Center for Information Management, Integration and Connectivity
line.jpg (716
bytes)  Home
line.jpg (716
bytes)  About Us
line.jpg (716
bytes)  Contact Us
cimic-noimages.jpg (4076 bytes)

Personnel

Activities

Research

Journals

Publications

Meadowlands Env. Research Inst.

E Commerce




"Content-Based Image Retrieval for Digital Libraries"

The seminar was conducted by James D Carswell
Department of Spatial Information Science and Engineering
University of Maine

on May 26, 2000, 10:00am

This presentation addressed the problem of content-based image retrieval using queries on image-object shape, completely in the raster domain. It focuses on the particularities of digital images encountered in typical libraries of geo-spatial data and presents the development of an environment for visual information management that enables such queries. The query consists of a user-provided raster sketch of the shape of an imaged object. The objective of the search is to retrieve images that contain an object sufficiently similar to the one specified in the query. The new contribution of this work combines the design of a comprehensive digital image database on-line query access strategy through the development of a feature library, image library and metadata library and the necessary matching tools. The matching algorithm is inspired by least-squares matching (lsm), and represents an extension of lsm to function with a variety of raster representations. The image retrieval strategy makes use of a hierarchical organization of linked feature (image-object) shapes within the feature library. The query results are ranked according to statistical scores and the user can subsequently narrow or broaden his/her search according to the previously obtained results and the purpose of the search




"Progressive Transmission of Vector Map Data: Design of Multi-Resolution Digital Libraries"

The seminar was conducted by Michela Bertolotto
National Center for Geographic Information and Analysis Department of Spatial Information Science and Engineering
University of Maine

on May 25, 2000, 10:00am

The exponential network growth and the global connectivity reached in the last few years have had a great impact on the requirements of contemporary and next-generation geographic information systems. Fundamental characteristics include (1) the efficient data access and delivery over the World Wide Web, and (2) the efficient organization of geographic data into structured digital libraries to help users extract the information they need from the available data. The capability to access and deliver data efficiently can be achieved by enabling a geographic information system to transmit data progressively. Progressive transmission of raster images over the World Wide Web has been successfully applied to provide the user with coarser versions of the data before downloading a complete image. This process is particularly useful when trying to access remote data through a slow communication link or in the case of wireless communication, and when datasets are very large. Progressive transmission is still a challenging topic in the vector domain, however. Increasing or decreasing the level of detail of a vector dataset is a complex and time-consuming process. A possible solution relies on the pre-computation of multiple coarser versions of the data to be stored on the server and delivered progressively upon request. The generation of pre-computation of multiple coarser versions of the data to be stored on the server and delivered progressively upon request. The generation of such versions must be done on the basis of sound cartographic principles to preserve consistency. This presentation will show how multiresolution map representations can be utilized as structured (hierarchical) digital libraries of generalized data to enable efficient access and manipulation of geographic data in vector format over the WWW. Within this framework, we have developed a model for generating less detailed representations of maps in vector format within a spatial database. Such a model is built on the basis of a set of generalization operators that perform atomic topological changes on maps. The defined operators guarantee the preservation of topological consistency, an essential property for the usability of generalized data. We will describe the model and its employment for progressive transmission of vector map data. The applicability of progressive transmission techniques for server-based computing and e-commerce will also be of vector map data. The applicability of progressive transmission techniques for server-based computing and e-commerce will also be outlined.




"A WWW-based Environmental Decision Support System"

The seminar will be conducted by Francisco Artigas and Nabil Adam,
CIMIC Center Rutgers University

on March 16, 2000, 12.30 - 2.30pm

Room J11-160, 701 18th Street NW, Washington DC

The speakers will present a WWW-based environmental decision support system designed for planners and decision makers. The system is currently being field tested by the Hackensack Development Commission, a New Jersey state agency with permitting authority over 80,000 hectares of degraded urban wetlands in north New Jersey. The system integrates near-real-time satellite images with real time field environmental parameters to provide insight into the actual state of the environment at any given time. A comprehensive electronic catalog and metadata system provides quick access to historical and current technical reports and planning documents. Interactive maps provide access to administrative information at the parcel level which include parcel boundaries, block and lot numbers, tax assessment, land use and zoning designations and regulations. The system supports interactive attribute data manipulation for modeling run-off under different development scenarios and a visualization module for digital fly-by of the area. The methods and technologies are scalable which will allow to address regions within countries and entire countries in the near future. 

Sponsored by BioNode
POC - Gunars Platais - 202 473 2627
Sharon Esumei -202 473 1906




"Fault Tolerant and Re-configurable Discrete Event Systems"

The seminar was conducted by Houshang Darabi
Industrial Engineering
Rutgers University

on March 10, 2000, 10:00am

Supervisory Control of Discrete Event Systems (SCDES) has been recently investigated by many researchers. Due to their discrete-event nature, a number of systems such as manufacturing systems, workflow management systems (information systems), and computer operating systems are considered to be discrete event systems. Thus, the control of these systems is SCDES problem. Typically, SCDES consists of three main elements: a Discrete Event System (DES), a supervisory controller and an observation network. The observation network is used to report the events of DES to the supervisory controller. Some examples of observation networks are sensory networks in automated manufacturing systems, communication channels in communication systems and on-line shared databases in Internet-based information systems. While the research in SCDES has mainly focused on the synthesis and design of the supervisory controller element of SCDES, there has been little work done on the observation network related problems. In this talk, some of the speaker's recent theoretical extensions on the analysis of observation networks and re-configurable DES are presented. Finite Automata Analysis, partial observation and language normality of SCDES are utilized to derive the theoretical results. It is briefly described how the research objectives are consistent with the current industrial needs and Federal agencies' funding trends. It is also mentioned how these results can be used to provide solutions to the following practical problems:

  • Recognizing redundant control specification of an information system
  • Handling sensor unavailability in manufacturing systems, communication channels failure in communication systems and database information unavailability in on-line computer information systems
  • Optimizing utilization of observation resources (e.g.. sensors in automated manufacturing, human experts in management information systems, etc.) Finally, the future research potentials related to the observation network of SCDES are discussed.

  • Friday Oct 29th, 1999
    10:00 - 11:30
    MEC 203

    Speaker:
    Dr. Eunmi Chang
    Samsung SDS
    Seoul, Korea

    Title:
    The Strategies and Stages of Environmental Information System in Korea


    October 1st, 1999, 11:00 - 12:00 noon.
    Speaker:
    Dr. Nathan S. Netanyahu
    Dept. of Math and Computer Science
    Bar-Ilan University, Israel
    and
    Center for Automation Research
    University of Maryland at College Park

    Title:

    Enhanced Metadata Extraction for NASA's Regional Application Center System


    July 16th, 1999, 10:00a.m.
    Speaker:
    Snehamay Banerjee
    Assoc. Prof. of MIS and Computer Science;
    Rutgers - Camden

    Title:

    Internet EDI: Implementation Strategies & Collaborative Planning Models


    June 9th, 1999, 10:00a.m.
    Speaker:
    Vasilis Megalooikonomou
    Johns Hopkins University
    Division of Neuroradiology

    Title:

    Seeing through the Forest: Mining Lesion-Deficit Associations in a Brain Image Database




    "Revolutionizing Decision Support for the Hubble Space Telescope"

    The seminar was conducted by Ms. Beryl Hoscak & Mr. Ralph Reitan
    Computer Sciences Corporation

    on April 23,1999 , 10:30am - 12:00pm

    Ms. Beryl Hosack and Mr Ralph Reitan will be speaking, in conjunction with Himanshu Parab and Igg Adiwijaya of Rutgers University about the work they have accomplished in the development of an Engineering Data Warehouse supporting the NASA Hubble Space Telescope Program, one of NASAs' "showcase" missions in terms of producing compelling scientific results supported by the implementation of leading edge technologies. It is one of these leading edge technologies that will be addressed within this colloquiem: Data Warehousing of spacecraft telemetry data and associated orbital events. NASA needed to re-design their ground system supporting the Hubble Space Telescope in order to reduce extensive operating costs, thereby extending the life of the Hubble an additional 8 years (through 2010). It was anticipated that operating costs could be reduced by ~ 50% and the implementation of a data warehouse has helped NASA to meet this goal by allowing collaborative on-line problem solving via the WWW to resolve problems with the spacecraft and it's instrument complement. This 2 Terabyte Data Warehouse makes available to its' users worldwide the entire history of engineering data (over the 9 year life of the mission) in near real time. At the time of implementation it was the only data warehouse in the United States to store engineering data. It represented a completely new paradigm for NASA in both the storage of telemetry data and the means by which this data could be queried in a complex, ad-hoc fashion, adding tremendous value to the work of spacecraft and instrument engineers. The effort which will be discussed has applicability far beyond the NASA community, reaching into many other commercial, and federal sectors such as: E-commerce (Interactive TV, WWW, manufacturing environments e.g. computer chips, disk drives and quality management); the Department of Defense (spacecraft programs, remote sensing) Environmental monitoring as well as Power Plant operations. Rutgers University has provided support to this research and development through NASA's CESDIS program which seeks to work in partnership with Universities to solve compelling technical problems and move the results into private industry. Specifically, CIMIC has supported this program by performing a study of commercial ad-hoc query tools which can be used to help Hubble's engineering users to move from simple, non-analytical data extractions to complex ad-hoc queries allowing information to be combined in ways it never could before. In addition to a review of this activity by the CESDIS participants, an overview of the program will be given by Mr. Ralph Reitan, chief Data Warehouse designer and Ms. Beryl Hosack who served as Integration Manager and Historic Load manager for the Data Warehouse. In addition, Mr. Reitan will give a tutorial on developing Data Warehouse Schema.