Matching Items (61)
Filtering by
- Genre: Masters Thesis
- Creators: Li, Baoxin
Description
Currently Java is making its way into the embedded systems and mobile devices like androids. The programs written in Java are compiled into machine independent binary class byte codes. A Java Virtual Machine (JVM) executes these classes. The Java platform additionally specifies the Java Native Interface (JNI). JNI allows Java code that runs within a JVM to interoperate with applications or libraries that are written in other languages and compiled to the host CPU ISA. JNI plays an important role in embedded system as it provides a mechanism to interact with libraries specific to the platform. This thesis addresses the overhead incurred in the JNI due to reflection and serialization when objects are accessed on android based mobile devices. It provides techniques to reduce this overhead. It also provides an API to access objects through its reference through pinning its memory location. The Android emulator was used to evaluate the performance of these techniques and we observed that there was 5 - 10 % performance gain in the new Java Native Interface.
ContributorsChandrian, Preetham (Author) / Lee, Yann-Hang (Thesis advisor) / Davulcu, Hasan (Committee member) / Li, Baoxin (Committee member) / Arizona State University (Publisher)
Created2011
Description
With the introduction of compressed sensing and sparse representation,many image processing and computer vision problems have been looked at in a new way. Recent trends indicate that many challenging computer vision and image processing problems are being solved using compressive sensing and sparse representation algorithms. This thesis assays some applications of compressive sensing and sparse representation with regards to image enhancement, restoration and classication. The first application deals with image Super-Resolution through compressive sensing based sparse representation. A novel framework is developed for understanding and analyzing some of the implications of compressive sensing in reconstruction and recovery of an image through raw-sampled and trained dictionaries. Properties of the projection operator and the dictionary are examined and the corresponding results presented. In the second application a novel technique for representing image classes uniquely in a high-dimensional space for image classification is presented. In this method, design and implementation strategy of the image classification system through unique affine sparse codes is presented, which leads to state of the art results. This further leads to analysis of some of the properties attributed to these unique sparse codes. In addition to obtaining these codes, a strong classier is designed and implemented to boost the results obtained. Evaluation with publicly available datasets shows that the proposed method outperforms other state of the art results in image classication. The final part of the thesis deals with image denoising with a novel approach towards obtaining high quality denoised image patches using only a single image. A new technique is proposed to obtain highly correlated image patches through sparse representation, which are then subjected to matrix completion to obtain high quality image patches. Experiments suggest that there may exist a structure within a noisy image which can be exploited for denoising through a low-rank constraint.
ContributorsKulkarni, Naveen (Author) / Li, Baoxin (Thesis advisor) / Ye, Jieping (Committee member) / Sen, Arunabha (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
Description
Real-world environments are characterized by non-stationary and continuously evolving data. Learning a classification model on this data would require a framework that is able to adapt itself to newer circumstances. Under such circumstances, transfer learning has come to be a dependable methodology for improving classification performance with reduced training costs and without the need for explicit relearning from scratch. In this thesis, a novel instance transfer technique that adapts a "Cost-sensitive" variation of AdaBoost is presented. The method capitalizes on the theoretical and functional properties of AdaBoost to selectively reuse outdated training instances obtained from a "source" domain to effectively classify unseen instances occurring in a different, but related "target" domain. The algorithm is evaluated on real-world classification problems namely accelerometer based 3D gesture recognition, smart home activity recognition and text categorization. The performance on these datasets is analyzed and evaluated against popular boosting-based instance transfer techniques. In addition, supporting empirical studies, that investigate some of the less explored bottlenecks of boosting based instance transfer methods, are presented, to understand the suitability and effectiveness of this form of knowledge transfer.
ContributorsVenkatesan, Ashok (Author) / Panchanathan, Sethuraman (Thesis advisor) / Li, Baoxin (Committee member) / Ye, Jieping (Committee member) / Arizona State University (Publisher)
Created2011
Description
Typically, the complete loss or severe impairment of a sense such as vision and/or hearing is compensated through sensory substitution, i.e., the use of an alternative sense for receiving the same information. For individuals who are blind or visually impaired, the alternative senses have predominantly been hearing and touch. For movies, visual content has been made accessible to visually impaired viewers through audio descriptions -- an additional narration that describes scenes, the characters involved and other pertinent details. However, as audio descriptions should not overlap with dialogue, sound effects and musical scores, there is limited time to convey information, often resulting in stunted and abridged descriptions that leave out many important visual cues and concepts. This work proposes a promising multimodal approach to sensory substitution for movies by providing complementary information through haptics, pertaining to the positions and movements of actors, in addition to a film's audio description and audio content. In a ten-minute presentation of five movie clips to ten individuals who were visually impaired or blind, the novel methodology was found to provide an almost two time increase in the perception of actors' movements in scenes. Moreover, participants appreciated and found useful the overall concept of providing a visual perspective to film through haptics.
ContributorsViswanathan, Lakshmie Narayan (Author) / Panchanathan, Sethuraman (Thesis advisor) / Hedgpeth, Terri (Committee member) / Li, Baoxin (Committee member) / Arizona State University (Publisher)
Created2011
Description
Once considered an abundant species in the eastern United States, local populations of red-shouldered hawks, Buteo lineatus, have declined due to habitat destruction. This destruction has created suitable habitat for red-tailed hawks, Buteo jamaicensis, and therefore increased competition between these two raptor species. Since suitable habitat is the main limiting factor for raptors, a computer model was created to simulate the effect of habitat loss in central Maryland and the impact of increased competition between the more aggressive red-tailed hawk. These simulations showed urban growth contributed to over a 30% increase in red-tailed hawk habitat as red-shouldered hawk habitat decreased 62.5-70.1% without competition and 71.8-76.3% with competition. However there was no significant difference seen between the rate of available habitat decline for current and predicted development growth.
ContributorsMurillo, Crystal (Author) / Whysong, Gary (Thesis advisor) / Alford, Eddie (Committee member) / Miller, William (Committee member) / Arizona State University (Publisher)
Created2011
Description
Modified and artificial water sources can be used as a management tool for game and non-game wildlife species. State, federal, and private agencies allocate significant resources to install and maintain artificial water sources (AWS) annually. Capture mark recapture methods were used to sample small mammal communities in the vicinity of five AWS and five paired control sites (treatments) in the surrounding Sonoran desert from October 2011 to May 2012. I measured plant species richness, density, and percent cover in the spring of 2012. A Multi-response Permutation Procedure was used to identify differences in small mammal community abundance, biomass, and species richness by season and treatment. I used Principle Component Analysis to reduce 11 habitat characteristics to five habitat factors. I related rodent occurrence to habitat characteristics using multiple and logistic regression. A total of 370 individual mammals representing three genera and eight species of rodents were captured across 4800 trap nights. Desert pocket mouse (Chaetodipus penicillatus) was the most common species in both seasons and treatments. Whereas rodent community abundance, biomass, and richness were similar between seasons, community variables of AWS were greater than CS. Rodent diversity was similar between treatments. Desert pocket mouse abundance and biomass were twice as high at AWS when compared to controls. Biomass of white-throated woodrat (Neotoma albigula) was five times greater at AWS. Habitat characteristics were similar between treatments. Neither presence of water nor distance to water explained substantial habitat variation. Occurrence of rodent species was associated with habitat characteristics. Desert rodent communities are adapted for arid environments (i.e. Heteromyids) and are not dependent on "free water". Higher abundances of desert pocket mouse at AWS were most likely related to increased disturbance and debris and not the presence of water. The results of this study and previous studies suggest that more investigation is needed and that short term studies may not be able to detect interactions (if any) between AWS and desert small mammal communities.
ContributorsSwitalski, Aaron (Author) / Bateman, Heather L (Thesis advisor) / Miller, William (Committee member) / Alford, Eddie (Committee member) / Arizona State University (Publisher)
Created2013
Description
Livestock-grazing, in particular cattle grazing, is a common use of public and private lands in western North America. As a result, the effects of grazing on both plants and animals are widely studied. Few studies, however, look directly at the long-term effects that cattle grazing may have on a particular species. The goal of this experiment was to continue research begun in 1988, to determine if the effects of cattle grazing are still seen in the age structure of two populations of saguaros (Carnegiea gigantea [Engelm.] Britton & Rose) at Saguaro National Park - Rincon Mountain District (SNP-RMD). The null hypothesis stated that enough time has elapsed since the cessation of grazing, and there is no difference in the age distribution of the saguaros of the two populations. The study area was comprised of a former fence line where grazing ceased on the western side of the fence in 1958 and the eastern side in 1978. Belt transects were laid on each side of the fence line and height was measured for each saguaro encountered in a transect. Approximate age of the individual was then calculated using an age-height correlation for SNP-RMD. Individuals were then placed into age classes of 10 year increments and a Log-Likelihood test was performed. The resulting calculated P value of 0.12 meant the null hypothesis was not rejected and there was no statistical difference between the age structure of the two populations. After 34 and 54 years rest from grazing, the negative effects of cattle grazing on the retention and recruitment of saguaro seedlings have ended, and replenishment of the populations is now dependent upon factors such as temperature and precipitation. Other factors such as climate change, increasing fire frequency, encroachment by invasive species, and poaching are sources of concern and increased mortality for these and other saguaros.
ContributorsKrone, Elizabeth AnnMarie Stewart (Author) / Alford, Eddie (Thesis advisor) / Brady, Ward (Committee member) / Green, Douglas (Committee member) / Arizona State University (Publisher)
Created2013
Description
The Southwestern Willow Flycatcher (Empidonax traillii extimus) has been studied for over two decades and listed as endangered for most of that time. Though the flycatcher has been granted protected status since 1995, critical habitat designation for the flycatcher has not shared the same history. Critical habitat designation is essential for achieving the long-term goals defined in the flycatcher recovery plan where emphasis is on both the protection of this species and "the habitats supporting these flycatchers [that] must be protected from threats and loss" (U.S. Fish and Wildlife Service 2002). I used a long-term data set of habitat characteristics collected at three study areas along the Lower Colorado River to develop a method for quantifying habitat quality for flycatcher. The data set contained flycatcher nest observations (use) and habitat availability (random location) from 2003-2010 that I statistically analyzed for flycatcher selection preferences. Using both Pearson's Chi-square test and SPSS Principal Component Analysis (PCA) I determined that flycatchers were selecting 30 habitat traits significantly different among an initial list of 127 habitat characteristics. Using PCA, I calculated a weighted value of influence for each significant trait per study area and used those values to develop a habitat classification system to build predictive models for flycatcher habitat quality. I used ArcGIS® Model Builder to develop three habitat suitability models for each of the habitat types occurring in western riparian systems, native, mixed exotic and exotic dominated that are frequented by breeding flycatchers. I designed a fourth model, Topock Marsh, to test model accuracy on habitat quality for flycatchers using reserved accuracy assessment points of previous nest locations. The results of the fourth model accurately predicted a decline in habitat at Topock Marsh that was confirmed by SWCA survey reports released in 2011 and 2012 documenting a significant decline in flycatcher productivity in the Topock Marsh study area.
ContributorsChenevert-Steffler, Ann (Author) / Miller, William (Thesis advisor) / Bateman, Heather (Committee member) / Alford, Eddie (Committee member) / Arizona State University (Publisher)
Created2013
Description
Mycorrhizal fungi form symbiotic relationships with plant roots, increasing nutrient and water availability to plants and improving soil stability. Mechanical disturbance of soil has been found to reduce mycorrhizal inoculum in soils, but findings have been inconsistent. To examine the impact of restoration practices on riparian mycorrhizal inoculum potential, soil samples were collected at the Tres Rios Ecosystem Restoration and Flood Control Project located at the confluence of the Salt, Gila, and Agua Fria rivers in central Arizona. The project involved the mechanical removal of invasive Tamarix spp.( tamarisk, salt cedar) and grading prior to revegetation. Soil samples were collected from three stages of restoration: pre-restoration, soil banks with chipped vegetation, and in areas that had been graded in preparation for revegetation. Bioassay plants were grown in the soil samples and roots analyzed for arbuscular mycorrhizal (AM) and ectomycorrhizal (EM) infection percentages. Vegetations measurements were also taken for woody vegetation at the site. The mean number of AM and EM fungal propagules did not differ between the three treatment area, but inoculum levels did differ between AM and EM fungi with AM fungal propagules detected at moderate levels and EM fungi at very low levels. These differences may have been related to availability of host plants since AM fungi form associations with a variety of desert riparian forbs and grasses and EM fungi only form associations with Populus spp. and Salix spp. which were present at the site but at low density and canopy cover. Prior studies have also found that EM fungi may be more affected by tamarisk invasions than AM fungi. Our results were similar to other restoration projects for AM fungi suggesting that it may not be necessary to add AM fungi to soil prior to planting native vegetation because of the moderate presence of AM fungi even in soils dominated by tamarisk and exposed to soil disturbance during the restoration process. In contrast when planting trees that form EM associations, it may be beneficial to augment soil with EM fungi collected from riparian areas or to pre-inoculate plants prior to planting.
ContributorsArnold, Susanne (Author) / Stutz, Jean (Thesis advisor) / Alford, Eddie (Committee member) / Green, Douglas (Committee member) / Arizona State University (Publisher)
Created2012