Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158033
A.S. Bahora, T. Collins, S. Davis, S. Goknur, J. Kearns, T. Liễu, T. Nguyen, J. Zeng, B. Horowitz, S. Patek
We present a study of the technical feasibility of applying peer-to-peer (P2P) technology to the domain of emergency response. The conceptual feasibility of such an endeavor is described in "Integrated Peer-to-Peer Applications for Advanced Emergency Response Systems Part I: Concept of Operations" (Bahora, et al.), which argues that current emergency response technologies do not adequately support the dynamic nature of emergency response. To provide support, peer-to-peer networks have been identified as an integrating architecture. We explore the integration of the peer-to-peer architecture, specifically the HyperCast peer-to-peer communications framework, with streaming video communication, GPS-based location awareness, and information access management control to better support emergency response. Upon conclusion of this research, these functionalities were implemented as prototypes.
{"title":"Integrated peer-to-peer applications for advanced emergency response systems. Part II. Technical feasibility","authors":"A.S. Bahora, T. Collins, S. Davis, S. Goknur, J. Kearns, T. Liễu, T. Nguyen, J. Zeng, B. Horowitz, S. Patek","doi":"10.1109/SIEDS.2003.158033","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158033","url":null,"abstract":"We present a study of the technical feasibility of applying peer-to-peer (P2P) technology to the domain of emergency response. The conceptual feasibility of such an endeavor is described in \"Integrated Peer-to-Peer Applications for Advanced Emergency Response Systems Part I: Concept of Operations\" (Bahora, et al.), which argues that current emergency response technologies do not adequately support the dynamic nature of emergency response. To provide support, peer-to-peer networks have been identified as an integrating architecture. We explore the integration of the peer-to-peer architecture, specifically the HyperCast peer-to-peer communications framework, with streaming video communication, GPS-based location awareness, and information access management control to better support emergency response. Upon conclusion of this research, these functionalities were implemented as prototypes.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126944596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.157999
Li Li, Dong-hai Zhai, F. Jin
Frequent itemsets mining plays an essential role in data mining, but it often generates a large number of redundant itemsets that reduce the efficiency of the mining task. Frequent closed itemsets are subset of frequent itemsets, but they contain all information of frequent itemsets. The most existing methods of frequent closed itemset mining are apriori-based. The efficiency of those methods is limited to the repeated database scan and the candidate set generation. We propose a graph-based algorithm for mining frequent closed itemsets called GFCG (graph-based frequent closed itemset generation). The new algorithm constructs an association graph to represent the frequent relationship between items, and recursively generates frequent closed itemset based on that graph. It scans the database for only two times, and avoids candidate set generation. GFCG outperforms a priori-based algorithm in experiment study and shows good performance both in speed and scale up properties.
{"title":"A graph-based algorithm for frequent closed itemsets mining","authors":"Li Li, Dong-hai Zhai, F. Jin","doi":"10.1109/SIEDS.2003.157999","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.157999","url":null,"abstract":"Frequent itemsets mining plays an essential role in data mining, but it often generates a large number of redundant itemsets that reduce the efficiency of the mining task. Frequent closed itemsets are subset of frequent itemsets, but they contain all information of frequent itemsets. The most existing methods of frequent closed itemset mining are apriori-based. The efficiency of those methods is limited to the repeated database scan and the candidate set generation. We propose a graph-based algorithm for mining frequent closed itemsets called GFCG (graph-based frequent closed itemset generation). The new algorithm constructs an association graph to represent the frequent relationship between items, and recursively generates frequent closed itemset based on that graph. It scans the database for only two times, and avoids candidate set generation. GFCG outperforms a priori-based algorithm in experiment study and shows good performance both in speed and scale up properties.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121196144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158019
Christopher Florentine, M. Isenstein, Jared Q. Libet, Steve Neece, Jim Zeng, Y. Haimes, B. Horowitz
We develop a methodology for analyzing risk to the meat supply, with a specific focus on slaughterhouses. The safety of the nation's meat supply from terrorist attack has become a major concern of the USDA after the September 11/sup th/ attacks on the Pentagon and World Trade Center. The methodology performs a risk assessment through hierarchical holographic modelling (HHM). The goal of the HHM is to capture all observables of a terrorist attack as well as identify all of the vulnerabilities of the target. A 2-player game using a blue team and red team aided in completing a more thorough HHM diagram. This process identified key observables such as researching of toxins and access to meat processing machines as possible indicators of an attack. Bayes' theorem serves as the key tool for analyzing the observables identified through the HHM diagram to assess the likelihood of terrorist attack.
{"title":"A risk-based methodology for combating terrorism","authors":"Christopher Florentine, M. Isenstein, Jared Q. Libet, Steve Neece, Jim Zeng, Y. Haimes, B. Horowitz","doi":"10.1109/SIEDS.2003.158019","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158019","url":null,"abstract":"We develop a methodology for analyzing risk to the meat supply, with a specific focus on slaughterhouses. The safety of the nation's meat supply from terrorist attack has become a major concern of the USDA after the September 11/sup th/ attacks on the Pentagon and World Trade Center. The methodology performs a risk assessment through hierarchical holographic modelling (HHM). The goal of the HHM is to capture all observables of a terrorist attack as well as identify all of the vulnerabilities of the target. A 2-player game using a blue team and red team aided in completing a more thorough HHM diagram. This process identified key observables such as researching of toxins and access to meat processing machines as possible indicators of an attack. Bayes' theorem serves as the key tool for analyzing the observables identified through the HHM diagram to assess the likelihood of terrorist attack.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134018720","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158023
D. Myers, N. Vincent, K. O'Loughlin, D. Marks, C. Snyder, K. P. White, R. Fairbrother, W. Terry
Lockheed Martin Distribution Technologies (LMDT) develops mail-sorting systems for the United States Postal Service. LMDT's development process is extensive and typically several prototypes are implemented before the best solution is found. This process can be both time consuming and costly to the company. Our Capstone team developed a program to help alleviate some design inefficiencies. Specifically we automated the requirements tracking between a simulation program and the company's contracts (Statements of Work, or SOW). In the past at LMDT, simulations have been used after the final system was designed to show the customer the functionality of the ultimate product. Our goal was to increase the usefulness of simulations by utilizing them before physical prototyping. Our product consists of three separate parts. First, a simulation must be built. Next, system requirements must be extracted from the SOW. Finally, an interface must be designed that will automatically check how the simulation results compare to the actual requirements.
{"title":"Simulation-based requirements testing","authors":"D. Myers, N. Vincent, K. O'Loughlin, D. Marks, C. Snyder, K. P. White, R. Fairbrother, W. Terry","doi":"10.1109/SIEDS.2003.158023","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158023","url":null,"abstract":"Lockheed Martin Distribution Technologies (LMDT) develops mail-sorting systems for the United States Postal Service. LMDT's development process is extensive and typically several prototypes are implemented before the best solution is found. This process can be both time consuming and costly to the company. Our Capstone team developed a program to help alleviate some design inefficiencies. Specifically we automated the requirements tracking between a simulation program and the company's contracts (Statements of Work, or SOW). In the past at LMDT, simulations have been used after the final system was designed to show the customer the functionality of the ultimate product. Our goal was to increase the usefulness of simulations by utilizing them before physical prototyping. Our product consists of three separate parts. First, a simulation must be built. Next, system requirements must be extracted from the SOW. Finally, an interface must be designed that will automatically check how the simulation results compare to the actual requirements.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"341 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124211261","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158020
D. Arrington, S. Fairgrieve, J. Silva, M. Wong, D.E. Brown, J. Dalton, D. Ponikvar, B. Gardner, J. Fox
The Capstone team has worked in conjunction with Defense Group Inc. (DGI) to develop a software tool called radiological assessment device (RAD). RAD includes seven tools that perform the following functions: 1) Searches for information regarding sealed sources. 2) Searches for information about historical radiological accidents/incidences. 3) Provides relevant assessment recommendations. 4) Allows for conversion between mass and activity. 5) Provides a mechanism for equivalent radiological unit conversions. 6) Returns critical exposure output based on user inputs relative to the type of scenario encountered. Through the above functionality, the goal is to support first responder decision making during radiological emergency response.
{"title":"Design of a radiological response software tool for use by first responders","authors":"D. Arrington, S. Fairgrieve, J. Silva, M. Wong, D.E. Brown, J. Dalton, D. Ponikvar, B. Gardner, J. Fox","doi":"10.1109/SIEDS.2003.158020","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158020","url":null,"abstract":"The Capstone team has worked in conjunction with Defense Group Inc. (DGI) to develop a software tool called radiological assessment device (RAD). RAD includes seven tools that perform the following functions: 1) Searches for information regarding sealed sources. 2) Searches for information about historical radiological accidents/incidences. 3) Provides relevant assessment recommendations. 4) Allows for conversion between mass and activity. 5) Provides a mechanism for equivalent radiological unit conversions. 6) Returns critical exposure output based on user inputs relative to the type of scenario encountered. Through the above functionality, the goal is to support first responder decision making during radiological emergency response.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130835646","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158014
C.K.H. Lee, H. Lau, K. Yu
Today's manufacturers provide integrated product process development from product design, production and finally to marketing and sales. The whole product life cycle contains lots of data where product data management is crucial for making use of update information to facilitate collaborative product commerce. A product data exchange standard, product information markup language, which is composed a set of well defined XML schema, is introduced to provide a flexible way to encode structured data into a format. In order to response to the rapid change of production environment or market trend, the intelligent operation is embedded in PIML such that it can provide advice for the design engineers to resolve the engineering problem and minimize the ripple effect caused by sudden changes. The intelligent operation is an artificial intelligent problem solving technique, case based reasoning, which retrieves the past most similar case by weighting various problem parameter and then adapts the past solution for the novel problem. A prototype system is developed and is proposed for applying in electroplating industry.
{"title":"Web-based concurrent design with product information markup language","authors":"C.K.H. Lee, H. Lau, K. Yu","doi":"10.1109/SIEDS.2003.158014","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158014","url":null,"abstract":"Today's manufacturers provide integrated product process development from product design, production and finally to marketing and sales. The whole product life cycle contains lots of data where product data management is crucial for making use of update information to facilitate collaborative product commerce. A product data exchange standard, product information markup language, which is composed a set of well defined XML schema, is introduced to provide a flexible way to encode structured data into a format. In order to response to the rapid change of production environment or market trend, the intelligent operation is embedded in PIML such that it can provide advice for the design engineers to resolve the engineering problem and minimize the ripple effect caused by sudden changes. The intelligent operation is an artificial intelligent problem solving technique, case based reasoning, which retrieves the past most similar case by weighting various problem parameter and then adapts the past solution for the novel problem. A prototype system is developed and is proposed for applying in electroplating industry.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123082063","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.157996
M. Brinn, J. M. Fleming, F.M. Hannaka, C. Thomas, P. Beling
Patent citation counts have been used extensively as a patent valuation technique because the information necessary for conducting analysis can be easily gathered and applied to mathematical models. Dominant theory asserts that patents with higher forward citation counts have greater economic value relative to other patents. We focus on evaluating the validity of the patent citation count theory. Three hypotheses that examine the relationship of forward citations with other aspects of patents are presented. In these hypothesis, three relationships are examined: the relationship between a patent's forward citation count and the average number of forward citations for patents in its innovation space, the size of the patent's portfolio, and the patent's second generation forward citation counts. While most of the results proved to be inconclusive, we found evidence that use of forward citations may be problematic.
{"title":"Investigation of forward citation count as a patent analysis method","authors":"M. Brinn, J. M. Fleming, F.M. Hannaka, C. Thomas, P. Beling","doi":"10.1109/SIEDS.2003.157996","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.157996","url":null,"abstract":"Patent citation counts have been used extensively as a patent valuation technique because the information necessary for conducting analysis can be easily gathered and applied to mathematical models. Dominant theory asserts that patents with higher forward citation counts have greater economic value relative to other patents. We focus on evaluating the validity of the patent citation count theory. Three hypotheses that examine the relationship of forward citations with other aspects of patents are presented. In these hypothesis, three relationships are examined: the relationship between a patent's forward citation count and the average number of forward citations for patents in its innovation space, the size of the patent's portfolio, and the patent's second generation forward citation counts. While most of the results proved to be inconclusive, we found evidence that use of forward citations may be problematic.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"91 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126267149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158043
Carolyn Quil, I. Young, J. Reeder, J. Burns, E.J. Bass
The icing branch at NASA Glenn Research Center has funded an exploratory effort to identify requirements for a flight dispatcher-centered Web-based icing training program. Our capstone team has followed an instructional systems design approach, focusing on the analysis stage. We gathered information from several sources, including the federal aviation regulations and aviation personnel at a major and a regional airline to complete this analysis. Our approach included identifying the stakeholder constraints, investigating icing-related weather sources, and developing a scenario framework for training. To aid our analysis process, we developed a low fidelity prototype. These steps led to the identification of recommendations for the development of a Web-based icing training system for flight dispatchers.
{"title":"Toward a Web-based flight dispatcher training tool on icing","authors":"Carolyn Quil, I. Young, J. Reeder, J. Burns, E.J. Bass","doi":"10.1109/SIEDS.2003.158043","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158043","url":null,"abstract":"The icing branch at NASA Glenn Research Center has funded an exploratory effort to identify requirements for a flight dispatcher-centered Web-based icing training program. Our capstone team has followed an instructional systems design approach, focusing on the analysis stage. We gathered information from several sources, including the federal aviation regulations and aviation personnel at a major and a regional airline to complete this analysis. Our approach included identifying the stakeholder constraints, investigating icing-related weather sources, and developing a scenario framework for training. To aid our analysis process, we developed a low fidelity prototype. These steps led to the identification of recommendations for the development of a Web-based icing training system for flight dispatchers.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129239284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.158036
Dave Bunt, G. Lamm, J. Buckingham
Weather and other environmental data are critical information that affects the decision-making abilities of the warfighter and commanders on the battlefield. Disposable, air-droppable, meteorological tower array (DAMTA) consists of a sensor-based platform that will be dispersed over an area of interest. The data (wind speed, direction, temperature, humidity and barometric pressure) and images from DAMTA are used to enhance the accuracy of the current Army forecast models, provide information on remote areas of the battlefield, provide forecasting for future battlefield operations and gather additional information about out of sector areas. We explored the capabilities imagery brings to DAMTA in future military operations and the integration of DAMTA into future combat systems. We focused on: 1) how imagery can assist specific military operations; 2) the trade-offs, attributes and relationships of imagery; 3) the configuration of the imagery-capturing device and 4) the vulnerabilities of imagery-based devices on sensors in several types of environments. We focus on the system integration tasks and tools used to identify the benefits of imagery collection for enhancing intelligence gathering.
{"title":"Benefits of imagery collection for enhancing intelligence gathering","authors":"Dave Bunt, G. Lamm, J. Buckingham","doi":"10.1109/SIEDS.2003.158036","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.158036","url":null,"abstract":"Weather and other environmental data are critical information that affects the decision-making abilities of the warfighter and commanders on the battlefield. Disposable, air-droppable, meteorological tower array (DAMTA) consists of a sensor-based platform that will be dispersed over an area of interest. The data (wind speed, direction, temperature, humidity and barometric pressure) and images from DAMTA are used to enhance the accuracy of the current Army forecast models, provide information on remote areas of the battlefield, provide forecasting for future battlefield operations and gather additional information about out of sector areas. We explored the capabilities imagery brings to DAMTA in future military operations and the integration of DAMTA into future combat systems. We focused on: 1) how imagery can assist specific military operations; 2) the trade-offs, attributes and relationships of imagery; 3) the configuration of the imagery-capturing device and 4) the vulnerabilities of imagery-based devices on sensors in several types of environments. We focus on the system integration tasks and tools used to identify the benefits of imagery collection for enhancing intelligence gathering.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130647318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2003-04-24DOI: 10.1109/SIEDS.2003.157997
Jinshan Tang, S. Acton
We describe the development of a decentralized image retrieval system for education (DIRECT). With DIRECT, content based image retrieval (CBIR) is a service that will run autonomously; the constituent collections will not need to be modified to accommodate DIRECT. DIRECT applies a consistent CBIR mechanism to the entire system and gives CBIR functionality to collections without effort by the collection providers. Collections that do not have their own CBIR system or are not able to invest in a CBIR system (such as small collections or collections that do not contain a large proportion of images) will gain greater distribution of their holdings via DIRECT's CBIR functionality. The DIRECT service finds the collections in the library and requests the images using the same method that the portal uses to search and request images. DIRECT will perform feature extraction on the images and store the features in its own database. A portal or client application can query DIRECT, and it will use the features of the desired image to search the library for images with the same features. DIRECT will automatically monitor collections in order to keep the DIRECT index up-to-date.
{"title":"A decentralized image retrieval system for education","authors":"Jinshan Tang, S. Acton","doi":"10.1109/SIEDS.2003.157997","DOIUrl":"https://doi.org/10.1109/SIEDS.2003.157997","url":null,"abstract":"We describe the development of a decentralized image retrieval system for education (DIRECT). With DIRECT, content based image retrieval (CBIR) is a service that will run autonomously; the constituent collections will not need to be modified to accommodate DIRECT. DIRECT applies a consistent CBIR mechanism to the entire system and gives CBIR functionality to collections without effort by the collection providers. Collections that do not have their own CBIR system or are not able to invest in a CBIR system (such as small collections or collections that do not contain a large proportion of images) will gain greater distribution of their holdings via DIRECT's CBIR functionality. The DIRECT service finds the collections in the library and requests the images using the same method that the portal uses to search and request images. DIRECT will perform feature extraction on the images and store the features in its own database. A portal or client application can query DIRECT, and it will use the features of the desired image to search the library for images with the same features. DIRECT will automatically monitor collections in order to keep the DIRECT index up-to-date.","PeriodicalId":256790,"journal":{"name":"IEEE Systems and Information Engineering Design Symposium, 2003","volume":"331 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2003-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133273550","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：