View All Items
(1 - 11 of 11)
- Title
- WITHOUT A CAMERA.
- Creator
-
Kulbaba, Brian, Robinson, E. Brady, University of Central Florida
- Abstract / Description
-
The method for creating my art is a matter of experimental process, manipulation of photographic elements, and time spent. I am a photographer in a digital age that does not use a camera. My moment of creativity occurs without the snap of a shutter, but relies on my understanding and control of the chemical components of photography. My work deconstructs the notion of duplication commonly found in photography. The procedure can be repeated but the results are variable. The process of creating...
Show moreThe method for creating my art is a matter of experimental process, manipulation of photographic elements, and time spent. I am a photographer in a digital age that does not use a camera. My moment of creativity occurs without the snap of a shutter, but relies on my understanding and control of the chemical components of photography. My work deconstructs the notion of duplication commonly found in photography. The procedure can be repeated but the results are variable. The process of creating my work often results in a multitude of prints, but the pieces that I select as art capture a number of instinctive characteristics which convey an emotion or message to me. When I present my photographs I offer the viewer an experience--an opportunity to see the work through my mind's eye as it makes sense to me. It is within this open dialogue that the work is complete: part process, part intuitive participation.
Show less - Date Issued
- 2008
- Identifier
- CFE0002100, ucf:47554
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0002100
- Title
- A CASE-STUDY OF THE AFRICAN LEOPARD (PANTHERA PARDUS PARDUS) POPULATION ON THE NAMBITI PRIVATE GAME RESERVE.
- Creator
-
Castaneda, Erica, Borgon, Robert, University of Central Florida
- Abstract / Description
-
The Nambiti Private Game Reserve in KwaZulu-Natal, South Africa is a nature reserve that aids in the conservation of some of the world's most renown species. This includes members of the "Big Five," which is comprised of the African lion (Panthera leo), the African elephant (Loxidonta africana), the Cape buffalo (Syncerus caffer), the black and white rhinoceroses (Diceros bicornis and Ceratotherium simum, respectively), and the African leopard (Panthera pardus pardus). These animals represent...
Show moreThe Nambiti Private Game Reserve in KwaZulu-Natal, South Africa is a nature reserve that aids in the conservation of some of the world's most renown species. This includes members of the "Big Five," which is comprised of the African lion (Panthera leo), the African elephant (Loxidonta africana), the Cape buffalo (Syncerus caffer), the black and white rhinoceroses (Diceros bicornis and Ceratotherium simum, respectively), and the African leopard (Panthera pardus pardus). These animals represent the top five African animals desired by trophy hunters and by tourists hoping to view wildlife (Caro and Riggio, 2014). While studies concerning the African leopard population status have been completed on surrounding game reserves (Balme et al., 2009; Chapman and Balme, 2010), there have not been any studies done investigating the African leopard population on Nambiti. It is important that the population on Nambiti be identified since conservation management of leopards is largely influenced by their population numbers. For example, southern African countries rely on population estimates to establish trophy hunting quotas (Balme et al., 2010). Furthermore, knowledge on the reserve's leopard population can also lead to ecotourism benefits by attracting tourists to visit areas of known leopard activity (Lindsey et al., 2007). This case study investigated baited camera trapping footage, obtained by Nambiti rangers between May 2015 - May 2017, to determine the African leopard population on Nambiti. Camera footage results revealed that there were four leopards identified in six different locations on the reserve between May 2015 - May 2017. Baited Location J in the Western region of the reserve showed the greatest amount of leopard activity, indicating that it is the baited location most likely to provide ecotourism benefits. Furthermore, 23 non-target species were identified from the camera trapping footage, providing insight into the reserve's biodiversity, prey availability, and competition among predators.
Show less - Date Issued
- 2018
- Identifier
- CFH2000285, ucf:45908
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFH2000285
- Title
- ACADEMIC ADVISING IN HIGHER EDUCATION: DISTANCE LEARNERS AND LEVELS OF SATISFACTION USING WEB CAMERA TECHNOLOGY.
- Creator
-
Hernandez, Terri, Tubbs, Levester, University of Central Florida
- Abstract / Description
-
The purpose of this study was to determine the efficacy of in-seat face-to-face advising in contrast to web camera advising of College of Arts and Sciences psychology majors in the 2005-2006 academic year. Satisfaction levels were determined and analyzed based on random assignment to either the control group (in-seat face-to face) or the experimental group (web camera) advising. The data collected for this study consisted of participants' responses to the Academic Advising Inventory (AAI)...
Show moreThe purpose of this study was to determine the efficacy of in-seat face-to-face advising in contrast to web camera advising of College of Arts and Sciences psychology majors in the 2005-2006 academic year. Satisfaction levels were determined and analyzed based on random assignment to either the control group (in-seat face-to face) or the experimental group (web camera) advising. The data collected for this study consisted of participants' responses to the Academic Advising Inventory (AAI) administered to undergraduate psychology majors (N = 102). Overall, students were satisfied with advising services regardless of the advising group to which they were randomly assigned. Although there was not a statistically significant difference between students who were advised in-seat face-to-face and those advised via web camera advising, the data reflected a slight preference for advisement via web camera.
Show less - Date Issued
- 2007
- Identifier
- CFE0001773, ucf:47250
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001773
- Title
- MULTIPLE VIEW GEOMETRY FOR VIDEO ANALYSIS AND POST-PRODUCTION.
- Creator
-
Cao, Xiaochun, Foroosh, Hassan, University of Central Florida
- Abstract / Description
-
Multiple view geometry is the foundation of an important class of computer vision techniques for simultaneous recovery of camera motion and scene structure from a set of images. There are numerous important applications in this area. Examples include video post-production, scene reconstruction, registration, surveillance, tracking, and segmentation. In video post-production, which is the topic being addressed in this dissertation, computer analysis of the motion of the camera can replace the...
Show moreMultiple view geometry is the foundation of an important class of computer vision techniques for simultaneous recovery of camera motion and scene structure from a set of images. There are numerous important applications in this area. Examples include video post-production, scene reconstruction, registration, surveillance, tracking, and segmentation. In video post-production, which is the topic being addressed in this dissertation, computer analysis of the motion of the camera can replace the currently used manual methods for correctly aligning an artificially inserted object in a scene. However, existing single view methods typically require multiple vanishing points, and therefore would fail when only one vanishing point is available. In addition, current multiple view techniques, making use of either epipolar geometry or trifocal tensor, do not exploit fully the properties of constant or known camera motion. Finally, there does not exist a general solution to the problem of synchronization of N video sequences of distinct general scenes captured by cameras undergoing similar ego-motions, which is the necessary step for video post-production among different input videos. This dissertation proposes several advancements that overcome these limitations. These advancements are used to develop an efficient framework for video analysis and post-production in multiple cameras. In the first part of the dissertation, the novel inter-image constraints are introduced that are particularly useful for scenes where minimal information is available. This result extends the current state-of-the-art in single view geometry techniques to situations where only one vanishing point is available. The property of constant or known camera motion is also described in this dissertation for applications such as calibration of a network of cameras in video surveillance systems, and Euclidean reconstruction from turn-table image sequences in the presence of zoom and focus. We then propose a new framework for the estimation and alignment of camera motions, including both simple (panning, tracking and zooming) and complex (e.g. hand-held) camera motions. Accuracy of these results is demonstrated by applying our approach to video post-production applications such as video cut-and-paste and shadow synthesis. As realistic image-based rendering problems, these applications require extreme accuracy in the estimation of camera geometry, the position and the orientation of the light source, and the photometric properties of the resulting cast shadows. In each case, the theoretical results are fully supported and illustrated by both numerical simulations and thorough experimentation on real data.
Show less - Date Issued
- 2006
- Identifier
- CFE0001014, ucf:46840
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001014
- Title
- SCENE MONITORING WITH A FOREST OF COOPERATIVE SENSORS.
- Creator
-
Javed, Omar, Shah, Mubarak, University of Central Florida
- Abstract / Description
-
In this dissertation, we present vision based scene interpretation methods for monitoring of people and vehicles, in real-time, within a busy environment using a forest of co-operative electro-optical (EO) sensors. We have developed novel video understanding algorithms with learning capability, to detect and categorize people and vehicles, track them with in a camera and hand-off this information across multiple networked cameras for multi-camera tracking. The ability to learn prevents the...
Show moreIn this dissertation, we present vision based scene interpretation methods for monitoring of people and vehicles, in real-time, within a busy environment using a forest of co-operative electro-optical (EO) sensors. We have developed novel video understanding algorithms with learning capability, to detect and categorize people and vehicles, track them with in a camera and hand-off this information across multiple networked cameras for multi-camera tracking. The ability to learn prevents the need for extensive manual intervention, site models and camera calibration, and provides adaptability to changing environmental conditions. For object detection and categorization in the video stream, a two step detection procedure is used. First, regions of interest are determined using a novel hierarchical background subtraction algorithm that uses color and gradient information for interest region detection. Second, objects are located and classified from within these regions using a weakly supervised learning mechanism based on co-training that employs motion and appearance features. The main contribution of this approach is that it is an online procedure in which separate views (features) of the data are used for co-training, while the combined view (all features) is used to make classification decisions in a single boosted framework. The advantage of this approach is that it requires only a few initial training samples and can automatically adjust its parameters online to improve the detection and classification performance. Once objects are detected and classified they are tracked in individual cameras. Single camera tracking is performed using a voting based approach that utilizes color and shape cues to establish correspondence in individual cameras. The tracker has the capability to handle multiple occluded objects. Next, the objects are tracked across a forest of cameras with non-overlapping views. This is a hard problem because of two reasons. First, the observations of an object are often widely separated in time and space when viewed from non-overlapping cameras. Secondly, the appearance of an object in one camera view might be very different from its appearance in another camera view due to the differences in illumination, pose and camera properties. To deal with the first problem, the system learns the inter-camera relationships to constrain track correspondences. These relationships are learned in the form of multivariate probability density of space-time variables (object entry and exit locations, velocities, and inter-camera transition times) using Parzen windows. To handle the appearance change of an object as it moves from one camera to another, we show that all color transfer functions from a given camera to another camera lie in a low dimensional subspace. The tracking algorithm learns this subspace by using probabilistic principal component analysis and uses it for appearance matching. The proposed system learns the camera topology and subspace of inter-camera color transfer functions during a training phase. Once the training is complete, correspondences are assigned using the maximum a posteriori (MAP) estimation framework using both the location and appearance cues. Extensive experiments and deployment of this system in realistic scenarios has demonstrated the robustness of the proposed methods. The proposed system was able to detect and classify targets, and seamlessly tracked them across multiple cameras. It also generated a summary in terms of key frames and textual description of trajectories to a monitoring officer for final analysis and response decision. This level of interpretation was the goal of our research effort, and we believe that it is a significant step forward in the development of intelligent systems that can deal with the complexities of real world scenarios.
Show less - Date Issued
- 2005
- Identifier
- CFE0000497, ucf:46362
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000497
- Title
- TOWARDS A SELF-CALIBRATING VIDEO CAMERA NETWORK FOR CONTENT ANALYSIS AND FORENSICS.
- Creator
-
Junejo, Imran, Foroosh, Hassan, University of Central Florida
- Abstract / Description
-
Due to growing security concerns, video surveillance and monitoring has received an immense attention from both federal agencies and private firms. The main concern is that a single camera, even if allowed to rotate or translate, is not sufficient to cover a large area for video surveillance. A more general solution with wide range of applications is to allow the deployed cameras to have a non-overlapping field of view (FoV) and to, if possible, allow these cameras to move freely in 3D space....
Show moreDue to growing security concerns, video surveillance and monitoring has received an immense attention from both federal agencies and private firms. The main concern is that a single camera, even if allowed to rotate or translate, is not sufficient to cover a large area for video surveillance. A more general solution with wide range of applications is to allow the deployed cameras to have a non-overlapping field of view (FoV) and to, if possible, allow these cameras to move freely in 3D space. This thesis addresses the issue of how cameras in such a network can be calibrated and how the network as a whole can be calibrated, such that each camera as a unit in the network is aware of its orientation with respect to all the other cameras in the network. Different types of cameras might be present in a multiple camera network and novel techniques are presented for efficient calibration of these cameras. Specifically: (i) For a stationary camera, we derive new constraints on the Image of the Absolute Conic (IAC). These new constraints are shown to be intrinsic to IAC; (ii) For a scene where object shadows are cast on a ground plane, we track the shadows on the ground plane cast by at least two unknown stationary points, and utilize the tracked shadow positions to compute the horizon line and hence compute the camera intrinsic and extrinsic parameters; (iii) A novel solution to a scenario where a camera is observing pedestrians is presented. The uniqueness of formulation lies in recognizing two harmonic homologies present in the geometry obtained by observing pedestrians; (iv) For a freely moving camera, a novel practical method is proposed for its self-calibration which even allows it to change its internal parameters by zooming; and (v) due to the increased application of the pan-tilt-zoom (PTZ) cameras, a technique is presented that uses only two images to estimate five camera parameters. For an automatically configurable multi-camera network, having non-overlapping field of view and possibly containing moving cameras, a practical framework is proposed that determines the geometry of such a dynamic camera network. It is shown that only one automatically computed vanishing point and a line lying on any plane orthogonal to the vertical direction is sufficient to infer the geometry of a dynamic network. Our method generalizes previous work which considers restricted camera motions. Using minimal assumptions, we are able to successfully demonstrate promising results on synthetic as well as on real data. Applications to path modeling, GPS coordinate estimation, and configuring mixed-reality environment are explored.
Show less - Date Issued
- 2007
- Identifier
- CFE0001743, ucf:47296
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001743
- Title
- The paths less traveled: Movement of Gopher Tortoises (Gopherus polyphemus) along roads and railways.
- Creator
-
Rautsaw, Rhett, Parkinson, Christopher, Mansfield, Kate, Seigel, Richard, University of Central Florida
- Abstract / Description
-
Urbanization and an expanding human population have led to a large degree of habitat destruction and fragmentation. These, in turn, reduce biodiversity and wildlife population sizes on a global scale. Transportation infrastructure, such as roads and railways, are some of the largest contributors to habitat fragmentation. Roads are well-established to negatively impact wildlife, but some studies suggest a potential use in habitat connectivity by functioning as wildlife corridors to connect...
Show moreUrbanization and an expanding human population have led to a large degree of habitat destruction and fragmentation. These, in turn, reduce biodiversity and wildlife population sizes on a global scale. Transportation infrastructure, such as roads and railways, are some of the largest contributors to habitat fragmentation. Roads are well-established to negatively impact wildlife, but some studies suggest a potential use in habitat connectivity by functioning as wildlife corridors to connect distant populations. Railways are similarly known to impact wildlife by increasing mortality rates as well as provide unique risks such as electrocution and entrapment when compared to roads. However, the influence of railways on the movement and behavior of most taxa remains understudied. Here, I used Gopher Tortoises (Gopherus polyphemus) at the John F. Kennedy Space Center as a model system to (1) determine whether roadsides are or could be used as a wildlife corridor to connect distant habitats and (2) evaluate the impacts of railways on tortoise movement and behavior while providing management implications for both roads and railways.To examine the use of roadsides as wildlife corridors, I tracked the movement of individuals found along roadsides using radio-telemetry to determine if tortoises used the roadsides to move between inland and coastal habitat. In addition, I compared home range sizes of tortoises along roads to those of inland and coastal habitats to examine differences in spatial use patterns with regards to roads. I translocated tortoises from distant habitats into the roadside corridors to determine whether they would use the roadsides as a connective route to return to their original capture location. Overall, I determined that roadsides do not function as movement pathways, as even translocated tortoises remained along roads throughout the duration of the study. Instead, roads appear to function as long-term residential areas and potentially suitable habitat. I suggest management of roadsides to reduce mortality and further studies to examine the potential of roadsides acting as ecological traps.To study the impact of railways on tortoise movement and behavior I first used radio-telemetry to track the movement of tortoises found less than 100 m from railways. I simulated movement by using 1000 correlated random walks per tortoise to determine if the number of observed crossing events were significantly less than what would be expected by chance. Second, I measured behavior via continuous focal sampling for one hour to determine railway crossing ability and test for behavioral differences associated with the familiarity of the railways using a principal component analysis. Lastly, I tested if trenches dug underneath the rails could be used as a management strategy to alleviate the impact of railways on tortoises. I found that tortoises are unlikely to cross the railways and the number of observed crossing events were significantly less than what we would expect by chance. Additionally, familiarity with the railway did not have any influence on a tortoise's ability to cross nor their behavior. Trenches were frequently used to move from one side of the railway to the other and are, therefore, a valid management strategy to alleviate the impacts railways have on tortoise mortality, movement, and behavior.Overall, I conclude that transportation infrastructure and the expanding human population have immense impacts on wildlife, especially on turtles and tortoises. I recommend further research continue to identify unique management strategies as well as alternative barriers that may play a large role in a species' decline.
Show less - Date Issued
- 2017
- Identifier
- CFE0006954, ucf:51660
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006954
- Title
- Turtle Cam: Live Multimedia Interaction For Engaging Potential Visitor Population To Canaveral National Seashore.
- Creator
-
Tortorelli, Brian, Cabrera, Cheryl, Lindgren, Robb, Reedy, Robert, University of Central Florida
- Abstract / Description
-
This project expands the outreach of the Canaveral National Seashore to its visitors, potential visitors, and virtual visitors through its goals in conservancy and preservation of its natural resources. This paper is involved with the current iteration of a series of digital media projects, the Sea Turtle Nest Camera, also known as, Turtle Cam. It details how and why this project was designed to be an ongoing initiative to assist in those goals.
- Date Issued
- 2012
- Identifier
- CFE0004330, ucf:49446
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0004330
- Title
- Cross-Continental Insights into Jaguar (Panthera onca) Ecology and Conservation.
- Creator
-
Figel, Joseph, Noss, Reed, Quintana-Ascencio, Pedro, Jenkins, David, Quigley, Howard, University of Central Florida
- Abstract / Description
-
The jaguar (Panthera onca) is a widely distributed large carnivore and the focal species of a range-wide connectivity initiative known as the jaguar conservation network (JCN). Comprised of ~83 Jaguar Conservation Units (JCUs) and ~75 corridors from northern Mexico to Argentina, the JCN functions as a conduit for jaguar movement and gene flow. Key linkages in the network are imperiled by human population growth, large-scale agriculture, highway expansion, and other infrastructural development...
Show moreThe jaguar (Panthera onca) is a widely distributed large carnivore and the focal species of a range-wide connectivity initiative known as the jaguar conservation network (JCN). Comprised of ~83 Jaguar Conservation Units (JCUs) and ~75 corridors from northern Mexico to Argentina, the JCN functions as a conduit for jaguar movement and gene flow. Key linkages in the network are imperiled by human population growth, large-scale agriculture, highway expansion, and other infrastructural development. Labeled (")corridors of concern,(") these vulnerable linkages are imperative to the maintenance of connectivity and genetic diversity throughout jaguar distribution. I take a multi-faceted approach to analyze conservation issues and identify potential solutions in three of the most vulnerable connections of the JCN. I estimate densities and assess local residents' perceptions of jaguars in a fragmented JCU in western Mexico, analyze 3 years of data from 275 camera-trap sites to evaluate jaguar habitat use in a corridor of concern in Colombia, and quantify the umbrella value of jaguars for endemic herpetofauna in Nuclear Central America, a ~ 370,000 km(&)#178; sub-region of the Mesoamerican biodiversity hotspot. My research produces the first jaguar density estimate in a JCU containing human population densities (>)50 people/km2 and provides the strongest support for jaguar association with wetlands collected to date. In Nuclear Central America, one of the most important yet vulnerable areas of the JCN, I demonstrate the umbrella value of this wide-ranging felid. I conclude with a discussion on the need to reevaluate extirpation thresholds of jaguars in human-use landscapes, to direct more research on wetlands as keystone habitats for jaguars, and to further assess the utility of umbrella analyses using jaguars as focal species to support holistic conservation planning.
Show less - Date Issued
- 2017
- Identifier
- CFE0006591, ucf:51258
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006591
- Title
- Investigation of infrared thermography for subsurface damage detection of concrete structures.
- Creator
-
Hiasa, Shuhei, Catbas, Necati, Tatari, Omer, Nam, Boo Hyun, Zaurin, Ricardo, Xanthopoulos, Petros, University of Central Florida
- Abstract / Description
-
Deterioration of road infrastructure arises from aging and various other factors. Consequently, inspection and maintenance have been a serious worldwide problem. In the United States, degradation of concrete bridge decks is a widespread problem among several bridge components. In order to prevent the impending degradation of bridges, periodic inspection and proper maintenance are indispensable. However, the transportation system faces unprecedented challenges because the number of aging...
Show moreDeterioration of road infrastructure arises from aging and various other factors. Consequently, inspection and maintenance have been a serious worldwide problem. In the United States, degradation of concrete bridge decks is a widespread problem among several bridge components. In order to prevent the impending degradation of bridges, periodic inspection and proper maintenance are indispensable. However, the transportation system faces unprecedented challenges because the number of aging bridges is increasing under limited resources, both in terms of budget and personnel. Therefore, innovative technologies and processes that enable bridge owners to inspect and evaluate bridge conditions more effectively and efficiently with less human and monetary resources are desired. Traditionally, qualified engineers and inspectors implemented hammer sounding and/or chain drag, and visual inspection for concrete bridge deck evaluations, but these methods require substantial field labor, experience, and lane closures for bridge deck inspections. Under these circumstances, Non-Destructive Evaluation (NDE) techniques such as computer vision-based crack detection, impact echo (IE), ground-penetrating radar (GPR) and infrared thermography (IRT) have been developed to inspect and monitor aging and deteriorating structures rapidly and effectively. However, no single method can detect all kinds of defects in concrete structures as well as the traditional inspection combination of visual and sounding inspections; hence, there is still no international standard NDE methods for concrete bridges, although significant progress has been made up to the present.This research presents the potential to reduce a burden of bridge inspections, especially for bridge decks, in place of traditional chain drag and hammer sounding methods by IRT with the combination of computer vision-based technology. However, there were still several challenges and uncertainties in using IRT for bridge inspections. This study revealed those challenges and uncertainties, and explored those solutions, proper methods and ideal conditions for applying IRT in order to enhance the usability, reliability and accuracy of IRT for concrete bridge inspections. Throughout the study, detailed investigations of IRT are presented. Firstly, three different types of infrared (IR) cameras were compared under active IRT conditions in the laboratory to examine the effect of photography angle on IRT along with the specifications of cameras. The results showed that when IR images are taken from a certain angle, each camera shows different temperature readings. However, since each IR camera can capture temperature differences between sound and delaminated areas, they have a potential to detect delaminated areas under a given condition in spite of camera specifications even when they are utilized from a certain angle. Furthermore, a more objective data analysis method than just comparing IR images was explored to assess IR data. Secondly, coupled structural mechanics and heat transfer models of concrete blocks with artificial delaminations used for a field test were developed and analyzed to explore sensitive parameters for effective utilization of IRT. After these finite element (FE) models were validated, critical parameters and factors of delamination detectability such as the size of delamination (area, thickness and volume), ambient temperature and sun loading condition (different season), and the depth of delamination from the surface were explored. This study presents that the area of delamination is much more influential in the detectability of IRT than thickness and volume. It is also found that there is no significant difference depending on the season when IRT is employed. Then, FE model simulations were used to obtain the temperature differences between sound and delaminated areas in order to process IR data. By using this method, delaminated areas of concrete slabs could be detected more objectively than by judging the color contrast of IR images. However, it was also found that the boundary condition affects the accuracy of this method, and the effect varies depending on the data collection time. Even though there are some limitations, integrated use of FE model simulation with IRT showed that the combination can be reduce other pre-tests on bridges, reduce the need to have access to the bridge and also can help automate the IRT data analysis process for concrete bridge deck inspections. After that, the favorable time windows for concrete bridge deck inspections by IRT were explored through field experiment and FE model simulations. Based on the numerical simulations and experimental IRT results, higher temperature differences in the day were observed from both results around noontime and nighttime, although IRT is affected by sun loading during the daytime heating cycle resulting in possible misdetections. Furthermore, the numerical simulations show that the maximum effect occurs at night during the nighttime cooling cycle, and the temperature difference decreases gradually from that time to a few hours after sunrise of the next day. Thus, it can be concluded that the nighttime application of IRT is the most suitable time window for bridge decks. Furthermore, three IR cameras with different specifications were compared to explore several factors affecting the utilization of IRT in regards to subsurface damage detection in concrete structures, specifically when the IRT is utilized for high-speed bridge deck inspections at normal driving speeds under field laboratory conditions. The results show that IRT can detect up to 2.54 cm delamination from the concrete surface at any time period. This study revealed two important factors of camera specifications for high-speed inspection by IRT as shorter integration time and higher pixel resolution.Finally, a real bridge was scanned by three different types of IR cameras and the results were compared with other NDE technologies that were implemented by other researchers on the same bridge. When compared at fully documented locations with 8 concrete cores, a high-end IR camera with cooled detector distinguished sound and delaminated areas accurately. Furthermore, indicated location and shape of delaminations by three IR cameras were compared to other NDE methods from past research, and the result revealed that the cooled camera showed almost identical shapes to other NDE methods including chain drag. It should be noted that the data were collected at normal driving speed without any lane closures, making it a more practical and faster method than other NDE technologies. It was also presented that the factor most likely to affect high-speed application is integration time of IR camera as well as the conclusion of the field laboratory test.The notable contribution of this study for the improvement of IRT is that this study revealed the preferable conditions for IRT, specifically for high-speed scanning of concrete bridge decks. This study shows that IRT implementation under normal driving speeds has high potential to evaluate concrete bridge decks accurately without any lane closures much more quickly than other NDE methods, if a cooled camera equipped with higher pixel resolution is used during nighttime. Despite some limitations of IRT, the data collection speed is a great advantage for periodic bridge inspections compared to other NDE methods. Moreover, there is a high possibility to reduce inspection time, labor and budget drastically if high-speed bridge deck scanning by the combination of IRT and computer vision-based technology becomes a standard bridge deck inspection method. Therefore, the author recommends combined application of the high-speed scanning combination and other NDE methods to optimize bridge deck inspections.
Show less - Date Issued
- 2016
- Identifier
- CFE0006323, ucf:51575
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006323
- Title
- Robust Subspace Estimation Using Low-Rank Optimization. Theory and Applications in Scene Reconstruction, Video Denoising, and Activity Recognition.
- Creator
-
Oreifej, Omar, Shah, Mubarak, Da Vitoria Lobo, Niels, Stanley, Kenneth, Lin, Mingjie, Li, Xin, University of Central Florida
- Abstract / Description
-
In this dissertation, we discuss the problem of robust linear subspace estimation using low-rank optimization and propose three formulations of it. We demonstrate how these formulations can be used to solve fundamental computer vision problems, and provide superior performance in terms of accuracy and running time.Consider a set of observations extracted from images (such as pixel gray values, local features, trajectories...etc). If the assumption that these observations are drawn from a...
Show moreIn this dissertation, we discuss the problem of robust linear subspace estimation using low-rank optimization and propose three formulations of it. We demonstrate how these formulations can be used to solve fundamental computer vision problems, and provide superior performance in terms of accuracy and running time.Consider a set of observations extracted from images (such as pixel gray values, local features, trajectories...etc). If the assumption that these observations are drawn from a liner subspace (or can be linearly approximated) is valid, then the goal is to represent each observation as a linear combination of a compact basis, while maintaining a minimal reconstruction error. One of the earliest, yet most popular, approaches to achieve that is Principal Component Analysis (PCA). However, PCA can only handle Gaussian noise, and thus suffers when the observations are contaminated with gross and sparse outliers. To this end, in this dissertation, we focus on estimating the subspace robustly using low-rank optimization, where the sparse outliers are detected and separated through the `1 norm. The robust estimation has a two-fold advantage: First, the obtained basis better represents the actual subspace because it does not include contributions from the outliers. Second, the detected outliers are often of a specific interest in many applications, as we will show throughout this thesis. We demonstrate four different formulations and applications for low-rank optimization. First, we consider the problem of reconstructing an underwater sequence by removing the turbulence caused by the water waves. The main drawback of most previous attempts to tackle this problem is that they heavily depend on modelling the waves, which in fact is ill-posed since the actual behavior of the waves along with the imaging process are complicated and include several noise components; therefore, their results are not satisfactory. In contrast, we propose a novel approach which outperforms the state-of-the-art. The intuition behind our method is that in a sequence where the water is static, the frames would be linearly correlated. Therefore, in the presence of water waves, we may consider the frames as noisy observations drawn from a the subspace of linearly correlated frames. However, the noise introduced by the water waves is not sparse, and thus cannot directly be detected using low-rank optimization. Therefore, we propose a data-driven two-stage approach, where the first stage (")sparsifies(") the noise, and the second stage detects it. The first stage leverages the temporal mean of the sequence to overcome the structured turbulence of the waves through an iterative registration algorithm. The result of the first stage is a high quality mean and a better structured sequence; however, the sequence still contains unstructured sparse noise. Thus, we employ a second stage at which we extract the sparse errors from the sequence through rank minimization. Our method converges faster, and drastically outperforms state of the art on all testing sequences. Secondly, we consider a closely related situation where an independently moving object is also present in the turbulent video. More precisely, we consider video sequences acquired in a desert battlefields, where atmospheric turbulence is typically present, in addition to independently moving targets. Typical approaches for turbulence mitigation follow averaging or de-warping techniques. Although these methods can reduce the turbulence, they distort the independently moving objects which can often be of great interest. Therefore, we address the problem of simultaneous turbulence mitigation and moving object detection. We propose a novel three-term low-rank matrix decomposition approach in which we decompose the turbulence sequence into three components: the background, the turbulence, and the object. We simplify this extremely difficult problem into a minimization of nuclear norm, Frobenius norm, and L1 norm. Our method is based on two observations: First, the turbulence causes dense and Gaussian noise, and therefore can be captured by Frobenius norm, while the moving objects are sparse and thus can be captured by L1 norm. Second, since the object's motion is linear and intrinsically different than the Gaussian-like turbulence, a Gaussian-based turbulence model can be employed to enforce an additional constraint on the search space of the minimization. We demonstrate the robustness of our approach on challenging sequences which are significantly distorted with atmospheric turbulence and include extremely tiny moving objects. In addition to robustly detecting the subspace of the frames of a sequence, we consider using trajectories as observations in the low-rank optimization framework. In particular, in videos acquired by moving cameras, we track all the pixels in the video and use that to estimate the camera motion subspace. This is particularly useful in activity recognition, which typically requires standard preprocessing steps such as motion compensation, moving object detection, and object tracking. The errors from the motion compensation step propagate to the object detection stage, resulting in miss-detections, which further complicates the tracking stage, resulting in cluttered and incorrect tracks. In contrast, we propose a novel approach which does not follow the standard steps, and accordingly avoids the aforementioned difficulties. Our approach is based on Lagrangian particle trajectories which are a set of dense trajectories obtained by advecting optical flow over time, thus capturing the ensemble motions of a scene. This is done in frames of unaligned video, and no object detection is required. In order to handle the moving camera, we decompose the trajectories into their camera-induced and object-induced components. Having obtained the relevant object motion trajectories, we compute a compact set of chaotic invariant features, which captures the characteristics of the trajectories. Consequently, a SVM is employed to learn and recognize the human actions using the computed motion features. We performed intensive experiments on multiple benchmark datasets, and obtained promising results.Finally, we consider a more challenging problem referred to as complex event recognition, where the activities of interest are complex and unconstrained. This problem typically pose significant challenges because it involves videos of highly variable content, noise, length, frame size ... etc. In this extremely challenging task, high-level features have recently shown a promising direction as in [53, 129], where core low-level events referred to as concepts are annotated and modeled using a portion of the training data, then each event is described using its content of these concepts. However, because of the complex nature of the videos, both the concept models and the corresponding high-level features are significantly noisy. In order to address this problem, we propose a novel low-rank formulation, which combines the precisely annotated videos used to train the concepts, with the rich high-level features. Our approach finds a new representation for each event, which is not only low-rank, but also constrained to adhere to the concept annotation, thus suppressing the noise, and maintaining a consistent occurrence of the concepts in each event. Extensive experiments on large scale real world dataset TRECVID Multimedia Event Detection 2011 and 2012 demonstrate that our approach consistently improves the discriminativity of the high-level features by a significant margin.
Show less - Date Issued
- 2013
- Identifier
- CFE0004732, ucf:49835
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0004732
