This thesis is concerned with the material processing, photophysical and third-order nonlinear optical responses, and applications of a set of conjugated polymers in the telecommunication regions. Polyacetylene-based third-order nonlinear optical materials were chosen as candidates for all-optical signal and image processing. Substituted polyacetylenes were obtained using ring-opening metathesis polymerization of mono-substituted cyclooctatetraenes. Polymerization and processing conditions have been developed to generate thick, large-area films possessing large third-order nonlinearities in the telecommunication bands. The good optical quality of a 200 μm thick substituted polyacetylene film allowed for image correlation via off-resonant degenerated four-wave mixing with improved diffraction efficiency. Poly(2-methoxy-5-(2-ethyl-hexyloxy)-(phenylene vinylene)) (MEH-PPV) and (6,6)-phenyl-C61-butyric acid methyl ester (PCBM) composites showed strong nonlinear absorption and potential as optical limiters in the region of 700-900 nm. High optical quality, thick film of MEH-PPV:PCBM with the plasticizer dioctylphthalate (DOP) were made. Optical limiting of femtosecond and nanosecond pulses in the near infrared on these composites showed strong power suppression over a broad temporal regime. Femtosecond and nanosecond transient studies on the same thick MEH-PPV:PCBM:DOP composite films and the experimental results showed evidence for the photogeneration of radical ions as being responsible for the enhanced nonlinear absorption and strong optical suppression in the near infrared. Dithienopyrrole-based donor-acceptor copolymers with narrow bandgap showed strong nonlinear absorption and potential as optical limiters in the telecommunication wavelengths. Molecular engineering was applied to manipulate the spectral overlap of two-photon absorption and subsequent nonlinear absorptions. Femtosecond transient spectroscopy showed near infrared transient absorption and 22 - 61% yields of photogenerated charge-transfer species depending on donor-acceptor coupling strength. Torsional fluctuations of the backbone structure potentially affected the excited state behavior. Evidence suggests that ultrafast relaxation occurs to ground state and to long-lived charge-transfer state from the initially excited state. The dispersion of nonlinear absorption measured using the Z-scan method revealed large two-photon absorption cross sections of these polymers in the telecommunication region. Large suppression of nanosecond pulses at 1064 nm was achieved.

The Internet of Things (IoTs) has triggered rapid advances in sensors, surveillance devices, wearables and body area networks with advanced Human-Computer Interfaces (HCI). Neural Networks optimized algorithmically for high accuracy and high representation power are very deep and require tremendous storage and processing capabilities leading to higher area and power costs. For developing smart front-ends for ‘always on’ sensor nodes we need to optimize for power and area. This requires considering trade-offs with respect to various entities such as resource utilization, processing time, area, power, accuracy etc. Our experimental results show that there is presence of a network configuration with minimum energy given the input constraints of an application in consideration. This presents the need for a hardware-software co-design approach. We present a highly parameterized hardware design on an FPGA allowing re-configurability and the ability to evaluate different design choices in a short amount of time. We also describe the capability of extending our design to offer run time configurability. This allows the design to be altered for different applications based on need and also allows the design to be used as a cascaded classifier beneficial for continuous sensing for low power applications. This thesis aims to evaluate the use of Restricted Boltzmann Machines for building such re-configurable low power front ends. We develop the hardware architecture for such a system and provide experimental results obtained for the case study of Posture detection for body worn cameras used for law enforcement.

隨著科技的進步以及網際網路的普及，影像資訊的傳遞已經漸漸取代文字的表達，人們對於影像的需求也越來越多元，使得影像處理技術以及影像資訊分析也就越來越重要。然而，影像中其中一項重要的資訊為特徵描述子，強而有力的描述子能使得影像在辨識、分類等應用上有較佳的回饋，描述子的建構方式根據編碼原則分為：基於區域梯度統計、基於點對關係以及基於點群關係。其中，基於點群關係的編碼方式因為點群的選取及排序過程中，可能會產生過多的關係表示方法數，以至於不利於計算，因此過去較少有利用點群關係的編碼方式所建構而成的特徵描述子。 本論文提出描述子建構方式-LIOR，是以點群排序關係為基礎的編碼方式，相較於LIOP方法隨著點群內的點數增加，元素關係數大幅度的成長，造成描述子維度過大，計算時間和空間皆可能需要大量的消耗，而本研究方法足以改善計算維度的問題，重新定義點群關係的排名機制，並以像素值為基準加入權重分配，以區別加權排序之間不同大小差值所造成的影響程度。 實驗結果顯示本研究方法對於不同影像劣化效果的資料集，不僅能提升選取多點為一群的影像比對評估效能，同時也能改善點群內元素關係過多的排名表示法，降低以多點為群集的特徵描述子維度，節省了影像比對的計算時間以及空間，仍可維持整體影像配對之效能。

Τα τελευταία χρόνια η αναμφισβήτητα εντυπωσιακή τεχνολογική πρόοδος έχει δώσει τεράστια ώθηση σε ένα ευρύ φάσμα επιστημονικών πεδίων τα οποία βασίζονται σε αυτή, με χαρακτηριστικό παράδειγμα τον τομέα των Τηλεπικοινωνιών. Απόλυτα λογικό αποτέλεσμα είναι η παγκόσμια αγορά να έχει κατακλυστεί από δισεκατομμύρια κινητές συσκευές, η χρήση των οποίων διευκολύνει απίστευτα και την ανάπτυξη της έρευνας στους αντίστοιχους κλάδους. Μία από τις χαρακτηριστικές τεχνολογίες που εκμεταλλευόμενη πλήρως αυτή τη αλματώδη πρόοδο έχει επιφέρει εντυπωσιακά αποτελέσματα είναι η Επαυξημένη Πραγματικότητα. Βασίζεται στην γεφύρωση του χάσματος μεταξύ της εικονικότητας και της πραγματικότητας με ένα μοναδικό τρόπο ενίσχυσης της αντίληψής μας ως προς την τελευταία και με εφαρμογές από το Στρατό και τη Διαφήμιση ως τα ερευνητικά πεδία της Αναγνώρισης Εικόνας και της Εκπαίδευσης. Αυτή της όμως η σημαντικότατη ανάπτυξη γεννά αναπόφευκτα πέρα από την παραγωγή γνώσης και βασικά ερωτήματα που απασχολούν την ερευνητική κοινότητα, ορισμένα εκ των οποίων είναι αν μπορεί η χρήση τεχνικών Επαυξημένης Πραγματικότητας μέσω της αναγνώρισης εικόνας, πέρα από την εξέλιξη του ευρύτερου κλάδου της Υπολογιστικής Όρασης, να συμβάλλει στη βελτίωση μαθησιακών εμπειριών και αν ναι κατά πόσο είναι εφικτή η αντιμετώπιση θεμελιωδών εκπαιδευτικών προβλημάτων, όπως η διάσπαση προσοχής και η έλλειψη διαδραστικότητας. Στα πλαίσια της παρούσας εργασίας προσπαθώντας να δώσουμε ουσιαστικές απαντήσεις στα παραπάνω ερωτήματα δημιουργήσαμε τη διαδραστική εφαρμογή «AugMentor», ειδικά σχεδιασμένη για τις ανάγκες της ξενάγησης των επισκεπτών του Μουσείου Επιστημών του Πανεπιστημίου Πατρών. Έχοντας ενσωματώσει ένα ολοκληρωμένο περιβάλλον Επαυξημένης Πραγματικότητας εντός της εφαρμογής αναπτύξαμε ένα παιχνίδι κρυμμένου θησαυρού κατά το οποίο οι επισκέπτες του μουσείου δέχονται ένα συγκεκριμένο αριθμό ερωτήσεων οι οποίες βασίζονται στα μουσειακά εκθέματα και αμέσως μετά καλούνται να περιηγηθούν οι ίδιοι στους εσωτερικούς χώρους του. Με βάση την επιτυχή αναγνώριση του κάθε εκθέματος και την προσωποποιημένη απεικόνιση των εκάστοτε απεικονιζόμενων πληροφοριών προσπαθούν να δώσουν απαντήσεις στα ερωτήματα αυτά. Η εφαρμογή AugMentor υλοποιήθηκε για κινητές συσκευές που υποστηρίζουν το λειτουργικό σύστημα Android και πλέον αποτελεί μέρος της μόνιμης ξενάγησης του Μουσείου Επιστημών και Τεχνολογίας. Παρουσιάζουμε μια περιγραφική και αναλυτική μελέτη περίπτωσης χρήσης της εφαρμογής, υλοποιώντας πειραματικά στο Μουσείο Επιστημών μια κανονική ξενάγηση. Χρησιμοποιώντας την εφαρμογή AugMentor οδηγούμαστε σε ασφαλή συμπεράσματα ως προς την επίτευξη του κύριου στόχου υλοποίησής της, ο οποίος είναι το κατά πόσο οι επισκέπτες του μουσείου στην προσπάθειά τους να απαντήσουν τις ερωτήσεις που τους ανατέθηκαν θα μπορούσαν να μάθουν και επιπρόσθετες πληροφορίες για τα εκθέματα, βελτιώνοντας τη μάθηση μέσω της χρήσης κινητών συσκευών. Το πείραμα που πραγματοποιήσαμε χωρίστηκε σε δύο μέρη, το πρώτο βασίζεται στα πρότυπα μιας κανονικής ξενάγησης των επισκεπτών, χωρίς τη χρήση κάποιας εφαρμογής και το δεύτερο στην ξενάγηση μέσω της εφαρμογής AugMentor. Στους επισκέπτες δόθηκαν αντίστοιχα ερωτηματολόγια και στις δύο περιπτώσεις και στη συνέχεια αξιολογήθηκε συγκριτικά η συμβολή της εφαρμογής AugMentor στη συνολική εμπλούτιση της μουσειακής εμπειρίας ενός επισκέπτη. / Over the last few years the indisputably outstanding technological advances have given a huge boost to a wide range of scientific areas of study that are based on it. A typical example can definitely be considered the one associated with Telecommunications. An utterly rational outcome constitutes the fact that global market is overwhelmed by billions of mobile devices, the use of which facilitates to a large extent the whole research development towards those fields. An iconic technology that fully exploiting this rapid progress has brought about impressive results is Augmented Reality. Augmented Reality is based on bridging the gap between virtuality and reality in a unique way to enhance our perception as to the latter and its applications vary from Army and Advertisement to the research fields of Image Recognition and Education. It is exactly the impressive growth of Augmented Reality which inevitably generates essential questions that research community has to deal with. Some of these questions are whether the use of Augmented Reality techniques, through Image Recognition, apart from the wider development of Computer Vision, can help improving learning experiences and if so to which extent it is possible to address fundamental educational problems, such as distractibility and lack of interactivity. In this master thesis, trying to give substantial answers to these questions we have created the interactive application «AugMentor», specially designed for the tour needs of Museum of Science and Technology, located in University of Patras. Having incorporated an integrated Augmented Reality environment inside this application, we developed a treasure hunt game in which visitors receive a certain number of questions closely related to museum’s exhibits and afterwards they have to wander within its interiors. Based on the successful recognition of each exhibit and personalized display of each respective information, visitors attempt to answer these questions. AugMentor is implemented for mobile devices that support the Android operating system and is now part of permanent museum guidance. Our main objective is improving Mobile Learning, which is learning through the use of mobile devices. We present a sufficient case study, selecting Museum of Science and Technology as an experimental space so as to determine the degree to which AugMentor application can practically contribute to this purpose. One of our key points towards this perspective was estimating the amount of totally imparted knowledge, since involving students with a significant pool of knowledge provides them with a motivation to have an active role during their guidance. The experiment we conducted was divided into two parts, the first one was based on the standards of a normal tour of visitors without using any application and the second one was based on a tour using AugMentor application. Guests are respectively given questionnaires in both cases and then we assessed AugMentor’s relative contribution to the overall implementation regarding enriching the museum experience of a visitor.

O avanço tecnológico recente tem atraído tanto a comunidade acadêmica quanto o mercado para a investigação de novos métodos, técnicas e linguagens formais para a área de Projeto de Engenharia. A principal motivação é o atendimento à demanda para desenvolver produtos e sistemas cada vez mais completos e que satisfaçam as necessidades do usuário final. Necessidades estas que podem estar ligadas, por exemplo, à análise e reconhecimento de objetos que compõe uma imagem pela sua textura, um processo essencial na automação de uma enorme gama de aplicações como: visão robótica, monitoração industrial, sensoriamento remoto, segurança e diagnóstico médico assistido. Em vista da relevância das inúmeras aplicações envolvidas e pelo fato do domínio de aplicação ser muito próximo do contexto do desenvolvedor, é apresentada uma proposta de um processo de design baseado no Projeto Axiomático como sendo o mais indicado para esta situação. Especificamente, se espera que no estudo de caso da análise de textura haja uma convergência mais rápida para a solução - se esta existir. No estudo de caso, se desenvolve uma nova concepção de arquitetura de rede neural artificial (RNA), auto-organizável, com a estrutura espacial bidimensional da imagem de entrada preservada, tendo a extração e reconhecimento/classificação de textura em uma única fase de aprendizado. Um novo conceito para o paradigma da competição entre os neurônios também é estabelecida. O processo é original por permitir que o desenvolvedor assuma concomitantemente o papel do cliente no projeto, e especificamente por estabelecer o processo de sistematização e estruturação do raciocínio lógico do projetista para a solução do problema a ser desenvolvido e implementado em RNA. / The recent technological advance has attracted the industry and the academic community to research and propose methods, seek for new techniques, and formal languages for engineering design in order to respond to the growing demand for sophisticated product and systems that fully satisfy customers needs. It can be associated, for instance, with an application of object recognition using texture features, essential to a variety of applications domains, such as robotic vision, industrial inspection, remote sensing, security and medical image diagnosis. Considering the importance of the large number of applications mentioned before, and due to their characteristic where both application and developer domain are very close to each other, this work aims to present a design process based on ideas extracted from axiomatic design to accelerate the development for the classical approach to texture analysis. Thus, a case study is accomplished where a new conception of neural network architecture is specially designed for the following proposal: preserving the two-dimensional spatial structure of the input image, and performing texture feature extraction and classification within the same architecture. As a result, a new mechanism for neuronal competition is also developed as specific knowledge for the domain. In fact, the process proposed has some originality because it does take into account that the developer assumes also the customers role on the project, and establishes the systematization process and structure of logical reasoning of the developer in order to develop and implement the solution in neural network domain.

This is a B.Sc. thesis within the Computer Science programme at the Mid Sweden University. The purpose of this project has been to investigate the possibility of using image based surveillance in smart applications on the Internet-of-Things. The goals involved investigating relevant technologies and designing, implementing and evaluating an application that can perform image recognition. A number of image recognition techniques have been investigated and the use of color histograms has been chosen for its simplicity and low resource requirement. The main source of study material has been the Internet. The solution has been developed in the Java programming language, for use on the Android operating system and using the MediaSense platform for communication. It consists of a camera application that produces image data and a monitor application that performs image recognition and handles user interaction. To evaluate the solution a number of tests have been performed and its pros and cons have been identified. The results show that the solution can differentiate between simple colored stick figures in a controlled environment. Variables such as lighting and the background are significant. The application can reliably send images from the camera to the monitor at a rate of one image every four seconds. The possibility of using streaming video instead of images has been investigated but found to be difficult under the given circumstances. It has been concluded that while the solution cannot differentiate between actual people it has shown that image based surveillance is possible on the IoT and the goals of this project have been satisfied. The results were expected and hold little newsworthiness. Suggested future work involves improvements to the MediaSense platform and infrastructure for processing and storing data. / MediaSense

To overcome the hurdles associated with space management and security controls in a housing system, research was projected to study and analyze the necessary factors of accomplishment. Over time, different processes were observed and reviewed to make this a possible deal. Various residents were interviewed on the daily constraints in parking and managing their vehicles within their housing premises. The reported daunting concern was majorly the gate access and personal hunts for the space to keep the individual resident’s cars. Every resident would always have to stop and hoot at the housing gate for the assigned personnel to check and open the gate. While this would waste every resident’s time, the visitors even face more delay often time. Hitherto, car access and parking constraint become a thing of worry that no one would want to engage the housing service anymore. The interest has got dwindled. And to re-awaken the high patronage of the housing system, a gap must be bridged with an immediate solution to space management with a gating system. These were subsequently given a classical thought, while a prototype solution was demonstrated and reviewed with the various residents of some selected housing. This received a high welcoming embracement and was beckoned to be made real by the logical heuristic. At this point, nothing was further considered than using the Internet of things (IoT) technology to implement Vehicular Access Management for the control and integration of intended space provisioning in any housings. Consequently, the number plate of every vehicle becomes the automatic access tag and would be used for security control within the housing location. Vehicles’ numbers would be captured and used to manage the residents passing through the automated gating system. With it, records would be made for all permitted residents and the visitors that own a car. Thus, a proper arrangement would be allotted accordingly, as provisioned by the gating system administrator. However, to allegories the above-proffered solution, this project work is divided into six sections. The introductory section introduces the project rationale, lists the objectives, explores related works, and introduces how IoT and vehicular systems can be merged. The second section delves into these vehicular systems. It introduces the Automatic License Plate Recognition System (ALRP) and the Raspberry Pi and highlights the merits of the Integrated Vehicular Access Security System. Open-CV and machine learning are also introduced. Section three covers the solution design, while section four is the implementation phase. Section five covers the testing and implementation of the solution. The final section summarizes the project. The project successfully models an automated solution for the security of tenants and vehicle users against unauthorized access to residential estates and buildings.

The 2019 novel coronavirus has been proven to present several unique features on chest X-rays and CT-scans that distinguish it from imaging of other pulmonary diseases such as bacterial pneumonia and viral pneumonia unrelated to COVID-19. However, the key characteristics of a COVID-19 infection have been proven challenging to detect with the human eye. The aim of this project is to explore if it is possible to distinguish a patient with COVID-19 from a patient who is not suffering from the disease from posteroanterior chest X-ray images using synthetic minority over-sampling and transfer learning. Furthermore, the report will also present the mechanics of COVID-19, the used dataset and models and the validity of the results.

Where there has been a steady development of Optical Character Recognition (OCR) techniques for printed documents, the instruments that provide good quality for hand-written manuscripts by Hand-written Text Recognition  methods (HTR) and transcriptions are still some steps behind. With the main focus on historical ciphers (i.e. encrypted documents from the past with various types of symbol sets), this thesis examines the performance of two machine learning architectures developed within the DECRYPT project framework, a clustering based unsupervised algorithm and a semi-supervised few-shot deep-learning model. Both models are tested on seen and unseen scribes to evaluate the difference in performance and the shortcomings of the two architectures, with the secondary goal of determining the influences of the datasets on the performance. An in-depth analysis of the transcription results is performed with particular focus on the Alchemic and Zodiac symbol sets, with analysis of the model performance relative to character shape and size. The results show the promising performance of Few-Shot architectures when compared to Clustering algorithm, with a respective SER average of 0.336 (0.15 and 0.104 on seen data / 0.754 on unseen data) and 0.596 (0.638 and 0.350 on seen data / 0.8 on unseen data).

This master‘s thesis deals with the Deep Learning methods for image recognition tasks from the first methods to the modern ones. The main focus is on convolutional neural nets based models for classification, detection and image segmentation. These methods are used for practical implemetation – counting passing cars on video from traffic camera. After several test of available models, the YOLOv2 architecture was chosen and retrained on own dataset. The application also includes the addition of the SORT tracking algorithm.