Consistency and accuracy in memory: Effects of schema-typicality, retention interval and question-format

This study examined the effect of schemas on consistency and accuracy of memory across interviews, providing theoretical hypotheses explaining why inconsistencies may occur. The design manipulated schema-typicality of items (schema-typical and atypical), question format (free-recall, cued-recall and recognition) and retention interval (immediate/2 week and 2 week/4 week). Consistency, accuracy and experiential quality of memory were measured. ^ All independent variables affected accuracy and experiential quality of memory while question format was the only variable affecting consistency. These results challenge the commonly held notion in the legal arena that consistency is a proxy for accuracy. The study also demonstrates that other variables, such as item-typicality and retention interval have different effects on consistency and accuracy in memory. ^

Ensemble fuzzy belief intrusion detection design

With the rapid growth of the Internet, computer attacks are increasing at a fast pace and can easily cause millions of dollar in damage to an organization. Detecting these attacks is an important issue of computer security. There are many types of attacks and they fall into four main categories, Denial of Service (DoS) attacks, Probe, User to Root (U2R) attacks, and Remote to Local (R2L) attacks. Within these categories, DoS and Probe attacks continuously show up with greater frequency in a short period of time when they attack systems. They are different from the normal traffic data and can be easily separated from normal activities. On the contrary, U2R and R2L attacks are embedded in the data portions of the packets and normally involve only a single connection. It becomes difficult to achieve satisfactory detection accuracy for detecting these two attacks. Therefore, we focus on studying the ambiguity problem between normal activities and U2R/R2L attacks. The goal is to build a detection system that can accurately and quickly detect these two attacks. In this dissertation, we design a two-phase intrusion detection approach. In the first phase, a correlation-based feature selection algorithm is proposed to advance the speed of detection. Features with poor prediction ability for the signatures of attacks and features inter-correlated with one or more other features are considered redundant. Such features are removed and only indispensable information about the original feature space remains. In the second phase, we develop an ensemble intrusion detection system to achieve accurate detection performance. The proposed method includes multiple feature selecting intrusion detectors and a data mining intrusion detector. The former ones consist of a set of detectors, and each of them uses a fuzzy clustering technique and belief theory to solve the ambiguity problem. The latter one applies data mining technique to automatically extract computer users’ normal behavior from training network traffic data. The final decision is a combination of the outputs of feature selecting and data mining detectors. The experimental results indicate that our ensemble approach not only significantly reduces the detection time but also effectively detect U2R and R2L attacks that contain degrees of ambiguous information.

Thermal imaging as a biometrics approach to facial signature authentication

This dissertation develops an image processing framework with unique feature extraction and similarity measurements for human face recognition in the thermal mid-wave infrared portion of the electromagnetic spectrum. The goals of this research is to design specialized algorithms that would extract facial vasculature information, create a thermal facial signature and identify the individual. The objective is to use such findings in support of a biometrics system for human identification with a high degree of accuracy and a high degree of reliability. This last assertion is due to the minimal to no risk for potential alteration of the intrinsic physiological characteristics seen through thermal infrared imaging. The proposed thermal facial signature recognition is fully integrated and consolidates the main and critical steps of feature extraction, registration, matching through similarity measures, and validation through testing our algorithm on a database, referred to as C-X1, provided by the Computer Vision Research Laboratory at the University of Notre Dame. Feature extraction was accomplished by first registering the infrared images to a reference image using the functional MRI of the Brain’s (FMRIB’s) Linear Image Registration Tool (FLIRT) modified to suit thermal infrared images. This was followed by segmentation of the facial region using an advanced localized contouring algorithm applied on anisotropically diffused thermal images. Thermal feature extraction from facial images was attained by performing morphological operations such as opening and top-hat segmentation to yield thermal signatures for each subject. Four thermal images taken over a period of six months were used to generate thermal signatures and a thermal template for each subject, the thermal template contains only the most prevalent and consistent features. Finally a similarity measure technique was used to match signatures to templates and the Principal Component Analysis (PCA) was used to validate the results of the matching process. Thirteen subjects were used for testing the developed technique on an in-house thermal imaging system. The matching using an Euclidean-based similarity measure showed 88% accuracy in the case of skeletonized signatures and templates, we obtained 90% accuracy for anisotropically diffused signatures and templates. We also employed the Manhattan-based similarity measure and obtained an accuracy of 90.39% for skeletonized and diffused templates and signatures. It was found that an average 18.9% improvement in the similarity measure was obtained when using diffused templates. The Euclidean- and Manhattan-based similarity measure was also applied to skeletonized signatures and templates of 25 subjects in the C-X1 database. The highly accurate results obtained in the matching process along with the generalized design process clearly demonstrate the ability of the thermal infrared system to be used on other thermal imaging based systems and related databases. A novel user-initialization registration of thermal facial images has been successfully implemented. Furthermore, the novel approach at developing a thermal signature template using four images taken at various times ensured that unforeseen changes in the vasculature did not affect the biometric matching process as it relied on consistent thermal features.

Adsorption and desorption of phosphate on limestone in experiments simulating seawater intrusion

This study investigates the potential release of from carbonate aquifers exposed to seawater intrusion. Adsorption and desorption of in the presence of deionized water (DIW) and seawater were conducted on a large block of Pleistocene age limestone to simulate the effects of seawater intrusion into a coastal carbonate aquifer at the laboratory scale. The limestone showed strong adsorption of in DIW, while adsorption was significantly less in the presence of seawater. Dissolution of CaCO3 was found to prevent adsorption at salinities less than 30 psu. Adsorption of was limited at higher salinities (30–33 psu), due to competition with ions for adsorption sites. At a salinity3 precipitated. Concentrations of between 2 and 5 μmol/L were released by desorption when the limestone was exposed to seawater. The results of this study suggest that as seawater intrudes into an originally freshwater coastal aquifer, adsorbed may be released into the groundwater. Consequently, adsorbed is expected to be released from coastal carbonate aquifers world-wide as sea level continues to rise exposing more of the freshwater aquifer to seawater.

Determination of signature volatile odor chemicals emanating from novel biological specimens by non-invasive analytical techniques for the potential use in forensic identifications

Human scent, or the volatile organic compounds (VOCs) produced by an individual, has been recognized as a biometric measurement because of the distinct variations in both the presence and abundance of these VOCs between individuals. In forensic science, human scent has been used as a form of associative evidence by linking a suspect to a scene/object through the use of human scent discriminating canines. The scent most often collected and used with these specially trained canines is from the hands because a majority of the evidence collected is likely to have been handled by the suspect. However, the scents from other biological specimens, especially those that are likely to be present at scenes of violent crimes, have yet to be explored. Hair, fingernails and saliva are examples of these types of specimens. ^ In this work, a headspace solid phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS) technique was used for the identification of VOCs from hand odor, hair, fingernails and saliva. Sixty individuals were sampled and the profiles of the extracted VOCs were evaluated to assess whether they could be used for distinguishing individuals. Preliminary analysis of the biological specimens collected from an individual (intra-subject) showed that, though these materials have some VOCs in common, their overall chemical profile is different for each specimen type. Pair-wise comparisons, using Spearman Rank correlations, were made between the chemical profiles obtained from each subject, per a specimen type. Greater than 98.8% of the collected samples were distinguished from the subjects for all of the specimen types, demonstrating that these specimens can be used for distinguishing individuals. ^ Additionally, field trials were performed to determine the utility of these specimens as scent sources for human scent discriminating canines. Three trials were conducted to evaluate hair, fingernails and saliva in comparison to hand odor, which was considered the standard source of human odor. It was revealed that canines perform similarly to these alternative human scent sources as they do to hand odor implying that, though there are differences in the chemical profiles released by these specimens, they can still be used for the discrimination of individuals by trained canines.^

Development of test-based wind-driven rain intrusion model for hurricane-induced building interior and contents damage

Major portion of hurricane-induced economic loss originates from damages to building structures. The damages on building structures are typically grouped into three main categories: exterior, interior, and contents damage. Although the latter two types of damages, in most cases, cause more than 50% of the total loss, little has been done to investigate the physical damage process and unveil the interdependence of interior damage parameters. Building interior and contents damages are mainly due to wind-driven rain (WDR) intrusion through building envelope defects, breaches, and other functional openings. The limitation of research works and subsequent knowledge gaps, are in most part due to the complexity of damage phenomena during hurricanes and lack of established measurement methodologies to quantify rainwater intrusion. This dissertation focuses on devising methodologies for large-scale experimental simulation of tropical cyclone WDR and measurements of rainwater intrusion to acquire benchmark test-based data for the development of hurricane-induced building interior and contents damage model. Target WDR parameters derived from tropical cyclone rainfall data were used to simulate the WDR characteristics at the Wall of Wind (WOW) facility. The proposed WDR simulation methodology presents detailed procedures for selection of type and number of nozzles formulated based on tropical cyclone WDR study. The simulated WDR was later used to experimentally investigate the mechanisms of rainwater deposition/intrusion in buildings. Test-based dataset of two rainwater intrusion parameters that quantify the distribution of direct impinging raindrops and surface runoff rainwater over building surface — rain admittance factor (RAF) and surface runoff coefficient (SRC), respectively —were developed using common shapes of low-rise buildings. The dataset was applied to a newly formulated WDR estimation model to predict the volume of rainwater ingress through envelope openings such as wall and roof deck breaches and window sill cracks. The validation of the new model using experimental data indicated reasonable estimation of rainwater ingress through envelope defects and breaches during tropical cyclones. The WDR estimation model and experimental dataset of WDR parameters developed in this dissertation work can be used to enhance the prediction capabilities of existing interior damage models such as the Florida Public Hurricane Loss Model (FPHLM).^

Uniquely Identifiable Tamper-Evident Device Using Coupling between Subwavelength Gratings

Reliability and sensitive information protection are critical aspects of integrated circuits. A novel technique using near-field evanescent wave coupling from two subwavelength gratings (SWGs), with the input laser source delivered through an optical fiber is presented for tamper evidence of electronic components. The first grating of the pair of coupled subwavelength gratings (CSWGs) was milled directly on the output facet of the silica fiber using focused ion beam (FIB) etching. The second grating was patterned using e-beam lithography and etched into a glass substrate using reactive ion etching (RIE). The slightest intrusion attempt would separate the CSWGs and eliminate near-field coupling between the gratings. Tampering, therefore, would become evident. Computer simulations guided the design for optimal operation of the security solution. The physical dimensions of the SWGs, i.e. period and thickness, were optimized, for a 650 nm illuminating wavelength. The optimal dimensions resulted in a 560 nm grating period for the first grating etched in the silica optical fiber and 420 nm for the second grating etched in borosilicate glass. The incident light beam had a half-width at half-maximum (HWHM) of at least 7 µm to allow discernible higher transmission orders, and a HWHM of 28 µm for minimum noise. The minimum number of individual grating lines present on the optical fiber facet was identified as 15 lines. Grating rotation due to the cylindrical geometry of the fiber resulted in a rotation of the far-field pattern, corresponding to the rotation angle of moiré fringes. With the goal of later adding authentication to tamper evidence, the concept of CSWGs signature was also modeled by introducing random and planned variations in the glass grating. The fiber was placed on a stage supported by a nanomanipulator, which permitted three-dimensional displacement while maintaining the fiber tip normal to the surface of the glass substrate. A 650 nm diode laser was fixed to a translation mount that transmitted the light source through the optical fiber, and the output intensity was measured using a silicon photodiode. The evanescent wave coupling output results for the CSWGs were measured and compared to the simulation results.

Ensemble Fuzzy Belief Intrusion Detection Design

With the rapid growth of the Internet, computer attacks are increasing at a fast pace and can easily cause millions of dollar in damage to an organization. Detecting these attacks is an important issue of computer security. There are many types of attacks and they fall into four main categories, Denial of Service (DoS) attacks, Probe, User to Root (U2R) attacks, and Remote to Local (R2L) attacks. Within these categories, DoS and Probe attacks continuously show up with greater frequency in a short period of time when they attack systems. They are different from the normal traffic data and can be easily separated from normal activities. On the contrary, U2R and R2L attacks are embedded in the data portions of the packets and normally involve only a single connection. It becomes difficult to achieve satisfactory detection accuracy for detecting these two attacks. Therefore, we focus on studying the ambiguity problem between normal activities and U2R/R2L attacks. The goal is to build a detection system that can accurately and quickly detect these two attacks. In this dissertation, we design a two-phase intrusion detection approach. In the first phase, a correlation-based feature selection algorithm is proposed to advance the speed of detection. Features with poor prediction ability for the signatures of attacks and features inter-correlated with one or more other features are considered redundant. Such features are removed and only indispensable information about the original feature space remains. In the second phase, we develop an ensemble intrusion detection system to achieve accurate detection performance. The proposed method includes multiple feature selecting intrusion detectors and a data mining intrusion detector. The former ones consist of a set of detectors, and each of them uses a fuzzy clustering technique and belief theory to solve the ambiguity problem. The latter one applies data mining technique to automatically extract computer users’ normal behavior from training network traffic data. The final decision is a combination of the outputs of feature selecting and data mining detectors. The experimental results indicate that our ensemble approach not only significantly reduces the detection time but also effectively detect U2R and R2L attacks that contain degrees of ambiguous information.

Determination of Signature Volatile Odor Chemicals Emanating from Novel Biological Specimens by Non-invasive Analytical Techniques for the Potential use in Forensic Identifications

Human scent, or the volatile organic compounds (VOCs) produced by an individual, has been recognized as a biometric measurement because of the distinct variations in both the presence and abundance of these VOCs between individuals. In forensic science, human scent has been used as a form of associative evidence by linking a suspect to a scene/object through the use of human scent discriminating canines. The scent most often collected and used with these specially trained canines is from the hands because a majority of the evidence collected is likely to have been handled by the suspect. However, the scents from other biological specimens, especially those that are likely to be present at scenes of violent crimes, have yet to be explored. Hair, fingernails and saliva are examples of these types of specimens. In this work, a headspace solid phase microextraction gas chromatography-mass spectrometry (HS-SPME-GC-MS) technique was used for the identification of VOCs from hand odor, hair, fingernails and saliva. Sixty individuals were sampled and the profiles of the extracted VOCs were evaluated to assess whether they could be used for distinguishing individuals. Preliminary analysis of the biological specimens collected from an individual (intra-subject) showed that, though these materials have some VOCs in common, their overall chemical profile is different for each specimen type. Pair-wise comparisons, using Spearman Rank correlations, were made between the chemical profiles obtained from each subject, per a specimen type. Greater than 98.8% of the collected samples were distinguished from the subjects for all of the specimen types, demonstrating that these specimens can be used for distinguishing individuals. Additionally, field trials were performed to determine the utility of these specimens as scent sources for human scent discriminating canines. Three trials were conducted to evaluate hair, fingernails and saliva in comparison to hand odor, which was considered the standard source of human odor. It was revealed that canines perform similarly to these alternative human scent sources as they do to hand odor implying that, though there are differences in the chemical profiles released by these specimens, they can still be used for the discrimination of individuals by trained canines.