CARVALHO, E. A. Segmentação de imagens de radar de abertura sintética por crescimento e fusão estatística de regiões. 2005. 121 f. Dissertação (Mestrado em Engenharia de Teleinformática) – Centro de Tecnologia, Universidade Federal do Ceará, Fortaleza, 2005. / Submitted by Marlene Sousa (mmarlene@ufc.br) on 2016-04-04T13:56:03Z No. of bitstreams: 1 2005_dis_eacarvalho.pdf: 7092051 bytes, checksum: 64dfe0e028ed1c09118f6dd20b218029 (MD5) / Approved for entry into archive by Marlene Sousa(mmarlene@ufc.br) on 2016-04-06T18:44:18Z (GMT) No. of bitstreams: 1 2005_dis_eacarvalho.pdf: 7092051 bytes, checksum: 64dfe0e028ed1c09118f6dd20b218029 (MD5) / Made available in DSpace on 2016-04-06T18:44:18Z (GMT). No. of bitstreams: 1 2005_dis_eacarvalho.pdf: 7092051 bytes, checksum: 64dfe0e028ed1c09118f6dd20b218029 (MD5) Previous issue date: 2005-05-23 / The regular coverage of the planet surface by spaceborne synthetic aperture radar (SAR)and also airborne systems have provided alternative means to gather remote sensing information of various regions of the planet, even of inaccessible areas. This work deals with the digital processing of synthetic aperture radar imagery, where segmentation is the main subject. It consists of isolating or partitioning relevant objects in a scene, aiming at improving image interpretation and understanding in subsequent tasks. SAR images are contaminated by coherent noise, known as speckle, which masks small details and transition zones among the objects. Such a noise is inherent in radar image generation process, making difficult tasks like automatic segmentation of the objects, as well as their contour identification. To segment radar images, one possible way is to apply speckle filtering before segmentation. Another one, applied in this work, is to perform noisy image segmentation using the original SAR pixels as input data, without any preprocessing,such as filtering. To provide segmentation, an algorithm based on region growing and statistical region merging has been developed, which requires some parameters to control the process. This task presents some advantages, as long as it eliminates preprocessing steps and favors the detection of the image structures, since original pixel information is exploited. A qualitative and quantitative performance evaluation of the segmented images is also executed, under different situations, by applying the proposed technique to simulated images corrupted with multiplicative noise. This segmentation method is also applied to real SAR images and the produced results are promising. / A cobertura regular de quase todo o planeta por sistemas de radar de abertura sintética (synthetic aperture radar - SAR) orbitais e o uso de sistemas aerotransportados têm propiciado novos meios para obter informações através do sensoriamento remoto de várias regiões de nosso planeta, muitas delas inacessíveis. Este trabalho trata do processamento de imagens digitais geradas por radar de abertura sintética, especificamente da segmentação, que consiste do isolamento ou particionamento dos objetos relevantes presentes em uma cena. A segmentação de imagens digitais visa melhorar a interpretação das mesmas em procedimentos subseqüentes. As imagens SAR são corrompidas por ruído coerente, conhecido por speckle, que mascara pequenos detalhes e zonas de transição entre os objetos. Tal ruído é inerente ao processo de formação dessas imagens e dificulta tarefas como a segmentação automática dos objetos existentes e a identificação de seus contornos. Uma possibilidade para efetivar a segmentação de imagens SAR consiste na filtragem preliminar do ruído speckle, como etapa de tratamento dos dados. A outra possibilidade, aplicada neste trabalho, consiste em segmentar diretamente a imagem ruidosa, usando seus pixels originais como fonte de informação. Para isso, é desenvolvida uma metodologia de segmentação baseada em crescimento e fusão estatística de regiões, que requer alguns parâmetros para controlar o processo. As vantagens da utilização dos dados originais para realizar a segmentação de imagens de radar são a eliminação de etapas de pré-processamento e o favorecimento da detecção das estruturas presentes nas mesmas. É realizada uma avaliação qualitativa e quantitativa das imagens segmentadas, sob diferentes situações, aplicando a técnica proposta em imagens de teste contaminadas artificialmente com ruído multiplicativo. Este segmentador é aplicado também no processamento de imagens SAR reais e os resultados são promissores.

Imagens de radar de abertura sintética (SAR) são tipicamente corrompidas pelo ruído \"speckle\" que também degrada imagens geradas por ultra-som, laser, etc. Esta tese propõe algoritmos de filtragem baseados na abordagem \"maximum a posteriori\" (MAP) para redução de \"speckle\" em imagens SAR. Na derivação dos filtros MAP, para imagens obtidas por detecção linear, são utilizadas as distribuições (condicionais) Rayleigh e raiz quadrada de gama na regra de Bayes como modelos para o ruído \"speckle\" em imagens SAR obtidas em amplitudes com 1 e múltiplas visadas, respectivamente, e usadas várias distribuições para o modelo \"a priori\". Toda a formulação dos algoritmos tem por base o modelo multiplicativo que constitui o modelo mais adequado ao \"speckle\". Propõe-se ainda neste trabalho a combinação dos filtros MAP formulados com o algoritmo k-médias e com a técnica de crescimento de regiões, como forma de melhoria da abordagem de filtragem proposta. Os resultados de filtragem foram avaliados segundo critérios (medidas) de melhoria da relação sinal-ruído e perda de resolução. O primeiro critério avalia a redução da intensidade do ruído \"speckle\" sobre regiões homogêneas e para avaliar a perda de resolução decorrente da filtragem é proposta uma nova técnica baseada na transformada de Hough. Os algoritmos foram testados em imagens artificialmente contaminadas por ruído \"speckle\" e em imagens SAR reais apresentando estatísticas Rayleigh e raiz de gama. Os resultados obtidos mostram a melhoria que proporcionam os algoritmos de filtragem MAP, especialmente quando combinados com o classificador k-médias e com a técnica de crescimento de região. O uso da técnica de crescimento de região reforça a conclusão de que o uso de vizinhança estatisticamente mais semelhante ao pixel ruidoso melhora a estimação dos parâmetros de filtragem. As medidas de desempenho e validação dos algoritmos MAP permitiram concluir que os filtros com distribuições \"a priori\" Gaussiana, gama, chi-quadrado e beta apresentaram melhores resultados de filtragem em relação aos demais modelos \"a priori\" quando comparados ao filtro de Kuan e com a técnica de \"wavelets\" para a classe de imagens utilizadas / Synthetic aperture radar (SAR) images are typically corrupted by speckle noise, which also degrade images produced by laser beams, ultrasound, etc. This thesis proposes filtering algorithms based on the \"maximum a posteriori\" (MAP) approach, to reduce speckle in SAR images. To derive the MAP filters for linearly detected images we assumed the multiplicative model for the speckle and used the conditional density functions in the Bayes rule following a Rayleigh and square root of gamma for one-look and N-looks images, respectively, and several different \"a priori\" densities. The MAP filters are combined with the k-means classifier and region growing tools to improve the proposed filtering approach. Measures evaluating both the signal-to-noise improvement and resolution loss due to filtering are computed. To assess the improvement brought by the proposed algorithms we evaluate them with respect to signal to noise ratio and edge preservation. The former is a classical way to evaluate the speckle strenght reduction over homogeneous areas and the latter is a new proposed technique based on the Hough transform that measures distortions at the edges produced by the speckle MAP filtering algorithms. The qualitative analysis of the MAP proposed algorithms includes the methods based on the curvature and wavelets . The algorithms were applied to simulated noisy speckled images and real SAR images with statistics of linearly detected images with one-look and N-looks. The obtained results demonstrated the improvement brought by the speckle MAP filtering algorithms, specially when combined with the k-means clustering algorithm and with the region growing approach. This region growing approach reinforces the conclusion that the use of a neighborhood whose pixels have statistics similar to the noisy pixel provides a better estimation for filtering. The evaluating measures point out that the MAP filters whose \"a priori\" models are the Gaussian, gamma, chi-square and beta presented better results than the other \"a priori\" models proposed in this thesis, the Kuan filter and the wavelets filter, for the class of images that were tested

This thesis work provides a new way to detect and track the displacement of flexor tendons within the human arm, using a non-invasive, ultrasound-based, speckle tracking technique. By tracking the tendons in the arm, it provides a way to monitor a person’s intention to move their hands and fingers. This has application to hand prosthetic control, as well as tendon injury assessment, which has significant contributions to the medical and rehabilitation community. The system works by capturing and processing a sequence of B-scan ultrasound images, to detect and track the flexor tendon motion (excursion) in the wrist, as the user flexes their muscles. Given the biomechanics of the hand, tendon displacement is correlated to the user’s intention to move their finger. Several speckle tracking techniques using B-scan ultrasound image sequences are developed in this work, including: auto-location of the tendon, a stationary ROI (region of interest), and novel use of similarity measures such as FT (Fisher Tippett), and hybrid methods. As well, work is done to investigate various speckle tracking parameters, and their effects on tracking accuracy. The different speckle tracking techniques are developed using data obtained from cadaver hands, and human volunteers undergoing regular surgery. The tracking techniques are compared in terms of successfully detecting the tendon, accurately tracking tendon displacement, successfully tracking multiple tendons, successfully detecting and tracking the onset of low tendon displacement, and computational efficiency of the algorithms. Another major aspect of this work is the design of a novel quad-array transducer that can collect image sequences from up to four tendons simultaneously. This transducer is instrumental to the motivation for controlling an advanced prosthesis. As well, specialized hardware is designed for the cadaver-based studies. Overall, this thesis successfully demonstrated the proposed tracking algorithms and newly designed hardware, for tracking the displacement of single and multiple flexor tendons. It has provided several important contributions to the field. / Graduate / 0548 / 0986 / 0760

Residual speckles in adaptive optics (AO) images represent a well-known limitation on the achievement of the contrast needed for faint source detection. Speckles in AO imagery can be the result of either residual atmospheric aberrations, not corrected by the AO, or slowly evolving aberrations induced by the optical system. We take advantage of the high temporal cadence (1 ms) of the data acquired by the System for Coronagraphy with High-order Adaptive Optics from R to K bands-VIS forerunner experiment at the Large Binocular Telescope to characterize the AO residual speckles at visible wavelengths. An accurate knowledge of the speckle pattern and its dynamics is of paramount importance for the application of methods aimed at their mitigation. By means of both an automatic identification software and information theory, we study the main statistical properties of AO residuals and their dynamics. We therefore provide a speckle characterization that can be incorporated into numerical simulations to increase their realism and to optimize the performances of both real-time and postprocessing techniques aimed at the reduction of the speckle noise. (C) 2017 Society of PhotoOptical Instrumentation Engineers (SPIE).

Ingeniero Civil Eléctrico / Taote es un dispositivo de ultrasonido portátil de bajo costo completamente diseñado y construido en la Universidad de Chile. Su portabilidad, fácil uso y bajo costo lo hacen una herramienta idónea para el pre diagnostico conducido por médicos generales, dando solución al problema de congestión de la red de salud pública al evitar derivar casos innecesarios a otros centros de atención de mayor complejidad y dando un mayor acceso a este tipo de exámenes en zonas donde su acceso es restringido. Como producto en desarrollo, aún enfrenta diversos desafíos que deben resolverse, entre ellos, el tratamiento de imágenes para su realce. Los dispositivos de ultrasonido son afectos a un fenómeno particular llamado speckle, generado por la interferencia de la señal exploradora con los elementos pequeños en las superficies examinadas, lo que se traduce en artefactos granulares, zonas de alta y baja intensidad que dificultan el diagnóstico médico y automatizado. El objetivo principal de este trabajo consiste en diseñar e implementar distintas técnicas de procesamiento considerando las restricciones que impone Taote, tal que permitan realzar las imágenes de ultrasonido que provienen de él. Para ello, se estudian algunas técnicas existentes en el estado del arte , se simulan y evalúan para su posterior implementación. Al respecto, el operador lineal con ganancia adaptable de Lee y el operador no lineal de mediana multicapa, entregan los mejores desempeños seguidos por los operadores fundamentales (promedio y mediana convencional) para los índices de transformación de imágenes. Éstos entregan una métrica de la calidad de la transformación y no de la calidad de la imagen por sí misma (la cual es subjetiva y depende del observador) . Debido a la complejidad computacional, se diseñan circuitos que implementan estos operadores con ganancia fija, es decir, representándolos como la interpolación constante de operadores fundamentales en Verilog, para su posterior síntesis e implementación en la FPGA del equipo. En su implementación en el equipo, las capturas de las nuevas imágenes procesadas demuestran transformaciones similares a las obtenidas en las simulaciones, aunque con métricas de transformación diferente. Aún así , éstas logran validar las transformaciones en la imagen original de Taote. Finalmente, estas capturas fueron sometidas a la evaluación del ojo experto, donde el operador no lineal con ventana de tamaño 5 X 5 fue la elegida para ser incorporada en el proyecto. Si bien fue posible incorporar técnicas que modifican métricas de la imagen, un examen exhaustivo debe hacerse al obtener el índice de eficiencia de diagnostico de los operadores del producto y compararlos con los del producto sin la técnica de procesamiento. Hacer estas pruebas escapa a los alcances de este trabajo, pero hacerlo llevará a validar objetivamente las técnicas de procesamiento y abre las puertas a nuevos desarrollos, como el diagnóstico automatizado, segmentación de imágenes y otro tipo de desarrollos que acercarán al proyecto para lograr su cometido; incorporar un nuevo elemento en el diagnóstico médico general. / 03/10/2021

The thesis deals with speckle suppression in images acquired by optical coherence tomograph. It is divided into four parts. The first part describes basic information about the medical imaging method. It also deals with principle of measurement. The second section discusses the formation of image speckle and selected methods to reduce them. Next part is practical and consists of data acquisition, determination of the evaluation methodology and the implementation of speckle reduction methods. The last part is focused on testing and reviews of algorithms and discussion about their results.

Hälsenan har en viktig roll för motoriken av underbenet samt rörelse vid gång och löpning. I och med detta sker det en stor belastning på hälsenan vid gång och löpning vilket kan leda till bristningar och skador på senan. Dessa bristningar kan ske när hälsenan deformeras i samband med fysisk aktivitet. Deformation av hälsenan är differensen mellan relativ displacement i senan och är ett viktigt koncept inom förståelsen om hälseneskadors etiologi. Syftet med detta examensarbete var att öka kunskapen om displacement i hälsenan och hur det skiljer sig vid habituell löpning med olika löparskor inklusive barfotalöpning. Vidare felsöktes en speckle tracking-algoritm som lägger grund för analysen av displacement i hälsenan. Arbetet är en utökning på ett pågående forskningssamarbete mellan KTH, GIH och KI. Det första steget i arbetet var att MATLAB-programmet för speckle tracking-algoritmen felsöktes och korrigerades. Detta gjordes genom att ändra initiala startvärden i algoritmen såsom centralfrekvensen och värdet för hastigheten av displacement i hälsenan. Efter att hastigheten valts till 2 cm/s fungerade algoritmen felfritt. Nästa steg var att exekvera programmet på den ultraljudsdata för hälsenan som tidigare samlats in från forskningssamarbetet. Algoritmen kördes för tre studiepersoner för habituell löpstil med tre olika löparskor samt barfota. Detta resulterade i att medelvärdeskurvor för ytlig och djup displacement i hälsenan kunde plottas där dess minimum- och maximumvärden kunde jämföras. Efter att differensen för minimum- och maximumvärdet för ytlig och djup displacement hade jämförts mellan samtliga löparskor samt barfota, kunde följande konstateras. Löpning med traditionella skor medförde högst djup displacement följt av skor med kolfibersula, minimalistiska skor och barfotalöpning. Den sko som medförde högst ytlig displacement var skor med kolfibersula följt av traditionella skor, minimalistiska skor samt löpning barfota. Jämförelsen har ej baserats på någon statistisk grund då endast tre studiepersoner analyserades. Detta medför en viss osäkerhet gällande vilken skotyp som medför högst respektive lägst displacement. Bortsätt från denna osäkerhet kan det konstateras att resultat från tidigare studier tyder på liknande insikter. / The Achilles tendon plays a crucial role in the motor function of the lower leg, as well as in the motion involved in walking and running. As a result, the Achilles tendon undergoes significant stress during walking and running, which can lead to tears and injuries to the tendon. These tears can occur when the Achilles tendon is deformed associated with physical activity. The deformation of the Achilles tendon is the difference between the relative displacement in the tendon and is an important concept in understanding the etiology of Achilles tendon injuries. The purpose of this thesis is to increase knowledge about the displacement in the Achilles tendon and how it differs during habitual running with different running shoes, including barefoot running. Furthermore, a speckle tracking algorithm that forms the basis for analyzing the displacement in the Achilles tendon is debugged. The work is an extension of an ongoing research collaboration between KTH, GIH, and KI. The first step in the work was to debug and correct the MATLAB program for the speckle tracking algorithm. This was done by changing initial starting values in the algorithm such as the central frequency and the value for the speed of displacement in the Achilles tendon. After choosing a speed of 2 cm/s, the algorithm worked flawlessly. The next step was to execute the program on the ultrasound data for the Achilles tendon previously collected from the research collaboration. The algorithm was run for test subjects 17, 18, and 21 for habitual running style with three different running shoes and barefoot. This resulted in mean value curves for superficial and deep displacement in the Achilles tendon being plotted, with their minimum and maximum values being compared. After comparing the difference between the minimum and maximum values of superficial and deep displacement across all running shoes as well as barefoot, the following observations were made. Running with traditional shoes caused the highest deep displacement, followed by shoes with carbon fiber soles, minimalist shoes, and barefoot running. The shoe that caused the highest superficial displacement was shoes with carbon fiber soles, followed by traditional shoes, minimalist shoes, and running barefoot. The comparison was not based on any statistical basis since only three test subjects were analyzed. This implies some uncertainty as to which type of shoe causes the highest and lowest displacement. Aside from this uncertainty, it can be noted that the results from previous studies have findings that suggest similar insights.

The interaction of optical waves with material systems often results in complex, seemingly random fields. Because the fluctuations of such fields are typically difficult to analyze, they are regarded as noise to be suppressed. Nevertheless, in many cases the fluctuations of the field result from a linear and deterministic, albeit complicated, interaction between the optical field and the scattering system. As a result, linear systems theory (LST) can be used to frame the scattering problem and highlight situations in which useful information can be extracted from the fluctuations of the scattered field. Three fundamental problems can be posed in LST regardless of the nature of the system: one direct and two inverse problems. The direct problem attempts to predict the response of a known system to a known input. The problem may be simple enough to admit analytical solutions as in the case of homogeneous materials, phase and amplitude screens, and weakly scattering materials; or the problem may require the use of numerical techniques. This dissertation will focus on the two inverse problems, namely the determination of either the excitation field or the scattering system. Traditionally, the excitation determination problem has relied on designing optical systems that respond to the property of interest in a simple, easily quantified way. For example, gratings can be used to map wavelength onto direction of propagation while waveplates and polarizers can map polarization properties onto intensity. The primary difficulty with directly applying the concepts of LST to scattering systems iv is that, while the outputs are still combinations of the inputs, they are not ``simple'' combinations such as Fourier transforms or spatially dispersed spectral components of the input spectrum. Instead, the scattered field can be thought of as a massive sampling and mixing of the excitation field. This dissertation will show that such complicated sampling functions can be characterized and that the corresponding scattering medium can then be used as an optical device such as a lens, polarimeter, or spectrometer. The second inverse problem, system determination, is often more difficult because the problem itself may be ill-posed. For scattering systems that are dominated by low-order scattering, the statistical properties of the scattered light may serve as a fingerprint for material discrimination; however, in many situations, the statistical properties of the output do not depend on the material properties. Rather than analyzing the scattered field from one realization of the random interaction, several measurement techniques have been developed that attempt to extract information about the material system from modifications of the scattered field in response to changes in either the excitation or the intrinsic dynamics of the medium itself. One such technique is dynamic light scattering. This dissertation includes an extension to this method that allows for a polarimetric measurement of the scattered light using a reference beam with controllable polarization. Another system determination problem relates to imaging the reflectivity of a target that is being randomly illuminated. It will be demonstrated that an approach based on the correlation between the integrated scattered intensity and the corresponding illumination intensity distribution can prove superior to standard imaging microscopy

No description available.

No description available.