Digital surveillance based on video CODEC System-on-a-Chip (SoC) platforms

Today, most conventional surveillance networks are based on analog system, which has a lot of constraints like manpower and high-bandwidth requirements. It becomes the barrier for today's surveillance network development. This dissertation describes a digital surveillance network architecture based on the H.264 coding/decoding (CODEC) System-on-a-Chip (SoC) platform. The proposed digital surveillance network architecture includes three major layers: software layer, hardware layer, and the network layer. The following outlines the contributions to the proposed digital surveillance network architecture. (1) We implement an object recognition system and an object categorization system on the software layer by applying several Digital Image Processing (DIP) algorithms. (2) For better compression ratio and higher video quality transfer, we implement two new modules on the hardware layer of the H.264 CODEC core, i.e., the background elimination module and the Directional Discrete Cosine Transform (DDCT) module. (3) Furthermore, we introduce a Digital Signal Processor (DSP) sub-system on the main bus of H.264 SoC platforms as the major hardware support system for our software architecture. Thus we combine the software and hardware platforms to be an intelligent surveillance node. Lab results show that the proposed surveillance node can dramatically save the network resources like bandwidth and storage capacity.

Color-based surface reflectance separation for scene illumination estimation and rendering

Given the importance of color processing in computer vision and computer graphics, estimating and rendering illumination spectral reflectance of image scenes is important to advance the capability of a large class of applications such as scene reconstruction, rendering, surface segmentation, object recognition, and reflectance estimation. Consequently, this dissertation proposes effective methods for reflection components separation and rendering in single scene images. Based on the dichromatic reflectance model, a novel decomposition technique, named the Mean-Shift Decomposition (MSD) method, is introduced to separate the specular from diffuse reflectance components. This technique provides a direct access to surface shape information through diffuse shading pixel isolation. More importantly, this process does not require any local color segmentation process, which differs from the traditional methods that operate by aggregating color information along each image plane. ^ Exploiting the merits of the MSD method, a scene illumination rendering technique is designed to estimate the relative contributing specular reflectance attributes of a scene image. The image feature subset targeted provides a direct access to the surface illumination information, while a newly introduced efficient rendering method reshapes the dynamic range distribution of the specular reflectance components over each image color channel. This image enhancement technique renders the scene illumination reflection effectively without altering the scene’s surface diffuse attributes contributing to realistic rendering effects. ^ As an ancillary contribution, an effective color constancy algorithm based on the dichromatic reflectance model was also developed. This algorithm selects image highlights in order to extract the prominent surface reflectance that reproduces the exact illumination chromaticity. This evaluation is presented using a novel voting scheme technique based on histogram analysis. ^ In each of the three main contributions, empirical evaluations were performed on synthetic and real-world image scenes taken from three different color image datasets. The experimental results show over 90% accuracy in illumination estimation contributing to near real world illumination rendering effects. ^

Estrogen and Lithium: Facilitating Factors Involved in Brain Cell Signaling Pathways

Learning and memory in adult females decline during menopause and estrogen replacement therapy is commonly prescribed during menopause. Post-menopausal women tend to suffer from depression and are prescribed antidepressants – in addition to hormone therapy. Estrogen replacement therapy is a topic that engenders debate since several studies contradict its efficacy as a palliative therapy for cognitive decline and neurodegenerative diseases. Signaling transduction pathways can alter brain cell activity, survival, and morphology by facilitating transcription factor DNA binding and protein production. The steroidal hormone estrogen and the anti-depressant drug lithium interact through these signaling transduction pathways facilitating transcription factor activation. The paucity of data on how combined hormones and antidepressants interact in regulating gene expression led me to hypothesize that in primary mixed brain cell cultures, combined 17beta-estradiol (E2) and lithium chloride (LiCl) (E2/LiCl) will alter genetic expression of markers involved in synaptic plasticity and neuroprotection. Results from these studies indicated that a 48 h treatment of E2/LiCl reduced glutamate receptor subunit genetic expression, but increased neurotrophic factor and estrogen receptor genetic expression. Combined treatment also failed to protect brain cell cultures from glutamate excitotoxicity. If lithium facilitates protein signaling pathways mediated by estrogen, can lithium alone serve as a palliative treatment for post-menopause? This question led me to hypothesize that in estrogen-deficient mice, lithium alone will increase episodic memory (tested via object recognition), and enhance expression in the brain of factors involved in anti-apoptosis, learning and memory. I used bilaterally ovariectomized (bOVX) C57BL/6J mice treated with LiCl for one month. Results indicated that LiCl-treated bOVX mice increased performance in object recognition compared with non-treated bOVX. Increased performance in LiCl-treated bOVX mice coincided with augmented genetic and protein expression in the brain. Understanding the molecular pathways of estrogen will assist in identifying a palliative therapy for menopause-related dementia, and lithium may serve this purpose by acting as a selective estrogen-mediated signaling modulator.

Efeitos da privação do sono em tarefas cognitivas

Learning and memory are important mechanism for species, since its allows to recognize conspecifics, routes and food place. Sleep is one of behaviors known by facilitate learning, it is a widespread phenomenon, present in most of vertebrates lives and highly investigated in many aspects. It is known that sleep deprivation modifies physiologic behavioral processes in animals, however, sleep function in organism is still debatable. Hypothesis range from energy conservation to memory consolidation, with different roles in animal’s evolution. The zebrafish (Danio rerio) emerg e in the last years as vertebrate model in genetics and developmental biology and quickly become popular in behavioral studies, as learning and memory. Despite the fact that zebrafish is a diurnal animal and have well characterized sleep behavior, zebrafish fish still has advantages due to its small size and low cost of maintenance, whichestablishes this species as interesting model for research on sleep. In this study we aimed to analyze the effects of partial and total sleep deprivation on learning acquisition, as well the concomitant administration of alcohol and melatonin. For this, the research was divided in three phases, each one with a different kind of conditioning: (1) object Recognition, (2) avoidance conditioning and (3) appetitive conditioning. The results showed the fish partially sleep deprived and totally sleep deprived + et hanol could perform the tasks just like the control group, however, fish totally sleep deprived and totally sleep deprived + melatonin showed impairments in attention and memory during the tests. Our results suggest that only one night of sleep deprivation is enough to harm the zebrafish performance in cognitive tasks. In addition, ethanol exposure on the night previously the test seems to suppress the negative effects of sleep deprivation, while the melatonin treatment seems not to be enough to promote sleep state, at least on the protocol applied here.

Efeitos da privação do sono em tarefas cognitivas

Learning and memory are important mechanism for species, since its allows to recognize conspecifics, routes and food place. Sleep is one of behaviors known by facilitate learning, it is a widespread phenomenon, present in most of vertebrates lives and highly investigated in many aspects. It is known that sleep deprivation modifies physiologic behavioral processes in animals, however, sleep function in organism is still debatable. Hypothesis range from energy conservation to memory consolidation, with different roles in animal’s evolution. The zebrafish (Danio rerio) emerg e in the last years as vertebrate model in genetics and developmental biology and quickly become popular in behavioral studies, as learning and memory. Despite the fact that zebrafish is a diurnal animal and have well characterized sleep behavior, zebrafish fish still has advantages due to its small size and low cost of maintenance, whichestablishes this species as interesting model for research on sleep. In this study we aimed to analyze the effects of partial and total sleep deprivation on learning acquisition, as well the concomitant administration of alcohol and melatonin. For this, the research was divided in three phases, each one with a different kind of conditioning: (1) object Recognition, (2) avoidance conditioning and (3) appetitive conditioning. The results showed the fish partially sleep deprived and totally sleep deprived + et hanol could perform the tasks just like the control group, however, fish totally sleep deprived and totally sleep deprived + melatonin showed impairments in attention and memory during the tests. Our results suggest that only one night of sleep deprivation is enough to harm the zebrafish performance in cognitive tasks. In addition, ethanol exposure on the night previously the test seems to suppress the negative effects of sleep deprivation, while the melatonin treatment seems not to be enough to promote sleep state, at least on the protocol applied here.

Progettazione e implementazione di una applicazione didattica interattiva per il riconoscimento di oggetti basata sull'algoritmo SIFT

Nell'elaborato viene introdotto l'ambito della Computer Vision e come l'algoritmo SIFT si inserisce nel suo panorama. Viene inoltre descritto SIFT stesso, le varie fasi di cui si compone e un'applicazione al problema dell'object recognition. Infine viene presentata un'implementazione di SIFT in linguaggio Python creata per ottenere un'applicazione didattica interattiva e vengono mostrati esempi di questa applicazione.

Adaptive Brain-Computer Interface Systems For Communication in People with Severe Neuromuscular Disabilities

Brain-computer interfaces (BCI) have the potential to restore communication or control abilities in individuals with severe neuromuscular limitations, such as those with amyotrophic lateral sclerosis (ALS). The role of a BCI is to extract and decode relevant information that conveys a user's intent directly from brain electro-physiological signals and translate this information into executable commands to control external devices. However, the BCI decision-making process is error-prone due to noisy electro-physiological data, representing the classic problem of efficiently transmitting and receiving information via a noisy communication channel.

This research focuses on P300-based BCIs which rely predominantly on event-related potentials (ERP) that are elicited as a function of a user's uncertainty regarding stimulus events, in either an acoustic or a visual oddball recognition task. The P300-based BCI system enables users to communicate messages from a set of choices by selecting a target character or icon that conveys a desired intent or action. P300-based BCIs have been widely researched as a communication alternative, especially in individuals with ALS who represent a target BCI user population. For the P300-based BCI, repeated data measurements are required to enhance the low signal-to-noise ratio of the elicited ERPs embedded in electroencephalography (EEG) data, in order to improve the accuracy of the target character estimation process. As a result, BCIs have relatively slower speeds when compared to other commercial assistive communication devices, and this limits BCI adoption by their target user population. The goal of this research is to develop algorithms that take into account the physical limitations of the target BCI population to improve the efficiency of ERP-based spellers for real-world communication.

In this work, it is hypothesised that building adaptive capabilities into the BCI framework can potentially give the BCI system the flexibility to improve performance by adjusting system parameters in response to changing user inputs. The research in this work addresses three potential areas for improvement within the P300 speller framework: information optimisation, target character estimation and error correction. The visual interface and its operation control the method by which the ERPs are elicited through the presentation of stimulus events. The parameters of the stimulus presentation paradigm can be modified to modulate and enhance the elicited ERPs. A new stimulus presentation paradigm is developed in order to maximise the information content that is presented to the user by tuning stimulus paradigm parameters to positively affect performance. Internally, the BCI system determines the amount of data to collect and the method by which these data are processed to estimate the user's target character. Algorithms that exploit language information are developed to enhance the target character estimation process and to correct erroneous BCI selections. In addition, a new model-based method to predict BCI performance is developed, an approach which is independent of stimulus presentation paradigm and accounts for dynamic data collection. The studies presented in this work provide evidence that the proposed methods for incorporating adaptive strategies in the three areas have the potential to significantly improve BCI communication rates, and the proposed method for predicting BCI performance provides a reliable means to pre-assess BCI performance without extensive online testing.

Using Image Processing Methods to Improve the Detection of Buried Explosive Threats in GPR Data

Current state of the art techniques for landmine detection in ground penetrating radar (GPR) utilize statistical methods to identify characteristics of a landmine response. This research makes use of 2-D slices of data in which subsurface landmine responses have hyperbolic shapes. Various methods from the field of visual image processing are adapted to the 2-D GPR data, producing superior landmine detection results. This research goes on to develop a physics-based GPR augmentation method motivated by current advances in visual object detection. This GPR specific augmentation is used to mitigate issues caused by insufficient training sets. This work shows that augmentation improves detection performance under training conditions that are normally very difficult. Finally, this work introduces the use of convolutional neural networks as a method to learn feature extraction parameters. These learned convolutional features outperform hand-designed features in GPR detection tasks. This work presents a number of methods, both borrowed from and motivated by the substantial work in visual image processing. The methods developed and presented in this work show an improvement in overall detection performance and introduce a method to improve the robustness of statistical classification.

Cognitive Component Skills of Reading Comprehension in Developing Readers

Die Fähigkeit, geschriebene Texte zu verstehen, d.h. eine kohärente mentale Repräsentation von Textinhalten zu erstellen, ist eine notwendige Voraussetzung für eine erfolgreiche schulische und außerschulische Entwicklung. Es ist daher ein zentrales Anliegen des Bildungssystems Leseschwierigkeiten frühzeitig zu diagnostizieren und mithilfe zielgerichteter Interventionsprogramme zu fördern. Dies erfordert ein umfassendes Wissen über die kognitiven Teilprozesse, die dem Leseverstehen zugrunde liegen, ihre Zusammenhänge und ihre Entwicklung. Die vorliegende Dissertation soll zu einem umfassenden Verständnis über das Leseverstehen beitragen, indem sie eine Auswahl offener Fragestellungen experimentell untersucht. Studie 1 untersucht inwieweit phonologische Rekodier- und orthographische Dekodierfertigkeiten zum Satz- und Textverstehen beitragen und wie sich beide Fertigkeiten bei deutschen Grundschüler(inne)n von der 2. bis zur 4. Klasse entwickeln. Die Ergebnisse legen nahe, dass beide Fertigkeiten signifikante und eigenständige Beiträge zum Leseverstehen leisten und dass sich ihr relativer Beitrag über die Klassenstufen hinweg nicht verändert. Darüber hinaus zeigt sich, dass bereits deutsche Zweitklässler(innen) den Großteil geschriebener Wörter in altersgerechten Texten über orthographische Vergleichsprozesse erkennen. Nichtsdestotrotz nutzen deutsche Grundschulkinder offenbar kontinuierlich phonologische Informationen, um die visuelle Worterkennung zu optimieren. Studie 2 erweitert die bisherige empirische Forschung zu einem der bekanntesten Modelle des Leseverstehens—der Simple View of Reading (SVR, Gough & Tunmer, 1986). Die Studie überprüft die SVR (Reading comprehension = Decoding x Comprehension) mithilfe optimierter und methodisch stringenter Maße der Modellkonstituenten und überprüft ihre Generalisierbarkeit für deutsche Dritt- und Viertklässler(innen). Studie 2 zeigt, dass die SVR einer methodisch stringenten Überprüfung nicht standhält und nicht ohne Weiteres auf deutsche Dritt- und Viertklässler(innen) generalisiert werden kann. Es wurden nur schwache Belege für eine multiplikative Verknüpfung von Dekodier- (D) und Hörverstehensfertigkeiten (C) gefunden. Der Umstand, dass ein beachtlicher Teil der Varianz im Leseverstehen (R) nicht durch D und C aufgeklärt werden konnte, deutet darauf hin, dass das Modell nicht vollständig ist und ggf. durch weitere Komponenten ergänzt werden muss. Studie 3 untersucht die Verarbeitung positiv-kausaler und negativ-kausaler Kohärenzrelationen bei deutschen Erst- bis Viertklässler(inne)n und Erwachsenen im Lese- und Hörverstehen. In Übereinstimmung mit dem Cumulative Cognitive Complexity-Ansatz (Evers-Vermeul & Sanders, 2009; Spooren & Sanders, 2008) zeigt Studie 3, dass die Verarbeitung negativ-kausaler Kohärenzrelationen und Konnektoren kognitiv aufwändiger ist als die Verarbeitung positiv-kausaler Relationen. Darüber hinaus entwickelt sich das Verstehen beider Kohärenzrelationen noch über die Grundschulzeit hinweg und ist für negativ-kausale Relationen am Ende der vierten Klasse noch nicht abgeschlossen. Studie 4 zeigt und diskutiert die Nützlichkeit prozess-orientierter Lesetests wie ProDi- L (Richter et al., in press), die individuelle Unterschiede in den kognitiven Teilfertigkeiten des Leseverstehens selektiv erfassen. Hierzu wird exemplarisch die Konstruktvalidität des ProDi-L-Subtests ‚Syntaktische Integration’ nachgewiesen. Mittels explanatorischer Item- Repsonse-Modelle wird gezeigt, dass der Test Fertigkeiten syntaktischer Integration separat erfasst und Kinder mit defizitären syntaktischen Fertigkeiten identifizieren kann. Die berichteten Befunde tragen zu einem umfassenden Verständnis der kognitiven Teilfertigkeiten des Leseverstehens bei, das für eine optimale Gestaltung des Leseunterrichts, für das Erstellen von Lernmaterialien, Leseinstruktionen und Lehrbüchern unerlässlich ist. Darüber hinaus stellt es die Grundlage für eine sinnvolle Diagnose individueller Leseschwierigkeiten und für die Konzeption adaptiver und zielgerichteter Interventionsprogramme zur Förderung des Leseverstehens bei schwachen Leser(inne)n dar.

Mechanisms underlying activation of neural stem cells in the adult central nervous system

À la fin du 19e siècle, Dr. Ramón y Cajal, un pionnier scientifique, a découvert les éléments cellulaires individuels, appelés neurones, composant le système nerveux. Il a également remarqué la complexité de ce système et a mentionné l’impossibilité de ces nouveaux neurones à être intégrés dans le système nerveux adulte. Une de ses citations reconnues : “Dans les centres adultes, les chemins nerveux sont fixes, terminés, immuables. Tout doit mourir, rien ne peut être régénérer” est représentative du dogme de l’époque (Ramón y Cajal 1928). D’importantes études effectuées dans les années 1960-1970 suggèrent un point de vue différent. Il a été démontré que les nouveaux neurones peuvent être générés à l’âge adulte, mais cette découverte a créé un scepticisme omniprésent au sein de la communauté scientifique. Il a fallu 30 ans pour que le concept de neurogenèse adulte soit largement accepté. Cette découverte, en plus de nombreuses avancées techniques, a ouvert la porte à de nouvelles cibles thérapeutiques potentielles pour les maladies neurodégénératives. Les cellules souches neurales (CSNs) adultes résident principalement dans deux niches du cerveau : la zone sous-ventriculaire des ventricules latéraux et le gyrus dentelé de l’hippocampe. En condition physiologique, le niveau de neurogenèse est relativement élevé dans la zone sous-ventriculaire contrairement à l’hippocampe où certaines étapes sont limitantes. En revanche, la moelle épinière est plutôt définie comme un environnement en quiescence. Une des principales questions qui a été soulevée suite à ces découvertes est : comment peut-on activer les CSNs adultes afin d’augmenter les niveaux de neurogenèse ? Dans l’hippocampe, la capacité de l’environnement enrichi (incluant la stimulation cognitive, l’exercice et les interactions sociales) à promouvoir la neurogenèse hippocampale a déjà été démontrée. La plasticité de cette région est importante, car elle peut jouer un rôle clé dans la récupération de déficits au niveau de la mémoire et l’apprentissage. Dans la moelle épinière, des études effectuées in vitro ont démontré que les cellules épendymaires situées autour du canal central ont des capacités d’auto-renouvellement et de multipotence (neurones, astrocytes, oligodendrocytes). Il est intéressant de noter qu’in vivo, suite à une lésion de la moelle épinière, les cellules épendymaires sont activées, peuvent s’auto-renouveller, mais peuvent seulement ii donner naissance à des cellules de type gliale (astrocytes et oligodendrocytes). Cette nouvelle fonction post-lésion démontre que la plasticité est encore possible dans un environnement en quiescence et peut être exploité afin de développer des stratégies de réparation endogènes dans la moelle épinière. Les CSNs adultes jouent un rôle important dans le maintien des fonctions physiologiques du cerveau sain et dans la réparation neuronale suite à une lésion. Cependant, il y a peu de données sur les mécanismes qui permettent l'activation des CSNs en quiescence permettant de maintenir ces fonctions. L'objectif général est d'élucider les mécanismes sous-jacents à l'activation des CSNs dans le système nerveux central adulte. Pour répondre à cet objectif, nous avons mis en place deux approches complémentaires chez les souris adultes : 1) L'activation des CSNs hippocampales par l'environnement enrichi (EE) et 2) l'activation des CSNs de la moelle épinière par la neuroinflammation suite à une lésion. De plus, 3) afin d’obtenir plus d’information sur les mécanismes moléculaires de ces modèles, nous utiliserons des approches transcriptomiques afin d’ouvrir de nouvelles perspectives. Le premier projet consiste à établir de nouveaux mécanismes cellulaires et moléculaires à travers lesquels l’environnement enrichi module la plasticité du cerveau adulte. Nous avons tout d’abord évalué la contribution de chacune des composantes de l’environnement enrichi à la neurogenèse hippocampale (Chapitre II). L’exercice volontaire promeut la neurogenèse, tandis que le contexte social augmente l’activation neuronale. Par la suite, nous avons déterminé l’effet de ces composantes sur les performances comportementales et sur le transcriptome à l’aide d’un labyrinthe radial à huit bras afin d’évaluer la mémoire spatiale et un test de reconnaissante d’objets nouveaux ainsi qu’un RNA-Seq, respectivement (Chapitre III). Les coureurs ont démontré une mémoire spatiale de rappel à court-terme plus forte, tandis que les souris exposées aux interactions sociales ont eu une plus grande flexibilité cognitive à abandonner leurs anciens souvenirs. Étonnamment, l’analyse du RNA-Seq a permis d’identifier des différences claires dans l’expression des transcripts entre les coureurs de courte et longue distance, en plus des souris sociales (dans l’environnement complexe). iii Le second projet consiste à découvrir comment les cellules épendymaires acquièrent les propriétés des CSNs in vitro ou la multipotence suite aux lésions in vivo (Chapitre IV). Une analyse du RNA-Seq a révélé que le transforming growth factor-β1 (TGF-β1) agit comme un régulateur, en amont des changements significatifs suite à une lésion de la moelle épinière. Nous avons alors confirmé la présence de cette cytokine suite à la lésion et caractérisé son rôle sur la prolifération, différentiation, et survie des cellules initiatrices de neurosphères de la moelle épinière. Nos résultats suggèrent que TGF-β1 régule l’acquisition et l’expression des propriétés de cellules souches sur les cellules épendymaires provenant de la moelle épinière.

Analysis of Local Search Methods for 3D Data

In questa tesi sono stati analizzati alcuni metodi di ricerca per dati 3D. Viene illustrata una panoramica generale sul campo della Computer Vision, sullo stato dell’arte dei sensori per l’acquisizione e su alcuni dei formati utilizzati per la descrizione di dati 3D. In seguito è stato fatto un approfondimento sulla 3D Object Recognition dove, oltre ad essere descritto l’intero processo di matching tra Local Features, è stata fatta una focalizzazione sulla fase di detection dei punti salienti. In particolare è stato analizzato un Learned Keypoint detector, basato su tecniche di apprendimento di machine learning. Quest ultimo viene illustrato con l’implementazione di due algoritmi di ricerca di vicini: uno esauriente (K-d tree) e uno approssimato (Radial Search). Sono state riportate infine alcune valutazioni sperimentali in termini di efficienza e velocità del detector implementato con diversi metodi di ricerca, mostrando l’effettivo miglioramento di performance senza una considerabile perdita di accuratezza con la ricerca approssimata.

Tinta Barroca: identidade e comunicação para uma marca de promoção de eventos hoteleiros

A ilustração aplicada ao branding resulta de um modo reflexivo por parte do autor. Esse modo é, por si só, o papel do ilustrador como designer gráfico. A identidade de uma marca nasce da sua história, contexto e sensações, as quais o autor adquire e transmite, segundo as suas vivências, de modo a responder às necessidades das pessoas que o rodeiam. O desenvolvimento de uma marca é um longo processo de análise e reflexão, contínuo e exigente. Aplicando a ilustração a este meio, como objeto visual principal, a língua deixa de ser um entrave e a identidade passa a ser comunicada aos olhos e memória de qualquer um, de forma imediata e eficaz. Conceptualmente, a Tinta Barroca absorve estes princípios, transformando-se numa marca de eventos culturais, embora bastante focada em eventos que podem abranger jantares bem portugueses ou provas de vinho. O projeto foi desenvolvido à base do experimentalismo. Todas as ilustrações da marca foram, numa primeira fase, produzidas manualmente e posteriormente tratadas digitalmente, testando diferentes formas, texturas e materiais. A excessividade ilustrativa é o ponto de partida para comunicar as ideologias da Tinta Barroca, baseando-se no barroquismo, erotismo e nos prazeres da vida. A identidade gráfica da marca misturase com uma decoração já pré-definida: uma mesa bem preenchida e recheada de flores, frutos, vinho e comidas divinais, que se aproximam, pelo excesso, dos princípios do barroco.

Mechanisms underlying activation of neural stem cells in the adult central nervous system

À la fin du 19e siècle, Dr. Ramón y Cajal, un pionnier scientifique, a découvert les éléments cellulaires individuels, appelés neurones, composant le système nerveux. Il a également remarqué la complexité de ce système et a mentionné l’impossibilité de ces nouveaux neurones à être intégrés dans le système nerveux adulte. Une de ses citations reconnues : “Dans les centres adultes, les chemins nerveux sont fixes, terminés, immuables. Tout doit mourir, rien ne peut être régénérer” est représentative du dogme de l’époque (Ramón y Cajal 1928). D’importantes études effectuées dans les années 1960-1970 suggèrent un point de vue différent. Il a été démontré que les nouveaux neurones peuvent être générés à l’âge adulte, mais cette découverte a créé un scepticisme omniprésent au sein de la communauté scientifique. Il a fallu 30 ans pour que le concept de neurogenèse adulte soit largement accepté. Cette découverte, en plus de nombreuses avancées techniques, a ouvert la porte à de nouvelles cibles thérapeutiques potentielles pour les maladies neurodégénératives. Les cellules souches neurales (CSNs) adultes résident principalement dans deux niches du cerveau : la zone sous-ventriculaire des ventricules latéraux et le gyrus dentelé de l’hippocampe. En condition physiologique, le niveau de neurogenèse est relativement élevé dans la zone sous-ventriculaire contrairement à l’hippocampe où certaines étapes sont limitantes. En revanche, la moelle épinière est plutôt définie comme un environnement en quiescence. Une des principales questions qui a été soulevée suite à ces découvertes est : comment peut-on activer les CSNs adultes afin d’augmenter les niveaux de neurogenèse ? Dans l’hippocampe, la capacité de l’environnement enrichi (incluant la stimulation cognitive, l’exercice et les interactions sociales) à promouvoir la neurogenèse hippocampale a déjà été démontrée. La plasticité de cette région est importante, car elle peut jouer un rôle clé dans la récupération de déficits au niveau de la mémoire et l’apprentissage. Dans la moelle épinière, des études effectuées in vitro ont démontré que les cellules épendymaires situées autour du canal central ont des capacités d’auto-renouvellement et de multipotence (neurones, astrocytes, oligodendrocytes). Il est intéressant de noter qu’in vivo, suite à une lésion de la moelle épinière, les cellules épendymaires sont activées, peuvent s’auto-renouveller, mais peuvent seulement ii donner naissance à des cellules de type gliale (astrocytes et oligodendrocytes). Cette nouvelle fonction post-lésion démontre que la plasticité est encore possible dans un environnement en quiescence et peut être exploité afin de développer des stratégies de réparation endogènes dans la moelle épinière. Les CSNs adultes jouent un rôle important dans le maintien des fonctions physiologiques du cerveau sain et dans la réparation neuronale suite à une lésion. Cependant, il y a peu de données sur les mécanismes qui permettent l'activation des CSNs en quiescence permettant de maintenir ces fonctions. L'objectif général est d'élucider les mécanismes sous-jacents à l'activation des CSNs dans le système nerveux central adulte. Pour répondre à cet objectif, nous avons mis en place deux approches complémentaires chez les souris adultes : 1) L'activation des CSNs hippocampales par l'environnement enrichi (EE) et 2) l'activation des CSNs de la moelle épinière par la neuroinflammation suite à une lésion. De plus, 3) afin d’obtenir plus d’information sur les mécanismes moléculaires de ces modèles, nous utiliserons des approches transcriptomiques afin d’ouvrir de nouvelles perspectives. Le premier projet consiste à établir de nouveaux mécanismes cellulaires et moléculaires à travers lesquels l’environnement enrichi module la plasticité du cerveau adulte. Nous avons tout d’abord évalué la contribution de chacune des composantes de l’environnement enrichi à la neurogenèse hippocampale (Chapitre II). L’exercice volontaire promeut la neurogenèse, tandis que le contexte social augmente l’activation neuronale. Par la suite, nous avons déterminé l’effet de ces composantes sur les performances comportementales et sur le transcriptome à l’aide d’un labyrinthe radial à huit bras afin d’évaluer la mémoire spatiale et un test de reconnaissante d’objets nouveaux ainsi qu’un RNA-Seq, respectivement (Chapitre III). Les coureurs ont démontré une mémoire spatiale de rappel à court-terme plus forte, tandis que les souris exposées aux interactions sociales ont eu une plus grande flexibilité cognitive à abandonner leurs anciens souvenirs. Étonnamment, l’analyse du RNA-Seq a permis d’identifier des différences claires dans l’expression des transcripts entre les coureurs de courte et longue distance, en plus des souris sociales (dans l’environnement complexe). iii Le second projet consiste à découvrir comment les cellules épendymaires acquièrent les propriétés des CSNs in vitro ou la multipotence suite aux lésions in vivo (Chapitre IV). Une analyse du RNA-Seq a révélé que le transforming growth factor-β1 (TGF-β1) agit comme un régulateur, en amont des changements significatifs suite à une lésion de la moelle épinière. Nous avons alors confirmé la présence de cette cytokine suite à la lésion et caractérisé son rôle sur la prolifération, différentiation, et survie des cellules initiatrices de neurosphères de la moelle épinière. Nos résultats suggèrent que TGF-β1 régule l’acquisition et l’expression des propriétés de cellules souches sur les cellules épendymaires provenant de la moelle épinière.

Generating Discriminative Objective Proposals via Submodular Ranking

Object recognition has long been a core problem in computer vision. To improve object spatial support and speed up object localization for object recognition, generating high-quality category-independent object proposals as the input for object recognition system has drawn attention recently. Given an image, we generate a limited number of high-quality and category-independent object proposals in advance and used as inputs for many computer vision tasks. We present an efficient dictionary-based model for image classification task. We further extend the work to a discriminative dictionary learning method for tensor sparse coding. In the first part, a multi-scale greedy-based object proposal generation approach is presented. Based on the multi-scale nature of objects in images, our approach is built on top of a hierarchical segmentation. We first identify the representative and diverse exemplar clusters within each scale. Object proposals are obtained by selecting a subset from the multi-scale segment pool via maximizing a submodular objective function, which consists of a weighted coverage term, a single-scale diversity term and a multi-scale reward term. The weighted coverage term forces the selected set of object proposals to be representative and compact; the single-scale diversity term encourages choosing segments from different exemplar clusters so that they will cover as many object patterns as possible; the multi-scale reward term encourages the selected proposals to be discriminative and selected from multiple layers generated by the hierarchical image segmentation. The experimental results on the Berkeley Segmentation Dataset and PASCAL VOC2012 segmentation dataset demonstrate the accuracy and efficiency of our object proposal model. Additionally, we validate our object proposals in simultaneous segmentation and detection and outperform the state-of-art performance. To classify the object in the image, we design a discriminative, structural low-rank framework for image classification. We use a supervised learning method to construct a discriminative and reconstructive dictionary. By introducing an ideal regularization term, we perform low-rank matrix recovery for contaminated training data from all categories simultaneously without losing structural information. A discriminative low-rank representation for images with respect to the constructed dictionary is obtained. With semantic structure information and strong identification capability, this representation is good for classification tasks even using a simple linear multi-classifier.

Behavioral traits predicting cocaine-conditioned place reference in mice: role of anxiety adn the basolateral amygdala

Aims. The individual susceptibility to cocaine addiction, a factor of interest in the understanding and prevention of this disorder, may be predicted by certain behavioral traits. However, these are not usually taken into account in research, making it difficult to identify whether they are a cause or a consequence of drug use. Methods. Male C57BL/6J mice underwent a battery of behavioral tests (elevated plus maze, hole-board, novelty preference in the Y maze, episodic-like object recognition memory and forced swimming test), followed by a cocaine-conditioned place preference (CPP) training to assess the reinforcing effect of the drug. In a second study, we aimed to determine the existence of neurobiological differences between the mice expressing high or low CPP by studying the number of neurons in certain addiction-related structures: the medial prefrontal cortex, the basolateral amygdala and the ventral tegmental area. Results. Anxiety-like behaviors in the elevated plus maze successfully predicted the cocaine-CPP behavior, so that the most anxious mice were also more likely to search for cocaine in a CPP paradigm. In addition, these mice exhibited an increased number of neurons in the basolateral amygdala, a key structure in emotional response including anxiety expression, without differences in the others regions analyzed. Conclusions. Our results suggest a relevant role of anxiety as a psychological risk factor for cocaine vulnerability, with the basolateral amygdala as potential common neural center for both anxiety and addiction.