Background of the invention

[0001] The invention relates to an anti-fraud system and method that is usable in a commercial environment, in particular in retailing.

[0002] Specifically, but not exclusively, the invention can help the seller (shop, supermarket, etc) that uses self-service electronic scales with which the consumer can weigh the item to be purchased and, by entering data identifying the item itself, obtain the receipt with the corresponding price.

[0003] In particular, the invention relates to a system and a method for controlling a purchase operation by a consumer.

[0004] Vision apparatuses with artificial intelligence are already known that are applied to scales to identify the weighed products, so as to avoid the customer having to key in the product on a keyboard, with resulting reduction of errors.

[0005] Apparatuses with artificial intelligence are also known for viewing the weighed product and then excluding from the set of products selectable by the customer those products that on the base of the reading made certainly to not correspond to the product weighed in order facilitate the choice by the customer.

[0006] Vision apparatuses with artificial intelligence of the prior art nevertheless have certain limits and drawbacks.

[0007] Firstly, known apparatuses can be easily subject to incorrect product evaluations, with resulting inconvenience for the customer and time wasting due to the need to release the purchase process.

[0008] Secondly, precisely in order to minimize incorrect readings and increase the reliability of the operation, known apparatuses comprise very sophisticated and precise devices and consequently have a cost that is such as not be advantageous when identifying fraud and lack of attention by the consumer in order to check the correctness of the purchase operation.

Summary of the invention

[0009] One object of the invention is to devise a system and/or a method that is able to overcome the aforesaid limits and drawbacks of the prior art.

[0010] One object is to provide an anti-fraud system and/or method that are alternative to those of the prior art.

[0011] One advantage is to give rise to an anti -fraud system and/or method that is usable for the sale of loose objects by number or weight.

[0012] One advantage is to provide an anti-fraud system and/or method able to automate the recognition of an anomalous situation (fraud or distraction), without however modifying the usual self-service purchasing process, i.e. with practically no impact on the habits of customers.

[0013] One advantage is to make an anti-fraud system available that is constructionally simple and cheap and is for practical and immediate use.

[0014] One advantage is to permit the use of detecting means (in particular vision means) with artificial intelligence that is not particularly sophisticated, precise and reliable because the task of the final check of possible fraud or distraction is anyway left to a human operator, in particular to an operator of the retail outlet, like for example a checkout operator or a supervisor.

[0015] Such objects and advantages and still others are achieved by the system and/or by the method according to one or more of the claims set out below.

[0016] In one embodiment, an anti-fraud system comprises scales, a user interface configured to enable data on an item weighed by the scales to be entered, sensor means for detecting at least one feature of the item weighed by the scales, a printer for issuing a document that shows the weight of the item and the data entry entered via the user interface, and (electronic and programmable) control means to receive the data entry and the feature detected by the sensor means, to check whether the data entry matches the detected feature, and to print an anomaly trace on the document if the data entry does not match the detected feature.

[0017] In one embodiment, an anti-fraud system comprises detecting means with artificial intelligence, in particular comprising a camera configured to view the pan of electronic scales, in order to recognize the type of item that is weighed each time, and with an electronic system configured to check whether the data entry identifying the item that is entered by the consumer (for example by keying in on the keyboard of the electronic scales), corresponds or does not correspond to the data actually detected by the detecting means. In the event of an anomaly, the electronic system can emit an attention signal that will not interrupt the purchase process (for example, will anyway enable a receipt to be issued). The detecting means may be able to recognize other anomalies of the purchase process, like, for example, the failure by the customer to close the bag that contains the item, where, as is known, this closure normally consists of knotting two ends of the bag. Also in this case, if the detecting means detects the fault (bag not knotted and thus not open), the purchase process is not interrupted.

[0018] If at least one anomaly is detected in a single sale, the electronic system reports the anomaly to a check-out operator and/or to another operator of the seller (for example a supervisor of the self-checkout or auto-scanner systems), who can thus perform a direct control of the reported product personally.

[0019] The invention in question thus enables the arrest of the sale process that is due to a false reading to be avoided, thus also decreasing inconvenience to the customer, and thus enables the checks by the operators assigned by the seller to be focused on the basis of specific (non-random) reports with a higher possibility of actual anomaly, thus increasing significantly the efficacy of the checks.

Brief description of the drawings

[0020] The invention can be better understood and implemented with reference to the attached drawings that illustrate one embodiment thereof by way of non-limiting example, in which:

Figure 1 is a diagram of an embodiment of an anti-fraud system according to the present invention.

Detailed description

[0021] With reference to the aforesaid figure, overall with 1 an anti -fraud system has been indicated that is usable in a commercial field, in particular in retailing. The anti-fraud system 1 can help, in particular, a seller (shop, supermarket, etc.) that uses self-service electronic scales owing to which a user (or consumer) can weigh the item to be purchased, enter data identifying the item and obtain the receipt with the corresponding price.

[0022] The anti-fraud system 1 in particular enables the correctness of the purchase operation to be checked by the consumer. An incorrect purchase operation may comprise, for example, keying in a data associated with an item different from the weighed item (for example by entering into the electronic system of the scales, via the user interface, the name of an item with a price lower than the item actually arranged on the pan of the scales). In general, the price of the entered item will be different from the price of the weighed item, so the commercial transaction will not be correct.

[0023] The anti-fraud system 1 may comprise, in particular, scales 2 (for example electronic) to weigh at least one item or group of items.

[0024] The anti-fraud system 1 may comprise, in particular, a user interface 3 configured to enable a user to enter data on an item weighed by the scales. The user interface 3 may comprise, in particular, an interface of known type, for example a keypad, in which each key may be associated with a respective item, connected to the (electronic) scales, for example in a fruit and vegetable section of a retail outlet, so that a type of fruit or vegetable product corresponds to each key.

[0025] The anti-fraud system 1 may comprise, in particular, detecting means configured to recognize at least one feature of an item arranged on the scales. The detecting means may be configured, in particular, to identify the type of fruit or vegetable product arranged on the scales 2 by a user.

[0026] Such detecting means may comprise, as in this specific embodiment, vision means 4. The vision means 4 may comprise, in particular, a vision system with artificial intelligence, for example of substantially known type. The vision system may comprise, in particular, a camera arranged for observing the scales and which is able to recognize the type of item arranged on the scales.

[0027] The anti-fraud system 1 may comprise, in particular, a printer 5 configured to issue a document (ticket, receipt, etc.) relating to the transaction. The printed document may show, in particular, the weight of the item and the corresponding data entry entered by the user. The printed document may show, in particular, the price or other data relating to the transaction. The printed document may show, in particular, data identifying the document, for example an identification code (in general a linear or bidmensional barcode).

[0028] The anti-fraud system 1 may comprise, in particular, control means 6 (for example an electronic control unit for managing and controlling the weighing station and/or for managing and controlling the retail outlet) configured to receive the data on the item entered by the user and to receive the feature of the item detected by the detecting means.

[0029] The control means 6 may be configured (for example may comprise computer programme instructions), in particular, to check whether the data entry entered by the user via the user interface 3 matches or does not match the feature actually detected by the detecting means.

[0030] The control means 6 may be configured, in particular, to control the printer 5 such that on the printed document, together with the transaction data, at least one anomaly trace is printed (in particular a trace that is suitable for being recognized subsequently by an operator visually and/or by an optical reader), in the event of detection of an anomaly, for example if the data entry entered by the user via the user interface 3 does not match the feature detected by the detecting means.

[0031] The anomaly trace could comprise, in particular, at least one trace that is visible to the naked eye, so as to be able to be easily identified by an operator of the retail outlet, for example by a check-out operator or a supervisor, or a trace recognizable by an optoelectronic reading device or by other reading means.

[0032] The detecting means may be configured, in particular, to recognize whether the item arranged on the scales 2 comprises a bag may be closed manually by a user. Such a bag may comprise, for example, a bag used normally in a fruit and vegetable section that can be closed by knotting two ends of the bag. The detecting means may be configured, in particular, to recognize whether the bag has actually been closed.

[0033] The detecting means (for example the vision means 4) may be configured, in particular, to detect the presence of a foreign and/or unexpected body at or near the scales (for example underneath the pan of the scales), in particular to recognize whether the user kept the pan of the scales lifted (with a finger or a hand or another object) to falsify the reading of the weight.

[0034] The detecting means may comprise, as said, vision means 4 that in turn may comprise, in particular, a camera with artificial intelligence, for example a camera of known type. The vision means 4 may comprise, in particular, an infrared and/or ultraviolet camera, configured to determine (with greater precision) the features of a product contained inside a bag.

[0035] Alternatively or in addition to the detecting means disclosed above, the detecting means may comprise, in particular, at least one remote detecting sensor, in particular of optical type, like for example an LIDAR (acronym of Light Detection and Ranging or Laser Imaging Detection and Ranging) sensor.

[0036] The control means 6 may be configured, in particular, to print an anomaly trace on the document if the detecting means recognizes that the bag has not been disclosed or that the weight reading has been falsified. It is possible for the anomaly trace to be printed only if there is no correspondence between the item entered by the user via the user interface 3 and the item actually detected (by the detecting means), or only in the case of detection (by the detecting means) of the fact that the bag has not been correctly closed, or only in the case of detection (by the detecting means) of a reading of the weight falsified by the user, or in at least any two of the aforesaid events, or in other combinations of the aforesaid events.

[0037] The aforesaid anomaly trace may comprise, in particular, the fact that at least one part of the document is printed in a different colour from the normal situation (absence of anomaly). It is possible to provide, for example, for a part of the document (the price and/or the barcode and/or the name of the item and/or a message or drawing suitably printed only on the document associated with an anomaly, etc) being written in red or in another colour other than the normal printing colour (which could be, as usual, black).

[0038] It is anyway possible to provide any other type of report that is printable on the document and detectable by an (electronic) reading device, or is also detectable visually, simply and immediately, by an operator of the retail outlet (for example a check-out operator or a supervisor). A possible report could be, for example, a written message or drawing or mark made purposely only on the document associated with an anomaly, also of the same colour as the rest of the print.

[0039] The anti-fraud system 1 may comprise, in particular, a payment station 7 (for example a check-out station of the retail outlet, far from the weighing station with the scales 2) provided with a reading device (for example a barcode reader) configured to read the documents issued by the printer 5. The reading device may be configured, in particular, so as to detect the aforesaid anomaly trace and, in the event of detection, issue a corresponding attention signal.

[0040] The payment station 7 may be configured, in particular, to receive an anomaly signal associated with a determined document printed on the occasion of detection of an anomaly during weighing. This anomaly signal may be issued, in particular, by the control means 6 if the data entry entered by the user does not match the feature detected by the detecting means.

[0041] The payment station 7 may be configured, in particular, to receive at least one piece of data identifying the document associated with the aforesaid anomaly, so as to recognize this document associated with the anomaly and, in the event of recognition, issue a corresponding attention signal, in particular an attention signal that is suitable for being received by an operator of the retail outlet, like for example a check-out operator or a controller, or suitable for being identified by the reading device used by an operator in the payment station 7.

[0042] In this manner, the operator in the payment station 7 is warned of a probable anomaly associated with the document (receipt) printed previously in the weighing station, so that the operator receives a reminder to examine with particular attention the item that is associated with the printed document to check the presence or absence of an anomaly.

[0043] The printer 5 may be configured, in particular, to print on the document an identification code (for example the usual linear or bidmensional barcode, which is usually printed on a receipt). The control means 6 may be configured, in particular, to place the aforesaid identification code on a blacklist, in particular if the data entry entered by the user does not match the feature detected by the detecting means. The attention signal may be issued each time the reading device of the payment station 7 reads an identification code that had been placed on the blacklist so as to warn the operator (check-out operator or supervisor) to perform a special check on the item associated with the printed document (receipt) placed on the blacklist.

[0044] The attention signal will have in particular, the task of alerting the operator, who can perform a definitive check in person. The decision relates to the actual blocking intervention, i.e. whether or not to block the transaction because of a possible anomaly, can be taken by the human element, in practice the operator or another person of the retail outlet, after an appropriate and rapid visual inspection. The attention signal will not in particular have the function of activating an anticipated block of the purchase so that for example, the signal must not be used to interrupt the printing of the document, but can be used to indicate, before the definitive completion of the transaction, a possible impropriety. [0045] As the definitive check will anyway be conducted after issuing of the printed document, in general at the moment of the move to the check-out at the payment station 7, the detecting means used may be of not outstanding quality and relatively low cost, because the valuation errors of the detecting means can be acceptable, because they will not be an immediate cause of an interruption to the transaction. The user can always conclude the preliminary weighing operation performed on the scales 2 without being interrupted by the anti-fraud system 1, inter alia by applying the habitual procedure of weighing the product and printing the receipt. Possible anomalies of the preliminary weighing operation can be investigated subsequently, at the payment station 7.

[0046] The operation of the anti-fraud system 1 will give rise to an anti-fraud method that may comprise the steps of providing scales 2 for weighing at least one item and providing a user interface 3 to enable a user to enter data relating to an item weighed by the scales 2. [0047] The anti-fraud method may comprise, in particular, the step of receiving (from the control means 6) a weight of an item weighed by the scales 2 and/or the step of receiving (by the control means 6) a data entry entered by a user via the user interface 3 and relating to the item weighed by the scales 2.

[0048] The anti-fraud method may comprise, in particular, the step of detecting, by the detecting means (vision means 4), at least one feature of the item weighed by the scales 2 (for example the type of fruit or vegetable product). The anti-fraud method may comprise, in particular, the step of checking whether the data entry entered by the user matches the feature detected by the detecting means.

[0049] The anti-fraud method may comprise, in particular, the step of printing a document that shows: (i) the weight of the item weighed by the scales 2, (ii) the data entry entered by the user and relating to the item weighed by the scales 2, and (iii) an anomaly trace if the detecting means has detected an anomaly, for example if it has detected that the data entry does not match the detected feature.

[0050] The anti-fraud method may comprise, in particular, the step of detecting by the detecting means whether the item arranged on the scales 2 comprises a bag that can be closed manually by a user and further of recognizing whether the bag has actually been closed or not by the user. The anti-fraud method may comprise, in particular, the step of recognizing by the detecting means, whether the user has falsified the reading of the weight by intervening on the scales.

[0051] The anti-fraud method may comprise, in particular, the step of printing an anomaly trace on the document if the detecting means recognizes that the bag has not actually been closed by the user, as the user should have done, or that the user has performed an anomalous intervention on the scales.

[0052] The step of printing the anomaly trace may comprise the step of printing at least one part of the document in a different printing colour from the normal situation (i.e. in the absence of an anomaly).

[0053] The anti-fraud method may comprise, in particular, the step of reading the document issued by the printer 5 by a reading device (for example a usual barcode reader) of a payment station 7. The anti-fraud method may comprise, in particular, the step of issuing an anomaly trace attention signal if the reading device actually detects the anomaly trace.

[0054] The anti-fraud method may comprise, in particular, the step of printing an identification code (for example a usual barcode) on the document. The anti-fraud method may comprise, in particular, the step of placing the identification code of the document on a blacklist in the event of an anomaly, for example if the data entry does not match the detected feature. The anti-fraud method may comprise, in particular, the step of issuing an attention signal (in particular receivable by an operator) when a reading device of a payment station 7 reads an identification code placed on the blacklist.