An overview of the advantages and constraints of coded pattern projection techniques for autonomous navigation

The absolute necessity of obtaining 3D information of structured and unknown environments in autonomous navigation reduce considerably the set of sensors that can be used. The necessity to know, at each time, the position of the mobile robot with respect to the scene is indispensable. Furthermore, this information must be obtained in the least computing time. Stereo vision is an attractive and widely used method, but, it is rather limited to make fast 3D surface maps, due to the correspondence problem. The spatial and temporal correspondence among images can be alleviated using a method based on structured light. This relationship can be directly found codifying the projected light; then each imaged region of the projected pattern carries the needed information to solve the correspondence problem. We present the most significant techniques, used in recent years, concerning the coded structured light method

The VAD connection : connecting ventricular assist devices to the aorta in silico

Les systèmes d'assistance ventriculaire sont apparus durant la dernière décade comme une approche thérapeutique efficace du traitement de l'insuffisance cardiaque terminale, en particulier dans le contexte de manque de donneurs d'organes. Néanmoins, et ceci malgré les progrès techniques majeurs, les taux de complications restent élevés et sont en partie liés à la configuration géométrique, en particulier le site d'implantation de la cannule de sortie à l'aorte thoracique. Bien que l'anastomose à l'aorte descendante permette une chirurgie moins invasive, les bénéfices de cette technique sont toujours controversés, comparée à la méthode standard de l'aorte ascendante, en raison du risque thrombo-embolique possiblement augmenté et des modifications hémodynamiques induites au niveau de l'arc aortique. Dans ce travail, nous comparons in silico en terme de débit et pression les deux possibilités anastomotiques. Nous développons un réseau de modèles mathématiques unidimensionnels, et l'appliquons à diverses situations cliniques, pour différents stades d'insuffisance cardiaque et de vitesses de rotation de la machine. Les données initiales sont obtenues grâce à un modèle OD (c'est-à-dire qui dépend uniquement du temps mais pas de l'espace) du système cardiovasculaire comprenant une assistance circulatoire, validé avec des données cliniques. Les simulations réalisées montrent que les deux méthodes sont similaires, en terme de débit et courbes de pression, ceci pour tous les cas cliniques étudiés. Ces résultats numériques soutiennent la possibilité d'utiliser la technique d'anastomose à l'aorte thoracique descendante, permettant une chirurgie moins invasive. Sur un plan plus fondamental, le système cardiovasculaire peut être simulé par le biais de multiples modèles de niveau de complexité différents, au prix d'un coût computationnel toujours plus élevé. Nous évaluons les avantages de modèles géométriques à plusieurs échelles (uni- et tridimensionnelle) avec données provenant de patients, comparés à des modèles simplifiés. Les résultats montrent que ces modèles de dimensions hétérogènes apportent un bénéfice important en terme de ressources de calcul, tout en conservant une précision acceptable. En conclusion, ces résultats encourageant montrent la relevance des études numériques dans le domaine médical, tant sur le plan fondamental et la compréhension des mécanismes physiopathologiques, que sur le plan applicatif et le développement de nouvelles thérapeutiques.

Matching stages of heavy-ion collision models

Heavy-ion reactions and other collective dynamical processes are frequently described by different theoretical approaches for the different stages of the process, like initial equilibration stage, intermediate locally equilibrated fluid dynamical stage, and final freeze-out stage. For the last stage, the best known is the Cooper-Frye description used to generate the phase space distribution of emitted, noninteracting particles from a fluid dynamical expansion or explosion, assuming a final ideal gas distribution, or (less frequently) an out-of-equilibrium distribution. In this work we do not want to replace the Cooper-Frye description, but rather clarify the ways of using it and how to choose the parameters of the distribution and, eventually, how to choose the form of the phase space distribution used in the Cooper-Frye formula. Moreover, the Cooper-Frye formula is used in connection with the freeze-out problem, while the discussion of transition between different stages of the collision is applicable to other transitions also. More recently, hadronization and molecular dynamics models have been matched to the end of a fluid dynamical stage to describe hadronization and freeze-out. The stages of the model description can be matched to each other on space-time hypersurfaces (just like through the frequently used freeze-out hypersurface). This work presents a generalized description of how to match the stages of the description of a reaction to each other, extending the methodology used at freeze-out, in simple covariant form which is easily applicable in its simplest version for most applications.

Increasing the efficiency of disparity computation by interpolation

Tässä diplomityössä tutkitaan dispariteettikartan laskennan tehostamista interpoloimalla. Kolmiomittausta käyttämällä stereokuvasta muodostetaan ensin harva dispariteettikartta, jonka jälkeen koko kuvan kattava dispariteettikartta muodostetaan interpoloimalla. Kolmiomittausta varten täytyy tietää samaa reaalimaailman pistettä vastaavat kuvapisteet molemmissa kameroissa. Huolimatta siitä, että vastaavien pisteiden hakualue voidaan pienentää kahdesta ulottuvuudesta yhteen ulottuvuuteen käyttämällä esimerkiksi epipolaarista geometriaa, on laskennallisesti tehokkaampaa määrittää osa dispariteetikartasta interpoloimalla, kuin etsiä vastaavia kuvapisteitä stereokuvista. Myöskin johtuen stereonäköjärjestelmän kameroiden välisestä etäisyydestä, kaikki kuvien pisteet eivät löydy toisesta kuvasta. Näin ollen on mahdotonta määrittää koko kuvan kattavaa dispariteettikartaa pelkästään vastaavista pisteistä. Vastaavien pisteiden etsimiseen tässä työssä käytetään dynaamista ohjelmointia sekä korrelaatiomenetelmää. Reaalimaailman pinnat ovat yleisesti ottaen jatkuvia, joten geometrisessä mielessä on perusteltua approksimoida kuvien esittämiä pintoja interpoloimalla. On myöskin olemassa tieteellistä näyttöä, jonkamukaan ihmisen stereonäkö interpoloi objektien pintoja.

Project for Active and Robot Vision Course

Tässä työssä raportoidaan harjoitustyön kehittäminen ja toteuttaminen Aktiivisen- ja robottinäön kurssille. Harjoitustyössä suunnitellaan ja toteutetaan järjestelmä joka liikuttaa kappaleita robottikäsivarrella kolmiuloitteisessa avaruudessa. Kappaleidenpaikkojen määrittämiseen järjestelmä käyttää digitaalisia kuvia. Tässä työssä esiteltävässä harjoitustyötoteutuksessa käytettiin raja-arvoistusta HSV-väriavaruudessa kappaleiden segmentointiin kuvasta niiden värien perusteella. Segmentoinnin tuloksena saatavaa binäärikuvaa suodatettiin mediaanisuotimella kuvan häiriöiden poistamiseksi. Kappaleen paikkabinäärikuvassa määritettiin nimeämällä yhtenäisiä pikseliryhmiä yhtenäisen alueen nimeämismenetelmällä. Kappaleen paikaksi määritettiin suurimman nimetyn pikseliryhmän paikka. Kappaleiden paikat kuvassa yhdistettiin kolmiuloitteisiin koordinaatteihin kalibroidun kameran avulla. Järjestelmä liikutti kappaleita niiden arvioitujen kolmiuloitteisten paikkojen perusteella.

Preferred Spatial Frequencies for Human Face Processing Are Associated with Optimal Class Discrimination in the Machine

Psychophysical studies suggest that humans preferentially use a narrow band of low spatial frequencies for face recognition. Here we asked whether artificial face recognition systems have an improved recognition performance at the same spatial frequencies as humans. To this end, we estimated recognition performance over a large database of face images by computing three discriminability measures: Fisher Linear Discriminant Analysis, Non-Parametric Discriminant Analysis, and Mutual Information. In order to address frequency dependence, discriminabilities were measured as a function of (filtered) image size. All three measures revealed a maximum at the same image sizes, where the spatial frequency content corresponds to the psychophysical found frequencies. Our results therefore support the notion that the critical band of spatial frequencies for face recognition in humans and machines follows from inherent properties of face images, and that the use of these frequencies is associated with optimal face recognition performance.

Reconstruction of 3D Planar Objects with Partial Occlusion

Kolmiulotteisten kappaleiden rekonstruktio on yksi konenäön haastavimmista ongelmista, koska kappaleiden kolmiulotteisia etäisyyksiä ei voida selvittää yhdestä kaksiulotteisesta kuvasta. Ongelma voidaan ratkaista stereonäön avulla, jossa näkymän kolmiulotteinen rakenne päätellään usean kuvan perusteella. Tämä lähestymistapa mahdollistaa kuitenkin vain rekonstruktion niille kappaleiden osille, jotka näkyvät vähintään kahdessa kuvassa. Piilossa olevien osien rekonstruktio ei ole mahdollista pelkästään stereonäön avulla. Tässä työssä on kehitetty uusi menetelmä osittain piilossa olevien kolmiulotteisten tasomaisten kappaleiden rekonstruktioon. Menetelmän avulla voidaan selvittää hyvällä tarkkuudella tasomaisista pinnoista koostuvan kappaleen muoto ja paikka käyttäen kahta kuvaa kappaleesta. Menetelmä perustuu epipolaarigeometriaan, jonka avulla selvitetään molemmissa kuvissa näkyvät kappaleiden osat. Osittain piilossa olevien piirteiden rekonstruointi suoritetaan käyttämäen stereonäköä sekä tietoa kappaleen rakenteesta. Esitettyä ratkaisua voitaisiin käyttää esimerkiksi kolmiulotteisten kappaleiden visualisointiin, robotin navigointiin tai esineentunnistukseen.

Measurement systems in wall production applications

Tässä tutkimuksessa kehitettiin prototyyppi betonielementin dimension mittaus järjestelmästä. Tämä järjestelmä mahdollistaa kolmiulotteisen kappaleen mittauksen. Tutkimuksessa kehitettiin myös stereonäköön perustuva kappaleen mittaus. Prototyyppiä testailin ja tulokset osoittautuivat luotettaviksi. Tutkimuksessa selvitetään ja vertaillaan myös muita lähestymistapoja ja olemassa olevia järjestelmiä kappaleen kolmiuloitteiseen mittaukseen, joita Suomalaiset yhtiöt käyttävät tällä alalla.

Fatigue resistance of denture teeth

AIMS: There is no standard test to determine the fatigue resistance of denture teeth. With the increasing number of patients with implant-retained dentures the mechanical strength of the denture teeth requires more attention and valid laboratory test set-ups. The purpose of the present study was to determine the fatigue resistance of various denture teeth using a dynamic load testing machine. METHODS: Four denture teeth were used: Bonartic II (Candulor), Physiodens (Vita), SR Phonares II (Ivoclar Vivadent) and Trubyte (Dentsply). For dynamic load testing, first upper molars with a similar shape and cusp inclination were selected. The molar teeth were embedded in cylindrical steel molds with denture base material (ProBase, Ivoclar Vivadent). Dynamic fatigue loading was carried out on the mesio-buccal cusp at a 45° angle using dynamic testing machines and 2,000,000 cycles at 2Hz in water (37°C). Three specimens per group and load were submitted to decreasing load levels (at least 4) until all the three specimens no longer showed any failures. All the specimens were evaluated under a stereo microscope (20× magnification). The number of cycles reached before observing a failure, and its dependence on the load and on the material, has been modeled using a parametric survival regression model with a lognormal distribution. This allowed to estimate the fatigue resistance for a given material as the maximal load for which one would observe less than 1% failure after 2,000,000 cycles. RESULTS: The failure pattern was similar for all denture teeth, showing a large chipping of the loaded mesio-buccal cusp. In our regression model, there were statistically significant differences among the different materials, with SR Phonares II and Bonartic II showing a higher resistance than Physiodens and Trubyte, the fatigue resistance being estimated at around 110N for the former two, and at about 60N for the latter two materials. CONCLUSION: The fatigue resistance may be a useful parameter to assess and to compare the clinical risk of chipping and fracture of denture tooth materials.

Sensor Fusion of Proprioception, Force and Vision in Estimation and Robot

Sensor-based robot control allows manipulation in dynamic environments with uncertainties. Vision is a versatile low-cost sensory modality, but low sample rate, high sensor delay and uncertain measurements limit its usability, especially in strongly dynamic environments. Force is a complementary sensory modality allowing accurate measurements of local object shape when a tooltip is in contact with the object. In multimodal sensor fusion, several sensors measuring different modalities are combined to give a more accurate estimate of the environment. As force and vision are fundamentally different sensory modalities not sharing a common representation, combining the information from these sensors is not straightforward. In this thesis, methods for fusing proprioception, force and vision together are proposed. Making assumptions of object shape and modeling the uncertainties of the sensors, the measurements can be fused together in an extended Kalman filter. The fusion of force and visual measurements makes it possible to estimate the pose of a moving target with an end-effector mounted moving camera at high rate and accuracy. The proposed approach takes the latency of the vision system into account explicitly, to provide high sample rate estimates. The estimates also allow a smooth transition from vision-based motion control to force control. The velocity of the end-effector can be controlled by estimating the distance to the target by vision and determining the velocity profile giving rapid approach and minimal force overshoot. Experiments with a 5-degree-of-freedom parallel hydraulic manipulator and a 6-degree-of-freedom serial manipulator show that integration of several sensor modalities can increase the accuracy of the measurements significantly.

Computational Evaluation of Print Unevenness According to Human Vision

Print quality and the printability of paper are very important attributes when modern printing applications are considered. In prints containing images, high print quality is a basic requirement. Tone unevenness and non uniform glossiness of printed products are the most disturbing factors influencing overall print quality. These defects are caused by non ideal interactions of paper, ink and printing devices in high speed printing processes. Since print quality is a perceptive characteristic, the measurement of unevenness according to human vision is a significant problem. In this thesis, the mottling phenomenon is studied. Mottling is a printing defect characterized by a spotty, non uniform appearance in solid printed areas. Print mottle is usually the result of uneven ink lay down or non uniform ink absorption across the paper surface, especially visible in mid tone imagery or areas of uniform color, such as solids and continuous tone screen builds. By using existing knowledge on visual perception and known methods to quantify print tone variation, a new method for print unevenness evaluation is introduced. The method is compared to previous results in the field and is supported by psychometric experiments. Pilot studies are made to estimate the effect of optical paper characteristics prior to printing, on the unevenness of the printed area after printing. Instrumental methods for print unevenness evaluation have been compared and the results of the comparison indicate that the proposed method produces better results in terms of visual evaluation correspondence. The method has been successfully implemented as ail industrial application and is proved to be a reliable substitute to visual expertise.

Comparison of Local Feature Detectors and Descriptors for Visual Object Categorization

Local features are used in many computer vision tasks including visual object categorization, content-based image retrieval and object recognition to mention a few. Local features are points, blobs or regions in images that are extracted using a local feature detector. To make use of extracted local features the localized interest points are described using a local feature descriptor. A descriptor histogram vector is a compact representation of an image and can be used for searching and matching images in databases. In this thesis the performance of local feature detectors and descriptors is evaluated for object class detection task. Features are extracted from image samples belonging to several object classes. Matching features are then searched using random image pairs of a same class. The goal of this thesis is to find out what are the best detector and descriptor methods for such task in terms of detector repeatability and descriptor matching rate.

Bag-of-Features Approach to Unsupervised Visual Object Categorisation

The large and growing number of digital images is making manual image search laborious. Only a fraction of the images contain metadata that can be used to search for a particular type of image. Thus, the main research question of this thesis is whether it is possible to learn visual object categories directly from images. Computers process images as long lists of pixels that do not have a clear connection to high-level semantics which could be used in the image search. There are various methods introduced in the literature to extract low-level image features and also approaches to connect these low-level features with high-level semantics. One of these approaches is called Bag-of-Features which is studied in the thesis. In the Bag-of-Features approach, the images are described using a visual codebook. The codebook is built from the descriptions of the image patches using clustering. The images are described by matching descriptions of image patches with the visual codebook and computing the number of matches for each code. In this thesis, unsupervised visual object categorisation using the Bag-of-Features approach is studied. The goal is to find groups of similar images, e.g., images that contain an object from the same category. The standard Bag-of-Features approach is improved by using spatial information and visual saliency. It was found that the performance of the visual object categorisation can be improved by using spatial information of local features to verify the matches. However, this process is computationally heavy, and thus, the number of images must be limited in the spatial matching, for example, by using the Bag-of-Features method as in this study. Different approaches for saliency detection are studied and a new method based on the Hessian-Affine local feature detector is proposed. The new method achieves comparable results with current state-of-the-art. The visual object categorisation performance was improved by using foreground segmentation based on saliency information, especially when the background could be considered as clutter.

Learning Grasp Affordances from Vision

Monimutkaisissa ja muuttuvissa ympäristöissä työskentelevät robotit tarvitsevat kykyä manipuloida ja tarttua esineisiin. Tämä työ tutkii robottitarttumisen ja robottitartuntapis-teiden koneoppimisen aiempaa tutkimusta ja nykytilaa. Nykyaikaiset menetelmät käydään läpi, ja Le:n koneoppimiseen pohjautuva luokitin toteutetaan, koska se tarjoaa parhaan onnistumisprosentin tutkituista menetelmistä ja on muokattavissa sopivaksi käytettävissä olevalle robotille. Toteutettu menetelmä käyttää intensititeettikuvaan ja syvyyskuvaan po-hjautuvia ominaisuuksi luokitellakseen potentiaaliset tartuntapisteet. Tämän toteutuksen tulokset esitellään.