Matching Items (1,478)
Filtering by
- All Subjects: Computer Science
- All Subjects: Materials Science
- Genre: Academic theses
- Member of: ASU Electronic Theses and Dissertations
Description
Advancements in computer vision and machine learning have added a new dimension to remote sensing applications with the aid of imagery analysis techniques. Applications such as autonomous navigation and terrain classification which make use of image classification techniques are challenging problems and research is still being carried out to find better solutions. In this thesis, a novel method is proposed which uses image registration techniques to provide better image classification. This method reduces the error rate of classification by performing image registration of the images with the previously obtained images before performing classification. The motivation behind this is the fact that images that are obtained in the same region which need to be classified will not differ significantly in characteristics. Hence, registration will provide an image that matches closer to the previously obtained image, thus providing better classification. To illustrate that the proposed method works, naïve Bayes and iterative closest point (ICP) algorithms are used for the image classification and registration stages respectively. This implementation was tested extensively in simulation using synthetic images and using a real life data set called the Defense Advanced Research Project Agency (DARPA) Learning Applied to Ground Robots (LAGR) dataset. The results show that the ICP algorithm does help in better classification with Naïve Bayes by reducing the error rate by an average of about 10% in the synthetic data and by about 7% on the actual datasets used.
ContributorsMuralidhar, Ashwini (Author) / Saripalli, Srikanth (Thesis advisor) / Papandreou-Suppappola, Antonia (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2011
Description
There is an inexorable link between structure and stress, both of which require study in order to truly understand the physics of thin films. To further our knowledge of thin films, the relationship between structure and stress development was examined in three separate systems in vacuum. The first was continued copper thin film growth in ultra-high vacuum after adsorption of a sub-monolayer quantity of oxygen. Results showed an increase in compressive stress generation, and theory was proposed to explain the additional compressive stress within the films. The second system explored was the adsorption of carbon monoxide on the platinum {111} surface in vacuum. The experiments displayed a correlation between known structural developments in the adsorbed carbon monoxide adlayer and the surface stress state of the system. The third system consisted of the growth and annealing stresses of ice thin films at cryogenic temperatures in vacuum. It was shown that the growth stresses are clearly linked to known morphology development from literature, with crystalline ice developing compressive and amorphous ice developing tensile stresses respectively, and that amorphous ice films develop additional tensile stresses upon annealing.
ContributorsKennedy, Jordan (Author) / Friesen, Cody (Thesis advisor) / Sieradzki, Karl (Committee member) / Crozier, Peter (Committee member) / Arizona State University (Publisher)
Created2011
Description
ABSTRACT The behavior of the fission products, as they are released from fission events during nuclear reaction, plays an important role in nuclear fuel performance. Fission product release can occur through grain boundary (GB) at low burnups; therefore, this study simulates the mass transport of fission gases in a 2-D GB network to look into the effects of GB characteristics on this phenomenon, with emphasis on conditions that can lead to percolation. A finite element model was created based on the microstructure of a depleted UO2 sample characterized by Electron Backscattering Diffraction (EBSD). The GBs were categorized into high (D2), low (D1) and bulk diffusivity (Dbulk) based on their misorientation angles and Coincident Site Lattice (CSL) types. The simulation was run using different diffusivity ratios (D2/Dbulk) ranging from 1 to 10^8. The model was set up in three ways: constant temperature case, temperature gradient effects and window methods that mimic the environments in a Light Water Reactor (LWR). In general, the formation of percolation paths was observed at a ratio higher than 10^4 in the measured GB network, which had a 68% fraction of high diffusivity GBs. The presence of temperature gradient created an uneven concentration distribution and decreased the overall mass flux. Finally, radial temperature and fission gas concentration profiles were obtained for a fuel pellet in operation using an approximate 1-D model. The 100 µm long microstructurally explicit model was used to simulate, to the scale of a real UO2 pellet, the mass transport at different radial positions, with boundary conditions obtained from the profiles. Stronger percolation effects were observed at the intermediate and periphery position of the pellet. The results also showed that highest mass flux happens at the edge of a pellet at steady state to accommodate for the sharp concentration drop.
ContributorsLim, Harn Chyi (Author) / Peralta, Pedro (Thesis advisor) / Dey, Sandwip (Committee member) / Sieradzki, Karl (Committee member) / Arizona State University (Publisher)
Created2011
Description
The behavior of a solid oxide fuel cell (SOFC) cermet (ceramic-metal composite) anode under reaction conditions depends significantly on the structure, morphology and atomic scale interactions between the metal and the ceramic components. In situ environmental transmission electron microscope (ETEM) is an important tool which not only allows us to perform the basic nanoscale characterization of the anode materials, but also to observe in real-time, the dynamic changes in the anode material under near-reaction conditions. The earlier part of this dissertation is focused on the synthesis and characterization of Pr- and Gd-doped cerium oxide anode materials. A novel spray drying set-up was designed and constructed for preparing nanoparticles of these mixed-oxides and nickel oxide for anode fabrication. X-ray powder diffraction was used to investigate the crystal structure and lattice parameters of the synthesized materials. Particle size distribution, morphology and chemical composition were investigated using transmission electron microscope (TEM). The nanoparticles were found to possess pit-like defects of average size 2 nm after subjecting the spray-dried material to heat treatment at 700 °C for 2 h in air. A novel electron energy-loss spectroscopy (EELS) quantification technique for determining the Pr and Gd concentrations in the mixed oxides was developed. Nano-scale compositional heterogeneity was observed in these materials. The later part of the dissertation focuses mainly on in situ investigations of the anode materials under a H2 environment in the ETEM. Nano-scale changes in the stand-alone ceramic components of the cermet anode were first investigated. Particle size and composition of the individual nanoparticles of Pr-doped ceria (PDC) were found to affect their reducibility in H2 gas. Upon reduction, amorphization of the nanoparticles was observed and was linked to the presence of pit-like defects in the spray-dried material. Investigation of metal-ceramic interactions in the Ni-loaded PDC nanoparticles indicated a localized reduction of Ce in the vicinity of the Ni/PDC interface at 420 °C. Formation of a reduction zone around the interface was attributed to H spillover which was observed directly in the ETEM. Preliminary results on the fabrication of model SOFCs and in situ behavior of Ni/Gd-doped ceria anodes have been presented.
ContributorsSharma, Vaneet (Author) / Crozier, Peter A. (Thesis advisor) / Sharma, Renu (Thesis advisor) / Adams, James B (Committee member) / Dey, Sandwip (Committee member) / Arizona State University (Publisher)
Created2011
Description
This thesis proposed a novel approach to establish the trust model in a social network scenario based on users' emails. Email is one of the most important social connections nowadays. By analyzing email exchange activities among users, a social network trust model can be established to judge the trust rate between each two users. The whole trust checking process is divided into two steps: local checking and remote checking. Local checking directly contacts the email server to calculate the trust rate based on user's own email communication history. Remote checking is a distributed computing process to get help from user's social network friends and built the trust rate together. The email-based trust model is built upon a cloud computing framework called MobiCloud. Inside MobiCloud, each user occupies a virtual machine which can directly communicate with others. Based on this feature, the distributed trust model is implemented as a combination of local analysis and remote analysis in the cloud. Experiment results show that the trust evaluation model can give accurate trust rate even in a small scale social network which does not have lots of social connections. With this trust model, the security in both social network services and email communication could be improved.
ContributorsZhong, Yunji (Author) / Huang, Dijiang (Thesis advisor) / Dasgupta, Partha (Committee member) / Syrotiuk, Violet (Committee member) / Arizona State University (Publisher)
Created2011
Description
This study presents the results of one of the first attempts to characterize the pore water pressure response of soils subjected to traffic loading under saturated and unsaturated conditions. It is widely known that pore water pressure develops within the soil pores as a response to external stimulus. Also, it has been recognized that the development of pores water pressure contributes to the degradation of the resilient modulus of unbound materials. In the last decades several efforts have been directed to model the effect of air and water pore pressures upon resilient modulus. However, none of them consider dynamic variations in pressures but rather are based on equilibrium values corresponding to initial conditions. The measurement of this response is challenging especially in soils under unsaturated conditions. Models are needed not only to overcome testing limitations but also to understand the dynamic behavior of internal pore pressures that under critical conditions may even lead to failure. A testing program was conducted to characterize the pore water pressure response of a low plasticity fine clayey sand subjected to dynamic loading. The bulk stress, initial matric suction and dwelling time parameters were controlled and their effects were analyzed. The results were used to attempt models capable of predicting the accumulated excess pore pressure at any given time during the traffic loading and unloading phases. Important findings regarding the influence of the controlled variables challenge common beliefs. The accumulated excess pore water pressure was found to be higher for unsaturated soil specimens than for saturated soil specimens. The maximum pore water pressure always increased when the high bulk stress level was applied. Higher dwelling time was found to decelerate the accumulation of pore water pressure. In addition, it was found that the higher the dwelling time, the lower the maximum pore water pressure. It was concluded that upon further research, the proposed models may become a powerful tool not only to overcome testing limitations but also to enhance current design practices and to prevent soil failure due to excessive development of pore water pressure.
ContributorsCary, Carlos (Author) / Zapata, Claudia E (Thesis advisor) / Wiczak, Matthew W (Thesis advisor) / Kaloush, Kamil (Committee member) / Sandra, Houston (Committee member) / Arizona State University (Publisher)
Created2011
Description
In the last few decades, the rapid development of electronic music technologies has changed the way society interacts with music, which in turn impacts the profession of music therapy. Except for a few cases, music therapy has not extensively explored the integration of new technology. However, current research trends show a willingness and excitement to explore the possibilities (Nagler, 2011; Ramsey, 2011; Magee, et al., 2011; Magee & Burland, 2008; Magee 2006). The project described in this paper intends to demonstrate one of these possibilities by combining modern technologies to create an interactive musical system with practical applications in music therapy. In addition to designing a practical tool, the project aims to question the role of technology in music therapy and to initiate dialogue between technologists and music therapists. The project, entitled MIST: A Musical Interactive Space for Therapy, uses modern gestural technology (the Microsoft® Kinect®) to capture body movements and turn them into music. It is intended for use in a clinical setting with children with mild to moderate disabilities. The system is a software/hardware package that is inexpensive, user-friendly, and portable. There are two functional modes of the system: the first sonifies specific movement tasks of reaching and balancing; the second is an interactive musical play space in which an entire room becomes responsive to presence and movement, creating a sonic playground. The therapeutic goals of the system are to motivate and train physical movement, encourage exploration of space and the body, and allow for musical expression, play, auditory perception, and social interaction.
ContributorsHeadlee, Kimberlee (Author) / Ingalls, Todd M (Thesis advisor) / Crowe, Barbara J. (Thesis advisor) / Stauffer, Sandra L (Committee member) / Arizona State University (Publisher)
Created2011
Description
Over the last decade copper electrodeposition has become the dominant process by which microelectronic interconnects are made. Replacing ultra-high vacuum evaporative film growth, the technology known as the Cu damascene process has been widely implemented in the microelectronics industry since the early 2000s. The transition from vacuum film growth to electrodeposition was enabled by solution chemistries that provide "bottom-up" or superfilling capability of vias and trenches. While the process has been and is used widely, the actual mechanisms responsible for superfilling remain relatively unknown. This dissertation presents and discusses the background and results of experimental investigations that have been done using in situ electrochemical surface stress monitoring techniques to study the evolution of stress on Cu{111} thin film electrodes. Because of its extreme sensitivity to the structure on both the electrode and solution sides of the interface, surface stress monitoring as analytical technique is well suited for the study of electrodeposition. These ultra-high resolution stress measurements reveal the dynamic response of copper electrodes to a number of electrochemical and chemical experimental variables. In the case of constant current pulsed deposition and stripping, the surface stress evolution depends not only on the magnitude of the current pulse, but also shows a marked response to plating bath composition. The plating bath chemistries used in this work include (1) additive free, (2) deposition suppressing solutions that include polyethylene glycol (PEG) and sodium chloride (NaCl) as well as (3) full additive solution combinations which contain PEG, NaCl, and a one of two deposition accelerating species (bis-(sodiumsulfopropyl)disulfide (SPS) or mercaptopropane sulfonic acid (MPS)). The development of thin film stress is further investigated through a series of solution exchange experiments that correlate the magnitude of electrode exchange current density and the stress state of the film. Remarkably, stress changes as large as ~8.5 N/m are observed during solution exchanges at the open circuit potential. Overall, this research demonstrates that solution chemistry can have a large impact on thin film stress evolution, even for very small deposition thicknesses (e.g. <10 ML) or in the absence of net addition or removal of material from the electrode.
ContributorsHeaton, Thomas Stanley (Author) / Friesen, Cody (Thesis advisor) / Buttry, Daniel (Committee member) / Sieradzki, Karl (Committee member) / Arizona State University (Publisher)
Created2011
Description
S-Taliro is a fully functional Matlab toolbox that searches for trajectories of minimal robustness in hybrid systems that are implemented as either m-functions or Simulink/State flow models. Trajectories with minimal robustness are found using automatic testing of hybrid systems against user specifications. In this work we use Metric Temporal Logic (MTL) to describe the user specifications for the hybrid systems. We then try to falsify the MTL specification using global minimization of robustness metric. Global minimization is carried out using stochastic optimization algorithms like Monte-Carlo (MC) and Extended Ant Colony Optimization (EACO) algorithms. Irrespective of the type of the model we provide as an input to S-Taliro, the user needs to specify the MTL specification, the initial conditions and the bounds on the inputs. S-Taliro then uses this information to generate test inputs which are used to simulate the system. The simulation trace is then provided as an input to Taliro which computes the robustness estimate of the MTL formula. Global minimization of this robustness metric is performed to generate new test inputs which again generate simulation traces which are closer to falsifying the MTL formula. Traces with negative robustness values indicate that the simulation trace falsified the MTL formula. Traces with positive robustness values are also of great importance because they indicate how robust the system is against the given specification. S-Taliro has been seamlessly integrated into the Matlab environment, which is extensively used for model-based development of control software. Moreover the toolbox has been developed in a modular fashion and therefore adding new optimization algorithms is easy and straightforward. In this work I present the architecture of S-Taliro and its working on a few benchmark problems.
ContributorsAnnapureddy, Yashwanth Singh Rahul (Author) / Fainekos, Georgios (Thesis advisor) / Lee, Yann-Hang (Committee member) / Gupta, Sandeep (Committee member) / Arizona State University (Publisher)
Created2011
Description
In many classication problems data samples cannot be collected easily, example in drug trials, biological experiments and study on cancer patients. In many situations the data set size is small and there are many outliers. When classifying such data, example cancer vs normal patients the consequences of mis-classication are probably more important than any other data type, because the data point could be a cancer patient or the classication decision could help determine what gene might be over expressed and perhaps a cause of cancer. These mis-classications are typically higher in the presence of outlier data points. The aim of this thesis is to develop a maximum margin classier that is suited to address the lack of robustness of discriminant based classiers (like the Support Vector Machine (SVM)) to noise and outliers. The underlying notion is to adopt and develop a natural loss function that is more robust to outliers and more representative of the true loss function of the data. It is demonstrated experimentally that SVM's are indeed susceptible to outliers and that the new classier developed, here coined as Robust-SVM (RSVM), is superior to all studied classier on the synthetic datasets. It is superior to the SVM in both the synthetic and experimental data from biomedical studies and is competent to a classier derived on similar lines when real life data examples are considered.
ContributorsGupta, Sidharth (Author) / Kim, Seungchan (Thesis advisor) / Welfert, Bruno (Committee member) / Li, Baoxin (Committee member) / Arizona State University (Publisher)
Created2011