Automatic Image Annotation Using SURF Descriptors

In recent years there is an apparent shift in research from content based image retrieval (CBIR) to automatic image annotation in order to bridge the gap between low level features and high level semantics of images. Automatic Image Annotation (AIA) techniques facilitate extraction of high level semantic concepts from images by machine learning techniques. Many AIA techniques use feature analysis as the first step to identify the objects in the image. However, the high dimensional image features make the performance of the system worse. This paper describes and evaluates an automatic image annotation framework which uses SURF descriptors to select right number of features and right features for annotation. The proposed framework uses a hybrid approach in which k-means clustering is used in the training phase and fuzzy K-NN classification in the annotation phase. The performance of the system is evaluated using standard metrics.

Preparation and Study of Optical Properties of CdS, CdS-TiO2 and CdS-Au Nanocomposites for Photonic Applications

Nanophotonics can be regarded as a fusion of nanotechnology and photonics and it is an emerging field providing researchers opportunities in fundamental science and new technologies. In recent times many new methodsand techniques have been developed to prepare materials at nanoscale dimensions. Most of these materials exhibit unique and interesting optical properties and behavior. Many of these have been found to be very useful to develop new devices and systems such as tracers in biological systems, optical limiters, light emitters and energy harvesters. This thesis presents a summary of the work done by the author in the field by choosing a few semiconductor systems to prepare nanomaterials and nanocomposites. Results of the study of linear and nonlinear optical properties of materials thus synthesized are also presented in the various chapters of this thesis. CdS is the material chosen here and the methods and the studies of the detailed investigation are presented in this thesis related to the optical properties of CdS nanoparticles and its composites. Preparation and characterization methods and experimental techniques adopted for the investigations were illustrated in chapter 2 of this thesis. Chapter 3 discusses the preparation of CdS, TiO2 and Au nanoparticles. We observed that the fluorescence behaviour of the CdS nanoparticles, prepared by precipitation technique, depends on excitation wavelength. It was found that the peak emission wavelength can be shifted by as much as 147nm by varyingthe excitation wavelengths and the reason for this phenomenon is the selective excitation of the surface states in the nanoparticles. This provided certain amount of tunability for the emission which results from surface states.TiO2 nanoparticle colloids were prepared by hydrothermal method. The optical absorption study showed a blue shift of absorption edge, indicating quantum confinement effect. The large spectral range investigated allows observing simultaneously direct and indirect band gap optical recombination. The emission studies carried out show four peaks, which are found to be generated from excitonic as well as surface state transitions. It was found that the emission wavelengths of these colloidal nanoparticles and annealed nanoparticles showed two category of surface state emission in addition to the excitonic emission. Au nanoparticles prepared by Turkevich method showed nanoparticles of size below 5nm using plasmonic absorption calculation. It was also found that there was almost no variation in size as the concentration of precursor was changed from 0.2mM to 0.4mM.We have observed SHG from CdS nanostructured thin film prepared onglass substrate by chemical bath deposition technique. The results point out that studied sample has in-plane isotropy. The relative values of tensor components of the second-order susceptibility were determined to be 1, zzz 0.14, xxz and 0.07. zxx These values suggest that the nanocrystals are oriented along the normal direction. However, the origin of such orientation remains unknown at present. Thus CdS is a promising nonlinear optical material for photonic applications, particularly for integrated photonic devices. CdS Au nanocomposite particles were prepared by mixing CdS nanoparticles with Au colloidal nanoparticles. Optical absorption study of these nanoparticles in PVA solution suggests that absorption tail was red shifted compared to CdS nanoparticles. TEM and EDS analysis suggested that the amount of Au nanoparticles present on CdS nanoparticles is very small. Fluorescence emission is unaffected indicating the presence of low level of Au nanoparticles. CdS:Au PVA and CdS PVA nanocomposite films were fabricated and optically characterized. The results showed a red-shift for CdS:Au PVA film for absorption tail compared to CdS PVA film. Nonlinear optical analysis showed a huge nonlinear optical absorption for CdS:Au PVA nanocomposite and CdS:PVA films. Also an enhancement in nonlinear optical absorption is found for CdS:Au PVA thin film compared to the CdS PVA thin film. This enhancement is due to the combined effect of plasmonic as well as excitonic contribution at high input intensity. Samples of CdS doped with TiO2 were also prepared and the linear optical absorption spectra of these nanocompositeparticles clearly indicated the influence of TiO2 nanoparticles. TEM and EDS studies have confirmed the presence of TiO2 on CdS nanoparticles. Fluorescence studies showed that there is an increase in emission peak around 532nm for CdS nanoparticles. Nonlinear optical analysis of CdS:TiO2 PVA nanocomposite films indicated a large nonlinear optical absorption compared to that of CdS:PVA nanocomposite film. The values of nonlinear optical absorption suggests that these nanocomposite particles can be employed for optical limiting applications. CdSe-CdS and CdSe-ZnS core-shell QDs with varying shell size were characterized using UV–VIS spectroscopy. Optical absorption and TEM analysis of these QDs suggested a particle size around 5 nm. It is clearly shown that the surface coating influences the optical properties of QDs in terms of their size. Fluorescence studies reveal the presence of trap states in CdSe-CdS and CdSe- ZnS QDs. Trap states showed an increase as a shell for CdS is introduced and increasing the shell size of CdS beyond a certain value leads to a decrease in the trap state emission. There is no sizeable nonlinear optical absorption observed. In the case of CdSe- ZnS QDs, the trap state emission gets enhanced with the increase in ZnS shell thickness. The enhancement of emission from trap states transition due to the increase in thickness of ZnS shell gives a clear indication of distortion occurring in the spherical symmetry of CdSe quantum dots. Consequently the nonlinear optical absorption of CdSe-ZnS QDs gets increased and the optical limiting threshold is decreased as the shell thickness is increased in respect of CdSe QDs. In comparison with CdSe-CdS QDs, CdSe-ZnS QDs possess much better optical properties and thereby CdSe-ZnS is a strong candidate for nonlinear as well as linear optical applications.

Information Exchange and Fusion in Dynamic and Heterogeneous Distributed Environments

Context awareness, dynamic reconfiguration at runtime and heterogeneity are key characteristics of future distributed systems, particularly in ubiquitous and mobile computing scenarios. The main contributions of this dissertation are theoretical as well as architectural concepts facilitating information exchange and fusion in heterogeneous and dynamic distributed environments. Our main focus is on bridging the heterogeneity issues and, at the same time, considering uncertain, imprecise and unreliable sensor information in information fusion and reasoning approaches. A domain ontology is used to establish a common vocabulary for the exchanged information. We thereby explicitly support different representations for the same kind of information and provide Inter-Representation Operations that convert between them. Special account is taken of the conversion of associated meta-data that express uncertainty and impreciseness. The Unscented Transformation, for example, is applied to propagate Gaussian normal distributions across highly non-linear Inter-Representation Operations. Uncertain sensor information is fused using the Dempster-Shafer Theory of Evidence as it allows explicit modelling of partial and complete ignorance. We also show how to incorporate the Dempster-Shafer Theory of Evidence into probabilistic reasoning schemes such as Hidden Markov Models in order to be able to consider the uncertainty of sensor information when deriving high-level information from low-level data. For all these concepts we provide architectural support as a guideline for developers of innovative information exchange and fusion infrastructures that are particularly targeted at heterogeneous dynamic environments. Two case studies serve as proof of concept. The first case study focuses on heterogeneous autonomous robots that have to spontaneously form a cooperative team in order to achieve a common goal. The second case study is concerned with an approach for user activity recognition which serves as baseline for a context-aware adaptive application. Both case studies demonstrate the viability and strengths of the proposed solution and emphasize that the Dempster-Shafer Theory of Evidence should be preferred to pure probability theory in applications involving non-linear Inter-Representation Operations.

Recycling urban waste as possible use for rooftop vegetable garden

Urban authorities in Europe are confronted with increasing demands by urban dwellers for allotment gardens, but vacant urban soil tends to be scarce and/or polluted by past industrial activities. A possible solution for local authorities could therefore be to promote rooftop gardening. However little technical information exists on certain forms of rooftop urban agriculture, called Z-Farming. In 2012, a pilot experiment was run in Paris (France). Simple and cheap systems of rooftop gardening were tested on a rooftop using as crop substrates only local urban organic waste so as to contribute to the urban metabolism. Production levels and heavy metal contents in cropping substrates and edible vegetables were measured. Available results show (i) high levels of crop production with limited inputs compared to land professional gardening, (ii) low levels of heavy metal pollutants in the edible parts of the crops, especially for Cd and Pb with respect to EU norms for vegetables and (iii) positive influence on yields on organizing the substrate in layers and enhancing the biological activity through earthworm inoculation. These encouraging results allow us to consider that rooftop gardening is feasible and seem to have a great potential to improve urban resiliency. It will nevertheless be necessary to identify more precisely the types of roof that can be used and to assess more fully the generic result of the low level of pollution, as well as the global sustainability of these cropping systems.

Diversity, local uses and availability of medicinal plants and wild yams in the Mahafaly region of south-western Madagascar

Short summary: This study was undertaken to assess the diversity of plant resources utilized by the local population in south-western Madagascar, the social, ecological and biophysical conditions that drive their uses and availability, and possible alternative strategies for their sustainable use in the region. The study region, ‘Mahafaly region’, located in south-western Madagascar, is one of the country’s most economically, educationally and climatically disadvantaged regions. With an arid steppe climate, the agricultural production is limited by low water availability and a low level of soil nutrients and soil organic carbon. The region comprises the recently extended Tsimanampetsotsa National Park, with numerous sacred and communities forests, which are threatened by slash and burn agriculture and overexploitation of forests resources. The present study analyzed the availability of wild yams and medicinal plants, and their importance for the livelihood of the local population in this region. An ethnobotanical survey was conducted recording the diversity, local knowledge and use of wild yams and medicinal plants utilized by the local communities in five villages in the Mahafaly region. 250 households were randomly selected followed by semi-structured interviews on the socio-economic characteristics of the households. Data allowed us to characterize sociocultural and socioeconomic factors that determine the local use of wild yams and medicinal plants, and to identify their role in the livelihoods of local people. Species-environment relationships and the current spatial distribution of the wild yams were investigated and predicted using ordination methods and a niche based habitat modelling approach. Species response curves along edaphic gradients allowed us to understand the species requirements on habitat conditions. We thus investigated various alternative methods to enhance the wild yam regeneration for their local conservation and their sustainable use in the Mahafaly region. Altogether, six species of wild yams and a total of 214 medicinal plants species from 68 families and 163 genera were identified in the study region. Results of the cluster and discriminant analysis indicated a clear pattern on resource, resulted in two groups of household and characterized by differences in livestock numbers, off-farm activities, agricultural land and harvests. A generalized linear model highlighted that economic factors significantly affect the collection intensity of wild yams, while the use of medicinal plants depends to a higher degree on socio-cultural factors. The gradient analysis on the distribution of the wild yam species revealed a clear pattern for species habitats. Species models based on NPMR (Nonparametric Multiplicative Regression analysis) indicated the importance of vegetation structure, human interventions, and soil characteristics to determine wild yam species distribution. The prediction of the current availability of wild yam resources showed that abundant wild yam resources are scarce and face high harvest intensity. Experiments on yams cultivation revealed that germination of seeds was enhanced by using pre-germination treatments before planting, vegetative regeneration performed better with the upper part of the tubers (corms) rather than the sets of tubers. In-situ regeneration was possible for the upper parts of the wild tubers but the success depended significantly on the type of soil. The use of manure (10-20 t ha¹) increased the yield of the D. alata and D. alatipes by 40%. We thus suggest the promotion of other cultivated varieties of D. alata found regions neighbouring as the Mahafaly Plateau.

Boundaries and Topological Algorithms

This thesis develops a model for the topological structure of situations. In this model, the topological structure of space is altered by the presence or absence of boundaries, such as those at the edges of objects. This allows the intuitive meaning of topological concepts such as region connectivity, function continuity, and preservation of topological structure to be modeled using the standard mathematical definitions. The thesis shows that these concepts are important in a wide range of artificial intelligence problems, including low-level vision, high-level vision, natural language semantics, and high-level reasoning.

A Compilation Strategy for Numerical Programs Based on Partial Evaluation

This work demonstrates how partial evaluation can be put to practical use in the domain of high-performance numerical computation. I have developed a technique for performing partial evaluation by using placeholders to propagate intermediate results. For an important class of numerical programs, a compiler based on this technique improves performance by an order of magnitude over conventional compilation techniques. I show that by eliminating inherently sequential data-structure references, partial evaluation exposes the low-level parallelism inherent in a computation. I have implemented several parallel scheduling and analysis programs that study the tradeoffs involved in the design of an architecture that can effectively utilize this parallelism. I present these results using the 9- body gravitational attraction problem as an example.

A Distributed Model for Mobile Robot Environment-Learning and Navigation

A distributed method for mobile robot navigation, spatial learning, and path planning is presented. It is implemented on a sonar-based physical robot, Toto, consisting of three competence layers: 1) Low-level navigation: a collection of reflex-like rules resulting in emergent boundary-tracing. 2) Landmark detection: dynamically extracts landmarks from the robot's motion. 3) Map learning: constructs a distributed map of landmarks. The parallel implementation allows for localization in constant time. Spreading of activation computes both topological and physical shortest paths in linear time. The main issues addressed are: distributed, procedural, and qualitative representation and computation, emergent behaviors, dynamic landmarks, minimized communication.

Virtual Model Control of a Biped Walking Robot

The transformation from high level task specification to low level motion control is a fundamental issue in sensorimotor control in animals and robots. This thesis develops a control scheme called virtual model control which addresses this issue. Virtual model control is a motion control language which uses simulations of imagined mechanical components to create forces, which are applied through joint torques, thereby creating the illusion that the components are connected to the robot. Due to the intuitive nature of this technique, designing a virtual model controller requires the same skills as designing the mechanism itself. A high level control system can be cascaded with the low level virtual model controller to modulate the parameters of the virtual mechanisms. Discrete commands from the high level controller would then result in fluid motion. An extension of Gardner's Partitioned Actuator Set Control method is developed. This method allows for the specification of constraints on the generalized forces which each serial path of a parallel mechanism can apply. Virtual model control has been applied to a bipedal walking robot. A simple algorithm utilizing a simple set of virtual components has successfully compelled the robot to walk eight consecutive steps.

Virtual Model Control of a Hexapod Walking Robot

Since robots are typically designed with an individual actuator at each joint, the control of these systems is often difficult and non-intuitive. This thesis explains a more intuitive control scheme called Virtual Model Control. This thesis also demonstrates the simplicity and ease of this control method by using it to control a simulated walking hexapod. Virtual Model Control uses imagined mechanical components to create virtual forces, which are applied through the joint torques of real actuators. This method produces a straightforward means of controlling joint torques to produce a desired robot behavior. Due to the intuitive nature of this control scheme, the design of a virtual model controller is similar to the design of a controller with basic mechanical components. The ease of this control scheme facilitates the use of a high level control system which can be used above the low level virtual model controllers to modulate the parameters of the imaginary mechanical components. In order to apply Virtual Model Control to parallel mechanisms, a solution to the force distribution problem is required. This thesis uses an extension of Gardner`s Partitioned Force Control method which allows for the specification of constrained degrees of freedom. This virtual model control technique was applied to a simulated hexapod robot. Although the hexapod is a highly non-linear, parallel mechanism, the virtual models allowed text-book control solutions to be used while the robot was walking. Using a simple linear control law, the robot walked while simultaneously balancing a pendulum and tracking an object.

A Reinforcement-Learning Approach to Power Management

We describe an adaptive, mid-level approach to the wireless device power management problem. Our approach is based on reinforcement learning, a machine learning framework for autonomous agents. We describe how our framework can be applied to the power management problem in both infrastructure and ad~hoc wireless networks. From this thesis we conclude that mid-level power management policies can outperform low-level policies and are more convenient to implement than high-level policies. We also conclude that power management policies need to adapt to the user and network, and that a mid-level power management framework based on reinforcement learning fulfills these requirements.

From First Contact to Close Encounters: A Developmentally Deep Perceptual System for a Humanoid Robot

This thesis presents a perceptual system for a humanoid robot that integrates abilities such as object localization and recognition with the deeper developmental machinery required to forge those competences out of raw physical experiences. It shows that a robotic platform can build up and maintain a system for object localization, segmentation, and recognition, starting from very little. What the robot starts with is a direct solution to achieving figure/ground separation: it simply 'pokes around' in a region of visual ambiguity and watches what happens. If the arm passes through an area, that area is recognized as free space. If the arm collides with an object, causing it to move, the robot can use that motion to segment the object from the background. Once the robot can acquire reliable segmented views of objects, it learns from them, and from then on recognizes and segments those objects without further contact. Both low-level and high-level visual features can also be learned in this way, and examples are presented for both: orientation detection and affordance recognition, respectively. The motivation for this work is simple. Training on large corpora of annotated real-world data has proven crucial for creating robust solutions to perceptual problems such as speech recognition and face detection. But the powerful tools used during training of such systems are typically stripped away at deployment. Ideally they should remain, particularly for unstable tasks such as object detection, where the set of objects needed in a task tomorrow might be different from the set of objects needed today. The key limiting factor is access to training data, but as this thesis shows, that need not be a problem on a robotic platform that can actively probe its environment, and carry out experiments to resolve ambiguity. This work is an instance of a general approach to learning a new perceptual judgment: find special situations in which the perceptual judgment is easy and study these situations to find correlated features that can be observed more generally.

Representation and Detection of Shapes in Images

We present a set of techniques that can be used to represent and detect shapes in images. Our methods revolve around a particular shape representation based on the description of objects using triangulated polygons. This representation is similar to the medial axis transform and has important properties from a computational perspective. The first problem we consider is the detection of non-rigid objects in images using deformable models. We present an efficient algorithm to solve this problem in a wide range of situations, and show examples in both natural and medical images. We also consider the problem of learning an accurate non-rigid shape model for a class of objects from examples. We show how to learn good models while constraining them to the form required by the detection algorithm. Finally, we consider the problem of low-level image segmentation and grouping. We describe a stochastic grammar that generates arbitrary triangulated polygons while capturing Gestalt principles of shape regularity. This grammar is used as a prior model over random shapes in a low level algorithm that detects objects in images.

Building Grounded Abstractions for Artificial Intelligence Programming

Most Artificial Intelligence (AI) work can be characterized as either ``high-level'' (e.g., logical, symbolic) or ``low-level'' (e.g., connectionist networks, behavior-based robotics). Each approach suffers from particular drawbacks. High-level AI uses abstractions that often have no relation to the way real, biological brains work. Low-level AI, on the other hand, tends to lack the powerful abstractions that are needed to express complex structures and relationships. I have tried to combine the best features of both approaches, by building a set of programming abstractions defined in terms of simple, biologically plausible components. At the ``ground level'', I define a primitive, perceptron-like computational unit. I then show how more abstract computational units may be implemented in terms of the primitive units, and show the utility of the abstract units in sample networks. The new units make it possible to build networks using concepts such as long-term memories, short-term memories, and frames. As a demonstration of these abstractions, I have implemented a simulator for ``creatures'' controlled by a network of abstract units. The creatures exist in a simple 2D world, and exhibit behaviors such as catching mobile prey and sorting colored blocks into matching boxes. This program demonstrates that it is possible to build systems that can interact effectively with a dynamic physical environment, yet use symbolic representations to control aspects of their behavior.

Geometric Structure of the Adaptive Controller of the Human Arm

The objects with which the hand interacts with may significantly change the dynamics of the arm. How does the brain adapt control of arm movements to this new dynamic? We show that adaptation is via composition of a model of the task's dynamics. By exploring generalization capabilities of this adaptation we infer some of the properties of the computational elements with which the brain formed this model: the elements have broad receptive fields and encode the learned dynamics as a map structured in an intrinsic coordinate system closely related to the geometry of the skeletomusculature. The low--level nature of these elements suggests that they may represent asset of primitives with which a movement is represented in the CNS.