Field programmable gate array based target detection and gesture recognition

The move from Standard Definition (SD) to High Definition (HD) represents a six times increases in data, which needs to be processed. With expanding resolutions and evolving compression, there is a need for high performance with flexible architectures to allow for quick upgrade ability. The technology advances in image display resolutions, advanced compression techniques, and video intelligence. Software implementation of these systems can attain accuracy with tradeoffs among processing performance (to achieve specified frame rates, working on large image data sets), power and cost constraints. There is a need for new architectures to be in pace with the fast innovations in video and imaging. It contains dedicated hardware implementation of the pixel and frame rate processes on Field Programmable Gate Array (FPGA) to achieve the real-time performance. ^ The following outlines the contributions of the dissertation. (1) We develop a target detection system by applying a novel running average mean threshold (RAMT) approach to globalize the threshold required for background subtraction. This approach adapts the threshold automatically to different environments (indoor and outdoor) and different targets (humans and vehicles). For low power consumption and better performance, we design the complete system on FPGA. (2) We introduce a safe distance factor and develop an algorithm for occlusion occurrence detection during target tracking. A novel mean-threshold is calculated by motion-position analysis. (3) A new strategy for gesture recognition is developed using Combinational Neural Networks (CNN) based on a tree structure. Analysis of the method is done on American Sign Language (ASL) gestures. We introduce novel point of interests approach to reduce the feature vector size and gradient threshold approach for accurate classification. (4) We design a gesture recognition system using a hardware/ software co-simulation neural network for high speed and low memory storage requirements provided by the FPGA. We develop an innovative maximum distant algorithm which uses only 0.39% of the image as the feature vector to train and test the system design. Database set gestures involved in different applications may vary. Therefore, it is highly essential to keep the feature vector as low as possible while maintaining the same accuracy and performance^

A model for continuous measurement of drilled shaft diameter during construction

Non-Destructive Testing (NDT) of deep foundations has become an integral part of the industry's standard manufacturing processes. It is not unusual for the evaluation of the integrity of the concrete to include the measurement of ultrasonic wave speeds. Numerous methods have been proposed that use the propagation speed of ultrasonic waves to check the integrity of concrete for drilled shaft foundations. All such methods evaluate the integrity of the concrete inside the cage and between the access tubes. The integrity of the concrete outside the cage remains to be considered to determine the location of the border between the concrete and the soil in order to obtain the diameter of the drilled shaft. It is also economic to devise a methodology to obtain the diameter of the drilled shaft using the Cross-Hole Sonic Logging system (CSL). Performing such a methodology using the CSL and following the CSL tests is performed and used to check the integrity of the inside concrete, thus allowing the determination of the drilled shaft diameter without having to set up another NDT device.^ This proposed new method is based on the installation of galvanized tubes outside the shaft across from each inside tube, and performing the CSL test between the inside and outside tubes. From the performed experimental work a model is developed to evaluate the relationship between the thickness of concrete and the ultrasonic wave properties using signal processing. The experimental results show that there is a direct correlation between concrete thicknesses outside the cage and maximum amplitude of the received signal obtained from frequency domain data. This study demonstrates how this new method to measuring the diameter of drilled shafts during construction using a NDT method overcomes the limitations of currently-used methods. ^ In the other part of study, a new method is proposed to visualize and quantify the extent and location of the defects. It is based on a color change in the frequency amplitude of the signal recorded by the receiver probe in the location of defects and it is called Frequency Tomography Analysis (FTA). Time-domain data is transferred to frequency-domain data of the signals propagated between tubes using Fast Fourier Transform (FFT). Then, distribution of the FTA will be evaluated. This method is employed after CSL has determined the high probability of an anomaly in a given area and is applied to improve location accuracy and to further characterize the feature. The technique has a very good resolution and clarifies the exact depth location of any void or defect through the length of the drilled shaft for the voids inside the cage. ^ The last part of study also evaluates the effect of voids inside and outside the reinforcement cage and corrosion in the longitudinal bars on the strength and axial load capacity of drilled shafts. The objective is to quantify the extent of loss in axial strength and stiffness of drilled shafts due to presence of different types of symmetric voids and corrosion throughout their lengths.^

Dynamic image precompensation for improving visual performance of computer users with ocular aberrations

With the progress of computer technology, computers are expected to be more intelligent in the interaction with humans, presenting information according to the user's psychological and physiological characteristics. However, computer users with visual problems may encounter difficulties on the perception of icons, menus, and other graphical information displayed on the screen, limiting the efficiency of their interaction with computers. In this dissertation, a personalized and dynamic image precompensation method was developed to improve the visual performance of the computer users with ocular aberrations. The precompensation was applied on the graphical targets before presenting them on the screen, aiming to counteract the visual blurring caused by the ocular aberration of the user's eye. A complete and systematic modeling approach to describe the retinal image formation of the computer user was presented, taking advantage of modeling tools, such as Zernike polynomials, wavefront aberration, Point Spread Function and Modulation Transfer Function. The ocular aberration of the computer user was originally measured by a wavefront aberrometer, as a reference for the precompensation model. The dynamic precompensation was generated based on the resized aberration, with the real-time pupil diameter monitored. The potential visual benefit of the dynamic precompensation method was explored through software simulation, with the aberration data from a real human subject. An "artificial eye'' experiment was conducted by simulating the human eye with a high-definition camera, providing objective evaluation to the image quality after precompensation. In addition, an empirical evaluation with 20 human participants was also designed and implemented, involving image recognition tests performed under a more realistic viewing environment of computer use. The statistical analysis results of the empirical experiment confirmed the effectiveness of the dynamic precompensation method, by showing significant improvement on the recognition accuracy. The merit and necessity of the dynamic precompensation were also substantiated by comparing it with the static precompensation. The visual benefit of the dynamic precompensation was further confirmed by the subjective assessments collected from the evaluation participants.

Digital signal processing implementation for near real-time wavelet transformation system for binary images

Communication has become an essential function in our civilization. With the increasing demand for communication channels, it is now necessary to find ways to optimize the use of their bandwidth. One way to achieve this is by transforming the information before it is transmitted. This transformation can be performed by several techniques. One of the newest of these techniques is the use of wavelets. Wavelet transformation refers to the act of breaking down a signal into components called details and trends by using small waveforms that have a zero average in the time domain. After this transformation the data can be compressed by discarding the details, transmitting the trends. In the receiving end, the trends are used to reconstruct the image. In this work, the wavelet used for the transformation of an image will be selected from a library of available bases. The accuracy of the reconstruction, after the details are discarded, is dependent on the wavelets chosen from the wavelet basis library. The system developed in this thesis takes a 2-D image and decomposes it using a wavelet bank. A digital signal processor is used to achieve near real-time performance in this transformation task. A contribution of this thesis project is the development of DSP-based test bed for the future development of new real-time wavelet transformation algorithms.

Simulação da destilação molecular de filme descendente para o petróleo

The heavy part of the oil can be used for numerous purposes, e.g. to obtain lubricating oils. In this context, many researchers have been studying alternatives such separation of crude oil components, among which may be mentioned molecular distillation. Molecular distillation is a forced evaporation technique different from other conventional processes in the literature. This process can be classified as a special distillation case under high vacuum with pressures that reach extremely low ranges of the order of 0.1 Pascal. The evaporation and condensation surfaces must have a distance from each other of the magnitude order of mean free path of the evaporated molecules, that is, molecules evaporated easily reach the condenser, because they find a route without obstacles, what is desirable. Thus, the main contribution of this work is the simulation of the falling-film molecular distillation for crude oil mixtures. The crude oil was characterized using UniSim® Design and R430 Aspen HYSYS® V8.5. The results of this characterization were performed in spreadsheets of Microsoft® Excel®, calculations of the physicochemical properties of the waste of an oil sample, i.e., thermodynamic and transport. Based on this estimated properties and boundary conditions suggested by the literature, equations of temperature and concentration profiles were resolved through the implicit finite difference method using the programming language Visual Basic® (VBA) for Excel®. The result of the temperature profile showed consistent with the reproduced by literature, having in their initial values a slight distortion as a result of the nature of the studied oil is lighter than the literature, since the results of the concentration profiles were effective allowing realize that the concentration of the more volatile decreases and of the less volatile increases due to the length of the evaporator. According to the transport phenomena present in the process, the velocity profile tends to increase to a peak and then decreases, and the film thickness decreases, both as a function of the evaporator length. It is concluded that the simulation code in Visual Basic® language (VBA) is a final product of the work that allows application to molecular distillation of petroleum and other similar mixtures.

Seabed photographs taken along OFOS profiles during POLARSTERN cruise PS96 (ANT-XXXI/2 FROSN)

Within the context of the overall ecological working programme Dynamics of Antarctic Marine Shelf Ecosystems (DynAMo) of the PS96 (ANT-XXXI/2) cruise of RV "Polarstern" to the Weddell Sea (Dec 2015 to Feb 2016), seabed imaging surveys were carried out along drift profiles by means of the Ocean Floor Observation System (OFOS) of the Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI) Bremerhaven. The setup and mode of deployment of the OFOS was similar to that described by Bergmann and Klages (2012, doi:10.1016/j.marpolbul.2012.09.018). OFOS is a surface-powered gear equipped with two downward-looking cameras installed side-by-side: one high-resolution, wide-angle still camera (CANON® EOS 5D Mark III; lens: Canon EF 24 f/1.4L II, f stop: 13, exposure time: 1/125 sec; in-air view angles: 74° (horizontal), 53° (vertical), 84° (diagonal); image size: 5760 x 3840 px = 21 MPix; front of pressure resistant camera housing consisting of plexiglass dome port) and one high-definition color video camera (SONY® FCB-H11). The system was vertically lowered over the stern of the ship with a broadband fibre-optic cable, until it hovers approximately 1.5 m above the seabed. It was then towed after the slowly sailing ship at a speed of approximately 0.5 kn (0.25 m/s). The ship's Global Acoustic Positioning System (GAPS), combining Ultra Short Base Line (USBL), Inertial Navigation System (INS) and satellite-based Global Positioning System (GPS) technologies, was used to gain highly precise underwater position data of the OFOS. During the profile, OFOS was kept hanging at the preferred height above the seafloor by means of the live video feed and occasional minor cable-length adjustments with the winch to compensate small-scale bathymetric variations in seabed morphology. Information on water depth and height above the seafloor were continuously recorded by means of OFOS-mounted sensors (GAPS transponder, Tritech altimeter). Three lasers, which are placed beside the still camera, emit parallel beams and project red light points, arranged as an equilateral triangle with a side length of 50 cm, in each photo, thus providing a scale that can be used to calculate the seabed area depicted in each image and/or measure the size of organisms or seabed features visible in the image. In addition, the seabed area depicted was estimated using altimeter-derived height above seafloor and optical characteristics of the OFOS still camera. In automatic mode, a seabed photo, depicting an area of approximately 3.45 m**2 (= 2.3 m x 1.5 m; with variations depending on the actual height above ground), was taken every 30 seconds to obtain series of "TIMER" stills distributed at regular distances along the profiles that vary in length depending on duration of the cast. At a ship speed of 0.5 kn, the average distance between seabed images was approximately 5 m. Additional "HOTKEY" photos were taken from interesting objects (organisms, seabed features, such as putative iceberg scours) when they appeared in the live video feed (which was also recorded, in addition to the stills, for documentation and possible later analysis). If any image from this collection is used, please cite the reference as given above.

Particle positions of numerical experiments 1 to 4

In mixed sediment beds, erosion resistance can change relative to that of beds composed of a uniform sediment because of varying textural and/or other grain-size parameters, with effects on pore water flow that are difficult to quantify by means of analogue techniques. To overcome this difficulty, a three-dimensional numerical model was developed using a finite difference method (FDM) flow model coupled with a distinct element method (DEM) particle model. The main aim was to investigate, at a high spatial resolution, the physical processes occurring during the initiation of motion of single grains at the sediment-water interface and in the shallow subsurface of simplified sediment beds under different flow velocities. Increasing proportions of very fine sand (D50=0.08 mm) were mixed into a coarse sand matrix (D50=0.6 mm) to simulate mixed sediment beds, starting with a pure coarse sand bed in experiment 1 (0 wt% fines), and proceeding through experiment 2 (6.5 wt% fines), experiment 3 (10.5 wt% fines), and experiment 4 (28.7 wt% fines). All mixed beds were tested for their erosion behavior at predefined flow velocities varying in the range of U 1-5=10-30 cm/s. The experiments show that, with increasing fine content, the smaller particles increasingly fill the spaces between the larger particles. As a consequence, pore water inflow into the sediment is increasingly blocked, i.e., there is a decrease in pore water flow velocity and, hence, in the flow momentum available to entrain particles. These findings are portrayed in a new conceptual model of enhanced sediment bed stabilization.

Stress Inversion Using LiDAR Tunnel Deformation Measurements

Far-field stresses are those present in a volume of rock prior to excavations being created. Estimates of the orientation and magnitude of far-field stresses, often used in mine design, are generally obtained by single-point measurements of stress, or large-scale, regional trends. Point measurements can be a poor representation of far-field stresses as a result of excavation-induced stresses and geological structures. For these reasons, far-field stress estimates can be associated with high levels of uncertainty. The purpose of this thesis is to investigate the practical feasibility, applications, and limitations of calibrating far-field stress estimates through tunnel deformation measurements captured using LiDAR imaging. A method that estimates the orientation and magnitude of excavation-induced principal stress changes through back-analysis of deformation measurements from LiDAR imaged tunnels was developed and tested using synthetic data. If excavation-induced stress change orientations and magnitudes can be accurately estimated, they can be used in the calibration of far-field stress input to numerical models. LiDAR point clouds have been proven to have a number of underground applications, thus it is desired to explore their use in numerical model calibration. The back-analysis method is founded on the superposition of stresses and requires a two-dimensional numerical model of the deforming tunnel. Principal stress changes of known orientation and magnitude are applied to the model to create calibration curves. Estimation can then be performed by minimizing squared differences between the measured tunnel and sets of calibration curve deformations. In addition to the back-analysis estimation method, a procedure consisting of previously existing techniques to measure tunnel deformation using LiDAR imaging was documented. Under ideal conditions, the back-analysis method estimated principal stress change orientations within ±5° and magnitudes within ±2 MPa. Results were comparable for four different tunnel profile shapes. Preliminary testing using plastic deformation, a rough tunnel profile, and profile occlusions suggests that the method can work under more realistic conditions. The results from this thesis set the groundwork for the continued development of a new, inexpensive, and efficient far-field stress estimate calibration method.

ThreeDimensional Epigraphic Recording at Stobi (Former Yugoslav Republic of Macedonia): Creating a Virtual Lapidarium

There is an increased need for 3D recording of archaeological sites and digital preservation of their artifacts. Digital photogrammetry with prosumer DSLR cameras is a suitable tool for recording epigraphy in particular, as it allows for the recording of inscribed surfaces with very high accuracy, often better than 2 mm and with only a short time spent in the field. When photogrammetry is fused with other computational photography techniques like panoramic tours and Reflectance Transformation Imaging, a workflow exists to rival traditional LiDAR­based methods. The difficulty however, arises in the presentation of 3D data. It requires an enormous amount of storage and end­user sophistication. The proposed solution is to use game­engine technology and high definition virtual tours to provide not only scholars, but also the general public with an uncomplicated interface to interact with the detailed 3D epigraphic data. The site of Stobi, located near Gradsko, in the Former Yugoslav Republic of Macedonia (FYROM) was used as a case study to demonstrate the effectiveness of RTI, photogrammetry and virtual tour imaging working in combination. A selection of nine sets of inscriptions from the archaeological site were chosen to demonstrate the range of application for the techniques. The chosen marble, sandstone and breccia inscriptions are representative of the varying levels of deterioration and degradation of the epigraphy at Stobi, in which both their rates of decay and resulting legibility is varied. This selection includes those which are treated and untreated stones as well as those in situ and those in storage. The selection consists of both Latin and Greek inscriptions with content ranging from temple dedication inscriptions to statue dedications. This combination of 3D modeling techniques presents a cost and time efficient solution to both increase the legibility of severely damaged stones and to digitally preserve the current state of the inscriptions.

Examination of body image among Canadian adolescents: Relations with physical activity and screen time

Background: Adolescence is a period of life associated with self-perceptions of negative body image. Physical activity levels are low and screen time levels are also high during this stage. These perceptions and behaviours are associated with poor health outcomes, making research on their determinants important. With adolescent populations, certain groups may be at higher risk of body dissatisfaction than others, and body dissatisfaction may influence individual physical activity and screen time levels. Objectives: The objectives of this thesis were to: 1) describe body image among young Canadians, examining possible health inequalities 2) estimate the strength and significance of associations between body satisfaction, physical activity and screen time, and 3) examine the potential etiological role of biological sex. Methods: Objective 1: The 2013/2014 Health Behaviour in School-aged Children study was employed. Sex-stratified Rao-Scott chi-square analyses were conducted to examine associations between socio-demographic factors and body satisfaction. Objective 2: The 2005/2006 and 2013/2014 cross-sectional and 2006 longitudinal HBSC data sets were used. Sex-stratified modified Poisson regressions were conducted and risk estimates and associated confidence intervals obtained. Results: Objective 1: Among males, being older, of East and Southeast Asian ethnicity, and reporting low SES all were associated with body dissatisfaction. Among females, being older, of Arab and West Asian or African ethnicity, being born in Canada, and reporting low SES were all associated with being body dissatisfied. Objective 2: Cross-sectionally, males who reported ‘too fat’ body dissatisfaction were more likely to be physically inactive. Adolescents of both sexes who reported ‘too fat’ body dissatisfaction were more likely to engage in high levels of screen time. Data from the longitudinal component supported the idea that male ‘too fat’ body dissatisfaction temporally leads to physical inactivity, but showed an inverse relationship between body dissatisfaction and screen time. Conclusions: Objective 1: Future prevention efforts in Canada should target subgroups to effectively help those at greatest risk of body dissatisfaction, and ameliorate potential inequalities at the population level. Objective 2: The presence of these relationships may inform future interventions as part of a multi-factorial etiology, in order to increase physical activity and decrease screen time among youth.

An exploration of students’ perceptions and attitudes towards creativity in engineering education

This study used a mixed methods approach to develop a broad and deep understanding of students’ perceptions towards creativity in engineering education. Studies have shown that students’ attitudes can have an impact on their motivation to engage in creative behavior. Using an ex-post facto independent factorial design, attitudes of value towards creativity, time for creativity, and creativity stereotypes were measured and compared across gender, year of study, engineering discipline, preference for open-ended problem solving, and confidence in creative abilities. Participants were undergraduate engineering students at Queen’s University from all years of study. A qualitative phenomenological methodology was adopted to study students’ understandings and experiences with engineering creativity. Eleven students participated in oneon- one interviews that provided depth and insight into how students experience and define engineering creativity, and the survey included open-ended items developed using the 10 Maxims of Creativity in Education as a guiding framework. The findings from the survey suggested that students had high value for creativity, however students in fourth year or higher had less value than those in other years. Those with preference for open-ended problem solving and high confidence valued creative more than their counterparts. Students who preferred open-ended problem solving and students with high confidence reported that time was less of a hindrance to their creativity. Males identified more with creativity stereotypes than females, however overall they were both low. Open-ended survey and interview results indicated that students felt they experienced creativity in engineering design activities. Engineering creativity definitions had two elements: creative action and creative characteristic. Creative actions were associated with designing, and creative characteristics were predominantly associated with novelty. Other barriers that emerged from the qualitative analysis were lack of opportunity, lack of assessment, and discomfort with creativity. It was concluded that a universal definition is required to establish clear and aligned understandings of engineering creativity. Instructors may want to consider demonstrating value by assessing creativity and establishing clear criteria in design projects. It is recommended that students be given more opportunities for practice through design activities and that they be introduced to design and creative thinking concepts early in their engineering education.

Differential contributions of the mIPS and PMd in the movement planning of reaches investigated through HD-tDCS

To reach for a target, we must formulate a movement plan - a difference vector of the target position with respect to the starting hand position. While it is known that the medial part of the intraparietal sulcus (mIPS) and the dorsal premotor (PMd) activity reflects aspects of a kinematic plan for a reaching movement, it is unclear whether or how the two regions may differ. We investigated the functional roles of the mIPS and PMd in the planning of reaching movements using high definition transcranial direct current stimulation (HD-tDCS) and examined changes in horizontal endpoint error when participants were subjected to anodal and cathodal stimulation. The left mIPS and PMd were functionally localized with fMRI in each participant using an interleaved center-out pointing and saccade task and mapped onto the scalp using Brainsight. We adopted a randomized, single-blind design and applied anodal and cathodal stimulation (2mA for 20 min; 3cm radius 4x1 electrode placement) during 4 separate visits scheduled at least a week apart. Each participant performed 250 baseline, stimulation, and post-stimulation memory-guided reaches starting from one of two initial hand positions (IHPs) to one of 4 briefly flashed targets (20 cm distant, 5 cm apart horizontally) while fixating on a straight-ahead cross located at the target line. Separate 2-way repeated measures ANOVAs of horizontal endpoint error difference after cathodal tDCS at each stimulation site revealed a significant IHP by target position interaction effect at the left mIPS, and significant IHP and target main effects at the left PMd. Behaviorally, these effects corresponded to IHP-dependent contractions after cathodal mIPS tDCS and IHP-independent contractions after cathodal PMd tDCS. These results suggest that the movement vector is not yet formed at the input level of mIPS, but is encoded at the input of PMd. These results also indicate that tDCS is a viable, useful method in investigating movement planning properties through temporary perturbations of the system.

A Bias Compensated Cross-Relation approach to Thermocouple Characterisation

The measurement of fast changing temperature fluctuations is a challenging problem due to the inherent limited bandwidth of temperature sensors. This results in a measured signal that is a lagged and attenuated version of the input. Compensation can be performed provided an accurate, parameterised sensor model is available. However, to account for the in influence of the measurement environment and changing conditions such as gas velocity, the model must be estimated in-situ. The cross-relation method of blind deconvolution is one approach for in-situ characterisation of sensors. However, a drawback with the method is that it becomes positively biased and unstable at high noise levels. In this paper, the cross-relation method is cast in the discrete-time domain and a bias compensation approach is developed. It is shown that the proposed compensation scheme is robust and yields unbiased estimates with lower estimation variance than the uncompensated version. All results are verified using Monte-Carlo simulations.

A Bias Compensated Cross-Relation approach to Thermocouple Characterisation

The measurement of fast changing temperature fluctuations is a challenging problem due to the inherent limited bandwidth of temperature sensors. This results in a measured signal that is a lagged and attenuated version of the input. Compensation can be performed provided an accurate, parameterised sensor model is available. However, to account for the influence of the measurement environment and changing conditions such as gas velocity, the model must be estimated in-situ. The cross-relation method of blind deconvolution is one approach for in-situ characterisation of sensors. However, a drawback with the method is that it becomes positively biased and unstable at high noise levels. In this paper, the cross-relation method is cast in the discrete-time domain and a bias compensation approach is developed. It is shown that the proposed compensation scheme is robust and yields unbiased estimates with lower estimation variance than the uncompensated version. All results are verified using Monte-Carlo simulations.

Ultrassonic Analysis of UO2 Pellets

Ceramic materials have been widely used for various purposes in many different industries due to certain characteristics, such as high melting point and high resistance to corrosion. In the nuclear area, ceramics are of great importance due to the process of fabrication of fuel pellets for nuclear reactors. Generally, high accuracy destructive techniques are used to characterize nuclear materials for fuel fabrication. These techniques usually require costly equipment and facilities, as well as experienced personnel. This paper aims at presenting an analysis methodology for UO2 pellets using a non-destructive ultrasonic technique for porosity measurement. This technique differs from traditional ultrasonic techniques in the sense it uses ultrasonic pulses in frequency domain instead of time domain. Therefore, specific characteristics of the analyzed material are associated with the obtained frequency spectrum. In the present work, four fuel grade UO2 pellets were analyzed and the corresponding results evaluated.