3D Computer Animation Through Anaglyph Stereo Projection. JUAN LEON (Bergen Community College Paramus, NJ 07652) ROBERT BENNETT (Brookhaven National Laboratory, Upton, NY, 11973)

Stereoscopic imaging is a process of taking a flat, two dimensional pictures, such as a photograph, and giving it a depth of field. This gives the viewer the illusion that the image is coming out towards them. The easiest way to do this is by taking two identical images and putting them side by side, and then having the viewer look at them through an enclosed viewer, such as the View-master, to get the effect. Though it is easy to get two identical images and a View-master, this process is for individual use only, but with today's advances in computing technology, it is easy to create images in stereo on a desktop pc. In order to create a stereoscopic animation, I had to use software that would allow me to render out stereoscopic images and that would also allow me to use real time stereoscopic renders. Softimage's XSI ended up being the program I used to create my animation, simply because it has an openGL renderer and it has many plug-ins capable of rendering stereoscopic images. I was able to create an electric plasma discharge "overview" in 3D. I created a plasma television, a fluorescent bulb, and a Bunsen burner. The modeling and camera animation were easy to create because I've already had experience working with the software. But because the technology of creating stereo images through a computer isn't set at a universal standard, most of the time in creating the 3D illusion was spent in figuring out the settings that the projector and software had to be in to give the best result.

Are You Centered? Automatic Sample Alignment for High-Throughput Protein Crystallography. ANUBHAV JAIN (Cornell University Ithaca, NY 14853) VIVIAN STOJANOFF (Brookhaven National Laboratory, Upton, NY, 11973)

High-throughput protein crystallography using industrial automation techniques have reduced the time needed to conduct protein structure experiments at many facilities around the world, including the X6A beam line at BNL. A key step in these experiments is protein crystal centering, which can take from ten to thirty minutes to perform manually. Automatic crystal centering would help reduce this time and open up the possibility of a fully computer-controlled crystallography experiment. However, crystal centering algorithms to date rely on human guidance or use heuristics which are not accurate enough to find the crystal in many cases. An improvement over these methods is developed using image processing and machine vision techniques. In addition, shape analysis is used to help distinguish which region corresponds to a crystal. Loop centering, which can center large crystals or perform a rough estimate of crystal position, is improved by analyzing the standard deviation information of the image data rather than pixel intensity. Tests are currently being conducted on a diverse array of image data sets to see how this alternative method compares with previous approaches. If successful, this research would mark an important step in fully automated protein structure experiments.

Biometrics Authentication System Face Recognition Using Eigenfaces. ELENA L (Brooklyn College Brooklyn, NY 11210) DR. U. S ROHATGI (Brookhaven National Laboratory, Upton, NY, 11973)

Identifying an individual from his or her face is one of the most non-intrusive modalities in biometrics, yet it is the most challenging one. The face recognition system is part of the Integrated Biometric Recognition System (IBRS). The goal for the project is to interpret eigenface algorithm and to enhance the clarity of the software documentations. The eigenface algorithm was used for the system, which is the combination of a face feature set. The face recognition system captures images, transforms images, reconstructs new images, compares images and match images with the data template. In order to improve the software documentation, we study and analyze the various algorithms for the clarity and ease of understanding to all users. This analysis reveals that we need to further simply the documentation to help users deploy the software. We conclude the algorithms were correctly used for a robust application, however the documentation still needs to be improve to complaint with software.

Construction of a Stereoscopic Model of the Solar System. JASON EMORD (Suffolk County Community College Selden, NY 11784) JONATHAN SMOLANSKY (Suffolk County Community College Selden, NY 11705) DOUGLAS TAMMANY (suffolk community college selden, NY 11784) GORDON SMITH (Brookhaven National Laboratory, Upton, NY, 11973)

The purpose of this venture is to construct a third-dimensional, stereoscopic model of the Solar System by means of Maya 6.5. Accurately scaled three-dimensional representations of the Solar System are virtually non-existent. Vast differences in size, orbital velocity, and semi-major axes present problems for researchers who are trying to realistically portray astronomical bodies. Professors, students, and any interested parties will be offered an up-close-and-personal journey through a small part of the Milky Way Galaxy. Various Internet resources were used to acquire the data necessary for the Solar System's creation. They include, but are not limited to, NASA.com, nationmaster.com, and similar web encyclopedia instruments. Maya 6.5, the program chosen for this project, is an award winning 3-D animation program used by Hollywood production studios in the creation of elaborate special effects scenes. The project was divided into three phases: The first phase involved creating a skeletal model of the Solar System. NURBS spheres represent the planets and their respective moons. Relative sizes were scaled with respect to Jupiter, which made it possible for all the planets to be visible in a single frame. Next, the rotational tilts, orbital inclinations, and orbital eccentricities of the planets were input. Although a method of scaling the relative semi-major axes was found, they were omitted from the final presentation to preserve the single-frame image of the Solar System. The second phase involved planet surface texture maps (skins), acquired from the Internet and wrapped around each sphere. A large sphere was created and shaded black to encompass the Solar System. Phase three involved exporting the model into StereoMovie Maker so that the presentation can be viewed in stereo. Maya's inability to output scenes in Stereo will necessitate its channeling through StereoMovie Maker, a similar program with Stereo capabilities. It will also allow for camera movement throughout the system. Ultimately, this project will function as a platform for those who aspire to build upon the existing model by adding other celestial objects such as galaxies, nebulae, comets, and asteroids.

Creating a New Plug-In for Gatan's Digital Micrograph and Jeol's Scanning Transmission Electron Microscope. KEVIN RYAN (SUNY Albany Albany, NY 12222) HUILIN LI (Brookhaven National Laboratory, Upton, NY, 11973)

Digital Micrograph (DM), written by Gatan Inc., allows remote control of the Scanning Transmission Electron Microscope (STEM), powerful data manipulation for analysis, and customization using their included Software Developers Kit (SDK) plus a high level scripting language. The new Plug-In was created to enable researchers who are working with fragile organic specimens to have full control of the microscope so they can preserve their sample. Many of the built in functions of DM were used, however, this was just a base to build on and improve. The Plug-In was created using Microsoft's' Visual Studio .NET 2003 to compile and revise the SDK while the scripting language built the user interface and incorporated the newly created functions into DM. Whenever a problem was encountered referring to Gatan's support pages or various websites, for example http://www.felmi-zfe.tugraz.at/dm_scripts/dm_scripts/index2.html, would always aid in resolving the situation. When the project was first started the script source code was quite small and basically provided the user interface with no functionality behind it. The script went through many changes during its maturity, like all computer programs in development. After much testing and debugging the program was completed, allowing the researchers to change the magnification and control the beam center and scanning speed. This lets them move and scan different areas digitally instead of manually moving the stage the specimen is on. The advantages of these digital movements are much more precise shifts, which are needed because the specimens are only several hundred nanometers long. Another feature of the program permits the researchers to save their settings in three different modes (focus, search and exposure) to easily locate and capture the image without damaging the frail specimen. At the end of development the program was a multiple file script to create a more powerful interface with a compiled Dynamic Link Library (DLL) behind it that can communicate with and manage the microscope with greater accuracy.

Development of Biometrics. VIRAJ MEHTA (State University of New York - Stony Brook Stony Brook, NY 11794) DR. UPENDRA K. ROHATGI (Brookhaven National Laboratory, Upton, NY, 11973)

Biometric Identification methods are the next level of security in a digital world because they are easy to use but are extremely difficult to compromise. In fact, email passwords and ATM pin numbers can be easily hacked, stolen or misused while biometrics is unique to every human being and uses fingerprints, hand scans, voice samples and iris scans, to identify the person. This technology helps to reduce identity theft and is also used to protect valuable resources of a company or individual. The biometric system can be designed for either individual characteristics such as finger, hand or face, or all of them can be integrated into one identification system. The advantage of integrated system is the increase in the accuracy of identification. An integrated biometric system was developed that combined fingerprints, face scans and hand scan to identify a person. The results obtained from the integrated hardware with the software were good. The software was able to identify authorized users / personnel 95% of the times. Out of the remaining percentage, 2% went towards giving permissions to unauthorized users, and 3% produced "Invalid Image" errors. This research project is the continuation of work that was done last summer. The biometric software was tested further under different Operating Systems other than Microsoft Windows. The desired Operating Systems selected to be used were Debian Linux and Red Hat Enterprise Linux Version 5. The database of existing users was also modified to create a larger database which is run by Oracle Database 9i. The purpose of having a different OS environment, gave us the ability to further test the software, discover any new bugs and continue to improve the efficiency and the stability of the program. Oracle Database is an industry accepted Database Software Application Suite and having the software use the resources of Oracle Database would allow us to increase the reliability, scalability of the software when deployed in an enterprise. The biometric software uses "generic" off the shelf hardware to make the software cost effective and not making it rely upon proprietary hardware. Digital Persona's U.are.U 4000S sensor is being used for fingerprint recognition, and 2 Logitech QuickCam 4000S webcams were being used for face and hand recognition. The conversion or porting of the biometric software from a Windows to a Linux based environment did cause errors but were resolved with the constant feedback with the help of the developers and other beta testers who had access to the software.

Further Development of VBNLTrak. EDWARD D'AZZO-CAISSER (St. Joseph's College Patchogue, NY 11772) AHOVIE KPONOU (Brookhaven National Laboratory, Upton, NY, 11973)

VBNLTrak, written by Andrew Siemion at Brookhaven National Laboratory, has been used by the Pre-injector group's Electron Beam Ion Source (EBIS) to display the results of tracking simulations for designing the Relativistic Heavy Ion Collider (RHIC) EBIS on a Windows compatible system. Originally designed to replace Field Precision's TriComp 2D^TM VTrak, VBNLTrak moved beyond limits in TriComp's visual implementation by handling multiple graph-types at once and by allowing complete graph configuration before a plot, in addition to a variety of customizable options and various output methods. Using Compaq Visual Fortran in conjunction with Interactive Software Services Ltd. (ISS)'s Winteracter Graphics libraries, we continued the progress made on the program by Siemion and resolved some stability issues. Most of the modifications were made in the Winteracter Development Environment (WiDE), with the aid of ISS's Graph Designer. VBNL Trak now has the option to save and load data locations from a control list (.TIN), simplifying a graph retrieval process which once required a series of cumbersome dialog boxes. There is also an option to take the files normally specified in the .TIN and save the data inside of a master file(.TRAK), lessening the number of output files cluttering a folder. As a result, program settings were expanded and augmented for these new changes, offering a wider versatility and more pleasant user experience over-all.

Interactive Stereoscopic 3-D Visualization of Time Resolved Plasma Emission Spectra. OLEKSIY FATYEYEV (Bergen Community College Paramus, NJ 07652) DANIEL MASTRANGELO (Bergen Community College Paramus, NJ 07649) ROBERT BENNETT (Brookhaven National Laboratory, Upton, NY, 11973)

Interactive stereoscopic data visualization is an area of computer graphics providing techniques and algorithms to display large datasets as two- and three-dimensional images, thereby giving scientists the ability to better analyze and discern details in their data. The goal of our effort was to develop and deploy a software package that provides the ability to interpret and visualize arbitrary sized data sets using interactive stereoscopic three-dimensional viewing techniques. Each raw data set consists of a series of frames (two dimensional arrays of emission intensities as a function of wavelength), where each frame is a spectral snapshot of the plasma at a precise unit of time after the initiation (voltage pulse) of the plasma. In order to streamline the processing of the raw data, we have developed a software tool using both open-source and commercial packages: Microsoft Visual Studio .NET 2003 was used as the IDE (Integrated Development Environment) and Visual C++ was the development programming language. We utilized the open-source Visualization Toolkit (vtk) to render the data in three dimensions and another freely available software package, Tool Command Language (Tcl) and its associated graphic user interface (GUI) toolkit (Tk) to create the GUI. This interactive tool will allow the user to study the formation and subsequent decay of active chemical species (excited state atoms, molecules and ions) over the duration of a short-lived plasma event, and should therefore yield greater insight into the underlying plasma processes.

Modeling for 3D Visualization. KYLE MCCOY (Alfred State College Alfred, NY 14802) WENDY SIMPSON (Alfred State College Alfred, NY 14802) MICHAEL MCGUIGAN (Brookhaven National Laboratory, Upton, NY, 11973)

Of particular interest to the design curricula at Alfred State College (ASC) is the expansion of their current capability to create 3D projections of objects on a 2D computer display using 3D software - specifically Autodesk VIZ 2005 and Maya 6.5. In architecture, using computers to model the interior of a building is much more efficient in giving the viewer a true sense of the space than building cardboard and wood models. The computer generated models are able to give a much better sense of the area than the scaled down physical models because they are created at true scale. Using the 3D visualization theatre on the ASC campus to help showcase student designs, both architectural and graphic, will be very useful in giving the viewer a true sense of the design. The focus of this work is primarily architectural based. Proposed was the transfer of models from the computer software Autodesk VIZ 2005 into the theatre projector. A benefit of this would be allowing students to follow the design process from beginning to end. The level of detail is restricted only by the resolution of the computer program and the practical speed of the theater's workstation. Being able to make this output of the model easily and effectively will be of great value to most of the design curricula offered by ASC. This accomplishment will be made available to any establishment with a 3D visualization theater that would like to teach and show any 3D subject more effectively and efficiently. In order to carry out this task, 3D models of a townhouse complex already planned and designed for ASC were modeled using Autodesk VIZ 2005. These models were then placed in their true environment on the terrain developed in Maya 6.5 and rendered into an animated movie file. This movie will be able to be viewed in stereo in the 3D visualization theatre on the ASC campus. A 'user's manual' was also developed in order to explain in greater detail the process taken to successfully complete this project at Brookhaven National Laboratory.

Stereoscopic Imaging of Scanning Electron Micrographs. VALERIE HALL (Nassau Community College Garden City, NY 11530) KARISSA MAGGIO (Nassau Community College Garden City, NY 11530) JOHN SPILETIC (Brookhaven National Laboratory, Upton, NY, 11973)

Scanning electron micrographs exhibit three-dimensional qualities due to the high depth of field inherent in the use of the scanning electron microscope (SEM). Using the program StereoPhoto Maker, we have attempted to render these micrographs viewable in three dimensions. For this to be accomplished, a routine procedure was developed that permitted movement from trial-and-error to a method that yields consistent results. As in conventional photography, creation of a stereoscopic image using the SEM requires a left-eye and right-eye image. SEM stage controls permit sample movement in the X and Y axis, 360 degree rotation, and a tilt of 90 degrees toward and 15 degrees away from the signal detector. Images were captured using successive 2.5 degree increments, horizontal and vertical shifting, and sample rotation. Since rotation could not be measured directly, the SEM-generated scale bar was used to measure increments depending upon magnification. The resultant two-dimensional images were opened in a software application known as Stereo Photo Maker and combined to form a stereoscopic image based on two images of differing perspectives. In addition to the routine method of generating stereo pairs through sample tilting, other methods were also tested. It was determined that lateral shifting does not allow for the production of a visually satisfying stereo image. Rotation, however, produced three dimensional images of equal quality to that of tilted samples. We find that the greater variation of surface structure a sample has, a higher initial degree of tilt is required. However, the study of many more samples will be needed to verify this hypothesis.

Stereoscopic Visualization and its Beneficial Use in the Classroom. SARAH MAINE (St. Joseph's College Patchogue, NY 11772) MELVYN MORRIS (Brookhaven National Laboratory, Upton, NY, 11973)

Visualization is an important component in the educational development of students of all ages. Often, a flat two-dimensional image is inadequate for illustrating the complexity of an object, such as protein structure, electron orbitals, or the human body. Stereoscopic visualization theaters provide a solution for educators in need of a powerful three-dimensional visualization tool. Using a polarized projection system, images that resemble three-dimensional holograms can be created to provide a comprehensive model that can be easily manipulated with the use of a computer. Research involved uncovering and exploring various computer programs capable of producing stereoscopic images. The programs were analyzed for ease of use, functionality, their potential for use in a classroom. Several programs were then selected as recommended programs. Using these programs, a general curriculum was developed for use in advanced high school level and college level courses. PyMOL and Visual Molecular Dynamics (VMD) allow the user to visualize molecular structures obtained from a Protein Data Bank (PDB). Other programs that assist in the implementation of these lessons include the Bayer College of Medicine (BCM) Search Launcher, ClustalW (an alignment program), and the Conserved Property Difference Locator (CPDL). These programs were utilized in developing lessons to enhance the general high school and college level biology curriculum. The stereoscopic theater and the developed lesson plans will provide the additional assistance students may require to understand the three-dimensional structures that would otherwise remain as abstract concepts.

Stereoscopic Visualization Laboratories and Tools for College Curricula. NICK TAYLOR (Jamestown Community College Jamestown, NY 14701) MEL MORRIS (Brookhaven National Laboratory, Upton, NY, 11973)

Quad-buffered stereoscopic visualization involves generating left and right eye images of an object and delivering each image to its designated eye. This is accomplished through denoting each image with light of opposite polarization. Special glasses are needed to allow light of specific polarization through so each eye receives its respective image. Stereo viewing produces depth (z-axis) in an object, which is not seen in a flat two-dimensional image, or even in a traditional three-dimensional image. As students of math and science often have difficulty visualizing fundamental concepts due to the limits of traditional 3D, stereoscopic rendering has firm academic uses. The goal of this project describes in detail potential uses of stereoscopy in a college curriculum. Some of the most beneficial topics to view in stereoscopy include atomic orbitals, chemical mechanisms, biochemistry, human anatomy, microchemistry, and engineering tools. PyMOL and VMD (Visual Molecular Dynamics) are both molecular viewers. These programs open up .pdb files (among others), which can easily be downloaded from the protein data bank. The primary molecule viewed is proteins, of which many are stored in the PDB. Atomic electron orbitals can be viewed using Orbital Viewer. Chemistry mechanisms, biology images, and many other things can be found by searching for VRML (Virtual Reality Modeling Language) files. A player can view any VRML file in stereoscopy and be able to rotate the object.

The Administrative Task of Adding a New Computer to an Existing Network. GABRIEL SANCHEZ (State University of New York - Stony Brook Stony Brook, NY 11790) CHEO TENG (Brookhaven National Laboratory, Upton, NY, 11973)

The National Synchrotron Light Source (NSLS) uses Hewlett-Packard UniX (HP-UX) machines to monitor the daily operations and data collection of particle accelerators. However, these machines must be maintained in order to safely continue the research carried out by the scientists that utilize the NSLS facility. In order to upgrade the NSLS network to a newer HP-UX operating system version, a test machine is securely configured on the existing network, and if there are no compatibility issues, it is safe to upgrade every computer on the network using the test machine's configuration. As part of the system administrative tasks, I developed small resident software designed to report idle users and tracking the location of every computer connected to the private network to maintain the security of the network. Whenever a new machine is configured onto a network, it must successfully initiate the Network Information Service (NIS), the Domain Name System (DNS), and the Network File System (NFS). When invoking NIS and DNS, this test machine acts as a client that connects to a server, then both the server and the client need specific configuration files edited so that the server can recognize the client as part of the network. The purpose of NFS, however, is to mount file systems from one machine onto another, so NIS and DNS must be carried out before this takes place.

VisCMAQ: Visualization Tool for EPA's Community Multiscale Air Quality (CMAQ) Modeling System. MICHAEL KHALIL (New Jersey City University JERSEY CITY, NJ 07304) DOUGLES WIRGHT (Brookhaven National Laboratory, Upton, NY, 11973)

Simulation models describe worldwide phenomena such as the transport and transformation of trace species in the Earth's atmosphere. Computer visualization plays an important role in the evaluation of model results with observations and in detailed studies of the physical and chemical processes represented in the model. The goal of this project is to enhance the VisCTM model, which was created last summer with additional flexibilities such as reading from NetCDF files. Also, the ability to process multiple time steps for multiple species at the same time. VisCTM would be able to read, and visualize different types of data that is written in C Language. These enhancements will allow the new VisCMAQ to accept an incorporation of data sets for various spatial domains and sets of chemical species. At this process the Visualization of the output from Multi scale Air Quality (VisCMAQ) model will be created. This VisMAQ project will link VisCTM tools with the Environmental Protection Agency's Community Multi-scale Air Quality (CMAQ) Model system; which will utilize VisCTM import files written by the IOAPI library that is built upon the NetCDF file format.