• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of sorting baggage at an airport in which baggage is transported by a conveyor system (301), wherein a digital image of a luggage identification tag (100) is acquired (110) and automatically recognizes, from said image, an aerial flight corresponding to said baggage to control said conveying system and orient the baggage in the airport, is characterized in that it comprises the following steps: said digital image is recovered from information printed on the label, this textual information is compared to data stored in a database (210) local to the airport and representative of a list of flights to or from said airport for detecting a match and in response to a match detection, a data representative of an airlifter number corresponding to the baggage (110) is automatically transmitted to the system conveying device (301).
    公开号:FR3081354A1
    申请号:FR1870600
    申请日:2018-05-24
    公开日:2019-11-29
    发明作者:Alexa Floury；Hicham El Bernoussi；Emmanuel Piegay
    申请人:Solystic SAS；
    IPC主号:
    专利说明:

    Description
    Title of the invention: Method of sorting baggage at an airport
    Technical Field [0001] The invention relates to a method of sorting baggage at an airport, in which baggage is transported by a conveyor system, and in which a digital image of a label attached to a baggage and edited for acquisition is acquired. identify said baggage, and an automatic flight is automatically recognized from said image to control said conveying system and orient the baggage in the airport.
    PRIOR ART Such a type of process is described in patent application WO 2017/137106.
    According to the method described in this patent application, baggage is provided with an identification label carrying textual information lines as well as a barcode, the latter allowing each baggage to be associated with a set of data. registered in a baggage database constructed as and when baggage is checked at a departure airport.
    The data recorded can represent a passenger name, a boarding number, a flight number, departure, arrival and, if applicable, transit airport codes, or even a flight plan. .
    Thus, reading the barcode makes it possible to identify and orient a piece of luggage in an automated conveyor system in the departure airport as in the arrival airport or, where appropriate, in an airport. transit, the relevant information from the database being transmitted to the airports concerned.
    In the event of difficulty in reading the barcode, for example in the case of a dirty or damaged label, patent application WO 2017/137106 suggests basing an identification of the baggage on the textual information by means of an analysis of the label by optical character recognition including a comparison between the recognized textual elements and the data recorded in the baggage database, in order to avoid using a human operator, a source of loss of efficiency and additional cost.
    On the other hand, this patent application does not respond to the problems of sorting luggage due to a lack in the database, to an interruption of access to it, or to a delay in the transmission of its data.
    In such situations, there are today no other solutions than to carry out manual sorting, an extremely expensive operation and which can lead to significant delays, when for example the baggage data of a Complete flights are unavailable when this flight arrives at an airport.
    Disclosure of the invention To this end, the invention relates to a method of sorting baggage at an airport in which the baggage is transported by a conveyor system, and in which at least one digital image is acquired d 'a label which is attached to a baggage and which has been edited to uniquely identify said baggage, and there is automatically recognized from said image an air flight corresponding to said baggage to control said conveyor system and orient the baggage in the airport, characterized in that it comprises the following steps: recovering from said digital image text information printed on the label, comparing this text information with data recorded in a local database at the airport and representative of '' a list of air flights to or from said airport to detect a match and, in response to a detect Ion of concordance, a data representative of an air flight number corresponding to the baggage is automatically transmitted to the conveying system.
    In this way, the sorting of baggage at an airport can be based on the association of the latter with a given flight on the basis of information from a local database whose access and completeness can be easily assured.
    The method according to the invention can advantageously have the following features:
    - we can save said data in said database before the label is edited to identify said baggage;
    - in automatic recognition of air flight, we can detect and identify a line of the largest characters on the label, we can isolate an area of the label including this line, we can detect another line of characters located in this zone, and a character recognition operation can be applied to this line;
    - The label may include an identification code readable by a machine, this code may be provided to correspond to a data set of a baggage database separate from said flight database and one can attempt to identify the baggage by reading this identification code;
    - we can compare the text information retrieved from the digital image with the data from said set of data in the baggage database to detect a match or a mismatch between them;
    - in the event of detection of mismatch, it is possible to operate a transmission of data representative of non-reading information to the automatic sorting unit to control the conveying unit;
    - It can also include a step of detecting a failure to decipher said identification code readable by a machine during the attempt to identify baggage by reading this identification code, and, in response to detection, text information retrieved from digital images can be used by the conveyor system to sort baggage;
    - It may further include a step of detecting a failure to associate said decrypted identification code with the data set of the baggage database in the absence of corresponding data in this database, and, in response to this detection, the text information retrieved from the digital images can be used by the conveyor system to sort the baggage; and it may further include a step of detecting a failure to associate said decrypted identification code with the data set of the baggage database in the event of an impossibility of connection to this baggage database, and in response to this detection, text information retrieved from digital images can be used by the conveyor system to sort baggage.
    Thus, the sorting of the baggage can be based on textual information recognized on the label by comparison with data from a database of air flights prepared in advance, this recognized data being able to replace or combine to information stored in a baggage database constantly updated and possibly distant from the airport where the baggage is located, and therefore whose exhaustiveness and access are likely to be problematic, which improves the identification of baggage and / or automatic sorting.
    Summary presentation of the drawings The present invention will be better understood and other advantages will appear on reading the detailed description of an embodiment taken by way of non-limiting example and illustrated by the accompanying drawings, wherein :
    - Figure IA illustrates airports and associated databases;
    - Figures IB and IC illustrate luggage tags;
    - Figure 1D illustrates a sorting system;
    - Figure 2A shows a method of sorting baggage and Figures 2B, 2C and 2D respectively illustrate a first, a second and a third sorting mode;
    - Figures 3A, 3B and 3C respectively represent a baggage database, a flight database, and a correspondence table; and [0028] - Figure 3D represents a sorting process by video coding.
    Description of an embodiment [0030] An airport baggage sorting method according to the invention is illustrated in Figures IA to 3D.
    When traveling by plane from airport A to airport B or from airport C to airport A, respectively V A b and V C a, each piece of luggage 110 transported is identified using a label 100 edited when the corresponding baggage is checked in and comprising an identification code such as a barcode 120 and textual information of different types which may be useful in tracking this baggage to its destination final.
    The identification code can be a machine identification code, such as QR codes or RFID chips, without being limited to barcodes only.
    The information indicated on the label includes at least one IATA airport code generally composed of a trigram 122 corresponding to an airport and a flight number 126 composed of a coded name 127 of an airline and d '' a flight number 128 internal to the company, but can also include, for example, a date 124 or a city 130.
    A baggage database 200 is dynamic, created as and when baggage is checked 110 so as to store this information, and possibly additional information, in association with the corresponding bar codes, and is shared between several airports using a computer network.
    Thus, a baggage database 200 can be made up of sets B (j) of data Dk (j) of different types k (owner, IATA code, flight number, date, etc.), each corresponding set to a piece of baggage (j), as illustrated in FIG. 3A, and has the function of identifying, tracking and sorting the baggage throughout their journey.
    In addition, a database of air flights 210 is established for each airport, listing the flights to or from the corresponding airport during a given activity period, for example one or more consecutive days of airport activity, including in particular the day of baggage handling according to the invention.
    These air flight databases 210 consist of sets E (i) of data each corresponding to an air flight (i), this data comprising at least one airport trigram (IATA code of destination airport ) T (i), and an air flight number consisting of a coded airline name C (i) and an internal airline flight number I (i), as illustrated in Figure 3B.
    Air flight databases can be prepared well in advance, before the start of a given period of activity at the corresponding airport and before the check-in of baggage carried on board the listed flights, and are not intended to be changed during this period; these databases can therefore be considered static at least during a given activity period, in particular with respect to dynamic baggage databases, refreshed as and when baggage is checked in or checked out. end of baggage check-in for a given flight, including during the airport's operating period.
    In addition, these flight databases can be located in and specific to the airport concerned rather than being remote and imposing long distance connections.
    Thus, flight databases imply better stability and can be known much more in advance than a baggage database, and their use therefore allows operational reliability much greater than the use of the baggage database.
    According to a method of the present embodiment illustrated in FIG. 1D, baggage 110 in a given airport is conveyed by a conveyor system 301, in which an automatic sorting unit 332 controls a conveyor unit 302, such as '' a set of conveyor belts, which passes the baggage to an automatic identification system 300 comprising a device 125 for reading the bar codes 120 carried by the labels 100 for sorting according to the data Dk (j) associated with this barcode in the baggage database 200, the reading device 125 being connected to the baggage database 200.
    The automatic identification system 300 further comprises a portal 310 equipped with several cameras 320 arranged so as to acquire digital images of the luggage from different angles of view, in order to maximize the chances of having exploitable images labels 100.
    In a baggage sorting process illustrated in FIG. 2A, baggage is passed under the gantry, so that at least one digital image of the label attached to said baggage is taken during an OCR preliminary step. -O of an optical character recognition software analysis relating to the label, known as OCR analysis.
    The automatic identification system 300 includes a computer processing unit 330 which is connected to the automatic sorting unit 332, to the reading device 125 and to the air flight database 210 of the given airport, and which subjects this digital image to OCR analysis so as to locate therein textual information or lines of text, present in the form of lines 130 and 131 of alphanumeric characters and to automatically compare them with the data T (i), C (i ) and I (i) of the sets E (i) of data from said air flight database 210, in order to detect a correspondence between the textual information and this data.
    The identification of relevant character lines, that is to say potentially corresponding to the data in the flight database, advantageously uses the fact that, by convention, the characters of the airport trigram 122 are the most close up of a luggage tag 100.
    Thus, the OCR recognition of these character lines can include, in this order and as illustrated in FIG. 2A, a single or multiple OCR-1 binarization of the digital image, an OCR-2 determination of aggregations 131 A pixel of the binarized image having the same color and an expected size of characters of an airport trigram 122 which corresponds to a line of the largest characters present on the image of the label, an OCR-3 calculation of a neighborhood graph of said aggregations 131 A , an OCR-4 determination that a line 131 of alphanumeric characters forms a triplet of neighboring aggregations from said neighborhood graph, this triplet of aggregations possibly plausibly corresponding to a trigram 122 of an airport, an OCR-5 determination and extraction of an area of interest 132 of the image substantially centered on said triplet of aggregations 131, an OCR-6 rectification and binarization of said area of interest of l 'picture , an OCR-7 location of lines of text 130 near said aggregation triplet in said area of interest of the rectified image, and an OCR-8 operation for analyzing the characters of said aggregation triplet and / or said lines of text during which hypotheses on the textual information of the lines of text 130 and 131 are produced and these hypotheses are verified by comparing them with the data T (i), C (i) and I (i) of the base of 210 flight data from the airport.
    In a first mode Ml, illustrated by FIG. 2B, in response to a detection of a concordance Y O cr of the hypotheses of the OCR analysis with at least one of the data T (i), C (i ) and I (i) of one of the sets E (i) of the database during a step D o of detection of agreement or non-agreement of the hypotheses, the computer processing unit operates a transmission TR-1 of a datum representative of an air flight corresponding to the baggage, such as an air flight number 126 corresponding, to the automatic sorting unit 332 which then controls, during a step COM-1, the conveying unit 302 on the basis of the information obtained by said OCR analysis of steps OCR-O to OCR-8 and / or data from said data set E (i) in accordance with the textual information of the label recovered from digital images, so as to sort by routing the luggage to a PC collection point.
    The sorting carried out by the automatic sorting unit 332 can for example be carried out by means of an assignment table 220 assigning a collection point CP (m) of luggage to each air flight (m) and therefore to each set E (m) of data from the air flight database 210 as illustrated in FIG. 3C.
    The collection points can be, for example, a conveyor entrance, a boarding gate, or even a baggage claim carousel.
    In the first mode Ml, the sorting is entirely based on the information obtained by means of the OCR analysis, and therefore on the air flight data stored in an air flight database, and the OCR analysis can represent an independent solution on which to base a sorting of luggage.
    A second mode M2, illustrated in FIG. 2C, differs from the first mode Ml in that the OCR analysis is not used to recognize data from an air flight corresponding to the baggage and to operate the transmission TR-1 to the conveying system that in the event of a failure Ni of the deciphering M2-1 of the bar code detected during a detection step Di of success or of failure to decipher the bar code, for example in the event of damaged and illegible barcode, or in the event of failure N 2 of association M2-2 of the decrypted code with a data set B (j) of the baggage database 200 during a step D 2 of detection of the association's success or failure, for example if it is impossible to connect to the baggage database or if there is no relevant data in it.
    In the M2 mode, in the event of success Y1 in decrypting the bar code and then in success Y2 in associating the decrypted code with a data set B (j) from the baggage database 200, an operation is carried out. transmission TR-2 of the baggage representative data to the automatic sorting system 332 which then controls, during a step COM-2, the conveyor unit 302 according to information from the data set of the baggage database associated with the baggage by its label, to sort the baggage.
    In a third mode M3, illustrated by FIG. 2D, a M3-1 reading of a barcode of the baggage label is carried out and the OCR analysis described above of the same label. , then a comparison M3-2 is carried out between the data obtained by the two methods, so as to cross the data obtained by means of these two methods and thus increase the reliability of the sorting of baggage.
    In the event of detection of a concordance Y C omp between the data obtained by the two methods during a step D 3 of detection of concordance or mismatch of the data, a transmission TR-3 of data representative of the baggage to the automatic sorting unit 332 which then controls in response the conveying unit 302 during a step COM-3 to proceed with the sorting of the baggage.
    In the event of detection N 3 of mismatch between the data obtained by the two methods during step D 3 , there is a transmission TR-4 of data representative of non-reading information at l 'automatic sorting unit 332 which then controls in response the conveying unit 302 to follow a particular treatment for baggage during a step E by placing it away from the automatic sorting system for example by bringing it to a pre-determined output and / or by applying to it a video coding operation or an entirely manual sorting, in order to increase the reliability of the sorting.
    For each of the modes Ml, M2 and M3, if it is impossible to arrive at a degree of certainty sufficient for sorting, for example if it is impossible N o to detect a coincidence between the hypotheses produced during OCR analysis and the data from the flight database detected during a coincidence detection step D o or in the event of a mismatch N 3 between the data obtained by reading M3-1 of a code- bars and the data obtained by the OCR analysis detected during step D 3 of concordance detection, a VCS video coding procedure can be used in which digital images IN of said baggage acquired by the cameras 320 are presented to an operator human O by means of a monitor M, operator which then manually searches for a correspondence between textual information visible in these images and said data from the flight database 210 by means of a connected computer terminal T the latter, as shown in Figure 3D.
    It goes without saying that the present invention cannot be limited to the embodiment as described above, which is capable of being modified without departing from the scope of the invention.
    权利要求:
    Claims (1)
    [1" id="c-fr-0001]
    claims [Claim 1] Method for sorting baggage at an airport (A) in which baggage is transported by a conveyor system (301), and in which at least one digital image of a label (100) is acquired which is attached to a baggage ( 110) and which has been edited to uniquely identify said baggage, and an automatic flight (V C a, Vab) corresponding to said baggage is automatically recognized from said image to control said conveying system and orient the baggage in the airport, characterized in that it comprises the following steps: text information (131, 130) printed on the label is recovered from said digital image, this text information is compared with recorded data (T (i), C ( i), I (i)) in a database (210) local to the airport and representative of a list of air flights to or from said airport (A) to detect a match and, in response to a conco detection rdance (Y 0C r), data representative of an air flight number (126) corresponding to the baggage (110) is automatically transmitted to the conveying system (301). [Claim 2] The method according to claim 1, characterized in that said data is recorded in said database (210) before the edition of the label (100) to identify said luggage. [Claim 3] The method according to any one of claims 1 to 2, characterized in that in the automatic recognition of air flight, a line (131) of the largest characters present on the label is detected and identified, an area ( 132) of the label comprising this line, another line of characters (130) located in this zone is detected, and a character recognition operation (OCR) is applied to this line. [Claim 4] The method according to any one of claims 1 to 3, characterized in that the label comprises an identification code (120) readable by a machine, this identification code being provided to correspond to a set (B (j )) of data (Di (j), D 2 (j)) of a baggage database (200) separate from said flight database (210) and in that one attempts to identify the baggage by reading this identification code (120). [Claim 5] The method according to claim 4, characterized in that the text information recovered from the digital image is compared with (M3-2)
    [Claim 6] [Claim 7] [Claim 8] [Claim 9] data from said set of data in the baggage database to detect (D 3 ) a match or a non-match between them.
    The method according to claim 5, characterized in that in the event of detection of a mismatch (N 3 ), a transmission (TR-4) of data representative of non-reading information is carried out to the unit of automatic sorting (332) to control the conveyor unit.
    The method according to claim 4, characterized in that it further comprises a step (Di) of detecting a failure (Ni) in deciphering (M2-1) of said identification code (120) readable by a machine at during the baggage identification attempt by reading this identification code, and in that, in response to the detection, the text information retrieved from the digital images is used by the conveyor system (301) to sort the luggage.
    The method according to claim 4, characterized in that it further comprises a step (D 2 ) of detecting a failure (N 2 ) of association (M2-2) of said identification code (120) deciphered with the data set B (j) of the baggage database (200) in the event of the absence of corresponding data in this database, and in that, in response to this detection, the textual information retrieved from the digital images are used by the conveyor system (301) to sort the baggage.
    The method according to claim 4, characterized in that it further comprises a step (D 2 ) of detecting a failure (N 2 ) of association (M2-2) of said identification code (120) deciphered with the data set (B (j)) of the baggage database (200) in the event of an impossibility of connection to this baggage database (200), and in that, in response to this detection , the text information retrieved from the digital images is used by the conveyor system (301) to sort the baggage.
    类似技术:
    公开号 | 公开日 | 专利标题
    US10839528B2|2020-11-17|Authentication-based tracking
    EP3572156B1|2020-07-22|Method for sorting luggage in an airport
    CN101765853B|2012-10-31|Transportation and sorting system for identifying transport goods, particularly luggage items
    US20160117631A1|2016-04-28|Orphaned item identification
    US9524486B2|2016-12-20|System and method for retail store promotional price tag detection and maintenance via heuristic classifiers
    US20220024700A1|2022-01-27|Conveyor belt assembly for identifying an asset sort location and methods of utilizing the same
    CA3016217C|2021-11-16|Method for making a description of a piece of luggage and luggage description system
    WO2018103486A1|2018-06-14|Label for positioning item subjected to security inspection and method for positioning item subjected to security inspection
    US10810437B2|2020-10-20|Security check system and method
    CA3014096C|2021-06-22|Method for sorting pieces of luggage and luggage sorting system
    US20120263350A1|2012-10-18|Method for identifying lost or unassignable luggage
    US20160371306A1|2016-12-22|Apparatus for a luggage conveying device, and method for operating a luggage conveying device
    CN106779599A|2017-05-31|A kind of data processing method and system for lost objects
    WO2014060227A1|2014-04-24|Pairing of an object and a holder of that object
    JPWO2013018615A1|2015-03-05|Collation / search system, collation / search server, image feature extraction device, collation / search method and program
    WO2021123790A1|2021-06-24|Image processing system and method
    EP3572157B1|2020-07-29|Method for sorting luggage in airports with optimised videocoding
    EP3786836A1|2021-03-03|Article identification and tracking
    US20190205899A1|2019-07-04|Method and system for identifying a product
    WO2021038218A1|2021-03-04|Article identification and tracking
    FR3094114A1|2020-09-25|LUGGAGE IDENTIFICATION KIT
    FR3094116A1|2020-09-25|LUGGAGE TRACKING PROCEDURE
    US20210166028A1|2021-06-03|Automated product recognition, analysis and management
    WO2010067132A2|2010-06-17|Baggage tracking system
    US11049234B2|2021-06-29|Baggage identification method
    同族专利:
    公开号 | 公开日
    ES2823559T3|2021-05-07|
    EP3572156B1|2020-07-22|
    US20190359429A1|2019-11-28|
    FR3081354B1|2021-06-11|
    EP3572156A1|2019-11-27|
    PT3572156T|2020-10-21|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    WO2017137106A1|2016-02-11|2017-08-17|Siemens Aktiengesellschaft|Method for sorting pieces of luggage, and luggage sorting system|
    EP3301624A1|2016-09-30|2018-04-04|Siemens Aktiengesellschaft|Baggage handling system for making a sorting decision|FR3081355B1|2018-05-24|2020-06-12|Solystic|AIRPORT LUGGAGE SORTING METHOD WITH OPTIMIZED VIDEOCODING|
    CN112193716A|2020-10-14|2021-01-08|安徽尚佰智能家居有限公司|Quality inspection and mark supplement system for paint plates|
    法律状态:
    2019-05-23| PLFP| Fee payment|Year of fee payment: 2 |
    2019-11-29| PLSC| Publication of the preliminary search report|Effective date: 20191129 |
    2020-05-22| PLFP| Fee payment|Year of fee payment: 3 |
    2021-05-03| PLFP| Fee payment|Year of fee payment: 4 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1870600A|FR3081354B1|2018-05-24|2018-05-24|LUGGAGE SORTING PROCESS IN AN AIRPORT|
    FR1870600|2018-05-24|FR1870600A| FR3081354B1|2018-05-24|2018-05-24|LUGGAGE SORTING PROCESS IN AN AIRPORT|
    PT191711423T| PT3572156T|2018-05-24|2019-04-25|Method for sorting luggage in an airport|
    ES19171142T| ES2823559T3|2018-05-24|2019-04-25|Baggage classification procedure at an airport|
    EP19171142.3A| EP3572156B1|2018-05-24|2019-04-25|Method for sorting luggage in an airport|
    US16/420,381| US20190359429A1|2018-05-24|2019-05-23|Method of sorting baggage at an airport|
    [返回顶部]