This PDF file contains the front matter associated with SPIE Proceedings Volume 7343, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

As this is a conference on mathematical tools for defense, I would like to dedicate this talk to the memory of Louis Auslander, who through his insights and visionary leadership, brought powerful new mathematics into DARPA, he has provided the main impetus to the development and insertion of wavelet based processing in defense. My goal here is to describe the evolution of a stream of ideas in Harmonic Analysis, ideas which in the past have been mostly applied for the analysis and extraction of information from physical data, and which now are increasingly applied to organize and extract information and knowledge from any set of digital documents, from text to music to questionnaires. This form of signal processing on digital data, is part of the future of wavelet analysis.

The "smile of a mother" is always recognized, whenever and wherever. But why is my PC always dumb and unable to recognize me or my needs, whoever or whatever? This paper postulates that such a 6 W's query and search system needs matching storage. Such a lament will soon be mended with a smarter PC, or a smarter Google engine, a network computer, working in the field of data retrieval, feature extraction, reduction, and knowledge precipitation. Specifically, the strategy of modern information storage and retrieval shall work like our brains, which are constantly overwhelmed by 5 pairs of identical tapes taken by eyes, ears, etc. 5 high fidelity sensors generate 5 pairs of high definition tapes which produce the seeing and hearing etc. in our perception. This amounts to 10 tapes recorded in a non-abridged fashion. How can we store and retrieve them when we need to? We must reduce the redundancy, enhancing the signal noise ratio, and fusing invariant features using a simple set of mathematical operations to write according to the union and read by the intersection in the higher dimensional vector space. For example, (see paper for equation) where the query must be phrased in terms of the union of imprecise or partial set of 6w's denoted by the union of lower case w's. The upper case W's are the archival storage of a primer tree. A simplified humanistic representation may be called the 6W space (who, what, where, when, why, how), also referred to as the Newspaper geometry. It seems like mapping the 6W to the 3W (World Wide Web) is becoming relatively easier. It may thus become efficient and robust by rapidly digging for knowledge through the set operations of union, writing, and intersection, reading, upon the design of 6 W query searching engine matched efficiently by the 6W vector index databases. In fact, Newspaper 6D geometry may be reduced furthermore by PCA (Principal Component Analysis) eigenvector mathematics and mapped into the 2D causality space comprised of the causes (What, How, Why) and the effects (Where, When and Who). If this hypothesis of brain strategy were true, one must then develop a 6W query language to support a 6Wordered set storage of linkage pointers in high D space. In other words, one can easily map the basic 1st Gen. Google Web, 1-D statistical PageRanking databases, to a nested 6W tree where each branch of sub-6-W is stemming from the prime 6 W tree, using a system of automated text mining assisted by syntactic semantics to discern the properties of the 6W for that query. Goehl et al. has demonstrated previously that such is doable, but one may need more tools to support the knowledge extraction and automated feature reduction. In this paper, we have set out to demonstrate lossless down sampling using the 2nd Gen wavelet transform, the so-called "1-D Cartesian lifting processing of Swelden" adopted by JPEG 2000. "The loss of statistics, if any (including PageRanking and 1-D lifting), is the loss of geometry insights," such as 2-D vector time series, video, whose 1-D lifting Cartesian product will loss the diagonal changes insights.

The signal generated by avalanche discharge circuitry that drives optical flash for cameras is ultra wideband and suitable as the excitation waveform for an Impulse Radar when transmitted through an appropriate antenna. In this paper, we experimentally and theoretically show that the Impulse Radar's radiated transmitted EM waves satisfy the admissibility condition of wavelets, i.e. (i) a finite power spectral density psd=<|S(f)|²>< and (ii) zero total area under the undulated wave amplitudes. If the radiated environment is linear (the natural scene), then the received signal also satisfies the admissibility condition.

Many imaging modalities, such as Synthetic Aperture Radar (SAR), can be described mathematically as collecting data in a Radon transform domain. The process of inverting the Radon transform to form an image can be unstable when the data collected contain noise so that the inversion needs to be regularized in some way. In this work, we develop a method for inverting the Radon transform using a shearlet-based decomposition, which provides a regularization that is nearly optimal for a general class of images. We then show through a variety of examples that this technique performs better than similar competitive methods based on the use of the wavelet and the curvelet transforms.

In this paper, we simulate sensor behavior by synthesizing different features of a sensor output under normal mode of operation. Any deviation from the normal behavior indicates a change. The synthesized features indicate behavior of different physical processes that a sensor is monitoring. The shapes of these features must be extracted from sensor output for sensor health management. The extracted features are compared to features of healthy sensor to monitor its performance. In this paper, we compare Fourier and Wavelet Transform methods for extraction of the sensor output features. The wavelet Transform Analysis is performed on the simulated data described above with Poisson distributed noise. The simulated data with Poisson distributed noise of SNRs ranging from 10 to 500 are generated. The data are analyzed using Discrete as well as Discretized Continuous Wavelet Transforms. The Short-Time Fourier Transform of the Signal is taken using the Hamming window. Three window widths are used. The DC value is removed from the windowed data prior to taking the FFT. The resulting three dimensional spectral plots provide good time frequency resolution. The results indicate distinct shapes corresponding to each process.

We show how to apply the Continuous Wavelet Transform (CWT) to discrete data. This is done by deriving analytical functions from the data that are n^th order integrable and differentiable. We also show how to make these Data Models compactly supported. Further, we show how to identify a stopping criteria for the data sampling process to initiate the wavelet transformation. We also suggest how the data interval can be exploited to obtain a fractal wavelet mother function from the sampled data. We compare this to classical techniques and note enhanced performance, and finally show how the number of terms in the analytical Data Model can be minimized by converting into a one-sided bi-spectral form using only cosine functions. From this bi-spectral form, we are able to forecast and backcast both the original data and the derived adaptive basis functions.

A hybrid algorithm of distinguishing rock from coal based on statistical analysis of Wavelet Transform (WT) is presented which can be used in the process of caving coal. First, eight groups of sound signals sampled with the speed 8192 samples/sec during caving are decomposed by db3 wavelet. Second, the WT results are analyzed by using the variance analytical method in the second-level details (D2). Third, the typical values i.e. the detail-level coefficient variances (Dvar) of the sound of the coal bumping the transporting coal armor plate, the rock bumping the armor plate and the mixing of coal and rock bumping the armor plate are calculated. Finally, the threshold value of distinguishing rock from coal is evaluated by the typical values and used to direct the opportunity for caving. We can learn by the experimental results that the proposed technique can depict effectively the different characteristics of the sampled signals. The experimental results also show that we can distinguish effectively different bumping sounds of coal, rock and the mixing of them by the characteristics when adjusting the appropriate threshold value. Meanwhile, the proposed method has strong ability to resist the noise occurred during mining. Therefore, the algorithm can be used to improve the miners' productivity and promote the construction of digital mine.

A new group analysis method to summarize the task-related BOLD responses based on independent component analysis (ICA) was presented. As opposite to the previously proposed group ICA (gICA) method, which first combined multi-subject fMRI data in either temporal or spatial domain and applied ICA decomposition only once to the combined fMRI data to extract the task-related BOLD effects, the method presented here applied ICA decomposition to the individual subjects' fMRI data to first find the independent BOLD effects specifically for each individual subject. Then, the task-related independent BOLD component was selected among the resulting independent components from the single-subject ICA decomposition and hence grouped across subjects to derive the group inference. In this new ICA group analysis (ICAga) method, one does not need to assume that the task-related BOLD time courses are identical across brain areas and subjects as used in the grand ICA decomposition on the spatially concatenated fMRI data. Neither does one need to assume that after spatial normalization, the voxels at the same coordinates represent exactly the same functional or structural brain anatomies across different subjects. These two assumptions have been problematic given the recent BOLD activation evidences. Further, since the independent BOLD effects were obtained from each individual subject, the ICAga method can better account for the individual differences in the task-related BOLD effects. Unlike the gICA approach whereby the task-related BOLD effects could only be accounted for by a single unified BOLD model across multiple subjects. As a result, the newly proposed method, ICAga, was able to better fit the task-related BOLD effects at individual level and thus allow grouping more appropriate multisubject BOLD effects in the group analysis.

Perception is the apparatus of many of the potential new sciences of the twenty-first century. We can imagine science as a symbolic language where there are concrete words which are taken from measurements while abstract words are defined by theory to construct new ways to think about more fuzzy concepts that occur in more human observation based sciences. By considering both semantics and meaning as part of establishing a theory we go beyond where syntax of mathematics has led us to in science. Thus, we argue that extending Shannon's model of communication to the semantic problem of communication is the problem of twenty-first century science and mathematics.

Precipitation is an important parameter in hydrologic and climate research. In addition to gauge and radar observations, there are several satellites providing spaceborne observations. Effectively merging these data products can improve the rainfall estimation accuracy. In this paper, we investigate the use of neural networks, i.e., multi-layer backpropagation neural network and radial basis function neural network in precipitation data merging. We also investigate the performance improvement from training sample selection via principal component analysis. The preliminary results show that our data merging approaches can outperform other linear methods such as weighted sum and the data preprocessing can also improve the performance.

We showed that a hard threshold for wavelet denoising based on higher order statistics is comparable to a second order soft threshold. The hard threshold can made adaptive by using a third order statistic as an estimate of the noise. In addition, the relationship between an adaptive hard threshold and retaining a fraction of wavelet coefficients is shown. Qualitative and quantitative metrics based on the mean-squared error are used to compare the hard thresholding and a soft-thresholding technique, BayesShrink.

Information-technology departments in large enterprises spend 40% of budget on information integration-combining information from different data sources into a coherent form. IDC, a market-intelligence firm, estimates that the market for data integration and access software (which includes the key enabling technology for information integration) was about $2.5 billion in 2007, and is expected to grow to $3.8 billion in 2012. This is only the cost estimate for structured or traditional database information integration. Just imagine the market for transforming text into structured information and subsequent fusion with traditional databases.

Classification performance of recognition tasks can be improved by selection of highly discriminative features from the low-dimensional linear representation of data. High-dimensional multivariate data can be represented in lower dimensions by unsupervised feature extraction techniques which attempts to remove the redundancy in the data and/or resolve the multivariate prediction problems. These extracted low-dimensional features of raw data may not ensure good class discrimination, therefore, supervised feature selection methods motivated by information-theoretic approaches can improve the recognition performance with lesser number of features. Proposed hybrid feature selection methods efficiently selects features with higher class discrimination in comparison to feature-class mutual information (MI), Fisher criterion or unsupervised selection using variance; thus, resulting in much improved recognition performance. Feature-class MI criterion and hybrid feature selection methods are computationally scalable and optimal selectors for statistically independent features.

Remote sensing techniques are well suited to quantify the spatial variability of coastal water quality. The correlation between remotely sensed data in the visible to near-infrared (VNIR) bands and in situ water measurements are well studied. Due to the high spatial variation and fine waterbody structure along shorelines, it may be beneficial to use remotely sensed images with higher spatial resolution, such as the Landsat data with 30m resolution. In this research, we investigate the traditional approaches, such as regression analysis, in the mapping of water quality (e.g., total suspended sediments (TSS), turbidity, and chlorophyll A). In particular, we also develop an approach based on neural network to generate additional bands, which can further improve the mapping accuracy.

The common methodology of biologically motivated concept of building of processing sensors systems with parallel input and picture operands processing and time-pulse coding are described in paper. Advantages of such coding for creation of parallel programmed 2D-array structures for the next generation digital computers which require untraditional numerical systems for processing of analog, digital, hybrid and neuro-fuzzy operands are shown. The optoelectronic time-pulse coded intelligent neural elements (OETPCINE) simulation results and implementation results of a wide set of neuro-fuzzy logic operations are considered. The simulation results confirm engineering advantages, intellectuality, circuit flexibility of OETPCINE for creation of advanced 2D-structures. The developed equivalentor-nonequivalentor neural element has power consumption of 10mW and processing time about 10...100us.

Particles flowing along with water largely affect safe drinking water, irrigation, aquatic life preservation and hydropower generation. This research describes activities that lead to development of fluvial particle characterization that includes detection of biological and non-biological particles and shape characterization using Image Processing and Artificial Neural Network (ANN). Fluvial particles are characterized based on multi spectral images processing using ANN. Images of wavelength of 630nm and 670nm are taken as most distinctive characterizing properties of biological and non-biological particles found in Bagmati River of Nepal. The samples were collected at pre-monsoon, monsoon and post-monsoon seasons. Random samples were selected and multi spectral images are processed using MATLAB 6.5. Thirty matrices were built from each sample. The obtained data of 42 rows and 60columns were taken as input training with an output matrix of 42 rows and 2 columns. Neural Network of Perceptron model was created using a transfer function. The system was first validated and later on tested at 18 different strategic locations of Bagmati River of Kathmandu Valley, Nepal. This network classified biological and non biological particles. Development of new non-destructive technique to characterize biological and non-biological particles from fluvial sample in a real time has a significance breakthrough. This applied research method and outcome is an attractive model for real time monitoring of particles and has many applications that can throw a significant outlet to many researches and for effective utilization of water resources. It opened a new horizon of opportunities for basic and applied research at Kathmandu University in Nepal.

Significant improvements have been achieved for Intellectual Property Protection (IPP) of IP cores at HDL level by means of our proposed watermarking techniques from the initial solution. The basic idea relies on spreading the bits of a digital signature at the HDL design level using combinational logic or look-up structures included within the original system. The techniques also include a secure and non-destructive signature extraction process. Recently, applicability has been extended due to the development of an automated tool for signature spreading purposes. Last advances refer to the development of new algorithms for the automated tool that result in important improvements and automatic generation of the VHDL code to add to the original core for signature extraction purposes. In addition, the most important metrics for watermarking techniques, such as the impact over the original core and the strength against attacks, allow the evaluation of the watermarking technique and illustrate the effectiveness of the proposal.

We present a simple approach for deriving ensembles of training data from notional belief networks. This is accomplished by specifying the belief variable interactions in the form of Bayes expert system or directed graph, where the node conditional and prior probabilities are specified heuristically from data or from subject matter expert (SME) heuristics. The resulting network is then sampled across parameter space and the associated input/output pairs retained for deriving a principal component Data Model using regression techniques. The method is general and the details of the algorithm are presented.

There is a lot of interest by society at large for reliable and affordable sensor and detection systems. Recent threats regarding public exposure to both biological and chemical agents have helped focus attention to the development of new sensor and detector technologies. Unfortunately, there is not even consensus in the scientific community on a uniform terminology of the subject. A large body of knowledge in sensor materials is available but most existing sensor materials are very limited in their ability to sense and discriminate small amounts of harmful agents embedded in large amounts of chemically inert but complex background. New materials are needed to meet the challenges ahead. Nanomaterials may be a solution to this problem. In this work we review the current state of the sensor lexicon. This is followed by a proposition of a consistent definition of terms used in the sensor and detection technology. Finally, we review a promising IR sensor system being developed using nanomaterials and several detector systems also developed using nanomaterials for nano solar cells and nano smart materials.

Using the characteristics of T cell mitogen called lectin protein from the jack-beam Canavalia ensiformis Concanavalin A (Con_A) with dual activities, cytotoxicity and immunomodulation, we have shown it has a therapeutic effect on hepatoma. Injection of Con_A can eradicate the established malign tumor, because Con_A can induce tumor cell autophagic, cell-programmed death, as well as activate the effector T cells. Combined, in this paper, with the absorption exceeding the Carbon NanoTube (CNT) band-gap (εbg=~1/CNT diameter) with an active short wave near-infrared (SWIR) (1.2~1.5 micron wavelengths), which happened to be translucent to the irradiation upon animal skin, similar to that used in hospital fingertip-clamped Pulse Oxymetry. Once the Con_ACNT is guided to hepatoma cells, it is bonded and internalized into the mitochondria (MC) compartment, the cellular energy factory. Con_A has the higher specificity for tumor cells useful for targeting because of the abnormal glycosylation on tumor cells. When CNT hitch hike with Con_A, they can t together like a laser-denotable chemical missile surgically targeting at the tumor cells precisely by Con_A-guidance. We switch on SWIR laser, when the Con_A-CNT conjugated complex has been bonded and internalized to MC of malign cells and already commenced cellular programmed death. Thus, it might appear to casual readers that we have initiated an overkill, chemical drugged autophage followed with physical laser ablation, but what if we can eradicate hepatoma totally if no blue print is left behind inadvertently in case of a partial failure. We conclude that using Con_A-CNT conjugated complex targeting specifically at malign tumor cells is a novel targeted-laser-radiation therapy for tumors in mice.

Nanotechnology is impacting the future of the military and aerospace. The increasing demands for high performance and property-specific applications are forcing the scientific world to take novel approaches in developing programs and accelerating output. CONTACT or Consortium for Nanomaterials for Aerospace Commerce and Technology is a cooperative nanotechnology research program in Texas building on an infrastructure that promotes collaboration between universities and transitioning to industry. The participants of the program include the US Air Force Research Laboratory (AFRL), five campuses of the University of Texas (Brownsville, Pan American, Arlington, Austin, and Dallas), the University of Houston, and Rice University. Through the various partnerships between the intellectual centers and the interactions with AFRL and CONTACT's industrial associates, the program represents a model that addresses the needs of the changing and competitive technological world. Into the second year, CONTACT has expanded to twelve projects that cover four areas of research: Adaptive Coatings and Surface Engineering, Nano Energetics, Electromagnetic Sensors, and Power Generation and Storage. This paper provides an overview of the CONTACT program and its projects including the research and development of new electrorheological fluids with nanoladen suspensions and composites and the potential applications.

Inspired by Asian rice-paddy and Firefighter spiraling steps staircase, we employ a nano-manipulator augmented with CAD as a nano-robot water-buffalo, promised to improve by an order of the magnitude the pioneer work of GE Solar voltaic cell (SVC) made of one Carbon NanoTube (CNT) enjoyed QE_CNT~5%. Our CNT was made of the semiconductor at NIR wavelength E_BG= 1.107 eV which can absorb any photon whose wavelength λ ≤ λ_NIR =1.11 μm. This EBG is chosen due to the cutoff of Pb-Crown glass which happened put us in the equal amount of solar energy spectrum as silicon p-n junction SVC. Nevertheless, the exceeding of the Shockley and Queisser efficiency limit 30% might be due to the fact that we have much compact one-dimensional building block CNT of a tiny diameter 0.66 nm. It allows us to construct 3D structure, called volume pixel, "voxel," in a much efficient spiraling steps staircase fashion to capture the solar spectral energy spreading naturally by a simple focusing lens without occlusion. For real-estate premium applications, in Space or Ocean, we designed a volume pixel (Voxel) housing a stack of 16 CNTs steps spiraling 22° each like the fire house staircase occupying the height of 16 x d_CNT =16 x 0.66nm= 10.56 nm and covering over 360°. The total SVC had the size 2x2 meter², consisting of 100×100 lenslet array. Each lens was made of Pb-Crown glass which was inexpensive simple spherical lens having the diameter of D_lens=2 cm and F#=0.7. It can focus the sunlight a millionth times stronger in a smallest possible focal spot size, λ_Yellow=0.635 μm< λ_{Max photons} <λ_Red=0.73 μm, where the largest number of solar photons, 68%, according to the Plank radiation spectrum at 6000°K and the Lord Rayleigh diffraction limit. The solar panel seals individually such an array of 3D cavities of SVC enjoying theoretically from the UV 12% (wasted in passing through) visible 68% to the infrared 20% at a total of 16x5%~80% total QECNT per cell. The solar panel is made of light-weight carbon composite tolerating about 20% inactive fill factor and 10% dead pixels.

It is not surprising that the earthquakes happened among clashing tectonic plate boundaries where numerous earthquake stations exist. Then, why do we need more? The significance of Sichuan and Tongshan earthquakes of China is a wakeup call that major earthquakes of logarithmic Richter scales beyond 7 could happen exactly within a single tectonic plate surprisingly. Thus, the previous border surveillance is broadened to areas coverage. Judging the success of archival survey of NASA with the gravitational potential by Liu et al., we review a unified earthquakes theory covering both the peripheral and the central plate in this paper, so that we can take seriously the need of a comprehensive global surveillance of natural calamity in the Space. The earth surface crust, like a kitchen kettle lid, covers tightly the melted mantle rock layer, like a pea soup cooking in the kettle. Given the time they will all become bubbling, rattling & shaking, known as the Bernard instability. This instability is ubiquitous for any liquid state matter being heated from below, if and only if it has a real positive thermal expansion coefficient. Likewise, the earth mantle is being cooked from below by an enormously hot fireball of the size of a moon. The heat comes from the radioactive decay confined within the core over eon's age. Due to the enormous gravitation attraction being always real positive and additive, the inner core is bifurcated into 2 regimes, a heat-melted liquid metal regime, where the earth magnetic field is produced and predicted by Faraday induction law. And further inside there exists a tightly squeezed solid metal ball regime, due to the gigantic weight compression, as confirmed by sonar experiments. The complexity of earth Bernard instability is due to the extra rotational Coriolis force that makes the up-down thermal convection side-way, creating the mass imbalance and permitting in-situ measurements feasible at a distance.

Since the world presents a dynamically changing environment, we need synthetic systems that can process and respond to motion. The main contribution of this work is the efficient implementation of a biologically inspired DSP architecture for gradient motion estimation that borrows nature templates as inspiration and makes use of an specific model of human visual motion perception: Multi Channel Gradient Model. This model can be enhanced using psychophysical and bioinspired properties according to biological vision in order to mimic the behavior and the performance of the mammalians. The architecture is designed with an asynchronous pipeline, chaining several signal processing stages. Experimental results and resource consumption are discussed, analyzing the associated customization of the system. This work concludes with several comparisons with the actual contributions and examples for synthetic sequences and real image applications.

Epilepsy is a chronic brain disorder characterized by recurrent seizures. The seizure is shot down by the surgical removal of the region which is so called "epileptogenc focus". However, the accuracy to detect the focus is not good (order of cm). Thus the extirpation of focus with significant margin causes the removal of normal brain and leads to the severe aftereffects such as restricted vision, motor dysfunction, disorder of memory, and so on. To cope with this problem, we should develop the technology of (1) detecting the epileptogenic focus, and (2) necrotizing the epileptogenic focus excluding normal brain by (a) colliquative necrosis with flash freezing and melting or (b) cauterizing by focused laser beam.

Biometrics are generally thought of as anatomical features that allow positive identification of a person. This paper describes biometrics that are also physiological in nature. The differences between anatomy and physiology have to do with the fact that physiology is dynamic, functioning, and changing with the state or actions of a person whereas anatomy is generally more stable. Biometrics in general usually refers to a trait, whereas the new type of biometrics discussed in this paper refer to a state, which is temporary, and often even transitory. By state, what is meant is the condition of a person at a particular time relative to their psychological, physical, medical, or physiological status. The present paper describes metrics that are cues to the state of a functioning individual observable through a thermal camera video system. An inferred state might then be tied to the positive identification of the person. Using thermal for this purpose is significant because the thermal signature of a human is dynamic and changes with physical and emotional state, while also revealing underlying anatomical structures. A new method involving the counting of open pores on the skin is discussed as a way of observing the Electrodermal Activity (EDA) of the skin, a primary component of the polygraph.

Previous work by the Army's Night Vision and Electronic Sensors Directorate extended the Army's target acquisition performance models to include the task of facial identification in the visible band. In this work, additional field data for facial identification is used to validate the direct view optic (DVO) performance model. The target acquisition model for direct view optics is based on the contrast threshold function of the eye with a modification for the optics modulation transfer function (MTF) and the optics magnification. We also show that the current Targeting Task Performance (TTP) metric can approximate the measured data, without using the full accuracy provided by the two-dimensional specific object method described in previous work.

NVESD is currently researching relationships between facial thermal signatures and underlying physiological stress. This paper describes some of the initial findings of a study involving participants watching emotion-eliciting video clips while being monitored by a high-resolution, Mid-Wave InfraRed (MWIR), cooled sensor system. Both global and region-specific temperature profiles of the face were obtained for each participant from thermal videos recorded during the experiment. Changes in the global temperature patterns of each participant were shown to be consistent with changes in the visual and auditory stimuli. In some cases, there were inconsistencies in thermal trends among local facial regions of individual participants. This indicates the possibility that different regions may exhibit different thermal patterns in response to a stimulus.

Advances in technologies can make them competitive in areas where they were not. Nevertheless, most commercial devices are designed with a particular application in mind. Attempting to use such a piece of equipment, notwithstanding technological improvements, may provide a suboptimal solution. However, simple and inexpensive modifications can sometimes be made allowing such a piece of equipment to have successful application is an area far removed from those intended in its original design. This paper describes such a situation, where a commercial thermography product designed for power line monitoring was modified to allow it to effectively function in a biometric application. Some of the approaches taken here could be applied to other technologies and other applications that could profit the community.

Earth's Field Nuclear Magnetic Resonance (⊕-NMR) is a novel detection technique that exploits the uniformity of the earth's magnetic field to measure precessing magnetizations of isotope-having objects such as human beings. The ability to provide standoff medical surveillance will increase the Biomedical Wellness (BMW) of not only aging American citizens but also persons we seek to track from healthy to diseased states. Taking advantage of the uniform field in the earth, spatio-temporal integration (3 orders due to time and 4 orders due to space) and Higher Order Statistics (HOS), namely, change detection via kurtosis (ΔK), SNR and the ability to make non-locality measurements may be increased to at least the level of the typical hospital NMR devices. Moreover, we present theoretical calculations of magnitude and consider time dynamics of the resonant/relaxing magnetic fields as a function of 3D human motion and distance.

AMM classification is an advanced version of the typical nearest neighbor classifier that allows one to minimize interclass dispersion while at the same time, maximizing intraclass separation. A technique based on the simple orthogonal feature space of Pentland eigenfaces, the combination of these two embodiments will become essential components to a non-cooperative standoff biometric system for military, medical and homeland security applications. The incorporation of robotic assistance further pushes the frontiers of possible surveillance and authentication that can be realized with such a system. The ability to perform out of the line of sight (OLS)-based surveillance adds an additional dimension, and thus novelty, to the already expanding methods to acquire and process environment-specific data.

Images taken of a scene at different times are often acquired from different angles under possibly different lighting conditions. Some of the macroscopic illumination changes between a source image and a target image can be captured by an affine illumination model with two parameterts, viz. a factor by which the contrast is changed and the offset of brightness. We propose an algorithm that embeds a photometric recovery in an iterative geometric recovery process. The convergence of this iteration gives a combined recovery of the geometric and photometric distortion.

Modularity Biometric System (MBS) is an approach to support AiTR of the cooperated and/or non-cooperated standoff biometric in an area persistent surveillance. Advanced active and passive EOIR and RF sensor suite is not considered here. Neither will we consider the ROC, PD vs. FAR, versus the standoff POT in this paper. Our goal is to catch the "most wanted (MW)" two dozens, separately furthermore ad hoc woman MW class from man MW class, given their archrivals sparse front face data basis, by means of various new instantaneous input called probing faces. We present an advanced algorithm: mini-Max classifier, a sparse sample realization of Cramer-Rao Fisher bound of the Maximum Likelihood classifier that minimize the dispersions among the same woman classes and maximize the separation among different man-woman classes, based on the simple feature space of MIT Petland eigen-faces. The original aspect consists of a modular structured design approach at the system-level with multi-level architectures, multiple computing paradigms, and adaptable/evolvable techniques to allow for achieving a scalable structure in terms of biometric algorithms, identification quality, sensors, database complexity, database integration, and component heterogenity. MBS consist of a number of biometric technologies including fingerprints, vein maps, voice and face recognitions with innovative DSP algorithm, and their hardware implementations such as using Field Programmable Gate arrays (FPGAs). Biometric technologies and the composed modularity biometric system are significant for governmental agencies, enterprises, banks and all other organizations to protect people or control access to critical resources.

Two very different research strategies and mathematical modeling cultures will be outlined: Bottom-up theory-driven modeling and top-down data-driven modeling. The former encapsulates prior knowledge and hypotheses, while the latter extracts relevant co-variation patterns within and between large data tables. The former relies on the scientist defining non-linear differential equations to model the dynamics of a process; the latter automatically finds and displays dominant latent structures by multivariate eigen-structure approximation. The two approaches have different strengths and weaknesses. Based on our experiences, we therefore here suggest a possible way for combining these approaches.

Biomedical Wellness (BMW) surveillance system may become indispensible in public health riding on four confluent trends: (i) The surge of retirement waves of WWII baby boomers; (ii) The longevity of the seniors, thanks to 3 decades steady NIH budgets~$23B per year; (iii) The economic melting down is exasperating resources for entitlements; (iv) The emergent Next Gen Internet having the last mile challenges solved entirely and friendly with IT wired and wireless delivery system, thanks to DoD/DARPA pioneer and numerous entrepreneurs visionary efforts. To be effective in job recreation in current economic slag, BMW needs a jump start, WH/HHS should establish a balanced investment policy in health care, not just to illness but to wellness, and a BMW blue ribbon panel (BMWbrp), which must be, however, independent of NIHbrp, recommending additional resources, say $5B budget for BMW infrastructure building that will surely create new jobs to prepare badly needed geriatric caretakers. Being governmental investment, the BMWbrp must come back with a upward-conversion compatible infrastructure Blue-Print, version#1, an open architecture of standard interfaces, workable with distributed data bases and programming languages. The infrastructure must be simple, transparent, scalable, to leverage with a win-win-win(senior) CRADA from the private investments worldwide, e.g. from the insurance companies and home security companies, etc. that can further attract senior community center BOT. To start whenever the rubber meets the road, we shall enlist the military infrastructure, their sensors suite located in DoD & DHS labs supported with their affiliated contractors and universities. However, the down selections should be done under open bids, oversight by BMWbrp, under HHS $5B budget, to translate their sophisticated military persistent and precision surveillance know-how technology to watch out, no longer the enemy of USA but also, the enemy of mankind, the malicious microorganisms and disorders. The degree of friendliness must be demanded and measured by the standoff methodology such as 4 nones: noninvasive, noncontact, and none-stop-to-measure, in this order.

Visualization of multi-dimensional data sets becomes a critical and significant area in modern medical image processing. To analyze such high dimensional data, novel nonlinear embedding approaches become increasingly important to show dependencies among these data in a two- or three-dimensional space. This paper investigates the potential of novel nonlinear dimensional data reduction techniques and compares their results with proven nonlinear techniques when applied to the differentiation of malignant and benign lesions described by high-dimensional data sets arising from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Two important visualization modalities in medical imaging are presented: the mapping on a lower-dimensional data manifold and the image fusion.

This study explores the use of Independent Component Analysis (ICA) applied to normalized logarithmic spectral changes in the activities of brain processes separated by spatial filters learned from electroencephalogram (EEG) data using a temporal ICA. EEG data were collected during 1-2 hour virtual-reality based driving experiments, in which subjects were instructed to maintain their cruising position and compensate for randomly induced drifts using the steering wheel. ICA was first applied to 30-channel EEG data to separate the recorded signals into a sum of maximally temporally independent components (ICs) for each of 15 subjects. Logarithmic spectra of IC activities were then submitted to PCA-ICA to find spectrally fixed and temporally independent modulator (IM) processes. The second ICA detected and modeled independent co-modulatory systems that multiplicatively affect the activities of spatially distinct IC processes. Across subjects, we found two consistent temporally independent modulators: theta-beta and alpha modulators that mediate spectral activations of the distinct cortical areas when the participants experience waves of alternating alertness and drowsiness during long hour simulated driving. Furthermore, the time courses of the theta-beta modulator were highly correlated with concurrent changes in subject driving error (a behavioral index of drowsiness).

In this paper, we present the design of electronic tongue system for multi-ion sensing applications. The ion-sensitive field-effect transistor (ISFET) detects the concentration of a particular ion in aqueous solution. However, when the given chemical solution contains two or more ions, the ISFET sensor can only provide the combined concentration of ions. In this end, our electronic tongue included a blind source separation (BSS) method of independent component analysis (ICA) to process the ISFET signals and to extract the concentrations of individual ions in the solution. The results of ISFET modeling based on fixed interference method (FIM) serve as a priori knowledge to help solve this blind source problem. Experiments are conducted on this electronic tongue system using aqueous solution containing hydrogen and sodium ions flowing through the array of dual H+ ISFET devices. The results of ICA processing successfully determined the concentration of hydrogen ions amidst the presence of sodium ions. This capability of ion separation allows us to move towards the development of smart electronic tongue systems for environmental and water quality monitoring.

AiTR is well developed field of R/D. Nonetheless. a relatively poor under-sampled infrared video may achieve a sharper imagery by smart pre-processing, similar to super-resolution attempts; the difference is in the details. We took a local adaptive contrast enhancement to exploit the pixel intensity correlation, such as smoothness, contrast, and continuity, among neighborhood pixels of variable region sizes, so-called adaptive local contrast enhancement. The final success or failure rate of AiTR will depend on the choice of cost function, such as LMS, etc. In that, we found that a sparse samples do not satisfy the usual underlying Gaussian assumption, of which the Maximum Likelihood, the Bayesian, the Fisher Rao criteria, etc. are usually depending on a priori assumption of dense sampling approaching the Gaussian statistics. Thus, in this paper, we have developed a sparse sampling classifier, called the min-Max classifier for Aided Target Recognition (AiTR), to minimize the intra-class dispersion and at the same to maximize the inter-class separation to select the optimum features vectors. As a standard test case, we choose Petland eigen-faces to benchmark our performance. We apply Szu's lossless divide and conquer theorem solving the NP Complete TSP solution to treat the multiple classes AiTR, in order to achieve min-Max classifier more efficiently than pair-wise SVM classifier.

During the course of routine Type II treatment of one of the authors, it was decided to derive predictive analytical Data Models of the daily sampled vital statistics: namely weight, blood pressure, and blood sugar, to determine if the covariance among the observed variables could yield a descriptive equation based model, or better still, a predictive analytical model that could forecast the expected future trend of the variables and possibly eliminate the number of finger stickings required to montior blood sugar levels. The personal history and analysis with resulting models are presented.