US20070143265A2 - Indexing of Digitized Entities - Google Patents

Indexing of Digitized Entities Download PDF

Info

Publication number
US20070143265A2
US20070143265A2 US11/401,129 US40112906A US2007143265A2 US 20070143265 A2 US20070143265 A2 US 20070143265A2 US 40112906 A US40112906 A US 40112906A US 2007143265 A2 US2007143265 A2 US 2007143265A2
Authority
US
United States
Prior art keywords
digitized
entities
server
client system
entity
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
US11/401,129
Other versions
US7539675B2 (en
US20060190445A1 (en
Inventor
Robert Risberg
Nils Andersson
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
picsearch AB
Original Assignee
picsearch AB
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by picsearch AB filed Critical picsearch AB
Priority to US11/401,129 priority Critical patent/US7539675B2/en
Publication of US20060190445A1 publication Critical patent/US20060190445A1/en
Publication of US20070143265A2 publication Critical patent/US20070143265A2/en
Application granted granted Critical
Publication of US7539675B2 publication Critical patent/US7539675B2/en
Adjusted expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/41Indexing; Data structures therefor; Storage structures
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99932Access augmentation or optimizing
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • Y10S707/99934Query formulation, input preparation, or translation
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10TECHNICAL SUBJECTS COVERED BY FORMER USPC
    • Y10STECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y10S707/00Data processing: database and file management or data structures
    • Y10S707/99931Database or file accessing
    • Y10S707/99933Query processing, i.e. searching
    • Y10S707/99935Query augmenting and refining, e.g. inexact access

Definitions

  • the present invention relates generally to indexing of digitized entities in a large and comparatively unstructured data collection such that a relevant search result can be obtained. More particularly the invention relates to a method of indexing digitized entities, such as images, video or audio files. The invention also relates to a computer program, a computer readable medium, a database and a server/client system for indexing digitized entries.
  • the U.S. Pat. No. 6,084,595 describes an indexing method for generating a searchable database from images, such that an image search engine can find content based information in images, which match a user's search query.
  • Feature vectors are extracted from visual data in the images. Primitives, such as color, texture and shape constitute parameters that can be distilled from the images.
  • a feature vector is based on at least one such primitive.
  • the feature vectors associated with the images are then stored in a feature database.
  • a query feature vector will be specified, as well as a distance threshold indicating the maximum distance that is of interest for the query. All images having feature vectors within that distance will be identified by the query. Additional information is computed from the feature vector being associated with each image, which can be used as a search index.
  • An alternative image and search retrieval system is disclosed in the international patent application WO99/22318.
  • the system includes a search engine, which is coupled to an image analyzer that in turn has access to a storage device.
  • Feature modules define particular regions of an image and measurements to make on pixels within the defined region as well as any neighboring regions. The feature modules thus specify parameters and characteristics which are important in a particular image match/search routine. As a result, a relatively rapid comparison of images is made possible.
  • the international patent application WO00/33575 describes a search engine for video and graphics.
  • the document proposes the creation and storage of identifiers by searching an area within a web page near a graphic file or a video file for searchable identification terms. Areas on web pages near links to graphic or video files are also searched for such identification terms. The identification terms found are then stored in a database with references to the corresponding graphic and video files. A user can find graphic or video files by performing a search in the database.
  • search result will, in general, still not be of sufficiently high quality, because the identification terms are not accurate enough.
  • relevant files may either end up comparatively far down in the hit list or be missed completely in the search.
  • the object is achieved by a method of indexing digitized entities as initially described, which is characterized by generating index information for a particular digitized entity on basis of at least one rank parameter.
  • the rank parameter is derived from basic information, which in turn pertains to at least one distinctive feature and at least one locator for each of the digitized entities.
  • the rank parameter indicates a degree of relevance for at least one distinctive feature with respect to each digitized entity.
  • a computer program directly loadable into the internal memory of a digital computer, comprising software for controlling the method described in the above paragraph when said program is run on a computer.
  • a computer readable medium having a program recorded thereon, where the program is to make a computer perform the method described in the penultimate paragraph above.
  • the object is achieved by a database for storing index information relating to digitized entities, which have been generated according to the proposed method.
  • a server/client system for searching for digitized entities in a data collection as initially described, which is characterized in that an index database in the server/client system is organized, such that index information contained therein, for a particular digitized entity comprises at least one rank parameter.
  • the rank parameter is indicative of a degree of relevance for at least one distinctive feature with respect to the digitized entity.
  • the invention provides an efficient tool for finding highly relevant non-text material on the Internet by means of a search query formulated in textual terms.
  • An advantage offered by the invention is that the web pages, or corresponding resources, where the material is located need not contain any text information to generate a hit.
  • a particular signature for each unique digitized entity utilized in the solution according to the invention makes it possible eliminate any duplicate copies of digitized entities in a hit list obtained by a search. Naturally, this further enhances the search quality.
  • FIG. 1 illustrates the generation of a first rank component in a proposed rank parameter according to an embodiment of the invention
  • FIG. 2 illustrates an enhancement of the first rank component according to an embodiment of the invention
  • FIG. 3 illustrates the generation of a second rank component in the proposed rank parameter according to an embodiment of the invention
  • FIG. 4 demonstrates an exemplary structure of a search result according to an embodiment of the invention
  • FIG. 5 shows a block diagram over a server/client system according to an embodiment of the invention.
  • FIG. 6 illustrates, by means of a flow diagram, an embodiment of the method according to the invention.
  • the invention aims at enhancing the relevancy of any distinctive features, for instance keywords, being related to digitized entities and thereby improving the chances of finding relevant entities in future searches.
  • at least one rank parameter is allocated to each distinctive feature that is related to a digitized entity.
  • the embodiment of the invention described below refers to digitized entities in the form of images. However, the digitized entities may equally well include other types of entities that are possible to identify uniquely, such as audio files or video sequences. Moreover, the digitized entities may either constitute sampled representations of analog signals or be purely computer-generated entities.
  • FIG. 1 shows four copies c a -c d of one and the same image n that are stored at different locations in a data collection, for instance in an internetwork, like the Internet.
  • the identity of the image n can be assessed by means of a so-called image signature, which may be determined from a total sum of all pixel values contained in the image.
  • a corresponding identity may, of course, be assessed also for an audio file or a video file.
  • the copies c a -c d of the image n are logically grouped together in a cluster C n .
  • Each copy c a -c d is presumed to be associated with at least one distinctive feature in the form of a keyword.
  • the keywords are data that are not necessarily being shown jointly with the image.
  • the keywords may be collected from data fields which are normally hidden to a visitor of a certain web page.
  • a first copy c a of the image n is associated with the keywords k 1 , k 2 , k 3 , k 4 up to k ja
  • a second copy c b is associated with the keywords k 3 , k 4 , k 7 , k 19 up to k jb
  • a third copy c c is associated with the keywords k 1 , k 3 , k 4 , k 5 up to k jc
  • a fourth copy c d is associated with the keywords k 2 , k 4 , k 9 , k 12 up to k jd .
  • ⁇ n (k 3 ) ⁇ i ⁇ k i , 3 ⁇ C n ⁇
  • ⁇ i ⁇ k i , 3 represents a sum of all occurrences of the keyword k 3 in the cluster C n
  • C n denotes a total number of keywords in the cluster C n , i.e. the sum of unique keywords plus any copies of the same.
  • a particular keyword for instance k 3
  • a first cluster C 1 contains nine copies c 11 -c 19 of a first image n 1
  • a second cluster C 2 contains four copies c 21 -c 24 of a second image n 2
  • a third cluster C 3 contains one copy c 31 of a third image n 3
  • the keyword k 3 occurs twice (affiliated with c 11 and c 12 ) in the first cluster C 1 , three times (affiliated with c 21 , c 22 and c 24 ) in the second cluster C 2 and once (affiliated with c 31 ) in the third cluster C 3
  • the copy c 12 occurs twice in the first cluster C 1 , on one hand associated with the keyword k 3 and on the other hand associated with a different keyword. In both cases, however, it is the same image.
  • the first rank component ⁇ for the keyword k 3 may now be improved by means of a figure reflecting the strength in linkage between the keyword k 3 and the images n 1 -n 3 (or clusters C 1 -C 3 ) to which it has been associated.
  • the keyword k 3 appears to have its strongest link to the second image n 2 , since it is associated with the largest number of copies of this image, namely c 21 , c 22 and c 24 .
  • the keyword k 3 has a second strongest link to the first image n 1 (where it occurs in two out of nine copies), and a third strongest link to the third image n 3 .
  • a normalization with respect to the largest cluster i.e. the cluster which includes the most copies may be used to model this aspect.
  • the largest cluster C 1 includes nine copies c 11 -c 19 . Therefore, a normalization of the keyword k 3 with respect to the images n 1 -n 3 is obtained by multiplying the first rank component ⁇ n (k 3 ) with the respective number of occurrences in each cluster C 1 -C 3 divided by nine.
  • the first rank component ⁇ is made more usable for automated processing if it is also normalized, such that the highest first rank component ⁇ for a particular keyword is equal to 1. This is accomplished by dividing the expression above with the following denominator: ( ⁇ i ⁇ k i , j ) ⁇ max , k j ⁇ C max ⁇ where ( ⁇ i ⁇ k i , j ) ⁇ max , k j denotes the number of occurrences of the keyword k j in the cluster, which includes most occurrences of this keyword k j . For instance, ( ⁇ i ⁇ k i , 3 ) ⁇ max , k 3 is equal to 3 in the present example, since the keyword k 3 occurs most times in the second cluster C 2 , namely three times.
  • the first rank component ⁇ n (k j ) itself constitutes a fair reflection of the relevancy of a keyword k j with respect to the image n.
  • a more accurate figure can be obtained by combining the first rank component ⁇ n (k j ) with a proposed second rank component ⁇ n (k j ), which will be described below.
  • FIG. 3 illustrates how the second rank component ⁇ n (k j ) may be generated according to an embodiment of the invention.
  • a digitized entity e.g. an image 301
  • distinctive features k 1 , k 2 and k 3 for instance in the form of keywords, which are found at various positions P in a descriptive field F.
  • Each distinctive feature k 1 -k 3 is estimated to have a relevance with respect to the digitized entity 301 that depends on the position P in the descriptive field F in which it is found.
  • a weight factor w 1 -w p for each position 1 -p in the descriptive field F reflects this.
  • a first distinctive feature k 1 and a second distinctive feature k 2 are located in a position 1 in the descriptive field F.
  • Both the distinctive feature k 1 and the distinctive feature k 2 occur a number ⁇ 1 times in this position. There are no distinctive features in a second position 2 . However, various distinctive features may be located in following positions 3 to p- 2 (not shown).
  • the field F contains ⁇ 2 copies of the first distinctive feature k 1 in a position p- 1 and ⁇ 1 copies of the second distinctive feature k 2 respective ⁇ 3 copies of a third distinctive feature k 3 in a position p.
  • the distinctive feature k 1 -k 3 is allocated a particular weight factor w 1 -w p . Furthermore, a relevancy parameter s 1 -s 4 is determined for every distinctive feature k 1 -k 3 , which depends on how many times ⁇ 1 - ⁇ 3 the distinctive feature k 1 -k 3 occurs in a position 1 -p relative a total number of distinctive features in this position 1 -p.
  • This parameter s 1 is further weighted with a weight factor w 1 in respect of the digitized entity 301 .
  • the same calculations are performed for all the positions 2 -p to obtain corresponding relevancy parameters s 1 -s 4 for these positions.
  • ⁇ i k i represents a “penalty” that decreases the relevancy for a distinctive feature k j in a position P, for each distinctive feature in this position, which is different from the distinctive feature k j .
  • other alternative formulas for determining the relevancy parameter s P are also conceivable.
  • the variable w i denotes the weight factor for a position i and the variable s i,j denotes the relevancy parameter for a distinctive feature k j in the position i.
  • the second rank component ⁇ is thus given a format k j : ⁇ n ⁇ , where the first component ⁇ n for a particular distinctive feature k i is always equal to 1.
  • Table 1 below shows an explicit example over weight factors w i certain positions P in a descriptive field F related to an image.
  • the second rank component ⁇ n (k j ) is used as an alternative to the first rank component ⁇ n (k j ).
  • the second rank component ⁇ n (k j ) is namely also a per se good descriptor of the relevancy of a keyword k j with respect to the image n.
  • any other range of the merge factors ⁇ ; ⁇ are likewise conceivable.
  • an index database is created, which also at least includes a field for identifying the respective digitized entity and a field containing one or more locators that indicate where the digitized entity can be retrieved.
  • the index database contains an intuitive representation of the digitized entity. If the digitized entity is an image, a thumbnail picture constitutes a suitable representation. If, however, the digitized entity is an audio file or multimedia file, other representations might prove more useful, for instance in the form of logotypes or similar symbols.
  • FIG. 4 demonstrates an exemplary structure of a search result according to an embodiment of the invention.
  • the search result is listed in a table 400, where a first column E contains the identity ID 1 -ID m of the entities that matched the search criteria sufficiently well.
  • a second column K contains an inventory of ranked distinctive features ⁇ (k 1 )- ⁇ (k 23 ) for each digitized entity.
  • a third column R includes a characterizing representation (or an illustrative element) r 1 -r m of the entity and a fourth column L contains at least one locator I 1 -I m to a corresponding “full version” of the entity.
  • the locator I 1 -I m is typically a URL (Universal Resource Locator).
  • URL Universal Resource Locator
  • search result structure may also include arbitrary additional fields.
  • a reduced set of fields may then be presented to a user. It could, for instance, be sufficient to display only the representation r 1 -r m and/or a limited number of the distinctive features, with or/without their respective ranking.
  • FIG. 5 shows a block diagram over a server/client system according to an embodiment of the invention, through which data may be both indexed, searched and retrieved.
  • Digitized entities are stored in a large and rather unstructured data collection 510 , for instance the Internet.
  • An indexing input device 520 gathers information ID n , ⁇ K ⁇ ; L from the data collection 510 with respect to digitized entities contained therein.
  • the information ID n , ⁇ K ⁇ ; L includes at least an identity field ID n that uniquely defines the digitized entity E, a set of distinctive features ⁇ K ⁇ and a locator L. Additional data, such as file size and file type may also be gathered by the indexing input device 520 .
  • an automatic data collector 521 for instance in the form of an web crawler, in the indexing input device 520 regularly accumulates C the information ID n , ⁇ K ⁇ ; L as soon as possible after addition of new items or after updating of already stored items.
  • An index generator 522 in the indexing input device 520 creates index information I E on basis of the information ID n , ⁇ K ⁇ ; L according to the methods disclosed above.
  • An index database 530 stores the index information I E in a searchable format, which is at least adapted to the operation of a search engine 540 .
  • One or more user client terminals 560 are offered a search interface towards the index information I E in the index database 530 via a user client interface 550 .
  • a user may thus enter a query phrase Q, for instance, orally via a voice recognition interface or by typing, via a user client terminal 560 .
  • the user client interface 550 re-formulates the query Q into a search directive, e.g. in the form of a search string S, which is adopted to the working principle of the search engine 540 .
  • the search engine 540 receives the search directive S and performs a corresponding search S′ in the index database 530 .
  • Any records in the database 530 that match the search directives S sufficiently well are sorted out and returned as a hit list ⁇ H ⁇ of digitized entities E to the user client interface 550 .
  • the user client interface 550 re-formats the hit list ⁇ H ⁇ into a search result having a structure H(R, L), which is better suited for human perception and/or adapted to the user client terminal 560 .
  • the hit list ⁇ H ⁇ preferably has the general structure shown in FIG. 4 .
  • the search result H(R, L) presented via the user client terminal 560 may have any other structure that is found appropriate for the specific application.
  • the search result H(R, L) has proven to demonstrate a desirable format when each search term in the hit list ⁇ H ⁇ is normalized before presentation to the user, such that a first combined rank parameter ⁇ n for each search term is equal to 1.
  • Any additional combined rank parameters ⁇ m for the respective search terms may, of course, have arbitrary lower value depending on the result of the search.
  • the signature associated with each unique digitized entity makes it possible eliminate any duplicate copies of digitized entities in the search result H(R, L). Such elimination produces a search result H(R, L) of very high quality and relevance.
  • the data sent to the user client terminal 560 includes a characteristic representation R of the digitized entities in the hit list ⁇ H ⁇ and corresponding locators L, e.g. URL, for indicating at least one storage location in the data collection 510 .
  • the latter gives the user at least a theoretical possibility to retrieve full versions of the digitized entities. In practice, however, the retrieval may be restricted in various ways, for instance by means of copyright protection and therefore require the purchase of the relevant rights.
  • the units 510 - 560 may either be physically separated from each other or be co-located in arbitrary combination.
  • a first step 601 involves input of basic information that contains one or more distinctive features being related to digitized entities in a data collection.
  • a following step 602 creates rank parameters for each of the digitized entities on the basis of the input information.
  • a step 603 generates a searchable index for the rank parameters and finally, the searchable index is stored in a searchable database in a step 604 .
  • All of the process steps, as well as any sub-sequence of steps, described with reference to the FIG. 6 above may be controlled by means of a computer program being directly loadable into the internal memory of a computer, which includes appropriate software for controlling the necessary steps when the program is run on a computer.
  • the computer program can likewise be recorded onto arbitrary kind of computer readable medium.

Abstract

The invention relates to indexing of digitized entities in a large and comparatively unstructured data collection, for instance the Internet, such that text-based searches with respect to the data collection can be ordered via a user client terminal. Index information is generated for each digitized entity, which contains distinctive features being ranked according to a rank parameter. The rank parameter indicates a degree of relevance of particular distinctive feature with respect to a given digitized entity and is derived from fields or tags associated with one or more copies of the digitized entity in the data collection. The index information is stored in a searchable database, which is accessible via a user client interface and a search engine. The derived distinctive features and the rank parameter thus provides a possibility to carry out text-based searches in respect of non-text digitized entities, such as images, audio files and video sequences and obtain a highly relevant search result.

Description

    CROSS-REFERENCE TO RELATED APPLICATIONS
  • This application is a divisional of U.S. application Ser. No. 10/471,105, filed Sep. 5, 2003, currently pending, which was a national stage filing under 35 U.S.C. 371 of PCT/SE02/00462 filed Mar. 13, 2002, which International Application was published by the International Bureau in English on Sep. 19, 2002, which is hereby incorporated herein in its entirety by reference.
  • BACKGROUND OF THE INVENTION
  • 1) Field of the Invention
  • The present invention relates generally to indexing of digitized entities in a large and comparatively unstructured data collection such that a relevant search result can be obtained. More particularly the invention relates to a method of indexing digitized entities, such as images, video or audio files. The invention also relates to a computer program, a computer readable medium, a database and a server/client system for indexing digitized entries.
  • 2) Description of Related Art
  • Search engines and index databases for automatically finding information in digitized text banks have been known for decades. In recent years the rapid growth of the Internet has intensified the development in this area. Consequently, there are today many examples of very competent tools for finding text information in large and comparatively unstructured data collections or networks, such as the Internet.
  • As the use of the Internet has spread to a widened group of users, the content of web pages and other resources has diversified to include not only text, but also other types of digitized entities, like graphs, images, video sequences, audio sequences and various other types of graphical or acoustic files. An exceptionally wide range of data formats may represent these files. However, they all have one feature in common, namely that they per se lack text information. Naturally, this fact renders a text search for the information difficult. Various attempts to solve this problem have nevertheless already been made.
  • For instance, the U.S. Pat. No. 6,084,595 describes an indexing method for generating a searchable database from images, such that an image search engine can find content based information in images, which match a user's search query. Feature vectors are extracted from visual data in the images. Primitives, such as color, texture and shape constitute parameters that can be distilled from the images. A feature vector is based on at least one such primitive. The feature vectors associated with the images are then stored in a feature database. When a query is submitted to the search engine, a query feature vector will be specified, as well as a distance threshold indicating the maximum distance that is of interest for the query. All images having feature vectors within that distance will be identified by the query. Additional information is computed from the feature vector being associated with each image, which can be used as a search index.
  • An alternative image and search retrieval system is disclosed in the international patent application WO99/22318. The system includes a search engine, which is coupled to an image analyzer that in turn has access to a storage device. Feature modules define particular regions of an image and measurements to make on pixels within the defined region as well as any neighboring regions. The feature modules thus specify parameters and characteristics which are important in a particular image match/search routine. As a result, a relatively rapid comparison of images is made possible.
  • The international patent application WO00/33575 describes a search engine for video and graphics. The document proposes the creation and storage of identifiers by searching an area within a web page near a graphic file or a video file for searchable identification terms. Areas on web pages near links to graphic or video files are also searched for such identification terms. The identification terms found are then stored in a database with references to the corresponding graphic and video files. A user can find graphic or video files by performing a search in the database.
  • However, the search result will, in general, still not be of sufficiently high quality, because the identification terms are not accurate enough. Hence, relevant files may either end up comparatively far down in the hit list or be missed completely in the search.
  • BRIEF SUMMARY OF THE INVENTION
  • It is therefore an object of the present invention to alleviate the problem above and thus provide an improved solution for finding relevant digitized entities, such as images, video files or audio files, by means of an automatic search being performed with respect to a large and relatively unstructured data collection, such as the Internet.
  • According to one aspect of the invention the object is achieved by a method of indexing digitized entities as initially described, which is characterized by generating index information for a particular digitized entity on basis of at least one rank parameter. The rank parameter is derived from basic information, which in turn pertains to at least one distinctive feature and at least one locator for each of the digitized entities. The rank parameter indicates a degree of relevance for at least one distinctive feature with respect to each digitized entity.
  • According to another aspect of the invention these objects are achieved by a computer program directly loadable into the internal memory of a digital computer, comprising software for controlling the method described in the above paragraph when said program is run on a computer.
  • According to yet another aspect of the invention these objects are achieved by a computer readable medium, having a program recorded thereon, where the program is to make a computer perform the method described in the penultimate paragraph above.
  • According to an additional aspect of the invention the object is achieved by a database for storing index information relating to digitized entities, which have been generated according to the proposed method.
  • According to yet an additional aspect of the invention the object is achieved by a server/client system for searching for digitized entities in a data collection as initially described, which is characterized in that an index database in the server/client system is organized, such that index information contained therein, for a particular digitized entity comprises at least one rank parameter. The rank parameter is indicative of a degree of relevance for at least one distinctive feature with respect to the digitized entity.
  • The invention provides an efficient tool for finding highly relevant non-text material on the Internet by means of a search query formulated in textual terms. An advantage offered by the invention is that the web pages, or corresponding resources, where the material is located need not contain any text information to generate a hit.
  • This is an especially desired feature, in comparison to the known solutions, since in many cases the non-text material may be accompanied by rather laconic, but counter intuitive text portions.
  • A particular signature for each unique digitized entity utilized in the solution according to the invention makes it possible eliminate any duplicate copies of digitized entities in a hit list obtained by a search. Naturally, this further enhances the search quality.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • The present invention is now to be explained more closely by means of preferred embodiments, which are disclosed as examples, and with reference to the attached drawings.
  • FIG. 1 illustrates the generation of a first rank component in a proposed rank parameter according to an embodiment of the invention,
  • FIG. 2 illustrates an enhancement of the first rank component according to an embodiment of the invention,
  • FIG. 3 illustrates the generation of a second rank component in the proposed rank parameter according to an embodiment of the invention,
  • FIG. 4 demonstrates an exemplary structure of a search result according to an embodiment of the invention,
  • FIG. 5 shows a block diagram over a server/client system according to an embodiment of the invention, and
  • FIG. 6 illustrates, by means of a flow diagram, an embodiment of the method according to the invention.
  • DETAILED DESCRIPTION OF THE INVENTION
  • The invention aims at enhancing the relevancy of any distinctive features, for instance keywords, being related to digitized entities and thereby improving the chances of finding relevant entities in future searches. In order to achieve this objective, at least one rank parameter is allocated to each distinctive feature that is related to a digitized entity. The embodiment of the invention described below refers to digitized entities in the form of images. However, the digitized entities may equally well include other types of entities that are possible to identify uniquely, such as audio files or video sequences. Moreover, the digitized entities may either constitute sampled representations of analog signals or be purely computer-generated entities.
  • FIG. 1 shows four copies ca-cd of one and the same image n that are stored at different locations in a data collection, for instance in an internetwork, like the Internet. The identity of the image n can be assessed by means of a so-called image signature, which may be determined from a total sum of all pixel values contained in the image. A corresponding identity may, of course, be assessed also for an audio file or a video file.
  • The copies ca-cd of the image n are logically grouped together in a cluster Cn. Each copy ca-cd is presumed to be associated with at least one distinctive feature in the form of a keyword. Typically, the keywords are data that are not necessarily being shown jointly with the image. On the contrary, the keywords may be collected from data fields which are normally hidden to a visitor of a certain web page. Thus, the keywords may be taken from HTML-tags such as Meta, Img or Title (HTML=HyperText Mark-up Language).
  • In this example a first copy ca of the image n is associated with the keywords k1, k2, k3, k4 up to kja, a second copy cb is associated with the keywords k3, k4, k7, k19 up to kjb, a third copy cc is associated with the keywords k1, k3, k4, k5 up to kjc, and a fourth copy cd is associated with the keywords k2, k4, k9, k12 up to kjd. In order to determine the relevance of a particular keyword, say k3, with respect to the image n a first rank component ⊖n(k3) is calculated according to the expression: Γ n ( k 3 ) = i k i , 3 C n
    where i k i , 3
    represents a sum of all occurrences of the keyword k3 in the cluster Cn and |Cn| denotes a total number of keywords in the cluster Cn, i.e. the sum of unique keywords plus any copies of the same.
  • However, it is also quite common that a particular keyword, for instance k3, is associated with many different images. This is illustrated in FIG. 2. Here, a first cluster C1 contains nine copies c11-c19 of a first image n1, a second cluster C2 contains four copies c21-c24 of a second image n2 and a third cluster C3 contains one copy c31 of a third image n3. The keyword k3 occurs twice (affiliated with c11 and c12) in the first cluster C1, three times (affiliated with c21, c22 and c24) in the second cluster C2 and once (affiliated with c31) in the third cluster C3. The copy c12 occurs twice in the first cluster C1, on one hand associated with the keyword k3 and on the other hand associated with a different keyword. In both cases, however, it is the same image.
  • The first rank component Γ for the keyword k3 may now be improved by means of a figure reflecting the strength in linkage between the keyword k3 and the images n1-n3 (or clusters C1-C3) to which it has been associated. The keyword k3 appears to have its strongest link to the second image n2, since it is associated with the largest number of copies of this image, namely c21, c22 and c24. Correspondingly, the keyword k3 has a second strongest link to the first image n1 (where it occurs in two out of nine copies), and a third strongest link to the third image n3. A normalization with respect to the largest cluster (i.e. the cluster which includes the most copies) may be used to model this aspect. In this example, the largest cluster C1 includes nine copies c11-c19. Therefore, a normalization of the keyword k3 with respect to the images n1-n3 is obtained by multiplying the first rank component Γn(k3) with the respective number of occurrences in each cluster C1-C3 divided by nine. Of course, the general expression becomes: Γ n ( k j ) = i k i , j C n · C n C max = i k i , j C max
    where |Cmax| is the largest number of keywords in a cluster for any image that includes the relevant keyword kj, for instance k3.
  • The first rank component Γ is made more usable for automated processing if it is also normalized, such that the highest first rank component Γ for a particular keyword is equal to 1. This is accomplished by dividing the expression above with the following denominator: ( i k i , j ) max , k j C max
    where ( i k i , j ) max , k j
    denotes the number of occurrences of the keyword kj in the cluster, which includes most occurrences of this keyword kj. For instance, ( i k i , 3 ) max , k 3
    is equal to 3 in the present example, since the keyword k3 occurs most times in the second cluster C2, namely three times.
  • Hence, the first rank component Γn(kj) for an image n with respect to keyword kj is preferably modelled by the simplified expression: Γ n ( k j ) = i k i , j ( i k i , j ) max , k j
    where i k i , j
    represents the sum of all occurrences of the keyword kj in the cluster Cn and ( i k i , j ) max , k j
    is the number of occurrences of the keyword kj in the cluster, which includes most occurrences of this keyword kj.
  • However, in order to the improve the search performance in a database containing indexed elements referring to the digitized entities, it is preferable to build an inverted index on keywords, such that a set of first rank components Γ is instead expressed for each keyword kj. Thus, according to a preferred embodiment of the invention, the format of the first rank component is kj:{Γn}. Consequently, the keyword k3 in the example above obtains the following set of first rank components:
    k 3:{Γ2=1; Γ1=2/3; Γ3=1/3}
  • The first rank component Γn(kj) itself constitutes a fair reflection of the relevancy of a keyword kj with respect to the image n. However, a more accurate figure can be obtained by combining the first rank component Γn(kj) with a proposed second rank component Πn(kj), which will be described below.
  • FIG. 3 illustrates how the second rank component Πn(kj) may be generated according to an embodiment of the invention.
  • A digitized entity, e.g. an image 301, is presumed to be associated with distinctive features k1, k2 and k3, for instance in the form of keywords, which are found at various positions P in a descriptive field F. Each distinctive feature k1-k3 is estimated to have a relevance with respect to the digitized entity 301 that depends on the position P in the descriptive field F in which it is found. A weight factor w1-wp for each position 1-p in the descriptive field F reflects this. In the illustrated example, a first distinctive feature k1 and a second distinctive feature k2 are located in a position 1 in the descriptive field F. Both the distinctive feature k1 and the distinctive feature k2 occur a number η1 times in this position. There are no distinctive features in a second position 2. However, various distinctive features may be located in following positions 3 to p-2 (not shown). The field F contains η2 copies of the first distinctive feature k1 in a position p-1 and η1 copies of the second distinctive feature k2 respective η3 copies of a third distinctive feature k3 in a position p.
  • Hence, depending on the position 1-p in which a certain distinctive feature k1-k3 is found, the distinctive feature k1-k3 is allocated a particular weight factor w1-wp. Furthermore, a relevancy parameter s1-s4 is determined for every distinctive feature k1-k3, which depends on how many times η13 the distinctive feature k1-k3 occurs in a position 1-p relative a total number of distinctive features in this position 1-p.
  • Thus, both the first distinctive feature k1 and the second distinctive feature k1 obtain the same relevancy parameter s1, which can be calculated as s11/(2η1)=1/2 in the position 1. This parameter s1 is further weighted with a weight factor w1 in respect of the digitized entity 301. The same calculations are performed for all the positions 2-p to obtain corresponding relevancy parameters s1-s4 for these positions.
  • Alternatively, the relevancy parameter sP can be determined as s P ( k j i ) = 1 - γ i k i ,
    where γΣiki represents a “penalty” that decreases the relevancy for a distinctive feature kj in a position P, for each distinctive feature in this position, which is different from the distinctive feature kj. Naturally, other alternative formulas for determining the relevancy parameter sP are also conceivable.
  • Nevertheless, a combined measure is determined, which fully captures the relationship between distinctive features kj and digitized entities n. The expression: ( n , k j ) = i = 1 p ( w i · s i , j ) 2 i = 1 p w i 2
    constitutes a reflection of the relevance of a distinctive feature kj with respect to a particular digitized entity n. The variable wi denotes the weight factor for a position i and the variable si,j denotes the relevancy parameter for a distinctive feature kj in the position i.
  • In analogy with the first rank component Γ, its is preferable also to normalize and build an inverted index on keywords. The second rank component Π is thus given a format kj:{Πn}, where the first component Πn for a particular distinctive feature ki is always equal to 1.
  • Table 1 below shows an explicit example over weight factors wi certain positions P in a descriptive field F related to an image.
    TABLE 1
    Position (P) Field (F) Weight factor (wP)
    1 pageSite 50
    2 pageDir 40
    3 pageName 50
    4 pageTitle 80
    5 pageDescription 90
    6 pageKeywords 90
    7 pageText 20
    8 imageSite 50
    9 imageDir 60
    10 imageName 100
    11 imageAlt 100
    12 imageAnchor 80
    13 imageCenterCaption 90
    14 imageCellCaption 90
    15 imageParagraphCaption 90
  • According to an embodiment of the invention, the second rank component Πn(kj) is used as an alternative to the first rank component Γn(kj). The second rank component Πn(kj) is namely also a per se good descriptor of the relevancy of a keyword kj with respect to the image n.
  • In a preferred embodiment of the invention, however, the first rank component Γ and the second rank component Π are merged into a combined rank parameter Δ according to the expression: Δ = ( αΓ ) 2 + ( β ) 2 α 2 + β 2
    where α is a first merge factor and β is a second merge factor. For instance, 0≦α≦1 and 0≦β≦1. However, any other range of the merge factors α;β are likewise conceivable.
  • Finally and in similarity with the first and second rank components Γ and Π respectively, it is preferable to normalize and build an inverted index on keywords, such that it obtains a format kj:{Δn}, where the first component Δn for a particular distinctive feature kj is always equal to 1.
  • When all, or at least a sufficiently large portion, of the digitized entities in the data collection have been related to at least one distinctive feature and a corresponding rank component/parameter (Γ, Π or Δ), an index database is created, which also at least includes a field for identifying the respective digitized entity and a field containing one or more locators that indicate where the digitized entity can be retrieved. Moreover, it is preferable if the index database contains an intuitive representation of the digitized entity. If the digitized entity is an image, a thumbnail picture constitutes a suitable representation. If, however, the digitized entity is an audio file or multimedia file, other representations might prove more useful, for instance in the form of logotypes or similar symbols.
  • FIG. 4 demonstrates an exemplary structure of a search result according to an embodiment of the invention. The search result is listed in a table 400, where a first column E contains the identity ID1-IDm of the entities that matched the search criteria sufficiently well. A second column K contains an inventory of ranked distinctive features Δ(k1)-Δ(k23) for each digitized entity. A third column R includes a characterizing representation (or an illustrative element) r1-rm of the entity and a fourth column L contains at least one locator I1-Im to a corresponding “full version” of the entity. In case the data collection is an internetwork the locator I1-Im is typically a URL (Universal Resource Locator). However, any other type of address is equally well conceivable.
  • Naturally, the search result structure may also include arbitrary additional fields. A reduced set of fields may then be presented to a user. It could, for instance, be sufficient to display only the representation r1-rm and/or a limited number of the distinctive features, with or/without their respective ranking.
  • FIG. 5 shows a block diagram over a server/client system according to an embodiment of the invention, through which data may be both indexed, searched and retrieved. Digitized entities are stored in a large and rather unstructured data collection 510, for instance the Internet. An indexing input device 520 gathers information IDn, {K}; L from the data collection 510 with respect to digitized entities contained therein. The information IDn, {K}; L includes at least an identity field IDn that uniquely defines the digitized entity E, a set of distinctive features {K} and a locator L. Additional data, such as file size and file type may also be gathered by the indexing input device 520. It is irrelevant exactly how the information IDn, {K}; L is entered into the indexing input device 520. However, according to a preferred embodiment of the invention, an automatic data collector 521, for instance in the form of an web crawler, in the indexing input device 520 regularly accumulates C the information IDn, {K}; L as soon as possible after addition of new items or after updating of already stored items. An index generator 522 in the indexing input device 520 creates index information IE on basis of the information IDn, {K}; L according to the methods disclosed above. An index database 530 stores the index information IE in a searchable format, which is at least adapted to the operation of a search engine 540.
  • One or more user client terminals 560 are offered a search interface towards the index information IE in the index database 530 via a user client interface 550. A user may thus enter a query phrase Q, for instance, orally via a voice recognition interface or by typing, via a user client terminal 560. Preferably, however not necessarily, the user client interface 550 re-formulates the query Q into a search directive, e.g. in the form of a search string S, which is adopted to the working principle of the search engine 540. The search engine 540 receives the search directive S and performs a corresponding search S′ in the index database 530.
  • Any records in the database 530 that match the search directives S sufficiently well are sorted out and returned as a hit list {H} of digitized entities E to the user client interface 550. If necessary, the user client interface 550 re-formats the hit list {H} into a search result having a structure H(R, L), which is better suited for human perception and/or adapted to the user client terminal 560. The hit list {H} preferably has the general structure shown in FIG. 4. However, the search result H(R, L) presented via the user client terminal 560 may have any other structure that is found appropriate for the specific application. If the query phrase Q comprises more than one search term (or distinctive feature), the search result H(R, L) has proven to demonstrate a desirable format when each search term in the hit list {H} is normalized before presentation to the user, such that a first combined rank parameter Δn for each search term is equal to 1. For instance, a hit list {H} resulting from a search query Q=“ferarri 550” is normalized such that the first combined rank parameter Δn=1 both with respect to “ferarri” and with respect to “550”. Any additional combined rank parameters Δm for the respective search terms may, of course, have arbitrary lower value depending on the result of the search.
  • The signature associated with each unique digitized entity makes it possible eliminate any duplicate copies of digitized entities in the search result H(R, L). Such elimination produces a search result H(R, L) of very high quality and relevance.
  • A minimum requirement is that the data sent to the user client terminal 560 includes a characteristic representation R of the digitized entities in the hit list {H} and corresponding locators L, e.g. URL, for indicating at least one storage location in the data collection 510. The latter gives the user at least a theoretical possibility to retrieve full versions of the digitized entities. In practice, however, the retrieval may be restricted in various ways, for instance by means of copyright protection and therefore require the purchase of the relevant rights.
  • The units 510-560 may either be physically separated from each other or be co-located in arbitrary combination.
  • In order to sum up, a method of generating a searchable index for digitized entities according to an embodiment of the invention will now be described with reference to a flow diagram in the FIG. 6.
  • A first step 601 involves input of basic information that contains one or more distinctive features being related to digitized entities in a data collection. A following step 602 creates rank parameters for each of the digitized entities on the basis of the input information. Then, a step 603 generates a searchable index for the rank parameters and finally, the searchable index is stored in a searchable database in a step 604.
  • All of the process steps, as well as any sub-sequence of steps, described with reference to the FIG. 6 above may be controlled by means of a computer program being directly loadable into the internal memory of a computer, which includes appropriate software for controlling the necessary steps when the program is run on a computer. The computer program can likewise be recorded onto arbitrary kind of computer readable medium.
  • The term “comprises/comprising” when used in this specification is taken to specify the presence of stated features, integers, steps or components. However, the term does not preclude the presence or addition of one or more additional features, integers, steps or components or groups thereof.
  • The invention is not restricted to the described embodiments in the figures, but may be varied freely within the scope of the claims.

Claims (11)

1. A server/client system for searching for digitized entities in a data collection comprising:
an indexing input device for collecting basic information pertaining to at least one distinctive feature and at least one locator for each digitized entity in a set of entities from the data collection,
an index database for storing index information relating to the digitized entities in the set,
a search engine for receiving search directives and in response thereto performing searches in the index database, and
a user client interface for receiving a search request from at least one user client terminal, forwarding the search request as a search directive to the search engine, receiving a hit list of digitized entities and returning a result of a corresponding search in the index database to the at least one user client terminal,
wherein the index database is organized such that the index information for a particular digitized entity comprises at least one rank parameter, which is indicative of a degree of relevance for at least one distinctive feature with respect to that digitized entity.
2. A server/client system according to claim 1, wherein the indexing input device includes an index generator for receiving the basic information and producing in response thereto the at least one rank parameter.
3. A server/client system according to claim 1, wherein the rank parameter includes a first rank component indicating a ranking of at least one individual distinctive feature related to the digitized entity on the basis of a relative occurrence of the at least one individual distinctive feature with respect to one or more copies of the digitized entity in the data collection.
4. A server/client system according to claim 1, wherein the rank parameter includes a second rank component indicating a ranking of at least one individual distinctive feature related to the digitized entity on basis of
a position of the least one individual distinctive feature in a descriptive field associated with the digitized entity, and
a relevancy parameter reflecting a distinctive feature's significance in relation to other distinctive features in a particular position in the descriptive field.
5. A server/client system according to claim 1, wherein the indexing input device includes an automatic data collector for finding relevant digitized entities in the data collection and creating therefrom the set of entities.
6. A server/client system according to claim 1, wherein each of the digitized entities in the hit list is associated with an identifier, at least one rank parameter and at least one locator for indicating a storage location in the data collection.
7. A server/client system according to claim 6, wherein each of the digitized entities in the hit list is also associated with an illustrative element to be displayed on the user client terminal together with the respective digitized entity.
8. A server/client system according to claim 7, wherein the illustrative element is a thumbnail picture.
9. A server/client system according to claim 5, wherein the data collection is an internetwork and the indexing input device includes a web crawler.
10. A server/client system according to claim 1, wherein the digitized entities include at least one of the file types: a text document, an image, a video sequence and an audio sequence.
11. A server/client system according to claim 10, wherein the digitized entities are stored in at least one of the formats: AIF, AIFC, AIFF, AU, AVI, BMP, DIVX, DOC, EPS, GIF, ICO, JPEG, JPG, MOV, MP3, MP4, MPEG, MPEG4, MPG, PDF, PNG, PPT, PS, QT, RA, RAM, RAS, SND, TIF, TIFF, VCD, WAV, XLS and XMP.
US11/401,129 2001-03-13 2006-04-10 Indexing of digitized entities Expired - Fee Related US7539675B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US11/401,129 US7539675B2 (en) 2001-03-13 2006-04-10 Indexing of digitized entities

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
SE0100856A SE520533C2 (en) 2001-03-13 2001-03-13 Method, computer programs and systems for indexing digitized devices
SE0100856-4 2001-03-13
PCT/SE2002/000462 WO2002073463A1 (en) 2001-03-13 2002-03-13 Indexing of digitised entities
US10/471,105 US7516129B2 (en) 2001-03-13 2002-03-13 Method of indexing digitized entities in a data collection to facilitate searching
SEPCT/SE02/00462 2002-03-13
US11/401,129 US7539675B2 (en) 2001-03-13 2006-04-10 Indexing of digitized entities

Related Parent Applications (3)

Application Number Title Priority Date Filing Date
US10/471,105 Division US7516129B2 (en) 2001-03-13 2002-03-13 Method of indexing digitized entities in a data collection to facilitate searching
PCT/SE2002/000462 Division WO2002073463A1 (en) 2001-03-13 2002-03-13 Indexing of digitised entities
US10471105 Division 2002-03-13

Publications (3)

Publication Number Publication Date
US20060190445A1 US20060190445A1 (en) 2006-08-24
US20070143265A2 true US20070143265A2 (en) 2007-06-21
US7539675B2 US7539675B2 (en) 2009-05-26

Family

ID=20283328

Family Applications (2)

Application Number Title Priority Date Filing Date
US10/471,105 Expired - Fee Related US7516129B2 (en) 2001-03-13 2002-03-13 Method of indexing digitized entities in a data collection to facilitate searching
US11/401,129 Expired - Fee Related US7539675B2 (en) 2001-03-13 2006-04-10 Indexing of digitized entities

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US10/471,105 Expired - Fee Related US7516129B2 (en) 2001-03-13 2002-03-13 Method of indexing digitized entities in a data collection to facilitate searching

Country Status (4)

Country Link
US (2) US7516129B2 (en)
EP (1) EP1377915A1 (en)
SE (1) SE520533C2 (en)
WO (1) WO2002073463A1 (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090292671A1 (en) * 2008-05-20 2009-11-26 Microsoft Corporation Motion-based data review and zoom
WO2010011344A1 (en) * 2008-07-23 2010-01-28 Ltu Technologies S.A.S. Frame based video matching
US20120030180A1 (en) * 2010-07-29 2012-02-02 Sap Ag Archive-system-independent archive-type objects
US20150169588A1 (en) * 2005-10-12 2015-06-18 Google Inc. Entity Display Priority in a Distributed Geographic Information System
US9152676B2 (en) * 2009-01-30 2015-10-06 Google Inc. Identifying query aspects

Families Citing this family (53)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020138246A1 (en) * 2001-03-08 2002-09-26 Czora Gregory J. System and method for simulating conciousness
US20020184196A1 (en) * 2001-06-04 2002-12-05 Lehmeier Michelle R. System and method for combining voice annotation and recognition search criteria with traditional search criteria into metadata
US7324990B2 (en) * 2002-02-07 2008-01-29 The Relegence Corporation Real time relevancy determination system and a method for calculating relevancy of real time information
US20060288006A1 (en) * 2003-10-23 2006-12-21 Xerox Corporation Methods and systems for attaching keywords to images based on database statistics
US8296304B2 (en) 2004-01-26 2012-10-23 International Business Machines Corporation Method, system, and program for handling redirects in a search engine
US7424467B2 (en) * 2004-01-26 2008-09-09 International Business Machines Corporation Architecture for an indexer with fixed width sort and variable width sort
US7293005B2 (en) * 2004-01-26 2007-11-06 International Business Machines Corporation Pipelined architecture for global analysis and index building
US7499913B2 (en) * 2004-01-26 2009-03-03 International Business Machines Corporation Method for handling anchor text
US20050246324A1 (en) * 2004-04-30 2005-11-03 Nokia Inc. System and associated device, method, and computer program product for performing metadata-based searches
US7788258B1 (en) * 2004-06-21 2010-08-31 Google Inc. Automatic determination of whether a document includes an image gallery
US7461064B2 (en) 2004-09-24 2008-12-02 International Buiness Machines Corporation Method for searching documents for ranges of numeric values
JPWO2006048998A1 (en) * 2004-11-05 2008-05-22 株式会社アイ・ピー・ビー Keyword extractor
US7574434B2 (en) * 2005-02-25 2009-08-11 Sony Corporation Method and system for navigating and selecting media from large data sets
US8694494B2 (en) * 2005-03-25 2014-04-08 Vistaprint Schweiz Gmbh Selecting images using associated keywords
US8145528B2 (en) 2005-05-23 2012-03-27 Open Text S.A. Movie advertising placement optimization based on behavior and content analysis
US9648281B2 (en) * 2005-05-23 2017-05-09 Open Text Sa Ulc System and method for movie segment bookmarking and sharing
US8755673B2 (en) 2005-05-23 2014-06-17 Open Text S.A. Method, system and computer program product for editing movies in distributed scalable media environment
US8141111B2 (en) 2005-05-23 2012-03-20 Open Text S.A. Movie advertising playback techniques
US8417693B2 (en) * 2005-07-14 2013-04-09 International Business Machines Corporation Enforcing native access control to indexed documents
US7526486B2 (en) 2006-05-22 2009-04-28 Initiate Systems, Inc. Method and system for indexing information about entities with respect to hierarchies
EP2030134A4 (en) 2006-06-02 2010-06-23 Initiate Systems Inc A system and method for automatic weight generation for probabilistic matching
US7801907B2 (en) * 2006-06-13 2010-09-21 Alamy Limited Assignment of a display order to images selected by a search engine
US7698268B1 (en) 2006-09-15 2010-04-13 Initiate Systems, Inc. Method and system for filtering false positives
US7685093B1 (en) 2006-09-15 2010-03-23 Initiate Systems, Inc. Method and system for comparing attributes such as business names
US8356009B2 (en) 2006-09-15 2013-01-15 International Business Machines Corporation Implementation defined segments for relational database systems
US20080092189A1 (en) * 2006-09-21 2008-04-17 Clipblast, Inc. Web video distribution system for e-commerce, information-based or services websites
US8359339B2 (en) 2007-02-05 2013-01-22 International Business Machines Corporation Graphical user interface for configuration of an algorithm for the matching of data records
US8515926B2 (en) 2007-03-22 2013-08-20 International Business Machines Corporation Processing related data from information sources
US8429220B2 (en) 2007-03-29 2013-04-23 International Business Machines Corporation Data exchange among data sources
US8423514B2 (en) 2007-03-29 2013-04-16 International Business Machines Corporation Service provisioning
US8321393B2 (en) 2007-03-29 2012-11-27 International Business Machines Corporation Parsing information in data records and in different languages
WO2008121700A1 (en) 2007-03-29 2008-10-09 Initiate Systems, Inc. Method and system for managing entities
US8086551B2 (en) 2007-04-16 2011-12-27 Blue Oak Mountain Technologies, Inc. Electronic system with simulated sense perception and method of providing simulated sense perception
US20090083332A1 (en) * 2007-09-21 2009-03-26 The Penn State Research Foundation Tagging over time: real-world image annotation by lightweight metalearning
WO2009042941A1 (en) 2007-09-28 2009-04-02 Initiate Systems, Inc. Method and system for analysis of a system for matching data records
EP2198374A4 (en) 2007-09-28 2013-07-17 Ibm Method and system for associating data records in multiple languages
US8713434B2 (en) * 2007-09-28 2014-04-29 International Business Machines Corporation Indexing, relating and managing information about entities
US8200025B2 (en) * 2007-12-07 2012-06-12 University Of Ottawa Image classification and search
US9501337B2 (en) 2008-04-24 2016-11-22 Adobe Systems Incorporated Systems and methods for collecting and distributing a plurality of notifications
US8589495B1 (en) 2009-01-13 2013-11-19 Adobe Systems Incorporated Context-based notification delivery
JP4811481B2 (en) * 2009-03-13 2011-11-09 富士ゼロックス株式会社 Discussion support device and discussion support program
JP4853535B2 (en) * 2009-03-13 2012-01-11 富士ゼロックス株式会社 Discussion support device and discussion support program
US8433136B2 (en) * 2009-03-31 2013-04-30 Microsoft Corporation Tagging video using character recognition and propagation
US8533598B2 (en) * 2009-04-30 2013-09-10 Apple Inc. Media editing with a segmented timeline
WO2011017746A1 (en) * 2009-08-11 2011-02-17 Someones Group Intellectual Property Holdings Pty Ltd Method, system and controller for searching a database
US9342582B2 (en) 2010-11-22 2016-05-17 Microsoft Technology Licensing, Llc Selection of atoms for search engine retrieval
US9195745B2 (en) 2010-11-22 2015-11-24 Microsoft Technology Licensing, Llc Dynamic query master agent for query execution
US9424351B2 (en) 2010-11-22 2016-08-23 Microsoft Technology Licensing, Llc Hybrid-distribution model for search engine indexes
US8620907B2 (en) * 2010-11-22 2013-12-31 Microsoft Corporation Matching funnel for large document index
US9529908B2 (en) 2010-11-22 2016-12-27 Microsoft Technology Licensing, Llc Tiering of posting lists in search engine index
US8953836B1 (en) 2012-01-31 2015-02-10 Google Inc. Real-time duplicate detection for uploaded videos
US11087206B2 (en) * 2017-04-28 2021-08-10 Intel Corporation Smart memory handling and data management for machine learning networks
US10353676B2 (en) 2017-11-13 2019-07-16 Bank Of America Corporation Intelligent software compiler dependency fulfillment

Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5638465A (en) * 1994-06-14 1997-06-10 Nippon Telegraph And Telephone Corporation Image inspection/recognition method, method of generating reference data for use therein, and apparatuses therefor
US5742816A (en) * 1995-09-15 1998-04-21 Infonautics Corporation Method and apparatus for identifying textual documents and multi-mediafiles corresponding to a search topic
US5875446A (en) * 1997-02-24 1999-02-23 International Business Machines Corporation System and method for hierarchically grouping and ranking a set of objects in a query context based on one or more relationships
US5893095A (en) * 1996-03-29 1999-04-06 Virage, Inc. Similarity engine for content-based retrieval of images
US5911139A (en) * 1996-03-29 1999-06-08 Virage, Inc. Visual image database search engine which allows for different schema
US5913205A (en) * 1996-03-29 1999-06-15 Virage, Inc. Query optimization for visual information retrieval system
US6084595A (en) * 1998-02-24 2000-07-04 Virage, Inc. Indexing method for image search engine
US6167397A (en) * 1997-09-23 2000-12-26 At&T Corporation Method of clustering electronic documents in response to a search query
US6182069B1 (en) * 1992-11-09 2001-01-30 International Business Machines Corporation Video query system and method
US6182063B1 (en) * 1995-07-07 2001-01-30 Sun Microsystems, Inc. Method and apparatus for cascaded indexing and retrieval
US6243713B1 (en) * 1998-08-24 2001-06-05 Excalibur Technologies Corp. Multimedia document retrieval by application of multimedia queries to a unified index of multimedia data for a plurality of multimedia data types
US20020057343A1 (en) * 2000-06-30 2002-05-16 Ronk Lawrence J. Image object ranking
US20020069218A1 (en) * 2000-07-24 2002-06-06 Sanghoon Sull System and method for indexing, searching, identifying, and editing portions of electronic multimedia files
US6574616B1 (en) * 2000-02-16 2003-06-03 Index Stock Imagery, Inc. Stochastic visually based image query and retrieval system
US6611834B1 (en) * 2000-01-12 2003-08-26 International Business Machines Corporation Customization of information retrieval through user-supplied code
US6718365B1 (en) * 2000-04-13 2004-04-06 International Business Machines Corporation Method, system, and program for ordering search results using an importance weighting
US7277928B2 (en) * 2000-12-22 2007-10-02 Canon Kabushiki Kaisha Method for facilitating access to multimedia content

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0596247A3 (en) * 1992-11-02 1994-10-12 Motorola Inc A full-text index creation, search, retrieval and display method.
DE4237150A1 (en) 1992-11-04 1994-05-05 Licentia Gmbh vacuum cleaner
WO1999022318A1 (en) 1997-10-27 1999-05-06 Massachusetts Institute Of Technology Image search and retrieval system
US6317740B1 (en) * 1998-10-19 2001-11-13 Nec Usa, Inc. Method and apparatus for assigning keywords to media objects
AU2034600A (en) * 1998-11-30 2000-06-19 Henry C. Yuen Search engine for video and graphics
US7099860B1 (en) * 2000-10-30 2006-08-29 Microsoft Corporation Image retrieval systems and methods with semantic and feature based relevance feedback
US6970860B1 (en) * 2000-10-30 2005-11-29 Microsoft Corporation Semi-automatic annotation of multimedia objects

Patent Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6182069B1 (en) * 1992-11-09 2001-01-30 International Business Machines Corporation Video query system and method
US5638465A (en) * 1994-06-14 1997-06-10 Nippon Telegraph And Telephone Corporation Image inspection/recognition method, method of generating reference data for use therein, and apparatuses therefor
US6182063B1 (en) * 1995-07-07 2001-01-30 Sun Microsystems, Inc. Method and apparatus for cascaded indexing and retrieval
US5742816A (en) * 1995-09-15 1998-04-21 Infonautics Corporation Method and apparatus for identifying textual documents and multi-mediafiles corresponding to a search topic
US5911139A (en) * 1996-03-29 1999-06-08 Virage, Inc. Visual image database search engine which allows for different schema
US5913205A (en) * 1996-03-29 1999-06-15 Virage, Inc. Query optimization for visual information retrieval system
US5893095A (en) * 1996-03-29 1999-04-06 Virage, Inc. Similarity engine for content-based retrieval of images
US5875446A (en) * 1997-02-24 1999-02-23 International Business Machines Corporation System and method for hierarchically grouping and ranking a set of objects in a query context based on one or more relationships
US6167397A (en) * 1997-09-23 2000-12-26 At&T Corporation Method of clustering electronic documents in response to a search query
US6084595A (en) * 1998-02-24 2000-07-04 Virage, Inc. Indexing method for image search engine
US6243713B1 (en) * 1998-08-24 2001-06-05 Excalibur Technologies Corp. Multimedia document retrieval by application of multimedia queries to a unified index of multimedia data for a plurality of multimedia data types
US6611834B1 (en) * 2000-01-12 2003-08-26 International Business Machines Corporation Customization of information retrieval through user-supplied code
US6574616B1 (en) * 2000-02-16 2003-06-03 Index Stock Imagery, Inc. Stochastic visually based image query and retrieval system
US6718365B1 (en) * 2000-04-13 2004-04-06 International Business Machines Corporation Method, system, and program for ordering search results using an importance weighting
US20020057343A1 (en) * 2000-06-30 2002-05-16 Ronk Lawrence J. Image object ranking
US20020069218A1 (en) * 2000-07-24 2002-06-06 Sanghoon Sull System and method for indexing, searching, identifying, and editing portions of electronic multimedia files
US7277928B2 (en) * 2000-12-22 2007-10-02 Canon Kabushiki Kaisha Method for facilitating access to multimedia content

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150169588A1 (en) * 2005-10-12 2015-06-18 Google Inc. Entity Display Priority in a Distributed Geographic Information System
US9715530B2 (en) * 2005-10-12 2017-07-25 Google Inc. Entity display priority in a distributed geographic information system
US9785648B2 (en) 2005-10-12 2017-10-10 Google Inc. Entity display priority in a distributed geographic information system
US9870409B2 (en) 2005-10-12 2018-01-16 Google Llc Entity display priority in a distributed geographic information system
US10592537B2 (en) 2005-10-12 2020-03-17 Google Llc Entity display priority in a distributed geographic information system
US11288292B2 (en) 2005-10-12 2022-03-29 Google Llc Entity display priority in a distributed geographic information system
US20090292671A1 (en) * 2008-05-20 2009-11-26 Microsoft Corporation Motion-based data review and zoom
US8395643B2 (en) 2008-05-20 2013-03-12 Microsoft Corporation Motion-based data review and zoom
WO2010011344A1 (en) * 2008-07-23 2010-01-28 Ltu Technologies S.A.S. Frame based video matching
US9152676B2 (en) * 2009-01-30 2015-10-06 Google Inc. Identifying query aspects
US20120030180A1 (en) * 2010-07-29 2012-02-02 Sap Ag Archive-system-independent archive-type objects
US9164998B2 (en) * 2010-07-29 2015-10-20 Sap Se Archive-system-independent archive-type objects

Also Published As

Publication number Publication date
SE520533C2 (en) 2003-07-22
US7516129B2 (en) 2009-04-07
SE0100856D0 (en) 2001-03-13
WO2002073463A1 (en) 2002-09-19
EP1377915A1 (en) 2004-01-07
SE0100856L (en) 2002-09-14
US7539675B2 (en) 2009-05-26
US20060190445A1 (en) 2006-08-24
US20040098399A1 (en) 2004-05-20

Similar Documents

Publication Publication Date Title
US7539675B2 (en) Indexing of digitized entities
US8276065B2 (en) System and method for classifying electronically posted documents
US5895470A (en) System for categorizing documents in a linked collection of documents
US5920856A (en) System for selecting multimedia databases over networks
US9031942B2 (en) Method and system for indexing information and providing results for a search including objects having predetermined attributes
US7548936B2 (en) Systems and methods to present web image search results for effective image browsing
US5835905A (en) System for predicting documents relevant to focus documents by spreading activation through network representations of a linked collection of documents
US7630973B2 (en) Method for identifying related pages in a hyperlinked database
US7231381B2 (en) Media content search engine incorporating text content and user log mining
JP4587512B2 (en) Document data inquiry device
JPH1166106A (en) Multimedia data base and method for extracting information from data base
JPH11328228A (en) Retrieved result fining method and device
JP2003167914A (en) Multimedia information retrieving method, program, recording medium and system therefor
JP2009539152A (en) Content-based image readout
US8732165B1 (en) Automatic determination of whether a document includes an image gallery
Shin et al. Document Image Retrieval Based on Layout Structural Similarity.
US8732187B1 (en) Link-based ranking of objects that do not include explicitly defined links
US20030018617A1 (en) Information retrieval using enhanced document vectors
Wen et al. A multi-paradigm querying approach for a generic multimedia database management system
Yu et al. Web search technology
Hatano et al. An interactive classification of Web documents by self-organizing maps and search engines
Mishra et al. KhojYantra: an integrated MetaSearch engine with classification, clustering and ranking
Zhuang et al. Web-based image retrieval: A hybrid approach
Brown et al. A prototype content-based retrieval system that uses virtual images to save space
Correlograms for Semistructured Data. International Journal on Digital Libraries, 1: 1, pp. 68-88, April 1997. 2] Gustavo O. Arocena, et al. Applications of a Web

Legal Events

Date Code Title Description
STCF Information on status: patent grant

Free format text: PATENTED CASE

CC Certificate of correction
FPAY Fee payment

Year of fee payment: 4

REMI Maintenance fee reminder mailed
FPAY Fee payment

Year of fee payment: 8

SULP Surcharge for late payment

Year of fee payment: 7

FEPP Fee payment procedure

Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY

LAPS Lapse for failure to pay maintenance fees

Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY

STCH Information on status: patent discontinuation

Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362

FP Lapsed due to failure to pay maintenance fee

Effective date: 20210526