WO2001084471A2 - Procede et systeme d'identification d'objets - Google Patents

Procede et systeme d'identification d'objets Download PDF

Info

Publication number
WO2001084471A2
WO2001084471A2 PCT/US2001/014348 US0114348W WO0184471A2 WO 2001084471 A2 WO2001084471 A2 WO 2001084471A2 US 0114348 W US0114348 W US 0114348W WO 0184471 A2 WO0184471 A2 WO 0184471A2
Authority
WO
WIPO (PCT)
Prior art keywords
objects
descriptor
color
primitives
descriptors
Prior art date
Application number
PCT/US2001/014348
Other languages
English (en)
Other versions
WO2001084471A3 (fr
Inventor
David Sonnenberg
Thomas Duncan Kemp
Original Assignee
Hunes B.V.
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 Hunes B.V. filed Critical Hunes B.V.
Priority to US10/275,028 priority Critical patent/US20030236795A1/en
Publication of WO2001084471A2 publication Critical patent/WO2001084471A2/fr
Publication of WO2001084471A3 publication Critical patent/WO2001084471A3/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/26—Visual data mining; Browsing structured data

Definitions

  • the present invention relates to a method and system for identifying objects. More specifically, the present invention provides a method and system for identifying objects of a selected type from a range of objects based upon the provisioning of at least one primitive identifying a type of object desired.
  • each classification palette includes a respective plurality of descriptors.
  • different classification palettes may be provided, for example, for different aspects of the nature of the objects or for different countries. Indeed, it would be possible for a user to define custom descriptors having user defined Primitives and to arrange these together in a custom classification palette.
  • Each classification palette may have an associated classification space including a respective plurality of dimensions wherein the characteristics of each descriptor of a classification palette are defined by ranges of Primitives in the dimensions of the respective classification space.
  • the slider may be one or more bars movable alongside the array of Primitives or may comprise one or more highlighted Primitives. In this way, the user is provided with a visual description of all of the
  • Figure 13 illustrates the structural relationships between a search engine and a search group for one embodiment of the present invention.
  • Figure 23 illustrates a screen shot of a Design (dimension) Search Selection page provided in conjunction with the Internet based embodiment identified in Figure 15.
  • Figure 29 illustrates a screen shot of a Full Repeat View page provided in conjunction with the Internet based embodiment identified in Figure 15.
  • the present invention supports the use of other Dimensions and Primitives such as a "formality" Dimension which may include the following Primitives: stately, formal, refined, whimsical, informal, casual and shabby. Since these Primitives all represent degrees of formality, it is preferable that they are arranged in the Dimension in the order of their degree of formality. In this way, an object may be defined by a range of the Primitives within the Dimension. Indeed, it is not always necessary to have individual Primitives, since they may often merely represent the "degree" of a property represented by their Dimension. Those skilled in the art appreciate that depending upon processing speeds and database availability, any object may be defined my numerous Dimensions and Primitives to any desired level of detail and/or precision.
  • a user might refer to a "Nictorian” design as generally consisting of beige colors printed on man-made fibers.
  • the terms "Nictorian,” “beige” and “man-made” all encompass ranges of specific characteristics of an object (for example, a fabric), but would nevertheless identify recognized types of such objects.
  • the present invention defines a series of Descriptors and stores them in a database 14 containing records which define the characteristics of the Descriptors. A possible arrangement for such a record is illustrated in Figure 4.
  • the characteristics of the Descriptors are defined in a manner similar to the characteristics of the objects. In particular, they are defined in terms of Primitives which may be arranged in Dimensions.
  • the sliders 32 include upper levels 34 and lower levels 36 which define the upper and lower extremes of the range of Primitives selected for the respective Dimension 28.
  • Each classification palette and its associated classification space is identified herein as a "Strand.”
  • the present invention is not limited to using the CIELAB color classification scheme, and may utilize other schemes, as desired.
  • a user may wish to utilize a Color classification scheme which recognizes four Dimensions (i.e., the "a,” “b,” and “L” dimensions used in CIELAB plus a fourth Dimension for intensity).
  • the user Upon determining the color classification scheme desired to be used, the user then must determine how to specify a color in an object, for example, with multiple shades of color. In the preferred embodiment, this classification is accomplished by identifying three "levels" of color, i.e., the dominant color, the secondary color and a highlight color. This embodiment assumes that a color is only recorded once (i.e., a color cannot be both the dominant and secondary color). Similarly, this embodiment assumes that a user is not concerned with minute differences between colors and thus utilizes a grouping scheme to classify colors.
  • a user may enter color information via one of two methods.
  • the first method is to capture an RGB image of the object using, for example, a digital camera, a digital scanner, or similar devices. Once the RGB image exists, the user then selects specific areas of the captured object to represent one of the three color levels. At this point the imaging software then measures the RGB values for the selected area, averages the value, and obtains a single RGB value which is then converted into a CIELAB value. The classifying software then calculates the proportion of the object which is the color first selected, including those colors which are considered to be the same as the selected color even though, in reality, they are slightly different.
  • this embodiment is preferred when one is implementing the most common uses of the fabric classification scheme of the present embodiment. Additionally, it is a more efficient embodiment than previous embodiments when one is dealing with subspaces which are cuboidal and where all dimensions are ordered, because a cuboidal subspace can be described as a list of pairs of maxima and minima. However, this embodiment may also be implemented in cases where subspaces consist of arbitrary sets of cuboidal subspaces and where the Dimensions may or may not be ordered.
  • the Cuboids table groups facts together into "cuboids.”
  • a cuboid can be constructed in n dimensions by giving n ranges of values, one for each of the Dimensions.
  • the cuboid consists of all the points which lie within all these ranges.
  • the cuboid for a rectangle in 2-dimensional space where a given range exists for the x axis and a given range exists for the y axis, includes all the points whose x and y values fall within both of these ranges.
  • Cuboids are preferred in this data model because via cuboids it is possible to classify an object as being composed of several such volumes. If a classifier selects Nictorian and Art Deco as Descriptors for a single object design then the classification of this object in the Design strand is the union of the two cuboids represented by Nictorian and Art Deco.
  • the Facts table as shown in Table 10, and as represented by the Facts block 904 in Figure 9, describes where objects belong on each relevant Dimension, based on an objective measurement (e.g., CIELAB L value of the dominant color measured between 30.6 and 35.2), or an expert opinion ("the formality of this object is between Whimsical and Casual"). Note that Facts records are grouped together into cuboids. For each cuboid there should be one Facts record per Dimension.
  • the Strands table as shown in Table 14 and represented by the Strands block 912 in Figure 9, provides the different Strands used to break up the classification of an object into sensible parts. As mentioned previously, currently there are four Strands used in the fabric classification embodiment: Color, Composition, Design, and Supply.
  • the search engine is utilized in conjunction with a classification scheme which designates a place in an N-dimensional space for each object.
  • These Dimensions are the characteristics that can be attributed to any object, such as primary color, depiction, lightness, etc. As discussed previously, these individual Dimensions are preferably grouped in one of three Strands: color, composition, and design.
  • color, composition, and design With the fourth Strand, supply, merely providing purchasing information after an object has been identified and selected. Further, each of these Strands is associated with its own Dimensions, such that the characteristics of an object for each Strand can be modeled as a set of one or more cuboids.
  • Figure 10 illustrates the associations, as previously described above, between Strands, objects, cuboids, ranges, and Dimensions.
  • the Cuboids are basically requests to the system to find those objects that satisfy (A OR B) AND C OR D). Additionally, the system preferably combines the SearchGroups as follows:
  • the final combined SearchGroup is formed by the operation AND(SG1, SG2). Which results in the following four groups: ⁇ A,C ⁇ , ⁇ A,D ⁇ , ⁇ B,C ⁇ , ⁇ B,D ⁇ , which are: (1) plain, informal, mini, busy, 1920, 1972; (2) plain, informal, mini, busy, 1920, Neo-Classical; (3) neutral, informal, mini, busy, 1920, Contemporary; and (4) neutral, informal, mini, busy, 1920, Neo-Classical.
  • Figure 13 illustrates the relationship between the search engine and a SearchSpace for the present invention.
  • the search engine 1302 accepts a SearchGroup object 1306 as a search criteria, fills the search table, executes the query, and returns the SearchResult 1304.
  • the SearchGroup object 1306 represents one or more SearchSpaces 1308 which are suitably derived from the subelements from which the SearchGroup is constructed.
  • a SearchSpace is a derived object and a set of
  • Figure 14 illustrates one embodiment of a structural design of a search engine 1400 based upon the foregoing descriptions.
  • the search engine 1400 preferably includes a SearchEngine 1402, which (upon establishing a connection and receiving a request) generates a search, builds a search table, exits queries, builds results, and outputs the SearchResult 1404 to a user.
  • the SearchEngine 1402 receives inputs from a SearchGroup 1406 which suitably constructs SearchSpaces 1408 based upon entries in a search table. Additionally, the SearchSpace 1408 utilizes
  • RangeGroups 1410 to determine which values to place in the SearchTable.
  • the RangeGroups 1410 receive information from SearchCuboids 1414 which include additional search ranges 1416 and SearchDescriptors 1412.
  • the SearchGroup 1406 may be configured, as particular searches require, to combine various other searches by performing logical Operations 1418 such as OrOperations 1420 and/or AndOperations 1422.
  • the structure of the search engine 1400 preferably resembles the structure by which data is entered into the system to define object requests and to identify objects.
  • the controller 16 can use various different types of comparison. Preferably, these are selectable by the user. They may also be selected individually for different Dimensions.
  • the controller 16 can look for a direct match between the Primitives defining the Descriptors and the Primitives defining the objects. Alternatively, the controller 16 could look for objects having Primitives within the ranges defined for the Descriptors or overlapping with the ranges defined for the Descriptors.
  • FIG. 15 one embodiment of a system implementing the features and functions of the present invention is depicted for a fabrics embodiment.
  • the present invention is hosted by an Internet Service Provider which is contacted by users via an Internet connection.
  • this embodiment requires a user to register as a member prior to gaining access to the features and functions provided therewith.
  • the present invention is not to be construed as requiring users to register before use is permitted and may be utilized in an embodiment where a user's identity remains anonymous.
  • this embodiment preferably uses a username and password to identify user members.
  • other recognition schemes such as audio and visual identifiers, coded representations and other information may be utilized.

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Image Analysis (AREA)

Abstract

L'invention concerne un système et un procédé d'identification d'objets sur la base de primitives. Ledit système fait intervenir un serveur (2) relié par l'intermédiaire d'un réseau (4) à des terminaux utilisateurs (6), des requêtes d'identification d'objets associés à des primitives identifiées étant entrées à partir desdits terminaux. Le système fait intervenir différentes bases de données pour l'identification des objets. Une base de données d'objets (12) stocke des résultats d'objets dans une gamme de recherche. Une base de données de descripteurs (14) stocke des associations de dimensions de descripteurs avec des primitives. Une base de données personnalisée (40) permet aux utilisateurs de décrire des descripteurs en termes de primitives personnalisées. Une base de données de « brins » (38) fournit des associations de descripteurs au travers d'une palette. Dans un mode de réalisation de l'invention, ledit procédé consiste à stocker des résultats d'objets avec des primitives définissant ces objets, à afficher une palette identifiant des types d'objets avec des descripteurs, à stocker avec chaque descripteur au moins une primitive définissant le descripteur, à déterminer une sélection d'un descripteur de la palette et à identifier des objets ayant au moins une primitive associée au descripteur sélectionné.
PCT/US2001/014348 2000-05-03 2001-05-03 Procede et systeme d'identification d'objets WO2001084471A2 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US10/275,028 US20030236795A1 (en) 2001-05-03 2001-05-03 Method and system for identifying objects

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
EP00303689.4 2000-05-03
EP00303689 2000-05-03
US25316700P 2000-11-27 2000-11-27
US60/253,167 2000-11-27

Publications (2)

Publication Number Publication Date
WO2001084471A2 true WO2001084471A2 (fr) 2001-11-08
WO2001084471A3 WO2001084471A3 (fr) 2002-08-15

Family

ID=26073146

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2001/014348 WO2001084471A2 (fr) 2000-05-03 2001-05-03 Procede et systeme d'identification d'objets

Country Status (1)

Country Link
WO (1) WO2001084471A2 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5193185A (en) * 1989-05-15 1993-03-09 David Lanter Method and means for lineage tracing of a spatial information processing and database system
US5497335A (en) * 1991-09-10 1996-03-05 Zellweger Luwa Ag System for creating a fault diagnosis on production machines and application of the system on textile machines
US5740425A (en) * 1995-09-26 1998-04-14 Povilus; David S. Data structure and method for publishing electronic and printed product catalogs
US5870771A (en) * 1996-11-15 1999-02-09 Oberg; Larry B. Computerized system for selecting, adjusting, and previewing framing product combinations for artwork and other items to be framed
US6243615B1 (en) * 1999-09-09 2001-06-05 Aegis Analytical Corporation System for analyzing and improving pharmaceutical and other capital-intensive manufacturing processes

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5193185A (en) * 1989-05-15 1993-03-09 David Lanter Method and means for lineage tracing of a spatial information processing and database system
US5497335A (en) * 1991-09-10 1996-03-05 Zellweger Luwa Ag System for creating a fault diagnosis on production machines and application of the system on textile machines
US5740425A (en) * 1995-09-26 1998-04-14 Povilus; David S. Data structure and method for publishing electronic and printed product catalogs
US5870771A (en) * 1996-11-15 1999-02-09 Oberg; Larry B. Computerized system for selecting, adjusting, and previewing framing product combinations for artwork and other items to be framed
US6243615B1 (en) * 1999-09-09 2001-06-05 Aegis Analytical Corporation System for analyzing and improving pharmaceutical and other capital-intensive manufacturing processes

Also Published As

Publication number Publication date
WO2001084471A3 (fr) 2002-08-15

Similar Documents

Publication Publication Date Title
US20190385210A1 (en) Automated color processing and selection platform
CA2890158C (fr) Plate-forme automatisee de traitement et de selection de couleurs
US8860748B2 (en) Computerized, personal-color analysis system
US6925196B2 (en) Method and system for product selection
CA2758870C (fr) Procede et appareil utilises pour l'acquisition numerisee d'un projet d'enduit
US20140089781A1 (en) Color-based designs
Cob et al. Colour forecasting for fashion
CN116420155A (zh) 基于图像数据库分析的商品推荐装置及方法
US20170140252A1 (en) Computerized, personal-scent analysis sytem
US20180374140A1 (en) Computerized, personal beauty product analysis system
US6842654B2 (en) System, and method for online color algorithm exchange
Park et al. Analysis of pairings of colors and materials of furnishings in interior design with a data-driven framework
Lee et al. The development of an emotion model based on colour combinations
GB2365152A (en) Information retrieval system
US20150039994A1 (en) Color-based designs
JP6195550B2 (ja) コーディネート提案装置及び方法
WO2001084471A2 (fr) Procede et systeme d'identification d'objets
Laaksonen et al. 6. Between automation and interpretation: Using data visualization in social media analytics companies
CN114402349A (zh) 用于推荐化妆调色板或头发着色方案中的至少一个的计算设备、方法和装置
JP2011043667A (ja) 各種デザイン作成用の嗜好感性価値観軸分類図表を使用するための見本を表示した嗜好感性診断ツール
WO2012125168A1 (fr) Conceptions en fonction de couleurs
Odabaşıoğlu Effect of area on color harmony in interior spaces
KR100343555B1 (ko) 이미지에 기반한 색채 정보 제공 시스템 및 컴퓨터 시스템을 이용한 색채 정보 제공 방법
Lee et al. Formulation of interior design color palette based on the regression analysis of digital color samples
Shamoi et al. Color Aesthetics: Fuzzy based User-driven Method for Harmony and Preference Prediction

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): CA CN JP US

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
WWE Wipo information: entry into national phase

Ref document number: 10275028

Country of ref document: US

122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase in:

Ref country code: JP