<p>Positron Emission Tomography, PET, can be used for dynamic studies in humans. In such studies a selected part of the body, often the whole brain, is imaged repeatedly after administration of a radiolabelled tracer. Such studies are performed to provide sequences of images reflecting the tracer’s kinetic behaviour, which may be related to physiological, biochemical and functional properties of tissues. This information can be obtained by analyzing the distribution and kinetic behaviour of the administered tracers in different regions, tissues and organs. Each image in the sequence thus contains part of the kinetic information about the administered tracer. </p><p>Several factors make analysis of PET images difficult, such as a high noise magnitude and correlation between image elements in conjunction with a high level of non-specific binding to the target and a sometimes small difference in target expression between pathological and healthy regions. It is therefore important to understand how these factors affect the derived quantitative measurements when using different methods such as kinetic modelling and multivariate image analysis.</p><p>In this thesis, a new method to explore the properties of the noise in dynamic PET images was introduced and implemented. The method is based on an analysis of the autocorrelation function of the images. This was followed by proposing and implementing three novel approaches for application of Principal Component Analysis, PCA, on dynamic human PET studies. The common underlying idea of these approaches was that the images need to be normalized before application of PCA to ensure that the PCA is signal driven, not noise driven. Different ways to estimate and correct for the noise variance were investigated. Normalizations were carried out Slice-Wise (SW), for the whole volume at once, and in both image domain and sinogram domain respectively. We also investigated the value of masking out and removing the area outside the brain for the analysis. </p><p>The results were very encouraging. We could demonstrate that for phantoms as well as for real image data, the applied normalizations allow PCA to reveal the signal much more clearly than what can be seen in the original image data sets. Using our normalizations, PCA can thus be used as a multivariate analysis technique that without any modelling assumptions can separate important kinetic information into different component images. Furthermore, these images contained optimized signal to noise ratio (SNR), low levels of noise and thus showed improved quality and contrast. This should allow more accurate visualization and better precision in the discrimination between pathological and healthy regions. Hopefully this can in turn lead to improved clinical diagnosis. </p>

Positron Emission Tomography, PET, can be used for dynamic studies in humans. In such studies a selected part of the body, often the whole brain, is imaged repeatedly after administration of a radiolabelled tracer. Such studies are performed to provide sequences of images reflecting the tracer’s kinetic behaviour, which may be related to physiological, biochemical and functional properties of tissues. This information can be obtained by analyzing the distribution and kinetic behaviour of the administered tracers in different regions, tissues and organs. Each image in the sequence thus contains part of the kinetic information about the administered tracer. Several factors make analysis of PET images difficult, such as a high noise magnitude and correlation between image elements in conjunction with a high level of non-specific binding to the target and a sometimes small difference in target expression between pathological and healthy regions. It is therefore important to understand how these factors affect the derived quantitative measurements when using different methods such as kinetic modelling and multivariate image analysis. In this thesis, a new method to explore the properties of the noise in dynamic PET images was introduced and implemented. The method is based on an analysis of the autocorrelation function of the images. This was followed by proposing and implementing three novel approaches for application of Principal Component Analysis, PCA, on dynamic human PET studies. The common underlying idea of these approaches was that the images need to be normalized before application of PCA to ensure that the PCA is signal driven, not noise driven. Different ways to estimate and correct for the noise variance were investigated. Normalizations were carried out Slice-Wise (SW), for the whole volume at once, and in both image domain and sinogram domain respectively. We also investigated the value of masking out and removing the area outside the brain for the analysis. The results were very encouraging. We could demonstrate that for phantoms as well as for real image data, the applied normalizations allow PCA to reveal the signal much more clearly than what can be seen in the original image data sets. Using our normalizations, PCA can thus be used as a multivariate analysis technique that without any modelling assumptions can separate important kinetic information into different component images. Furthermore, these images contained optimized signal to noise ratio (SNR), low levels of noise and thus showed improved quality and contrast. This should allow more accurate visualization and better precision in the discrimination between pathological and healthy regions. Hopefully this can in turn lead to improved clinical diagnosis.

Volume measurements of timber loads is done in conjunction with timber trade. When dealing with goods of major economic values such as these, it is important to achieve an impartial and fair assessment when determining price-based volumes. With the help of Saab’s missile targeting technology, CIND AB develops products for digital volume measurement of timber loads. Currently there is a system in operation that automatically reconstructs timber trucks in motion to create measurable images of them. Future iterations of the system is expected to fully automate the scaling by generating a volumetric representation of the timber and calculate its external gross volume. The first challenge towards this development is to separate the timber load from the truck. This thesis aims to evaluate and implement appropriate method for semantic pixel-wise segmentation of timber loads in real time. Image segmentation is a classic but difficult problem in computer vision. To achieve greater robustness, it is therefore important to carefully study and make use of the conditions given by the existing system. Variations in timber type, truck type and packing together create unique combinations that the system must be able to handle. The system must work around the clock in different weather conditions while maintaining high precision and performance.

[pt] Redes neurais profundas são responsáveis pelo grande progresso em diversas tarefas perceptuais, especialmente nos campos da visão computacional,reconhecimento de fala e processamento de linguagem natural. Estes resultados produziram uma mudança de paradigma nas técnicas de reconhecimentode padrões, deslocando a demanda do design de extratores de característicaspara o design de arquiteturas de redes neurais. No entanto, o design de novas arquiteturas de redes neurais profundas é bastante demandanteem termos de tempo e depende fortemente da intuição e conhecimento de especialistas,além de se basear em um processo de tentativa e erro. Neste contexto, a idea de automatizar o design de arquiteturas de redes neurais profundas tem ganhado popularidade, estabelecendo o campo da busca por arquiteturas neurais(NAS - Neural Architecture Search). Para resolver o problema de NAS, autores propuseram diversas abordagens envolvendo o espaço de buscas, a estratégia de buscas e técnicas para mitigar o consumo de recursos destes algoritmos. O Q-NAS (Quantum-inspired Neural Architecture Search) é uma abordagem proposta para endereçar o problema de NAS utilizando um algoritmo evolucionário com inspiração quântica como estratégia de buscas. Este método foi aplicado de forma bem sucedida em classificação de imagens, superando resultados de arquiteturas de design manual nos conjuntos de dados CIFAR-10 e CIFAR-100 além de uma aplicação de mundo real na área da sísmica. Motivados por este sucesso, propõe-se nesta Dissertação o SegQNAS (Quantum-inspired Neural Architecture Search applied to Semantic Segmentation), uma adaptação do Q-NAS para a tarefa de segmentação semântica. Diversos experimentos foram realizados com objetivo de verificar a aplicabilidade do SegQNAS em dois conjuntos de dados do desafio Medical Segmentation Decathlon. O SegQNAS foi capaz de alcançar um coeficiente de similaridade dice de 0.9583 no conjunto de dados de baço, superando os resultados de arquiteturas tradicionais como U-Net e ResU-Net e atingindo resultados comparáveis a outros trabalhos que aplicaram NAS a este conjunto de dados, mas encontrando arquiteturas com muito menos parãmetros. No conjunto de dados de próstata, o SegQNAS alcançou um coeficiente de similaridade dice de 0.6887 superando a U-Net, ResU-Net e o trabalho na área de NAS que utilizamos como comparação. / [en] Deep neural networks are responsible for great progress in performance for several perceptual tasks, especially in the fields of computer vision, speech recognition, and natural language processing. These results produced a paradigm shift in pattern recognition techniques, shifting the demand from feature extractor design to neural architecture design. However, designing novel deep neural network architectures is very time-consuming and heavily relies on experts intuition, knowledge, and a trial and error process. In that context, the idea of automating the architecture design of deep neural networks has gained popularity, establishing the field of neural architecture search (NAS). To tackle the problem of NAS, authors have proposed several approaches regarding the search space definition, algorithms for the search strategy, and techniques to mitigate the resource consumption of those algorithms. Q-NAS (Quantum-inspired Neural Architecture Search) is one proposed approach to address the NAS problem using a quantum-inspired evolutionary algorithm as the search strategy. That method has been successfully applied to image classification, outperforming handcrafted models on the CIFAR-10 and CIFAR-100 datasets and also on a real-world seismic application. Motivated by this success, we propose SegQNAS (Quantum-inspired Neural Architecture Search applied to Semantic Segmentation), which is an adaptation of Q-NAS applied to semantic segmentation. We carried out several experiments to verify the applicability of SegQNAS on two datasets from the Medical Segmentation Decathlon challenge. SegQNAS was able to achieve a 0.9583 dice similarity coefficient on the spleen dataset, outperforming traditional architectures like U-Net and ResU-Net and comparable results with a similar NAS work from the literature but with fewer parameters network. On the prostate dataset, SegQNAS achieved a 0.6887 dice similarity coefficient, also outperforming U-Net, ResU-Net, and outperforming a similar NAS work from the literature.

Technology has revamped every aspect of our life, one of those various facets is fashion industry. Plenty of deep learning architectures are taking shape to augment fashion experiences for everyone. There are numerous possibilities of enhancing the fashion technology with deep learning. One of the key ideas is to generate fashion style and recommendation using artificial intelligence. Likewise, another significant feature is to gather reliable information of fashion trends, which includes analysis of existing fashion related images and data. When specifically dealing with images, localisation and segmentation are well known to address in-depth study relating to pixels, objects and labels present in the image. In this master thesis a complete framework is presented to perform localisation and segmentation on fashionista images. This work is a part of an interesting research work related to Fashion Style detection and Recommendation. Developed solution aims to leverage the possibility of localising fashion items in an image by drawing bounding boxes and labelling them. Along with that, it also provides pixel-wise semantic segmentation functionality which extracts fashion item label-pixel data. Collected data can serve as ground truth as well as training data for the aimed deep learning architecture. A study related to localisation and segmentation of videos has also been presented in this work. The developed system has been evaluated in terms of flexibility, output quality and reliability as compared to similar platforms. It has proven to be fully functional solution capable of providing essential localisation and segmentation services while keeping the core architecture simple and extensible. / Tekniken har förnyat alla aspekter av vårt liv, en av de olika fasetterna är modeindustrin. Massor av djupa inlärningsarkitekturer tar form för att öka modeupplevelser för alla. Det finns många möjligheter att förbättra modetekniken med djup inlärning. En av de viktigaste idéerna är att skapa modestil och rekommendation med hjälp av artificiell intelligens. På samma sätt är en annan viktig egenskap att samla pålitlig information om modetrender, vilket inkluderar analys av befintliga moderelaterade bilder och data. När det specifikt handlar om bilder är lokalisering och segmentering väl kända för att ta itu med en djupgående studie om pixlar, objekt och etiketter som finns i bilden. I denna masterprojekt presenteras en komplett ram för att utföra lokalisering och segmentering på fashionista bilder. Detta arbete är en del av ett intressant forskningsarbete relaterat till Fashion Style detektering och rekommendation. Utvecklad lösning syftar till att utnyttja möjligheten att lokalisera modeartiklar i en bild genom att rita avgränsande lådor och märka dem. Tillsammans med det tillhandahåller det även pixel-wise semantisk segmenteringsfunktionalitet som extraherar dataelementetikett-pixeldata. Samlad data kan fungera som grundsannelse samt träningsdata för den riktade djuplärarkitekturen. En studie relaterad till lokalisering och segmentering av videor har också presenterats i detta arbete. Det utvecklade systemet har utvärderats med avseende på flexibilitet, utskriftskvalitet och tillförlitlighet jämfört med liknande plattformar. Det har visat sig vara en fullt fungerande lösning som kan tillhandahålla viktiga lokaliseringsoch segmenteringstjänster samtidigt som kärnarkitekturen är enkel och utvidgbar.

The dissertation investigates the issues of image authentication and tamper localization after general image processing operations – blurring, sharpening, rotation and JPEG compression. The dissertation consists of four parts including Introduction, 4 chapters, Conclusions, References. The introduction reveals the investigated problem, importance of the thesis and the object of research and describes the purpose and tasks of the paper, research methodology, scientific novelty, the practical significance of results examined in the paper and defended statements. The introduction ends in presenting the author’s publications on the subject of the defended dissertation, offering the material of made presentations in conferences and defining the structure of the dissertation. Chapter 1 revises used literature, analyzes competitive methods. The devastating effect of blur/sharpen methods on digital image matrices is shown. General pixel-wise tamper localization methods are completely inefficient after such casual processing. Block-wise methods demonstrate some resistance against blurring/sharpening, but no tamper localization with the resolution of up to one pixel is possible. There is clearly a need for a method, able to locate damaged pixels despite general image processing operations such as blurring or sharpening. Chapter 2 defines theoretical foundation for the proposed method. It is shown that phase of Fourier transform demonstrates invariance against blurring or sharpening... [to full text] / Disertacijoje nagrinėjamos atvaizdų apsaugos – autentiškumo užtikrinimo ir pažeidimų radimo, po bendrųjų vaizdo apdorojimo procedūrų – blukinimo, ryškinimo, pasukimo ar JPEG suspaudimo – problemos. Disertaciją sudaro įvadas, keturi skyriai, rezultatų apibendrinimas, naudotos literatūros ir autoriaus publikacijų disertacijos tema sąrašai. Įvadiniame skyriuje aptariama tiriamoji problema, darbo aktualumas, aprašomas tyrimų objektas, formuluojamas darbo tikslas bei uždaviniai, aprašoma tyrimų metodika, darbo mokslinis naujumas, darbo rezultatų praktinė reikšmė, ginamieji teiginiai. Įvado pabaigoje pristatomos disertacijos tema autoriaus paskelbtos publikacijos ir pranešimai konferencijose bei disertacijos struktūra. Pirmasis skyrius skirtas literatūros apžvalgai. Jame pateikta konkuruojančių metodų apžvalga. Parodomas globalus blukinimo/ryškinimo operacijų poveikis atvaizdo skaitmeninėms matricoms. Išsiaiškinama kodėl pikselių tikslumo metodai nėra atsparūs tokiam poveikiui. Kodėl blokiniai metodai demonstruoja dalinį atsparumą – bet jie nėra pajėgūs surasti vieno pažeisto pikselio. Parodomas aiškus poreikis metodo, tiek galinčio rasti pažeistus pikselius, tiek atsparaus bendroms vaizdo apdorojimo procedūroms, tokioms kaip blukinimas ar aštrinimas. Antrajame skyriuje pateiktas sukurto metodo teorinis pagrindimas. Įrodoma, kad Furjė fazė pasižymi atsparumu blukinimui ir ryškinimui. Tačiau iškyla sekanti problema dėl to, kad Furjė fazėje neįmanoma rasti konkretaus pikselio –... [toliau žr. visą tekstą]

[pt] A crescente disponibilidade de dados de sensoriamento remoto vem criando novas oportunidades e desafios em aplicações de monitoramento de processos naturais e antropogénicos em escala global. Nos últimos anos, as técnicas de aprendizado profundo tornaram-se o estado da arte na análise de dados de sensoriamento remoto devido sobretudo à sua capacidade de aprender automaticamente atributos discriminativos a partir de grandes volumes de dados. Um dos problemas chave em análise de imagens é a segmentação semântica, também conhecida como rotulação de pixels. Trata-se de atribuir uma classe a cada sítio de imagem. As chamadas redes totalmente convolucionais de prestam a esta função. Os anos recentes têm testemunhado inúmeras propostas de arquiteturas de redes totalmente convolucionais que têm sido adaptadas para a segmentação de dados de observação da Terra. O presente trabalho avalias cinco arquiteturas de redes totalmente convolucionais que representam o estado da arte em segmentação semântica de imagens de sensoriamento remoto. A avaliação considera dados provenientes de diferentes plataformas: veículos aéreos não tripulados, aeronaves e satélites. Cada um destes dados refere-se a aplicações diferentes: segmentação de espécie arbórea, segmentação de telhados e desmatamento. O desempenho das redes é avaliado experimentalmente em termos de acurácia e da carga computacional associada. O estudo também avalia os benefícios da utilização do Campos Aleatórios Condicionais (CRF) como etapa de pósprocessamento para melhorar a acurácia dos mapas de segmentação. / [en] The increasing availability of remote sensing data has created new opportunities and challenges for monitoring natural and anthropogenic processes on a global scale. In recent years, deep learning techniques have become state of the art in remote sensing data analysis, mainly due to their ability to learn discriminative attributes from large volumes of data automatically. One of the critical problems in image analysis is the semantic segmentation, also known as pixel labeling. It involves assigning a class to each image site. The so-called fully convolutional networks are specifically designed for this task. Recent years have witnessed numerous proposals for fully convolutional network architectures that have been adapted for the segmentation of Earth observation data. The present work evaluates five fully convolutional network architectures that represent the state of the art in semantic segmentation of remote sensing images. The assessment considers data from different platforms: unmanned aerial vehicles, airplanes, and satellites. Three applications are addressed: segmentation of tree species, segmentation of roofs, and deforestation. The performance of the networks is evaluated experimentally in terms of accuracy and the associated computational load. The study also assesses the benefits of using Conditional Random Fields (CRF) as a post-processing step to improve the accuracy of segmentation maps.

[pt] A eficiência de culturas pode ser aprimorada monitorando-se suas condições de forma contínua e tomando-se decisões baseadas em suas análises. Os dados para análise podem ser obtidos através de sensores de imagens e o processo de monitoramento pode ser automatizado utilizando-se algoritmos de reconhecimento de imagem com diferentes níveis de complexidade. Alguns dos algoritmos de maior êxito estão relacionados a abordagens supervisionadas de aprendizagem profunda (Deep Learning) as quais utilizam formas de Redes Neurais de Convolucionais (CNNs). Nesta dissertação de mestrado, empregaram-se modelos de aprendizagem profunda supervisionados para classificação, regressão, detecção de objetos e segmentação semântica em tarefas de monitoramento de culturas, utilizando-se amostras de imagens obtidas através de três níveis distintos: Satélites, Veículos Aéreos Não Tripulados (UAVs) e Robôs Terrestres Móveis (MLRs). Ambos satélites e UAVs envolvem o uso de imagens multiespectrais. Para o primeiro nível, implementou-se um modelo CNN baseado em Transfer Learning para a classificação de espécies vegetativas. Aprimorou-se o desempenho de aprendizagem do transfer learning através de um método de análise estatística recentemente proposto. Na sequência, para o segundo nível, implementou-se um algoritmo segmentação semântica multitarefa para a detecção de lavouras de cana-de-açúcar e identificação de seus estados (por exemplo, saúde e idade da cultura). O algoritmo também detecta a vegetação ao redor das lavouras, sendo relevante na busca por ervas daninhas. No terceiro nível, implementou-se um algoritmo Single Shot Multibox Detector para detecção de cachos de tomate. De forma a avaliar o estado dos cachos, utilizaram-se duas abordagens diferentes: uma implementação baseada em segmentação de imagens e uma CNN supervisionada adaptada para cálculos de regressão capaz de estimar a maturação dos cachos de tomate. De forma a quantificar cachos de tomate em vídeos para diferentes estágios de maturação, empregou-se uma implementação de Região de Interesse e propôs-se um sistema de rastreamento o qual utiliza informações temporais. Para todos os três níveis, apresentaram-se soluções e resultados os quais superam as linhas de base do estado da arte. / [en] Crop efficiency can be improved by continually monitoring their state and making decisions based on their analysis. The data for analysis can be obtained through images sensors and the monitoring process can be automated by using image recognition algorithms with different levels of complexity. Some of the most successful algorithms are related to supervised Deep Learning approaches which use a form of Convolutional Neural Networks (CNNs). In this master s dissertation, we employ supervised deep learning models for classification, regression, object detection, and semantic segmentation in crop monitoring tasks, using image samples obtained through three different levels: Satellites, Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs). Both satellites and UAVs levels involve the use of multispectral images. For the first level, we implement a CNN model based on transfer learning to classify vegetative species. We also improve the transfer learning performance by a newly proposed statistical analysis method. Next, for the second level, we implement a multi-task semantic segmentation algorithm to detect sugarcane crops and infer their state (e.g. crop health and age). The algorithm also detects the surrounding vegetation, being relevant in the search for weeds. In the third level, we implement a Single Shot Multibox detector algorithm to detect tomato clusters. To evaluate the cluster s state, we use two different approaches: an implementation based on image segmentation and a supervised CNN regressor capable of estimating their maturity. In order to quantify the tomato clusters in videos at different maturation stages, we employ a Region of Interest implementation and also a proposed tracking system which uses temporal information. For all the three levels, we present solutions and results that outperform state-of-the art baselines.

Vision based active safety systems have become more frequently occurring in modern vehicles to estimate depth of the objects ahead and for autonomous driving (AD) and advanced driver-assistance systems (ADAS). In this thesis a lightweight deep convolutional neural network performing real-time depth estimation on single monocular images is implemented and evaluated. Many of the vision based automatic brake systems in modern vehicles only detect pre-trained object types such as pedestrians and vehicles. These systems fail to detect general objects such as road debris and roadside obstacles. In stereo vision systems the problem is resolved by calculating a disparity image from the stereo image pair to extract depth information. The distance to an object can also be determined using radar and LiDAR systems. By using this depth information the system performs necessary actions to avoid collisions with objects that are determined to be too close. However, these systems are also more expensive than a regular mono camera system and are therefore not very common in the average consumer car. By implementing robust depth estimation in mono vision systems the benefits from active safety systems could be utilized by a larger segment of the vehicle fleet. This could drastically reduce human error related traffic accidents and possibly save many lives. The network architecture evaluated in this thesis is more lightweight than other CNN architectures previously used for monocular depth estimation. The proposed architecture is therefore preferable to use on computationally lightweight systems. The network solves a supervised regression problem during the training procedure in order to produce a pixel-wise depth estimation map. The network was trained using a sparse ground truth image with spatially incoherent and discontinuous data and output a dense spatially coherent and continuous depth map prediction. The spatially incoherent ground truth posed a problem of discontinuity that was addressed by a masked loss function with regularization. The network was able to predict a dense depth estimation on the KITTI dataset with close to state-of-the-art performance.

[pt] A partir de melhorias incrementais em uma conhecida rede neural convolucional (U-Net), diferentes técnicas são avaliadas quanto às suas performances na tarefa de segmentação semântica em imagens sísmicas. Mais especificamente, procura-se a identificação e delineamento de estruturas salinas no subsolo, o que é de grande relevância na indústria de óleo e gás para a exploração de petróleo em camadas pré-sal, por exemplo. Além disso, os desafios apresentados no tratamento destas imagens sísmicas se assemelham em muito aos encontrados em tarefas de áreas médicas como identificação de tumores e segmentação de tecidos, o que torna o estudo da tarefa em questão ainda mais valioso. Este trabalho pretende sugerir uma metodologia adequada de abordagem à tarefa e produzir redes neurais capazes de segmentar imagens sísmicas com bons resultados dentro das métricas utilizadas. Para alcançar estes objetivos, diferentes estruturas de redes, transferência de aprendizado e técnicas de aumentação de dados são testadas em dois datasets com diferentes níveis de complexidade. / [en] Through incremental improvements in a well-known convolutional neural network (U-Net), different techniques are evaluated regarding their performance on the task of semantic segmentation of seismic images. More specifically, the objective is the better identification and outline of subsurface salt structures, which is a task of great relevance for the oil and gas industry in the exploration of pre-salt layers, for example. Besides that application, the challenges imposed by the treatment of seismic images also resemble those found in medical fields like tumor detection and tissue segmentation, which makes the study of this task even more valuable. This work seeks to suggest a suitable methodology for the task and to yield neural networks that are capable of performing semantic segmentation of seismic images with good results regarding specific metrics. For that purpose, different network structures, transfer learning and data augmentation techniques are applied in two datasets with different levels of complexity.