1,721,038 research outputs found
Proceedings of the third "international Traveling Workshop on Interactions between Sparse models and Technology" (iTWIST'16)
No full textThe third edition of the "international - Traveling Workshop on Interactions between Sparse models and Technology" (iTWIST) took place in Aalborg, the 4th largest city in Denmark situated beautifully in the northern part of the country, from the 24th to 26th of August 2016. The workshop venue was at the Aalborg University campus. One implicit objective of this biennial workshop is to foster collaboration between international scientific teams by disseminating ideas through both specific oral/poster presentations and free discussions. For this third edition, iTWIST'16 gathered about 50 international participants and features 8 invited talks, 12 oral presentations, and 12 posters on the following themes, all related to the theory, application and generalization of the "sparsity paradigm": Sparsity-driven data sensing and processing (e.g., optics, computer vision, genomics, biomedical, digital communication, channel estimation, astronomy); Application of sparse models in non-convex/non-linear inverse problems (e.g., phase retrieval, blind deconvolution, self calibration); Approximate probabilistic inference for sparse problems; Sparse machine learning and inference; "Blind" inverse problems and dictionary learning; Optimization for sparse modelling; Information theory, geometry and randomness; Sparsity? What's next? (Discrete-valued signals; Union of low-dimensional spaces, Cosparsity, mixed/group norm, model-based, low-complexity models, ...); Matrix/manifold sensing/processing (graph, low-rank approximation, ...); Complexity/accuracy tradeoffs in numerical methods/optimization; Electronic/optical compressive sensors (hardware)
Proceedings of the third "international Traveling Workshop on Interactions between Sparse models and Technology" (iTWIST'16)
No full textThe third edition of the "international - Traveling Workshop on Interactions between Sparse models and Technology" (iTWIST) took place in Aalborg, the 4th largest city in Denmark situated beautifully in the northern part of the country, from the 24th to 26th of August 2016. The workshop venue was at the Aalborg University campus. One implicit objective of this biennial workshop is to foster collaboration between international scientific teams by disseminating ideas through both specific oral/poster presentations and free discussions. For this third edition, iTWIST'16 gathered about 50 international participants and features 8 invited talks, 12 oral presentations, and 12 posters on the following themes, all related to the theory, application and generalization of the "sparsity paradigm": Sparsity-driven data sensing and processing (e.g., optics, computer vision, genomics, biomedical, digital communication, channel estimation, astronomy); Application of sparse models in non-convex/non-linear inverse problems (e.g., phase retrieval, blind deconvolution, self calibration); Approximate probabilistic inference for sparse problems; Sparse machine learning and inference; "Blind" inverse problems and dictionary learning; Optimization for sparse modelling; Information theory, geometry and randomness; Sparsity? What's next? (Discrete-valued signals; Union of low-dimensional spaces, Cosparsity, mixed/group norm, model-based, low-complexity models, ...); Matrix/manifold sensing/processing (graph, low-rank approximation, ...); Complexity/accuracy tradeoffs in numerical methods/optimization; Electronic/optical compressive sensors (hardware)
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Advanced sampling and reconstruction of images in Atomic Force Microscopy
No full textI dette projekt, bliver det undersøgt hvorvidt det er muligt at forbedre afbildningstiden i forbindelse med Atomar kraft mikroskopi. Dette er forsøgt gjort med en kombination af nye billedrekonstruktions metoder som Deep-Image-Prior og som en adaptiv sample metode. Atomar kraft mikroskopi er en metode til at lave meget nøjagtige højdekort af små ting. Det er muligt at forstørre meget mere end i et almindeligt lys mikroskop. Dette er gjort ved at føre en meget skarp og lille probe over overfladen med høj nøjagtighed. Det tager dog lang tid da proben fysisk skal røre alle stederne på objektet man gerne vil afbillede. Deep-image-prior er ny forskning der beskriver hvordan det ikke er krævet at træne et neuralt netværk for at det kan være af nytte i en billedgenskabelses kontekst. Det vises at gode resultater kan opnås ved at bruge netværks strukturen som regulerings funktion. Dette kan blandt andet bruges til at fjerne støj fra billeder eller til at fylde information ind i områder hvor det mangler eller er korrupt. Dette minder meget om problemet som arbejdes med her, hvor man kun kigger på en del af billedet og forsøger at gætte sig frem til resten. Dette forsøges derfor brugt til at genskabe et under samplet billede hvor kun 5-20% af billedet er tilgængeligt. Dette viste sig dog ikke særligt brugbart da rekonstruktions tiden er urimeligt lang så der ikke rigtigt spares noget tid i forhold til bare at scanne alle punkterne på hele ens objekt. Der fremstilles også en algoritme der scanner billedet i to omgange. En gang med meget lav opløsning for at få en idé om hvor i billedet der kan være interessante områder. Derefter scannes billedet en ekstra gang med højere opløsning i disse områder. Dette resulterer i en rekonstruktion som over hele billedet er værre end hvis man bare havde lavet en raster scanning med samme længde men et bedre resultat hvis der kun kigges på de relevante dele af billedet. Hvis man ønsker en bestemt opløsning på ens afbildning kan man med denne metode være sikker på at det vigtige får den opløsning mens man ikke spilder tid på at afbillede baggrunden. Der er andre lignende metoder som f.eks. afbilleder et objekt ved at følge kanten rundt på det objektet. Dette kræver dog at man aktivt styrer proben i dens kontrol løkke. Dette kræver en ændring af ens AKM mikroskop. Derfor har metoden fra denne rapport stadig relevans da det kun kræver at man kan programmere en sti som proben skal følge.AFM is a very useful technique to make a topology map of a specimen at a large range of magnifications. However the scanning process can be very time consuming since the probe has to slide across the surface of the sample in order to take the measurements. The work in this report tries to make this scan time shorter based on cutting edge research in image reconstruction and sample pattern generation. For reconstruction, there has recently been proposed a way of utilizing the structure of a NN for image reconstruction without training. This method is called DIP.A two shot adaptive sample pattern is also proposed where at first a crude scan is performed on the specimen. Then the interesting regions are identified, and scanned with the wanted resolution.The DIP method for reconstruction was deemed infeasible for AFM due to its reconstruction performance being comparable to that of interpolation, while being much more computationally expensive.The adaptive sample pattern has promising results for the relevant parts of the image. It shows an average speedup of 10 times for a reconstruction with approximately 44 dB PSNR for the relevant parts of the image. This speedup can however vary greatly from image to image but the method will ensure that only the most relevant parts of the image is raster scanned.<br/
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Computational Considerations for Ultra-Reliable Low Latency Wireless Networks
No full textIndustrielle netværksbaserede kontrolsystemer er traditionelt byggede på trådede forbindelser,såsom Ethernet. En af de overvejende grunde til dette er de strenge krav til latenstid ogpålidelighed.Næste generation af trådløse mobilnetværk, 5G, forsøger at imødekomme dette behov medforbedringer på begge fronter. Disse forbedringer er imidlertid ikke nok til alle fremtidige behov.Wireless Isochronous Real Time communication (WIRT) er et nyligt forslået kommunikationssystem bygget til netværksbaserede kontrolsystemer. Bedre latenstid, pålidelighed og fokus påperiodisk kommunikation er dele af WIRT som gør det egnet til de mest krævende kontrolsystemer.Denne afhandling omhandler de beregningsmæssige overvejelser der skal laves for at implementeredet fysiske lag af WIRT. Det fysiske lag er baseret på Orthogonal Frequency Division Multiplexing(OFDM) til Ultra-WideBand (UWB) spektrum.I forbindelse med projektet er en prototype transceiver designet og implementeret. Prototypen erintegreret med UWB testudstyr for at virke som udviklingsplatform til videre arbejde på WIRT.Prototypen virker som afsæt til evaluering af hvilke dele af transceiver-designet der påvirkerlatenstiden mest. Delkomponenterne af systemet gennemgås for at finde den nedre grænse forlatenstid, og efterfølgende estimeres den beregningsmæssige kompleksitet af hvert komponent.Baseret på dette estimat diskuteres egnede computerarkitekturer.Dekodning af den anvendte Error Correction Code (ECC) viser sig at være den størsteberegningskompleksitet og dermed den største kilde til forsinkelse. Den anvendte algoritmerevurderes og ændres for at reducere dekodningslatenstiden til et acceptabelt niveau. Yderligereændringer laves for at passe systemet til en rekonfigurerbar platform.Projektet konkluderes med en diskussion af brugbarhed af prototypen og fremtidige studier.Industrial control networks have traditionally been implemented on wired connections due to latency and reliability constraints. Next-generation cellular networks include services with improvements in these areas, but the improvements are not sufficient for all targeted use cases. Wireless Isochronous Real Time communication (WIRT) is a newly proposed system that targets networked control systems with periodic transmissions and extreme latency and reliability requirements. This work highlights considerations when implementing the WIRT physical layer. The system architecture is based on Orthogonal Frequency Division Multiplexing (OFDM) for Ultra-WideBand (UWB) spectrum. A prototype implementation is made. The implementation integrates UWB testing equipment, to serve as a testbed for further WIRT development. The implementation is used to estimate which parts of the transceiver introduce the highest latency. The minimum possible latency is determined and the computational complexity of each component is evaluated. Based on this evaluation, architectures for a latencyaccurate implementation are discussed. The decoding of Error Correction Codes (ECCs) is found to be the largest single contributor to latency. Algorithmic alterations are made to reduce the minimum decoding latency to an acceptable level, along with other considerations required for implementation on a reconfigurable logic platform. In the end considerations and further work to prove the feasibility of the system are discussed
Generalized Sampling: From Fourier to Wavelet
No full textIn this project we investigate generalized sampling as a tool for signal reconstruction and compression. Generalized sampling is a relatively new method for recovering any element in a finite dimensional space given finitely many samples in an arbitrary frame. The focus is on Fourier frames as sampling space and Daubechies wavelets as reconstruction space. We investigate the subject both in theory and in practise by proving relevant theorems and implementing algorithms in Python. Most of the theory is already published by others. However, to the best of our knowledge, it has not been implemented in Python before. The method is tested on several different signals with overall positive results. Among the test signals are both continuous and discontinuous signals, signals in one and two dimensions, and uniformly and nonuniformly sampled signals. For most of the tested signals compression using generalized sampling results in smaller errors than compression directly in the Fourier frame
- …
