One of the oldest biometrics that has been used to uniquely identify a person is their fingerprint. Recent developments in research on fingerprint collection have made it possible to collect fingerprint data from a stand-off digital image. Each of the techniques developed so far have relied on either a very controlled capture environment to ensure only a single fingertip is collected or manual cropping of the image down to the fingertip. The main body of the research focuses on extracting the fingerprint itself. If fingerprint collection via digital image is ever to be fielded in the real world on such devices as smart phones or tablets it will be necessary for the software to automatically detect a single or multiple fingertips in an image and isolate them for extracting the fingerprint. We introduce an automatic fingertip detection algorithm that couples image processing techniques with a machine learning capability to successfully identify varying numbers of fingertips in digital images. Our algorithm proves that while it is difficult to remove all constraints from the capture environment it is achievable with the method we have developed and we can achieve a recall of 69.77% at a precision of 78.95%. This gives us the important capability to detect varying numbers of fingertips in an image and provide a crucial piece in what could be a complete automated fingerprint recognition system.

Un nouveau dispositif tactile capable d'enregistrer et de reproduire des scènes tactiles à l'aide de forces tangentielles a été conçu et réalisé sur la base des limites et des exigences du système somatosensoriel humain. Les micro-déformations tangentielles du doigt coulissant sur la texture sont mesurées par le transducteur dans une bande passante de 500Hz et sont reproduites par le déplacement d'une plaque de verre sous l'action contrôlée d'un moteur électrodynamique critiquement amorti. Dans le but d'évaluer la pertinence des signaux sensoriels pour la reproduction d'une scène tactile, les grandeurs physiques qui influent sur la perception tactile humaine ont été étudiées. En utilisant une méthode d'escalier, il a été démontré que des ondelettes avec différentes combinaisons d'amplitude et de durée sont ressenties égales en intensité. Ces résultats suggèrent qu'il existe des quantités physiques communes - des invariants - pour ces signaux auques le cerveau est sensible, ce qui pourrait se rapporter à une constance perceptuelle dans l'exploration d'aspérités. En analysant les forces de frottement créées par un doigt glissant sur une surface dotées de points braille à des vitesses différents, il a été constaté que, bien que la réponse mécanique instantanée varie dans son ensemble, l'intégrale des forces tangentielles locales au cours d'une période de déformation courte reste constante. Ces enregistrements ont ensuite été classés selon leur vitesse et utilisés comme stimuli dans une tâche de comparaison. Les participants devaient glisser leur index sur une plateforme de verre qui vibrait afin de reproduire les points brailles de hauteurs différentes enregistrés à la même. / A new tactile device able to record and reproduce tactile scenes using tangential forces was designed and realized based on the limits and requirements of our somatosensory system. The tangential micro-deformations of the finger sliding on a textured surface can be measured with 500 Hz-bandwidth and reproduced by vibrating a glass plate under the controlled action of a critically damped electrodynamic actuator. In an effort to identify what sensory cues are relevant to tactile sensations for the reproduction of a scene, the physical quantities that influence tactile perception were studied. Using a staircase method, it was demonstrated that tactile wavelets with different combinations of amplitude and duration could be felt perceptually equal in intensity. These results suggested that there are common physical quantities – invariants – for these signals that the brain is sensitive to, which could relate to a perceptual constancy in asperity exploration. By analyzing the friction forces of a finger exploring braille dots with different pressures and velocities profiles, we found that although the mechanical response at a highly localized stimulus varies as a whole, the integral of the local tangential forces during a short deformation period remained constant. These recordings were then categorized by velocity and used as stimuli in a comparison task in which participants explored virtual dots of different heights at different speeds. While sliding on a glass platform which vibrated tangentially to reproduce a braille dot recorded at the same exploration speed, they were asked to report which of the two stimuli was stronger (or ‘higher’), a task that they could successfull.

No description available.

Sensor technology is rapidly developing and, consequently, the generation of point cloud data is constantly increasing. Since the recent release of PointNet, it is possible to process this unordered 3-dimensional data directly in a neural network. The company TLT Screen AB, which develops cutting-edge tracking technology, seeks to optimize the localization of the fingertips of a hand in a point cloud. To do so, the identification of relevant 3D neural network models for modeling hands and detection of fingertips in various hand orientations is essential. The Hand PointNet processes point clouds of hands directly and generate estimations of fixed points (joints), including fingertips, of the hands. Therefore, this model was selected to optimize the localization of fingertips for TLT Screen AB and forms the subject of this research. The model has advantages over conventional convolutional neural networks (CNN). First of all, in contrast to the 2D CNN, the Hand PointNet can use the full 3-dimensional spatial information. Compared to the 3D CNN, moreover, it avoids unnecessarily voluminous data and enables more efficient learning. The model was trained and evaluated on the public dataset MRSA Hand. In contrast to previously published work, the main object of this investigation is the estimation of only 5 joints, for the fingertips. The behavior of the model with a reduction from the usual 21 to 11 and only 5 joints are examined. It is found that the reduction of joints contributed to an increase in the mean error of the estimated joints. Furthermore, the examination of the distribution of the residuals of the estimate for fingertips is found to be less dense. MC dropout to study the prediction uncertainty for the fingertips has shown that the uncertainty increases when the joints are decreased. Finally, the results show that the uncertainty is greatest for the prediction of the thumb tip. Starting from the tip of the thumb, it is observed that the uncertainty of the estimates decreases with each additional fingertip.

This master’s thesis deals with the issue of scanning human fingerprints, which is currently very topical and represents the most widespread biometric technology. The theoretical part of the work acquaints the reader with basics of dactyloscopy and biometrics and concerns technologies used for fingerprinting, image preprocessing methods and commercially available contactless optical scanners. The practical part is a realisation of a contactless optical scanner based on a Raspberry Pi minicomputer, implementation of preprocessing algorithms in Python and testing of the device from the perspective of image quality.

During dexterous manipulation with the fingertips, forces are applied to objects' surfaces. To achieve grasp stability, these forces must be appropriate given the properties of the objects and the skin of the fingertips, and the nature of the task. It has been demonstrated that tactile sensors in the fingertips provide crucial information about both object properties and mechanical events critical for the control of fingertip forces, while in certain tasks vision may also contribute to predictions of required fingertip actions. This thesis focuses on two specific aspects of the sensory control of manipulation: (i) how individual fingers are controlled for grasp stability when people restrain objects subjected to unpredictable forces tangential to the grasped surfaces, and (ii) how tactile sensors in the fingertips encode direction of fingertip forces and shape of surfaces contacted by the fingertips. When restraining objects with two fingers, subjects adjust the fingertip forces to the local friction at each digit-object interface for grasp stability. This is accomplished primarily by partitioning the tangential force between the digits in a way that reflects the local friction whereas the normal forces at the involved digits are scaled by the average friction and the total load. The neural control mechanisms in this task rely on tactile information pertaining to both the friction at each digit-object interface and the development of tangential load. Moreover, these mechanisms controlled the force application at individual digits while at the same time integrating sensory inputs from all digits involved in the task. Microneurographical recordings in awake humans shows that most SA-I, SA-II and FA-I sensors in the distal phalanx are excited when forces similar to those observed during actual manipulation are applied to the fingertip. Moreover, the direction of the fingertip force influences the impulse rates in most afferents and their responses are broadly tuned to a preferred direction. The preferred direction varies among the afferents and, accordingly, ensembles of afferents can encode the direction of fingertip forces. The local curvature of the object in contact with the fingertip also influenced the impulse rates in most afferents, providing a curvature contrast signals within the afferent populations. Marked interactions were observed in the afferents' responses to object curvature and force direction. Similar findings were obtained for the onset latency in individual afferents. Accordingly, for ensembles of afferents, the order by which individual afferents initially discharge to fingertip events effectively represents parameters of fingertip stimulation. This neural code probably represents the fastest possible code for transmission of parameters of fingertip stimuli to the CNS.

Fingertip amputation injuries are common in all ages. Conservatively treated fingertips can regenerate skin and soft tissues to form a functionally and cosmetically excellent new fingertip. Little is known about this ability that, in humans, is confined to the fingertips. Even less is known about the role of the bacteria that regularly colonize these wounds without negative impact on regeneration and healing. As an alternative to surgery, self-adhesive film dressings are commonly used to establish a wet chamber around the injury. These dressings leak malodorous wound fluid eventually until the wound is dry. Having that into consideration, we have therefore developed a silicone finger cap that forms a mechanically protected, wet chamber around the injury for optimal regeneration conditions. It contains a puncturable reservoir for excess wound fluid, which can be thus routinely analyzed for diagnostic and research purposes. This study protocol explains the first randomized controlled trial (RCT) on the semiocclusive treatment of fingertip amputations in both children and adults comparing traditional film dressings with the novel silicone finger cap. Being the first RCT using 2 medical devices not yet certified for this indication, it will gather valuable information for the understanding of fingertip regeneration and the design of future definitive studies.

Disposer d’un modèle numérique de doigt permettant de simuler le contact de façon réaliste serait un atout pour les domaines d’applications de la réalité virtuelle et de l’aide à la conception adaptée de matériel. De plus, savoir simuler un contact « biofidèle » entre les doigts et un objet permettrait de répondre à des questions de recherche fondamentale concernant la préhension (évaluation de la qualité d’une prise, choix de stratégies de préhension…).Les modèles existants dans la littérature sont variés en termes de propriétés matériaux et de géométrie mais aucun modèle ne s’impose pour prédire de façon satisfaisante le comportement mécanique de la pulpe des doigts. Cette thèse a pour objectif de proposer des lignes directrices pour le développement d’un modèle éléments finis de l’extrémité du doigt orienté vers la simulation de la manipulation fine.Dans un premier temps, la loi de comportement la plus simple possible mais rendant bien compte du comportement hyperélastique des tissus est identifiée. Ensuite, l’étude de modèles géométriques simplifiés permet de proposer un jeu de marqueurs géométriques permettant de construire un modèle idéalisé. Enfin, une campagne expérimentale innovante de chargements mécaniques sur le doigt sous IRM (8 sujets) permet d’enrichir la base de données des essais existants et de valider les hypothèses de modélisation faites en termes de loi de comportement et de géométrie / A fingertip model enabling realistic contact simulations would be an attractive feature in the virtual reality field and could help the design process of new products. Furthermore, such a tool would allow investigating fundamental research questions associated with prehension (assessment of the efficiency of a grasp, choice of a grasping strategy,…).Existent fingertip models exhibit various material properties and geometries but none of them stand out in the prediction of the mechanical behavior of the fingertip.The main topic of this PhD work is to propose general guidelines for the development of fingertip models dedicated to fine manipulation tasks. First a hyperslastic behavior law is identified, being both as simple as possible and enough complex to reproduce the non-linear behavior of the soft tissue. Then, a geometrical study permits to determine a set of geometric markers enabling the development of an idealized geometrical model. Finally, an MRI innovative experimental campaign of fingertip loading tests is performed (on 8 subjects) to expand the existent experimental database and validate the modeling assumptions made concerning the behavior law and the geometrical approach

Exploring surfaces is an essential ability for humans, allowing them to interact with a large variety of objects within their environment. This ability to explore surfaces is also of a major interest in the development of a new generation of humanoid robots, which requires the development of more efficient artificial tactile sensing techniques. The details perceived by statically touching different surfaces of objects not only improve robotic hand performance in force-controlled grasping tasks but also enables the feeling of vibrations on touched surfaces. This thesis presents an extensive experimental study of object surface exploration using biologically-Inspired tactile-enabled robotic fingers. A new multi-modal tactile sensor, embedded in both versions of the robotic fingertips (similar to the human distal phalanx) is capable of measuring the heart rate with a mean absolute error of 1.47 bpm through static explorations of the human skin. A two-phalanx articulated robotic finger with a new miniaturized tactile sensor embedded into the fingertip was developed in order to detect and classify surface textures. This classification is performed by the dynamic exploration of touched object surfaces. Two types of movements were studied: one-dimensional (1D) and two-dimensional (2D) movements. The machine learning techniques - Support Vector Machine (SVM), Multilayer Perceptron (MLP), Random Forest, Extra Trees, and k-Nearest Neighbors (kNN) - were tested in order to find the most efficient one for the classification of the recovered textured surfaces. A 95% precision was achieved when using the Extra Trees technique for the classification of the 1D recovered texture patterns. Experimental results confirmed that the 2D textured surface exploration using a hemispheric tactile-enabled finger was superior to the 1D exploration. Three exploratory velocities were used for the 2D exploration: 30 mm/s, 35 mm/s, and 40 mm/s. The best classification accuracy of the 2D recovered texture patterns was 99.1% and 99.3%, using the SVM classifier, for the two lower exploratory velocities (30 mm/s and 35mm/s), respectively. For the 40 mm/s velocity, the Extra Trees classifier provided a classification accuracy of 99.4%. The results of the experimental research presented in this thesis could be suitable candidates for future development.

Fingertip amputation injuries are common in all ages. Conservatively treated fingertips can regenerate skin and soft tissues to form a functionally and cosmetically excellent new fingertip. Little is known about this ability that, in humans, is confined to the fingertips. Even less is known about the role of the bacteria that regularly colonize these wounds without negative impact on regeneration and healing. As an alternative to surgery, self-adhesive film dressings are commonly used to establish a wet chamber around the injury. These dressings leak malodorous wound fluid eventually until the wound is dry. Having that into consideration, we have therefore developed a silicone finger cap that forms a mechanically protected, wet chamber around the injury for optimal regeneration conditions. It contains a puncturable reservoir for excess wound fluid, which can be thus routinely analyzed for diagnostic and research purposes. This study protocol explains the first randomized controlled trial (RCT) on the semiocclusive treatment of fingertip amputations in both children and adults comparing traditional film dressings with the novel silicone finger cap. Being the first RCT using 2 medical devices not yet certified for this indication, it will gather valuable information for the understanding of fingertip regeneration and the design of future definitive studies.