Sök i SwePub databas

  Utökad sökning

Träfflista för sökning "hsv:(NATURVETENSKAP) hsv:(Data och informationsvetenskap) hsv:(Datorseende och robotik autonoma system) "

Sökning: hsv:(NATURVETENSKAP) hsv:(Data och informationsvetenskap) hsv:(Datorseende och robotik autonoma system)

Sortera/gruppera träfflistan
  • Granlund, Gösta H. (författare)
  • A Nonlinear, Image-content Dependent Measure of Image Quality
  • 1977
  • Rapport (övrigt vetenskapligt)abstract
    • In recent years, considerable research effort has been devoted to the development of useful descriptors for image quality. The attempts have been hampered by i n complete understanding of the operation of the human visual system. This has made it difficult to relate physical measures and perceptual traits.A new model for determination of image quality is proposed. Its main feature is that it tries to invoke image content into consideration. The model builds upon a theory of image linearization, which means that the information in an image can well enough be represented using linear segments or structures within local spatial regions and frequency ranges. This implies a l so a suggestion that information in an image has to do with one- dimensional correlations. This gives a possibility to separate image content from noise in images, and measure them both.Also a hypothesis is proposed that the visual system of humans does in fact perform such a linearization.
  • Liu, Yang, et al. (författare)
  • Movement Status Based Vision Filter for RoboCup Small-Size League
  • 2012
  • Ingår i: Advances in Automation and Robotics, Vol. 2. - : Springer. - 9783642256455 - 9783642256462 ; , s. 79-86
  • Bokkapitel (övrigt vetenskapligt)abstract
    • Small-size soccer league is a division of the RoboCup (Robot world cup) competitions. Each team uses its own designed hardware and software to compete with othersunder defined rules. There are two kinds of data which the strategy system will receive from the dedicated server, one of them is the referee commands, and the other one is vision data. However, due to the network delay and the vision noise, we have to process the data before we can actually use it. Therefore, a certain mechanism is needed in this case.Instead of using some prevalent and complex algorithms, this paper proposes to solve this problem from simple kinematics and mathematics point of view, which can be implemented effectively by hobbyists and undergraduate students. We divide this problem by the speed status and deal it in three different situations. Testing results show good performance with this algorithm and great potential in filtering vision data thus forecasting actual coordinates of tracking objects.
  • Shirabe, Takeshi (författare)
  • Drawing with geography
  • 2015
  • Ingår i: Lecture Notes in Geoinformation and Cartography. - 9783319167862 ; , s. 327-341
  • Konferensbidrag (refereegranskat)abstract
    • A method is proposed to assist spatial planners in drawing with ‘geographic’ constraints. These constraints constrain graphic objects to have certain relationships that are not limited to be (Euclidean) geometric or topological but allowed to be dependent on the spatial variation of selected conditions (e.g., elevation and vegetation) characterizing an underlying geographic space. Just as in existing computer-aided design systems, the method accepts a manual change to a graphic object or constraint, and updates all affected graphic objects accordingly. The paper discusses how such a method is motivated and improves the graphic editing capability of geographic information systems, and identifies key issues for its implementation.
  • Liu, Fei, et al. (författare)
  • Infrared-visible image registration for augmented reality-based thermographic building diagnostics
  • 2015
  • Ingår i: Visualization in Engineering. - : Springer. - 2213-7459. ; 3:1
  • Tidskriftsartikel (refereegranskat)abstract
    • Background: In virtue of their capability to measure temperature, thermal infrared cameras have been widely used in building diagnostics for detecting heat loss, air leakage, water damage etc. However, the lack of visual details in thermal infrared images makes the complement of visible images a necessity. Therefore, it is often useful to register images of these two modalities for further inspection of architectures. Augmented reality (AR) technology, which supplements the real world with virtual objects, offers an ideal tool for presenting the combined results of thermal infrared and visible images. This paper addresses the problem of registering thermal infrared and visible façade images, which is essential towards developing an AR-based building diagnostics application. Methods: A novel quadrilateral feature is devised for this task, which models the shapes of commonly present façade elements, such as windows. The features result from grouping edge line segments with the help of image perspective information, namely, vanishing points. Our method adopts a forward selection algorithm to determine feature correspondences needed for estimating the transformation model. During the formation of the feature correspondence set, the correctness of selected feature correspondences at each step is verified by the quality of the resulting registration, which is based on the ratio of areas between the transformed features and the reference features. Results and conclusions: Quantitative evaluation of our method shows that registration errors are lower than errors reported in similar studies and registration performance is usable for most tasks in thermographic inspection of building façades.
  • Boström, Henrik (författare)
  • Maximizing the Area under the ROC Curve with Decision Lists and Rule Sets
  • 2007
  • Ingår i: Proceedings of the 7th SIAM International Conference on Data Mining. - : Society for Industrial and Applied Mathematics. - 9780898716306 ; , s. 27-34
  • Konferensbidrag (refereegranskat)abstract
    • Decision lists (or ordered rule sets) have two attractive properties compared to unordered rule sets: they require a simpler classi¯cation procedure and they allow for a more compact representation. However, it is an open question what effect these properties have on the area under the ROC curve (AUC). Two ways of forming decision lists are considered in this study: by generating a sequence of rules, with a default rule for one of the classes, and by imposing an order upon rules that have been generated for all classes. An empirical investigation shows that the latter method gives a significantly higher AUC than the former, demonstrating that the compactness obtained by using one of the classes as a default is indeed associated with a cost. Furthermore, by using all applicable rules rather than the first in an ordered set, an even further significant improvement in AUC is obtained, demonstrating that the simple classification procedure is also associated with a cost. The observed gains in AUC for unordered rule sets compared to decision lists can be explained by that learning rules for all classes as well as combining multiple rules allow for examples to be ranked according to a more fine-grained scale compared to when applying rules in a fixed order and providing a default rule for one of the classes.
  • Bååth, Rasmus, et al. (författare)
  • A prototype based resonance model of rhythm categorization
  • 2014
  • Ingår i: i-Perception. - : Pion Ltd. - 2041-6695. ; 5:6, s. 548-558
  • Tidskriftsartikel (refereegranskat)abstract
    • Categorization of rhythmic patterns is prevalent in musical practice, an example of this being the transcription of (possibly not strictly metrical) music into musical notation. In this article we implement a dynamical systems’ model of rhythm categorization based on the resonance theory of rhythm perception developed by Large (2010). This model is used to simulate the categorical choices of participants in two experiments of Desain and Honing (2003). The model accurately replicates the experimental data. Our results support resonance theory as a viable model of rhythm perception and show that by viewing rhythm perception as a dynamical system it is possible to model central properties of rhythm categorization.
  • Wallenberg, Marcus, et al. (författare)
  • Improving Random Forests by Correlation-Enhancing Projections and Sample-Based Sparse Discriminant Selection
  • 2016
  • Ingår i: Proceedings 13th Conference on Computer and Robot Vision CRV 2016. - : Institute of Electrical and Electronics Engineers (IEEE). - 9781509024919 - 9781509024926 ; , s. 222-227
  • Konferensbidrag (refereegranskat)abstract
    • Random Forests (RF) is a learning techniquewith very low run-time complexity. It has found a nicheapplication in situations where input data is low-dimensionaland computational performance is paramount. We wish tomake RFs more useful for high dimensional problems, andto this end, we propose two extensions to RFs: Firstly, afeature selection mechanism called correlation-enhancing pro-jections, and secondly sparse discriminant selection schemes forbetter accuracy and faster training. We evaluate the proposedextensions by performing age and gender estimation on theMORPH-II dataset, and demonstrate near-equal or improvedestimation performance when using these extensions despite aseventy-fold reduction in the number of data dimensions.
  • Kjellin, Andreas, et al. (författare)
  • Evaluating 2D and 3D Visualizations of Spatiotemporal Information
  • 2010
  • Ingår i: ACM Transactions on Applied Perception. - 1544-3558 .- 1544-3965. ; 7:3, s. 19:1-19:23
  • Tidskriftsartikel (refereegranskat)abstract
    • Time-varying geospatial data presents some specific challenges for visualization. Here, we report the results of three experiments aiming at evaluating the relative efficiency of three existing visualization techniques for a class of such data. The class chosen was that of object movement, especially the movements of vehicles in a fictitious landscape. Two different tasks were also chosen. One was to predict where three vehicles will meet in the future given a visualization of their past movement history. The second task was to estimate the order in which four vehicles arrived at a specific place. Our results reveal that previous findings had generalized human perception in these situations and that large differences in user efficiency exist for a given task between different types of visualizations depicting the same data. Furthermore, our results are in line with earlier general findings on the nature of human perception of both object shape and scene changes. Finally, the need for new taxonomies of data and tasks based on results from perception research is discussed.
  • Heintz, Fredrik, et al. (författare)
  • Linköping Humanoids : Application RoboCup 2016 Standard Platform League
  • 2016
  • Konferensbidrag (övrigt vetenskapligt)abstract
    • This is the application for the RoboCup 2016 Standard Platform League from the Linköping Humanoids team.Linköping Humanoids participated in RoboCup 2015. We didn’t do very well, but we learned a lot. When we arrived nothing worked. However, we fixed more and more of the open issues and managed to play a draw in our final game. We also participated in some of the technical challenges and scored some points. At the end of the competition we had a working team. This was both frustrating and rewarding. Analyzing the competition we have identified both what we did well and the main issues that we need to fix. One important lesson is that it takes time to develop a competitive RoboCup SPL team. Weare dedicated to improving our performance over time in order to be competitive in 2017.
Skapa referenser, mejla, bekava och länka
Typ av publikation
konferensbidrag (1340)
tidskriftsartikel (623)
bokkapitel (104)
doktorsavhandling (103)
rapport (63)
licentiatavhandling (41)
visa fler...
annan publikation (40)
proceedings (redaktörskap) (15)
forskningsöversikt (13)
samlingsverk (redaktörskap) (10)
bok (7)
patent (5)
konstnärligt arbete (1)
visa färre...
Typ av innehåll
refereegranskat (1849)
övrigt vetenskapligt (488)
populärvet., debatt m.m. (23)
Lindeberg, Tony, 196 ... (114)
Balkenius, Christian (109)
Lindblad, Joakim (75)
Bengtsson, Ewert (68)
Kragic, Danica (61)
Sladoje, Nataša (57)
visa fler...
Åström, Karl (48)
Johnsson, Magnus (46)
Heyden, Anders (44)
Felsberg, Michael, 1 ... (44)
Felsberg, Michael (40)
Borgefors, Gunilla (38)
Brun, Anders (34)
Kahl, Fredrik (33)
Li, Haibo (33)
Sintorn, Ida-Maria (33)
Johansson, Birger (33)
Khan, Fahad Shahbaz, ... (33)
Ahlberg, Jörgen, 197 ... (30)
Hotz, Ingrid (30)
Ek, Carl Henrik (28)
Nyström, Ingela (28)
Nalpantidis, Lazaros (28)
Oskarsson, Magnus (27)
Pham, Tuan D. (27)
Hast, Anders (25)
Strand, Robin (25)
Svensson, Stina (25)
Gasteratos, Antonios (25)
Seipel, Stefan (24)
Wählby, Carolina (23)
Maki, Atsuto (22)
Olsson, Carl (21)
Ballerini, L. (21)
Solem, Jan Erik (20)
Kragic, Danica, 1971 ... (20)
Kjellström, Hedvig (20)
Danelljan, Martin, 1 ... (20)
Bigun, Josef, 1961- (19)
Borgefors, G. (19)
Bengtsson, E (19)
Berg, Amanda, 1988- (18)
Lilienthal, Achim J. ... (17)
ur Réhman, Shafiq (17)
Eklundh, Jan-Olof (17)
Smedby, Örjan (15)
Josephson, Klas (15)
Lenz, Reiner (15)
Sminchisescu, Cristi ... (15)
Mester, Rudolf (15)
visa färre...
Uppsala universitet (587)
Kungliga Tekniska Högskolan (570)
Linköpings universitet (439)
Lunds universitet (426)
Örebro universitet (114)
Umeå universitet (90)
visa fler...
Högskolan i Halmstad (79)
Sveriges Lantbruksuniversitet (56)
Högskolan i Skövde (33)
Högskolan i Gävle (32)
Luleå tekniska universitet (26)
Blekinge Tekniska Högskola (26)
Mälardalens högskola (23)
Chalmers tekniska högskola (22)
Mittuniversitetet (13)
RISE (11)
Karolinska Institutet (11)
Linnéuniversitetet (10)
Stockholms universitet (8)
Jönköping University (8)
Högskolan Väst (7)
Göteborgs universitet (6)
Högskolan Dalarna (4)
Röda Korsets Högskola (2)
Malmö universitet (1)
Högskolan i Borås (1)
Karlstads universitet (1)
visa färre...
Engelska (2327)
Svenska (33)
Franska (3)
Spanska (1)
Forskningsämne (UKÄ/SCB)
Naturvetenskap (2364)
Teknik (246)
Medicin och hälsovetenskap (65)
Samhällsvetenskap (50)
Humaniora (48)
Lantbruksvetenskap (21)


Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

pil uppåt Stäng

Kopiera och spara länken för att återkomma till aktuell vy