WO2013018613A1 - Système de comparaison, serveur de comparaison, procédé de comparaison, dispositif d'extraction de caractéristique, dispositif de comparaison/recherche, et programme - Google Patents

Système de comparaison, serveur de comparaison, procédé de comparaison, dispositif d'extraction de caractéristique, dispositif de comparaison/recherche, et programme Download PDF

Info

Publication number
WO2013018613A1
WO2013018613A1 PCT/JP2012/068827 JP2012068827W WO2013018613A1 WO 2013018613 A1 WO2013018613 A1 WO 2013018613A1 JP 2012068827 W JP2012068827 W JP 2012068827W WO 2013018613 A1 WO2013018613 A1 WO 2013018613A1
Authority
WO
WIPO (PCT)
Prior art keywords
collation
contour
target
shape
verification
Prior art date
Application number
PCT/JP2012/068827
Other languages
English (en)
Japanese (ja)
Inventor
石山　塁
Original Assignee
日本電気株式会社
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 日本電気株式会社 filed Critical 日本電気株式会社
Publication of WO2013018613A1 publication Critical patent/WO2013018613A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5854—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using shape and object relationship
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30128—Food products

Definitions

  • the present invention relates to a collation system, a collation server, a collation method, a feature extraction device, a collation / retrieval device, and a program.
  • Patent Document 1 an individual information acquisition system for agricultural products that allows consumers and the like to easily know individual information on agricultural products.
  • Patent Document 1 The technology of Patent Document 1 is, for example, by assigning an ID number that is an identifier to agricultural products such as melons, such as a person other than a person (producer farmer or agricultural cooperative) involved in the production of these melons (for example, a consumer) ) Makes it possible to obtain individual information on the melon based on the ID number.
  • a label strip or sticker
  • an ID number is affixed to the produce.
  • a database system is constructed by storing individual information on agricultural products together with the above ID numbers in a computer system provided in a neutral organization (for example, a database corporation) provided separately.
  • the producer transmits the ID number to the computer system, and the individual information of the agricultural product, the production location, the producer name, the variety, the cultivation method, the harvest date, and the quality information (sugar content, etc.) ) And send.
  • At least one computer is installed in a retailer's store, a consumer's home, and the like, and is connected to the computer system through a general public line. Then, the retailer's store or the consumer obtains individual information regarding the agricultural product specified by the ID number by transmitting the ID number described on the label attached to the agricultural product from the computer to the computer system. can do.
  • a label (seal or sticker)
  • a label Is removed from agricultural, forestry and fishery products, or copied and imitated, and attached to other similar agricultural, forestry and fishery products, since there is no way to identify the individual agricultural, forestry and fishery products, so that counterfeiting such as the shipping source of agricultural, forestry and fishery products is prevented. I could't.
  • inventory management information for each individual item could not be retrieved when the label could not be attached or peeled off.
  • the present invention has been invented in view of the above-described problems, and its purpose is not to use a special device such as an IC tag, and a collation and retrieval system capable of collating and retrieving information on a collation target, collation and A search server, a collation and retrieval method, a feature extraction apparatus, a collation and retrieval apparatus, and a program are provided.
  • the present invention includes a first imaging unit that captures an image including a predetermined part and an outline of a verification target that is a regular product or a management target, and an image of the verification target captured by the first imaging unit.
  • a first feature calculation means for calculating the contour shape and position features of the object to be verified with reference to a predetermined part; and the contour shape of the regular object to be verified calculated by the first feature calculator;
  • Second feature calculation means for calculating the shape and position characteristics of the contour of the verification target based on the predetermined part from the image of the target; and the contour shape of the verification target that is the genuine product or the management target And the characteristics of the position and the authenticity
  • a matching and retrieval system and a collating means for collating the features of contour shape and position of the collation object focus.
  • the present invention provides a first feature calculation for calculating a feature of a contour shape and a position of a collation target with reference to the predetermined portion from an image including a predetermined portion and contour of a collation target that is a regular product or a management target. And second feature calculation means for calculating the shape and position features of the contour of the collation target with reference to the predetermined portion from the image including the predetermined portion and contour of the collation target to be verified.
  • a collation / retrieval server having collation means for collating the shape and position characteristics of the outline of the collation target object that is the genuine product or the management object with the outline shape and position characteristics of the collation target object of the authenticity collation It is.
  • the present invention calculates the feature of the shape and position of the contour of the collation target based on the predetermined portion from the image including the predetermined portion and contour of the collation target that is a regular product or a management target, and the calculated
  • the shape and position characteristics of the contour of the verification target object of the genuine product are stored, and from the image including the predetermined part and contour of the verification target object to be verified, the contour of the verification target object based on the predetermined part Collation that calculates shape and position features and collates the contour shape and position features of the genuine or managed object to be collated with the contour and shape features of the object to be collated ⁇ It is a search method.
  • the present invention relates to a process for calculating the shape and position characteristics of the contour of a verification target with reference to the predetermined portion from an image including the predetermined portion and contour of a verification target that is a regular product or a management target, and authenticity verification.
  • Processing for calculating the contour shape and position characteristics of the object to be verified with reference to the predetermined part from the image including the predetermined part and contour of the object to be verified, and verification that is the genuine product or the management target It is a program for causing a computer to execute a process of collating the shape and position characteristics of the contour of the object with the contour shape and position characteristics of the collation target of the verification.
  • the present invention relates to the shape of the outline of the verification object based on the predetermined part from the image including the predetermined part and the outline of the verification object that is a genuine product or a management target, or the verification target that is the target of authenticity verification. And a feature extraction device having feature calculation means for calculating the feature of the position.
  • the present invention provides a first feature calculation for calculating a feature of a contour shape and a position of a collation target with reference to the predetermined portion from an image including a predetermined portion and contour of a collation target that is a regular product or a management target. And second feature calculation means for calculating the shape and position features of the contour of the collation target with reference to the predetermined portion from the image including the predetermined portion and contour of the collation target to be verified.
  • a collation / retrieval device having collation means for collating the shape and position characteristics of the outline of the verification object that is the genuine product or the management target with the outline shape and position characteristics of the verification target object of the authenticity verification is there.
  • the present invention does not require a special device such as an IC tag, and can collate the object to be collated or search for management information.
  • FIG. 1 is a block diagram of an embodiment of the present invention.
  • FIG. 2 is a diagram for explaining the first embodiment.
  • FIG. 3 is a diagram for explaining the first embodiment.
  • FIG. 4 is a diagram for explaining the first embodiment.
  • FIG. 5 is a diagram for explaining the first embodiment.
  • FIG. 6 is a diagram for explaining the first embodiment.
  • FIG. 7 is a diagram for explaining the second embodiment.
  • FIG. 8 is a diagram for explaining the second embodiment.
  • FIG. 9 is a diagram for explaining the third embodiment.
  • FIG. 10 is a block diagram of an embodiment of the collation / search system.
  • FIG. 11 is a block diagram of the collation / search terminal device.
  • FIG. 1 is a block diagram of an embodiment of the present invention.
  • the first embodiment of the present invention detects an image capturing unit 1 that captures an image including an image including a predetermined region and an outline of a verification target object that is a genuine product, and detects a predetermined region from an image captured by the image capturing unit 1.
  • a feature calculation unit 2 that calculates the features of the contour and position of a matching object that is a genuine product normalized with respect to a predetermined part, and the contour of the genuine product that is calculated by the feature calculation unit 2
  • the recording unit 3 in which the features of the position and the shape are recorded, the imaging unit 4 that captures an image including a predetermined part and outline of the verification target object that is the target of authentication verification, and the verification target image captured by the imaging unit 4
  • a feature calculation unit 5 that detects a predetermined part from the image and calculates the feature of the position and shape of the contour of the object to be verified that is normalized with reference to the predetermined part, and a verification target that is a genuine product
  • a matching and retrieval system having a matching portion 6 for matching the characteristics of the position and shape of the contour of the collation object.
  • the present invention uses the fact that the contour shape of the verification target object determined based on a predetermined part such as a label or an exterior is different for each individual / lot, and the authenticity verification target object registered is registered. It is to check whether it is the same as the product.
  • the matching object to which the present invention is applied is not particularly limited as long as the contour is different for each individual, but examples of the contour being different for each individual are fruits, vegetables, meat, There are agricultural, forestry and fishery products such as fish and timber, and industrial products such as steel materials that differ from lot to lot.
  • the predetermined part of the verification target is a label affixed to the verification target, an exterior or package (box) that wraps the verification target, or a characteristic part that does not depend on the individual of the verification target It is.
  • wood is used as an object to be verified, and a label attached to wood as a predetermined part is explained as an example.
  • the verification object registered in the recording unit 3 by the producer or the distribution manager is described as a genuine product, and the verification object to be verified by the distribution manager or the consumer is described as the authenticity verification object. To do.
  • the imaging unit 1 captures an image of wood including a predetermined part (label 11) of the wood 10 that is a regular product or inventory management target.
  • a special device may be provided for imaging regular wood, but a digital camera equipped with a CCD, a CMOS, or a mobile phone may also be used.
  • FIG. 2 is a diagram showing a cross section of a regular wood 10 to be registered.
  • a label 11 is attached to the wood 10.
  • the sticking position of the label 11 is not specifically limited, when the cross section of the timber 10 is image
  • the imaging unit 1 captures an image so that an entire image of the wood 10 including the label 11 can be captured.
  • the object has a region with a visible feature that can specify the position and orientation regardless of the individual, such as the center of the annual ring or a mark engraved or written a predetermined figure, the part, A label can be substituted, and no label is required.
  • the predetermined area is referred to as a label.
  • the predetermined part is a region of appearance having characteristics that do not depend on an individual in the verification target, for example, literature (Toshinori Hosoi, Tetsuaki Suzuki, Satoshi Sato, face detection by generalized learning vector quantization, It is possible to automatically specify the location in the image by using a technology such as IEICE Technical Report, Volume 102 No. 651 (PRMU2002 201-210) pages 47-52), It is also possible to automatically take an image based on a predetermined part.
  • this technique is merely an example, and various known pattern detection techniques can be used. Further, JPEG or the like may be used regardless of the file format of the captured image.
  • the photograph of the genuine product is photographed by a person who is involved in production (for example, a production farmer, an agricultural cooperative, or a factory) or a person who performs quality control of an object to be verified in the distribution process, but is limited to these persons. It is not something.
  • the feature calculation unit 2 detects the label 11 from the image of the wood 10 captured by the imaging unit 1 and calculates the shape and position features of the contour of the genuine wood 10 normalized using the label 11 as a reference. .
  • the feature calculation unit 2 specifies a region of the wood 10 from the image of the wood 10 using a difference in color between the background and the wood 10. Further, the position and shape of the label 11 are extracted from the image, and the feature is calculated based on the reference coordinate system of the position and size.
  • the characteristics of the label 11 as registered data in advance, it is possible to use a label of a free design or an exterior pattern as a reference instead of a predetermined pattern such as a barcode. Specifically, features such as the shape, color, arrangement and size of the label 11, characters written on the label, and the like are stored in advance for detecting the label 11. In this way, the label 11 can be detected from the image.
  • the feature calculation unit 2 determines a reference coordinate system that serves as a reference for calculating the contour shape and position features of the wood 10 based on the label 11.
  • FIG. 3 is an example of a reference coordinate system in which the top end 12 of the label 11 is the origin, the horizontal direction of the label 11 is the X axis, and the vertical direction of the label 11 is the Y axis.
  • the feature calculation unit 2 detects the label 11 from the image of the wood 10 captured by the imaging unit 1, extracts the contour of the wood 10, and uses the contour as a reference coordinate system. And shape and position features such as contour distance ⁇ from the origin are calculated.
  • shape and position features such as contour distance ⁇ from the origin are calculated.
  • the top of a label is not necessarily image
  • the storage unit 3 stores the shape and position features of the contour based on the label 11 of the genuine wood 10 calculated by the feature calculation unit 2.
  • the production area or production site of the regular product producer name or manufacturer, variety, cultivation method, harvest date, production date, quality information (sugar content, etc.), eating time, raw materials, It may be stored in association with production information such as a lot number, distribution information such as a shipping location and a shipping date, and the like. By doing in this way, production management and distribution management of a regular article can be performed.
  • the imaging unit 4 exists on the side where a consumer or the like performs verification of an authentic verification target, and has the same configuration as the imaging unit 1.
  • a consumer or the like who requests collation captures an image including a label of wood, which is an authentic collation target.
  • the captured image is transmitted to the feature calculation unit 5.
  • the feature calculation unit 5 has the same configuration as the feature calculation unit 2 described above, detects the label 11 from the image of the wood that is the object of authenticity verification, and features the shape and position of the contour of the wood based on the label 11. Calculate Then, the calculated feature of the wood of the authentic verification target object is transmitted to the verification unit 6.
  • the collation unit 6 collates the wood feature of the authenticity verification target object from the feature calculation unit 5 with the feature of the regular product recorded in the recording unit 3, and identifies which of the regular products it is.
  • the position of the contour of the authenticity verification target with reference to the label or the exterior changes, so that it is possible to determine impersonation or an error.
  • the position and orientation of the outline of the object to be verified with reference to the label and the exterior are also different for each individual. It is possible to search for management information.
  • the label and the exterior are genuine, even if the genuine product itself is replaced with a fake, the contour of the individual itself is different, and the shape and position of the contour based on the label and the exterior also change. Therefore, the impersonation or error can be determined.
  • the feature calculation unit 2, the feature calculation unit 3 and the collation unit described above do not need to be on the producer or manager side for registering the genuine product, or on the manager or consumer side for authentic verification. You may provide in the collation process side in the form like a collation server.
  • the present invention can be applied even when the collation / search object is a set of a plurality of individuals.
  • the verification target is composed of a plurality of individuals.
  • the verification target object is composed of a plurality of individuals
  • a case where a plurality of individuals are collectively packaged together is conceivable.
  • strawberries and the like are in a form in which a plurality of strawberries are packed in one tray 20 and wrapped and sold.
  • the arrangement of strawberries differs individually, and the contours formed by a plurality of strawberries as shown in FIG. 7 are also different. Therefore, if such an object composed of a plurality of individuals is used as one object to be verified, the object to be verified and the contour formed by the plurality of individuals forming the object to be verified have a one-to-one relationship. It can be used for collation of collation objects.
  • the feature object calculation unit 2 sets a reference coordinate system in advance with a tray filled with strawberries as a predetermined part.
  • the upper left end point of the tray 20 is detected from the label 21 stuck on the upper left side of the tray, this end point is set as the origin, the horizontal direction of the tray is the X axis, and the vertical direction of the tray is the Y axis.
  • the reference coordinate system Define the reference coordinate system.
  • the feature calculation unit 2 detects the tray 20 and the label 21 from the image captured by the imaging unit 1, and calculates the feature of the contour shape and position composed of a plurality of strawberries normalized with the tray 20 as a reference. To do.
  • the calculated feature is registered in the storage unit 3.
  • the production area or production site of the regular product may be stored in association with production information such as a lot number, distribution information such as a shipping location and a shipping date, and the like. By doing in this way, production management and distribution management of a regular article can be performed.
  • the imaging unit 4 that exists on the side that performs verification of the authenticity verification target by the consumer or the like captures the entire image of the packaged strawberry that is the authenticity verification target. The captured image is transmitted to the feature calculation unit 5.
  • the feature calculation unit 5 has the same configuration as the feature calculation unit 2 described above, detects the tray 20 from the entire image of the packaged strawberry that is the object of authenticity verification, and is normalized with reference to the tray 20. Calculate the shape and position features of the contour consisting of multiple strawberries. Then, the calculated feature of the wood of the authentic verification target object is transmitted to the verification unit 6.
  • the collation unit 6 collates the packaged strawberry features of the authenticity verification object from the feature calculation unit 5 with the genuine product features recorded in the recording unit 3 to determine which regular product is the genuine product. Identify.
  • the contour formed by the plurality of individuals is used for collation, so that Similarly, collation processing can be performed.
  • the third embodiment is characterized in that, in addition to the feature of the contour shape and position of the verification target object, the surface pattern of the verification target object in the contour of the verification target object is also used for verification.
  • the matching object to which the present invention is applied is not particularly limited as long as the surface pattern of the matching object is characteristic.
  • Examples of the matching object having a characteristic surface pattern include melon and strawberry. , Apples, lemons, none, tangerines, seafood, wood.
  • the surface pattern of the object to be verified includes, for example, a surface pattern of melon, seed grains on the surface of a strawberry, fresh flower leaf veins, fish scales, wood grain and annual rings.
  • the feature unit 2 extracts a surface pattern in the extracted contour as a feature in addition to the operation of the second embodiment described above.
  • FIG. 9 is a diagram showing an example in which spots on the epidermis of the strawberry within the extracted contour are extracted as features of the surface pattern.
  • the feature extraction may be performed using a technique for extracting image features of a fingerprint such as a luminance value of an image, a feature point, an inflection point of a line of a binarized image, an intersection, an end point, and a center point.
  • the extracted regular product packaged strawberry outline shape, position, and surface pattern features in the outline are registered in the recording unit 3.
  • the same processing is performed on the authenticity verification side, and is transmitted to the verification unit 6.
  • the collation unit 6 performs collation of the surface pattern in addition to the collation described in the second embodiment.
  • a method for matching image features For example, the end points, intersections, and corners of epidermis pattern lines are used as feature points (in the case of strawberry epidermis patterns, the center of the epidermis spot is the feature point).
  • the arrangement of the points is collated, and when the number of pairs of feature points at the same position (range within a certain threshold) is equal to or greater than a predetermined number, the points are determined to be the same.
  • an existing fingerprint collation technique can be used.
  • the surface pattern of the verification target object in the contour of the verification target is also used for verification.
  • a highly accurate collation process can be performed.
  • FIG. 10 is a block diagram of an embodiment of the collation / search system.
  • the collation / retrieval system includes a feature extraction apparatus 100 installed on the production / management side of a collation target (for example, wood), a collation / search server 200 that performs collation / search, and a consumer.
  • a collation target for example, wood
  • a collation / search server 200 that performs collation / search
  • a consumer for example, a consumer.
  • Mobile phone 300 Mobile phone
  • the image feature extraction apparatus 100 includes an imaging unit 101, a feature calculation unit 102, and a transmission unit 103.
  • the imaging unit 101 and the feature calculation unit 102 have the same configuration as the imaging unit 1 and the feature calculation unit 2 described above.
  • the transmission unit 103 transmits the image feature of the verification target (for example, wood) calculated by the feature calculation unit 102 and the manufacturing information (production location, producer name, production date, quality information, etc.) of the verification. .
  • the collation / search server 200 includes a database 201, a feature calculation unit 202, and a collation / search unit 203.
  • the feature calculation unit 202 and the collation / retrieval unit 203 have the same configuration as the feature calculation unit 5 and the collation unit 6 described above.
  • the database 201 features of the shape and position of the contour based on the wood label transmitted from the feature extraction apparatus 100 on the producer side and the production information of the wood are stored in association with each other.
  • the mobile phone 300 includes an imaging unit 301 and a transmission / reception unit 302.
  • the imaging unit 301 has the same configuration as the imaging unit 4 described above.
  • the transmission unit 302 transmits an image including a wood label imaged by the imaging unit 301 to the collation / search server 200. Also, the determination result from the collation / search server 200 is received.
  • the producer registers the feature of the label 11 attached to the wood as shown in FIG. Then, the whole image of the wood including the label 11 is taken by the imaging unit 101.
  • the feature calculation unit 102 detects the label 11 from the captured image of the wood 10 and calculates the feature of the shape and position of the contour of the wood based on the detected label 11.
  • the producer inputs the photographed wood production information (production location, producer name, production date, quality information, etc.) to the feature extraction apparatus 100. Then, the transmission unit 103 transmits the characteristics of the wood 10 and the production information of the wood to the collation / search server 200.
  • the collation / search server 200 registers the characteristics of the wood from the producer's feature extraction apparatus 100 and the production information of the wood in the database 201.
  • the consumer captures the entire image of the wood 10 including the label 11 by the imaging unit 301 of the mobile phone 300 in order to collate the wood 10 to be collated.
  • the captured image is transmitted to the collation / search server 200 by the transmission / reception unit 302.
  • the image feature extraction apparatus 202 of the collation / search server 200 detects the label 11 of the wood 10 to be collated from the received image by the same method as described above, and the detected label 11 of the wood 10 is detected. Calculate contour shape and position features. The calculated feature is output to the collation / retrieval unit 203.
  • the collation / search unit 203 collates the received image feature with the feature registered in the database 201, and if there is a match, reads the production information associated with the feature and transmits it to the mobile phone 300. . On the other hand, if there is no match as a result of the verification, it is notified that the wallet to be verified is camouflaged.
  • the mobile phone 300 receives the authentication result from the verification / search server 200 and notifies the consumer.
  • the product can be verified, and as a result of the verification, production information of the product can also be acquired. Further, not only the consumer but also the side managing the distribution can obtain the production information, so the distribution of the product can be managed.
  • the feature calculation unit 102 that first extracts the feature of the matching target is configured to be installed in the producer's device, but also serves as the feature calculation unit 202 of the matching / search server 200. You can also.
  • the producer side is configured to transmit an image of the verification object.
  • the collation / search server 200 is provided, and each consumer performs the collation processing by communication.
  • the collation is performed from the database in which the contour feature of the collation target such as the database 201 is registered.
  • the outline feature of the object may be downloaded in advance, and the matching process may be performed without communication.
  • the collation / search terminal device as described above includes a storage unit 300 that stores outline features of a collation target registered in a producer, an imaging unit 301, a feature calculation unit 302, and a collation / search.
  • a unit 303 and a display unit 304 are provided.
  • the imaging unit 301, the feature calculation unit 302, and the collation / search unit 303 have the same configuration as the above-described imaging unit, feature calculation unit, and collation / search unit, and the display unit 304 includes the collation / search unit 303.
  • the collation result, production information of the collation target object when it is determined that they are identical, and the like are displayed.
  • the feature amount calculation unit and the collation unit can be configured by hardware, but can also be realized by a computer program.
  • functions and operations similar to those in the above-described embodiments and examples are realized by a processor that operates according to a program stored in the program memory.
  • the 1st imaging means which images the image containing the predetermined part and outline of the collation target object which is a regular article or management object, First feature calculation means for calculating features of the shape and position of the contour of the verification target with reference to the predetermined part from the image of the verification target image captured by the first imaging means; Recording means for recording the contour shape and position features of the regular verification object calculated by the first feature calculation means; A second image pickup means for picking up an image including a predetermined portion and a contour of a verification target object to be verified; Second feature calculation means for calculating features of the shape and position of the contour of the verification target based on the predetermined part from the image of the verification target image captured by the second imaging means; A collation / retrieval system comprising collation means for collating the shape and position characteristics of the genuine object or the object of collation to be managed with the outline shape and position characteristics of the object of collation for authenticity collation.
  • the said collation target object is an industrial product from which an outline differs because the fixed position and direction with respect to an exterior differ individually from agriculture, forestry and fishery products, an industrial product from which a shape of an outline differs for every individual, or an exterior.
  • the said 1st and 2nd characteristic calculation means normalize the shape and position of the outline of the said collation target based on the coordinate system on the basis of the said predetermined part, and additional remarks from the additional remark 1 4.
  • the verification system according to any one of 3.
  • the said 1st and 2nd characteristic calculation means in addition to the characteristic of the shape and position of the outline of the said collation target object WHEREIN: The characteristic of the surface pattern of the said collation target object in the outline of the said collation target object Calculate
  • the recording means stores the shape and position characteristics of the contour of the regular verification target object, and the characteristics of the surface pattern in the contour
  • the collation means includes a contour shape, a contour position, and a feature of a surface pattern in the contour of the genuine collation target object, a contour shape, a contour position, and a contour pattern of the collation target object in the authenticity collation. 6.
  • the collation system according to any one of appendix 1 to collation 5 for collating the characteristics of the surface pattern.
  • the 1st feature calculation which calculates the feature of the shape and position of the outline of a collation target on the basis of the predetermined part from the image containing the predetermined part and outline of the collation target which is a regular article or a management object Means, Second feature calculation means for calculating the shape and position characteristics of the contour of the collation target based on the predetermined portion, from an image including the predetermined portion and contour of the collation target to be verified.
  • a collation / retrieval server comprising collation means for collating the shape and position characteristics of the outline of the verification object that is the genuine product or the management target with the outline shape and position characteristics of the verification target object of the authenticity verification.
  • the supplementary note is an agricultural, forestry or fishery product, an industrial product that has a difference in contour shape for each individual, or an industrial product that has a different contour because the fixed position / orientation with respect to the exterior differs individually. 7.
  • part is the characteristic part which does not depend on the label affixed on the said collation target object, the exterior of the said collation target object, or the individual
  • the said 1st and 2nd characteristic calculation means normalizes the shape and position of the outline of the said collation target based on the coordinate system on the basis of the said predetermined part, and additional remarks from the additional remark 7 9.
  • the collation / search server according to any one of 9 above.
  • Additional remark 13 The said collation target is described in the additional remark 12 which is an agricultural product which differs in the shape of an outline for every agriculture, forestry and fishery product, an individual, or an industrial product from which an outline differs because the fixed position and direction with respect to an exterior differ individually. Collation / search method.
  • the predetermined part is any one of Supplementary Note 12 to Supplementary Note 13, which is a label affixed to the verification target, an exterior of the verification target, and a characteristic part that does not depend on an individual of the verification target.
  • a collation target includes a plurality of individuals, and a contour formed by the plurality of individuals is defined as a contour of the collation target.
  • the 1st feature calculation which calculates the feature of the shape and position of the outline of a collation target on the basis of the predetermined part from the picture containing the predetermined part and outline of the collation target which is a regular article or a management object Means, Second feature calculation means for calculating the shape and position characteristics of the contour of the collation target based on the predetermined portion, from an image including the predetermined portion and contour of the collation target to be verified.
  • a collation / retrieval device comprising collation means for collating the shape and position characteristics of the genuine object or management target contour object with the contour shape and position characteristics of the collation target object for authenticity verification.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Library & Information Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)

Abstract

La présente invention porte sur un système de comparaison/recherche qui comprend : un premier moyen d'imagerie qui obtient une image qui comprend un site prescrit et un contour d'un sujet de comparaison qui est un produit légitime ou un objet qui est géré ; un premier moyen de calcul de caractéristique qui calcule, à partir de l'image du sujet de comparaison obtenue par le premier moyen d'imagerie, les caractéristiques de la forme et de la position du contour du sujet de comparaison en utilisant comme référence le site prescrit ; un moyen d'enregistrement qui enregistre les caractéristiques de la forme et de la position du contour du sujet de comparaison qui est un produit légitime, les caractéristiques ayant été calculées par le premier moyen de calcul de caractéristique ; un second moyen d'imagerie qui obtient une image qui comprend un site prescrit et un contour d'un sujet de comparaison qui est un objet en train d'être authentifié ; un second moyen de calcul de caractéristique qui calcule, à partir de l'image du sujet de comparaison obtenue par le second d'imagerie, les caractéristiques de la forme et de la position du contour du sujet de comparaison en utilisant comme référence le site prescrit ; et un moyen de comparaison qui compare les caractéristiques de la forme et de la position du contour du sujet de comparaison qui est un produit légitime ou un objet en train d'être géré et les caractéristiques de la forme et de la position du contour du sujet de comparaison qui est un objet en train d'être authentifié.
PCT/JP2012/068827 2011-07-29 2012-07-25 Système de comparaison, serveur de comparaison, procédé de comparaison, dispositif d'extraction de caractéristique, dispositif de comparaison/recherche, et programme WO2013018613A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2011-166788 2011-07-29
JP2011166788 2011-07-29

Publications (1)

Publication Number Publication Date
WO2013018613A1 true WO2013018613A1 (fr) 2013-02-07

Family

ID=47629139

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2012/068827 WO2013018613A1 (fr) 2011-07-29 2012-07-25 Système de comparaison, serveur de comparaison, procédé de comparaison, dispositif d'extraction de caractéristique, dispositif de comparaison/recherche, et programme

Country Status (1)

Country Link
WO (1) WO2013018613A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020039567A1 (fr) * 2018-08-23 2020-02-27 日本電気株式会社 Dispositif de collationnement d'objet

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006146570A (ja) * 2004-11-19 2006-06-08 National Agriculture & Bio-Oriented Research Organization 個体認証システム並びにそれを用いた農産物の追跡方法及び農産物追跡システム
JP2007135501A (ja) * 2005-11-21 2007-06-07 Atom System:Kk 鼻特徴情報生成装置及び鼻特徴情報生成プログラム

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006146570A (ja) * 2004-11-19 2006-06-08 National Agriculture & Bio-Oriented Research Organization 個体認証システム並びにそれを用いた農産物の追跡方法及び農産物追跡システム
JP2007135501A (ja) * 2005-11-21 2007-06-07 Atom System:Kk 鼻特徴情報生成装置及び鼻特徴情報生成プログラム

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020039567A1 (fr) * 2018-08-23 2020-02-27 日本電気株式会社 Dispositif de collationnement d'objet
JPWO2020039567A1 (ja) * 2018-08-23 2021-08-10 日本電気株式会社 物体照合装置
JP7215485B2 (ja) 2018-08-23 2023-01-31 日本電気株式会社 物体照合装置
US11983864B2 (en) 2018-08-23 2024-05-14 Nec Corporation Object matching device

Similar Documents

Publication Publication Date Title
US11983725B2 (en) Collation/retrieval system, collation/retrieval server, image feature extraction apparatus, collation/retrieval method, and program
JP7001122B2 (ja) 流通管理方法、流通管理装置及び流通管理システム
US9436959B2 (en) Individual product identification system, individual product identification method, and device and program used by same
JP2019055828A (ja) 棚情報推定装置及び情報処理プログラム
JP6399281B2 (ja) 照合・検索システム、照合・検索サーバ、画像特徴抽出装置、照合・検索方法及びプログラム
US10769399B2 (en) Method for improper product barcode detection
JP6688520B2 (ja) 取引システム及び取引方法
WO2014112874A1 (fr) Procédé et système de fourniture d'informations d'identification de produit
WO2013018613A1 (fr) Système de comparaison, serveur de comparaison, procédé de comparaison, dispositif d'extraction de caractéristique, dispositif de comparaison/recherche, et programme
CN107908799A (zh) 一种个体对象产生过程视频鉴证方法
US20240249086A1 (en) Method for the identification, authentication, and tracking of physical assets
Vo Traceability in the southern rock lobster (SRL) export supply chain: investigating lobster grading and identification using low-cost image-based biometrics

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12820397

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12820397

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP