<p>In robot navigation, and image content searches reliable salient features are of pivotal importance. Also in biometric human recognition, salient features are increasingly used. </p><p>Regardless the application, image matching is one of the many problems in computer </p><p>vision, including object recognition. </p><p> </p><p>This report investigates some salient features to match sub-images of different images. </p><p>An underlying assumption is that sub-images, also called image objects, or objects, are </p><p>possible to recognize by the salient features that can be recognized independently. </p><p> </p><p>Since image objects are images of 3D objects, the salient features in 2D images must be </p><p>invariant to reasonably large viewing direction and distance (scale) changes. These </p><p>changes are typically due to 3D rotations and translations of the 3D object with respect to </p><p>the camera. Other changes that influence the matching of two 2D image objects is </p><p>illumination changes, and image acquisition noise. </p><p> </p><p>This thesis will discuss how to find the salient features and will compare them with </p><p>respect to their matching performance. Also it will explore how these features are </p><p>invariant to rotation and scaling.</p>

In robot navigation, and image content searches reliable salient features are of pivotal importance. Also in biometric human recognition, salient features are increasingly used. Regardless the application, image matching is one of the many problems in computer vision, including object recognition. This report investigates some salient features to match sub-images of different images. An underlying assumption is that sub-images, also called image objects, or objects, are possible to recognize by the salient features that can be recognized independently. Since image objects are images of 3D objects, the salient features in 2D images must be invariant to reasonably large viewing direction and distance (scale) changes. These changes are typically due to 3D rotations and translations of the 3D object with respect to the camera. Other changes that influence the matching of two 2D image objects is illumination changes, and image acquisition noise. This thesis will discuss how to find the salient features and will compare them with respect to their matching performance. Also it will explore how these features are invariant to rotation and scaling.

In two experiments the Weighted Salience Model (WSM) of conceptual combination was examined. Several of the hypotheses set forth in the WSM were evaluated, including the importance of salience of constituent features, differential interpretation strategies based on similarity, an initial reliance on the modifier as opposed to the head, and a context effect of salience reorganization. Results confirmed that the hierarchy of output dominance within constituent features was important in determining features in final combinations. Additionally, similar pairs were defined with property interpretations more frequently than were dissimilar pairs, and dissimilar pairs were defined with relation interpretations more frequently than were similar pairs. Context effects were demonstrated through the finding that target features were found more often in primed than unprimed pairs. The hypothesis of modifier superiority was not confirmed. These findings indicate that the WSM adds to the current understanding of conceptual combination through a reliance on output dominance and the importance of context. Despite these strengths, changes to the WSM may be necessary if future studies fail to support the importance of the modifier over the head noun.

In two experiments the Weighted Salience Model (WSM) of conceptual combination was examined. Several of the hypotheses set forth in the WSM were evaluated, including the importance of salience of constituent features, differential interpretation strategies based on similarity, an initial reliance on the modifier as opposed to the head, and a context effect of salience reorganization. Results confirmed that the hierarchy of output dominance within constituent features was important in determining features in final combinations. Additionally, similar pairs were defined with property interpretations more frequently than were dissimilar pairs, and dissimilar pairs were defined with relation interpretations more frequently than were similar pairs. Context effects were demonstrated through the finding that target features were found more often in primed than unprimed pairs. The hypothesis of modifier superiority was not confirmed. These findings indicate that the WSM adds to the current understanding of conceptual combination through a reliance on output dominance and the importance of context. Despite these strengths, changes to the WSM may be necessary if future studies fail to support the importance of the modifier over the head noun.

No description available.

The explosion of internet traffic, advent of social media sites such as Facebook and Twitter, and increased availability of digital cameras has saturated life with images and videos. Never before has it been so important to sift quickly through large amounts of digital information. Salient Object Detection (SOD) is a computer vision topic that finds methods to locate important objects in pictures. SOD has proven to be helpful in numerous applications such as image forgery detection and traffic sign recognition. In this thesis, I outline a novel SOD technique to automatically isolate important objects from the background in images.

Conventional surveillance systems present video to a user from more than one camera on a single display. Such a display allows the user to observe different part of the scene, or to observe the same part of the scene from different viewpoints. Each video is usually labeled by a fixed textual annotation displayed under the video segment to identify the image. With the growing number of surveillance cameras set up and the expanse of surveillance area, the conventional split-screen display approach cannot provide intuitive correspondence between the images acquired and the areas under surveillance. Such a system has a number of inherent flaws¡GLower relativity of split videos¡BThe difficulty of tracking new activities¡BLow resolution of surveillance videos¡BThe difficulty of total surveillance¡FIn order to improve the above defects, the ¡§Immersive Surveillance for Total Situational Awareness¡¨ use computer graphic technique to construct 3D model of buildings on the 2D satellite-images, the users can construct the floor platform by defining the information of each floor or building and the position of each camera. This information is combined to construct 3D surveillance scene, and the images acquired by surveillance cameras are pasted into the constructed 3D model to provide intuitively visual presentation. The users could also walk through the scene by a fixed-frequency , self-defined business model to perform a virtual surveillance. Multi-camera Human Tracking on Realtime 3D Immersive Surveillance System based on the ¡§Immersive Surveillance for Total Situational Awareness,¡¨ 1. Salient object detection¡GThe System converts videos to corresponding image sequences and analyze the videos provided by each camera. In order to filter out the foreground pixels, the background model of each image is calculated by pixel-stability-based background update algorithm. 2. Nighttime image fusion¡GUse the fuzzy enhancement method to enhance the dark area in nighttime image, and also maintain the saturation information. Then apply the Salient object detection Algorithm to extract salient objects of the dark area. The system divides fusion results into 3 parts: wall, ceiling, and floor, then pastes them as materials into corresponding parts of 3D scene. 3. Multi-camera human tracking¡GApply connected component labeling to filter out small area and save each block¡¦s infomation. Use RGB-weight percentage information in each block and 5-state status (Enter¡BLeave¡BMatch¡BOcclusion¡BFraction) to draw out the trajectory of each person in every camera¡¦s field of view on the 3D surveillance scene. Finally, fuse every camera together to complete the multi-camera realtime people tracking. Above all, we can track every human in our 3D immersive surveillance system without watching out each of thousand of camera views.

A new hybrid machine having variable reluctance and permanent magnets (PMs) is presented. The machine makes use of the features of a PM machine and variable reluctance machine. The resulting machine is doubly salient and has a structure free of flux reversals. Unlike conventional doubly salient permanent magnet machines (DSPMs), the one proposed in this report is driven by unipolar currents and uses an asymmetric converter which is used to drive switched reluctance machines. The reason to have a new hybrid machine without the drawbacks of conventional flux-reversal-free-stator SRMs and conventional DSPMs is also described. Conventional doubly salient permanent magnet machines which are driven by alternating currents, do not use reluctance torque and have flux reversals in the stator iron. Homopolar flux at the peak flux density lowers hysteresis and eddy-current loss, since the machine's core operates in only one magnetizing quadrant. Due to unbalanced forces in conventional stator-flux-reversal-free machines, their deployment in industrial and end-user applications has been hindered. The presented hybrid machine has balanced radial forces. Therefore, it maintains the advantages of conventional stator-flux-reversal-free machines while shedding its disadvantages. The proposed machine has significantly increased power density and is more electromechanically efficient than its predecessor. A experimental prototype motor has been designed and built. Its static torque characteristics correlated well with predicted data. Experimental operation of the drive under open loop speed control shows the efficiency to be 90.8% under non-ideal driving conditions. In the current energy conscious environment and market, this motor because of its high efficiency has a significant role in reducing the energy consumption in household, industrial and automotive applications requiring electric motors. / Ph. D.

The natural environment will become an increasingly important arena for economic competition in the future. The growing world population and the increased global consumption raise concerns about the sustainability of the current and future use of natural resources. Due to the growing global population the production of agricultural crops and food security is high on the global policy agenda. The purpose of this study has been to investigate the opportunities for a sustainable intensification in the Swedish cereals production and how the Lantmännen cooperative, a company active in the whole cereals value chain can act in order to enable the intensification. The research was conducted as a descriptive case study. Given the comprehensive nature of the purpose of this study, we have studied the Swedish agriculture from a system perspective in order to get the comprehensive understanding of its complexity. The findings presented in this report are based on data from multiple research methods such as interviews with stakeholder in the agriculture, database search, and archive searches. We show in this study that a sustainable intensification of the Swedish cereals production can be realized through utilizing the existing unused farmland registered by the farmer as fallow for a number of consecutive years in a row, which can be described as acreage that is not used for crop production but is still kept in farmable condition. Our study has identified four barriers hindering the intensification; Single payment system, Leasing contract, Logistics and Investment costs. To overcome these barriers we argue that the RO PSS business model is appropriate. By using this model Lantmännen will be able to deliver four different types of values to the farmers. The first value is the intangible value of identity, which let inactive farmers continue to live on their farms while being able to focus on other sources of main income. The second value we name ownerless consumption. This value offers farmers with little capital to start up or expand their production without having to invest in new machinery and equipment. The third value we call safety, which is connected to the barrier of lease contracts identified and the distrust between landowners and tenants. The fourth value, specialization, gives the farmer possibility to leverage overall cereals bulk production by including niche crops. We conclude by arguing that the opportunities for an intensification of the Swedish Cereals production lie in delivering these four values. / Naturen kommer i framtiden bli en alltmer viktig arena för ekonomisk konkurrens. En växande världsbefolkning och en ökad global konsumtion skapar oro kring hållbarheten i vårt användande av naturresurser. På grund av den ökande befolkningsmängden placerar sig frågor såsom produktionen av jordbruksgrödor och livsmedelsförsörjning högt upp på den globala politiska dagordningen. Studiens syfte är att undersöka möjligheter för en hållbar intensifiering av den svenska spannmålsodlingen och hur Lantmännen, ett kooperativt företag som verkar i spannmålets hela värdekedja ska agera för att möjliggöra intensifieringen. Med tanke på den övergripande karaktären av våra frågeställningar har vi studerat det svenska jordbruket ur ett systemperspektiv för att få en övergripande förståelse av dess komplexitet. Forskningen har genomförts i form av en beskrivande fallstudie. Våra resultat baseras på data från flera forskningsmetoder som intervjuer med intressenter inom jordbruket, databassökning, arkivsökning. Vår studie visar att en hållbar intensifiering av den svenska spannmålsodlingen kan realiseras genom att utnyttja befintlig oanvänd jordbruksmark som av jordbrukaren registrerats som träda i flera år i följd. Träda kan beskrivas som jordbruksareal som inte används till växtodling men som fortfarande hålls i odlingsbart skick. För att få denna mark i bruk och därmed möjliggöra en intensifiering krävs det att få bukt med fyra identifierade barriärer inom det svenska jordbrukssystemet; Gårdsstödet, Arrendekontrakt, Logistik och Investeringskostnader. Vi hävdar, baserat på våra resultat och analyser, att affärsmodellkonceptet RO PSS är lämplig att använda då Lantmännen ska lyckas få bukt med barriärerna. Denna modell kommer leverera fyra olika värden till jordbrukarna. Det första värdet har vi döpt till Identitet och är av en immateriell natur som vilket låter inaktiva jordbrukare fortsätta bo kvar på sina gårdar medan de inriktar sig på andra inkomstbaserade verksamheter. Det andra värdet gav vi namnet Ägandefri konsumtion. Detta ger värde till jordbrukaren i form av att jordbrukaren inte behöver låna pengar för höga  investeringskostnader i maskiner och infrastruktur vid uppstart eller expansion av verksamheten. Det tredje värdet kallar vi för Säkerhet, vilket är relaterat till barriären arrendekontrakt där vi har funnet en misstro mellan markägare och arrendatorer. Det fjärde värdet, Specialisering, ger jordbrukaren möjlighet att blanda in specialgrödor i sin produktion, och på så vis skapa en hävstång åt bulkproduktionen. Vår slutsats är att möjligheterna för en intensifiering av den Svenska spannmålsodlingen ligger i att kunna leverera dessa värden.

The central concern of this thesis is to discuss interpretations of learning in educational research. A point of departure is taken in core epistemological and ontological assumptions informing three major approaches to learning: behaviourism, cognitive constructivism and socioculturalism. It is argued that all three perspectives provide important insights into research on learning, but each alone runs the risk of reducing learning and interpretations of learning to single aspects. Specific attention is therefore given to Intentional Analysis, as it has been developed to account for sociocultural aspects that influence learning and individual cognition. It is argued that interpretations of learning processes face challenges, different kinds of holism, underdetermination and the complexity of intentionality, that need to be accounted for in order to make valid interpretations. Interpretation is therefore also discussed in light of philosopher Donald Davidson’s theories of knowledge and interpretation. It is suggested that his theories may provide aspects of an ontological and epistemological stance that can form the basis for interpretations of learning in educational research. A first brief sketch, referred to as ‘epistemic holism’, is thus drawn. The thesis also exemplifies how such a stance can inform empirical research. It provides a first formulation of research strategies – a so-called ‘interpretative holism’. The thesis discusses what such a stance may imply with regard to the nature and location of knowledge and the status of the learning situation. Ascribing meaning to observed behaviour, as it is described in this thesis, implies that an action is always an action under a specific description. Different descriptions may not be contradictory, but if we do not know the learner’s language use, we cannot know whether there is a difference in language or in beliefs. It is argued that the principle of charity and reference to saliency, that is, what appears as the figure for the learner, may help us decide. However, saliency does not only appear as a phenomenon in relation to physical objects and events, but also in the symbolic world, thus requires that the analysis extend beyond the mere transcription of an interview or the description of an observation. Hence, a conclusion to be drawn from this thesis is that the very question of what counts as data in the interpretation of complex learning processes is up for discussion. / <p>At the time of the doctoral defense, the following paper was unpublished and had a status as follows: Paper 4: Manuscript.</p>