core of the proposed approach is a Bayesian tracking framework using particle filters. Contributions are made on its three key elements: the inference algorithm, the dynamic model and the observation model. In particular, the use of a Markov chain Monte Carlo method is proposed for sampling, which circumvents the exponential complexity increase of traditional particle filters thus making joint multiple vehicle tracking affordable. On the other hand, the aforementioned rectified domain allows for the definition of a constant-velocity dynamic model since it preserves the smooth motion of vehicles in highways. Finally, a multiple-cue observation model is proposed that not only accounts for vehicle appearance but also integrates the available information from the analysis in the previous blocks. The proposed approach is proven to run near real-time in a general purpose PC and to deliver outstanding results compared to traditional methods.

Cloud Computing Service for Managing Large Medical Image Data-Sets Using Balanced Collaborative Agents

Managing large medical image collections is an increasingly demanding important issue in many hospitals and other medical settings. A huge amount of this information is daily generated, which requires robust and agile systems. In this paper we present a distributed multi-agent system capable of managing very large medical image datasets. In this approach, agents extract low-level information from images and store them in a data structure implemented in a relational database. The data structure can also store semantic information related to images and particular regions. A distinctive aspect of our work is that a single image can be divided so that the resultant sub-images can be stored and managed separately by different agents to improve performance in data accessing and processing. The system also offers the possibility of applying some region-based operations and filters on images, facilitating image classification. These operations can be performed directly on data structures in the database.

Level Set Segmentation with Shape and Appearance Models Using Affine Moment Descriptors

We propose a level set based variational approach that incorporates shape priors into edge-based and region-based models. The evolution of the active contour depends on local and global information. It has been implemented using an efficient narrow band technique. For each boundary pixel we calculate its dynamic according to its gray level, the neighborhood and geometric properties established by training shapes. We also propose a criterion for shape aligning based on affine transformation using an image normalization procedure. Finally, we illustrate the benefits of the our approach on the liver segmentation from CT images.

Identificación y reconocimiento de matrículas de automóviles con MATLAB

Este Proyecto Fin de Carrera trata sobre el reconocimiento e identificación de caracteres de matrículas de automóviles. Este tipo de sistemas de reconocimiento también se los conoce mundialmente como sistemas ANPR ("Automatic Number Plate Recognition") o LPR ("License Plate Recognition"). La gran cantidad de vehículos y logística que se mueve cada segundo por todo el planeta, hace necesaria su registro para su tratamiento y control. Por ello, es necesario implementar un sistema que pueda identificar correctamente estos recursos, para su posterior procesado, construyendo así una herramienta útil, ágil y dinámica. El presente trabajo ha sido estructurado en varias partes. La primera de ellas nos muestra los objetivos y las motivaciones que se persiguen con la realización de este proyecto. En la segunda, se abordan y desarrollan todos los diferentes procesos teóricos y técnicos, así como matemáticos, que forman un sistema ANPR común, con el fin de implementar una aplicación práctica que pueda demostrar la utilidad de estos en cualquier situación. En la tercera, se desarrolla esa parte práctica en la que se apoya la base teórica del trabajo. En ésta se describen y desarrollan los diversos algoritmos, creados con el fin de estudiar y comprobar todo lo planteado hasta ahora, así como observar su comportamiento. Se implementan varios procesos característicos del reconocimiento de caracteres y patrones, como la detección de áreas o patrones, rotado y transformación de imágenes, procesos de detección de bordes, segmentación de caracteres y patrones, umbralización y normalización, extracción de características y patrones, redes neuronales, y finalmente el reconocimiento óptico de caracteres o comúnmente conocido como OCR. La última parte refleja los resultados obtenidos a partir del sistema de reconocimiento de caracteres implementado para el trabajo y se exponen las conclusiones extraídas a partir de éste. Finalmente se plantean las líneas futuras de mejora, desarrollo e investigación, para poder realizar un sistema más eficiente y global. This Thesis deals about license plate characters recognition and identification. These kinds of systems are also known worldwide as ANPR systems ("Automatic Number Plate Recognition") or LPR ("License Plate Recognition"). The great number of vehicles and logistics moving every second all over the world, requires a registration for treatment and control. Thereby, it’s therefore necessary to implement a system that can identify correctly these resources, for further processing, thus building a useful, flexible and dynamic tool. This work has been structured into several parts. The first one shows the objectives and motivations attained by the completion of this project. In the second part, it’s developed all the different theoretical and technical processes, forming a common ANPR system in order to implement a practical application that can demonstrate the usefulness of these ones on any situation. In the third, the practical part is developed, which is based on the theoretical work. In this one are described and developed various algorithms, created to study and verify all the questions until now suggested, and complain the behavior of these systems. Several recognition of characters and patterns characteristic processes are implemented, such as areas or patterns detection, image rotation and transformation, edge detection processes, patterns and character segmentation, thresholding and normalization, features and patterns extraction, neural networks, and finally the optical character recognition or commonly known like OCR. The last part shows the results obtained from the character recognition system implemented for this thesis and the outlines conclusions drawn from it. Finally, future lines of improvement, research and development are proposed, in order to make a more efficient and comprehensive system.