Shoplifter prevention system

Abstract

To improve a rate of identifying a fixed person, and reduce a shoplift occurrence rate at each shop to suppress a damage amount caused by shoplifters at each shop by enhancing a disliking rate by a shoplifter addict.SOLUTION: According to a shoplifter prevention system, a server device 1 comprises: a registration part 16-1 that registers an account for specifying data terminals 2-1 to 2-N in a monitor staff memory; a recognition part 16-2 that recognizes propriety of a fixed person by receiving specified person information notified from the data terminals 2-1 to 2-N; and an issue part 16-3 that issues an alert to a shop terminal, based on information about the fixed person recognized by the recognition part 16-2.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an anti-shoplifting system that acquires video information captured by a camera installed in a store and distributes the acquired video information to a user monitoring the store to identify and report a person hiding. [Background technology]

[0002] Traditionally, stores have been equipped with cameras to prevent shoplifting, and staff in a separate room monitor the cameras to stare at the screen and identify customers whose behavior fits into empirically known patterns. They then respond by reporting the situation to store staff or sounding an alarm.

[0003] For example, Patent Document 1 below discloses that "in order to provide an information processing system or the like capable of preventing shoplifting without increasing the burden on store staff, etc., multiple imaging devices are distributed throughout a store. A server detects, based on image data acquired by each of the multiple imaging devices, whether or not the person captured in the image data has picked up a product displayed in the store. When the server detects that the person in the image data has picked up a product, it uses a trained model that has been trained to output information indicating whether the person captured in the image data has committed an act of theft when image data is input, to determine whether or not the person who picked up the product has committed an act of theft, based on the image data acquired by the imaging devices." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6573185 Summary of the Invention [Problem to be solved by the invention]

[0005] Security systems that use AI technology like the one described above require multiple high-performance cameras - for example, in an eight-story store, the same number of cameras must be installed on each floor - and to function as equipment, the initial investment is heavy and the effectiveness is sometimes considered to be not very high.

[0006] On the other hand, if a human wave tactic is used and the number of security guards monitoring the store is increased, this will result in higher labor costs (depending on the contract with the security company), and in many cases the success rate of identifying those behaving suspiciously is not proportional to the number of security guards.

[0007] The present invention has been made to solve the above-mentioned problems, and aims to provide an anti-shoplifting system that reduces the cost of the entire system configuration by distributing real-time video captured by cameras installed in contracted stores from a server device to the terminals of users who wish to act as monitors, and having them identify and report suspicious store visitors, and that can pay a reward to the monitoring user if the confirmed person is identified. [Means for solving the problem]

[0008] The shoplifting prevention system of the present invention is a shoplifting prevention system that allows communication between a store terminal, a network camera installed in a store, a server device that acquires video information captured by the network camera, and a data terminal operated by a user that monitors the video information distributed by the server device via a specified communication medium, and is characterized in that the server device comprises a registration means for registering an account for identifying the data terminal in a database, a certification means for receiving identified individual information reported from the data terminal and determining whether the identified individual is a confirmed individual, and an issuance means for issuing an alert to the store terminal based on information regarding the confirmed individual certified by the certification means. Effect of the Invention

[0009] According to the present invention, the cost of the entire system configuration can be reduced, and by paying a reward to the monitored user if a confirmed individual is identified, the rate at which confirmed individuals are identified can be improved, and the incidence of shoplifting in each store can be reduced, thereby increasing the rate at which stores are disliked by habitual shoplifters and reducing the amount of damage caused by shoplifting in each store. [Brief description of the drawings]

[0010] The drawings show specific embodiments of the present invention, including essential features of the invention as well as alternative and preferred embodiments. [Figure 1] 1 is a block diagram illustrating a configuration of an anti-shoplifting system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a block diagram illustrating the hardware configuration of the shoplifting prevention system shown in FIG. [Diagram 3] FIG. 2 is a block diagram illustrating a usage form of the shoplifting prevention system shown in FIG. 1. [Figure 4] FIG. 2 is a block diagram illustrating a usage form of the shoplifting prevention system shown in FIG. 1. [Diagram 5] FIG. 2 is a block diagram illustrating a distribution process of security video information in the shoplifting prevention system according to the embodiment. [Figure 6] 5 is a flowchart showing an example of security video information processing in the shoplifting prevention system according to the embodiment. [Figure 7] 5 is a flowchart showing an example of security video information processing in the shoplifting prevention system according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Next, the best mode for carrying out the present invention will be described with reference to the drawings.

[0012] <System configuration> [First embodiment] FIG. 1 is a block diagram illustrating the configuration of an anti-shoplifting system according to this embodiment.

[0013] In Fig. 1, reference numeral 1 denotes a server device, which is connected to stores BL1 to BLN and data terminals 2-1 to 2-N operated by registered users who perform monitoring work, via a network 4 using a communication unit 1-1, so as to be able to communicate bidirectionally. Note that each store has a store terminal (not shown) connected to the network 4.

[0014] Reference numeral 1-2 denotes a controller unit, which includes hardware resources described in detail below. Reference numeral 1-3 denotes a database, which includes a function for storing registered user information of each user who operates the data terminals 2-1 to 2-N and for managing reward rankings based on predetermined conditions.

[0015] The data terminals 2-1 to 2-N are linked to registered users Y1 to YN registered in the server device 1, and when access to the server device 1 is approved, security video information captured by network cameras (e.g., dome cameras) CA1-1 to CA1-6 installed on each floor of the stores BL1 to BLN is distributed from the server device 1 so that each registered user Y1 to YN can perform security video monitoring work. The data terminals 2-1 to 2-N are equipped with communication units 2-1-1 to 2-N-1 for connecting to the network 4.

[0016] It is assumed that a similar camera system is installed in each store and transmits security video information to server device 1.

[0017] The network cameras CA1-1 to CA1-6 transmit clear security video information to the server device 1 via the communication units TM1 to TMN. The security video information that the server device 1 distributes to each of the data terminals 2-1 to 2-N is subjected to image processing as described below, and privacy protection processing is performed so that the face of the monitored subject cannot be identified.

[0018] M1 to MN are monitors who are configured to operate data terminals 3-1 to 3-N to receive alerts sent from server device 1 and recognize any entrant behaving suspiciously on any floor of the store as a specific target for monitoring. In addition, on the floor where the network cameras CA1-1 to CA1-6 are arranged, separate sub-cameras may be set up on the evacuation stairs, toilets, and outside stairs in the direction of the exit.

[0019] FIG. 2 is a block diagram for explaining the hardware configuration of the shoplifting prevention system shown in FIG. 1, and the same components as those in FIG. 1 are given the same reference numerals and their explanations are omitted.

[0020] 2, reference numeral 13 denotes a CPU, which executes a control program stored in an external memory 18 to construct an anti-shoplifting platform 1A on a RAM 16 and realizes various data processing. Reference numeral 15 denotes a keyboard, which issues operational instructions to a user interface displayed on a display 17.

[0021] In the shoplifting prevention platform 1A, 16-1 is a registration unit that registers stores BL1 to BLN as monitoring targets, including the location, camera number, store name, emergency contact information, and contact information including email addresses of data terminals 3-1 to 3-N operated by the administrator. When registering a store, a specified registration fee (annual contract, monthly contract, or subscription contract) is charged.

[0022] Furthermore, the registration unit 16-1 registers user information for identifying the data terminals 2-1 to 2-N operated by each user who will be a security monitor, terminal identification information, and the user's desired monitoring schedule in the database 1-3. 16-2 is a certification unit that receives specific person information reported from the data terminals 2-1 to 2-N operated by each user and executes a process of certifying whether the person is a confirmed person or not.

[0023] 16-3 is an issuing unit that issues an alert to data terminals 3-1 to 3-N operated by store managers based on information about the confirmed person certified by the certification unit 16-2. The format of the alert is not important, but it is assumed that the alert is provided with information that allows each store's security guard to recognize which floor they are on. The issuing unit 16-3 is also configured to be able to issue alerts to multiple reporting destinations according to a specified alert level. This makes it possible to build a system that aggregates alerts without limiting the system to a single reporting point.

[0024] The data terminals 3-1 to 3-N operated by the monitors M1 to MN may be configured to share alert information by linking a sub-application with data terminals operated by floor staff in other stores.

[0025] 16-4 is a settlement unit that executes a process of paying remuneration to each user by electronic settlement based on answer information on the confirmed person notified from the data terminals 3-1 to 3-N. It is assumed that various electronic money systems are configured to be available for electronic settlement.

[0026] An image processor 16-5 performs privacy processing on the video information captured by the network cameras CA1-1 to CA1-6 to generate a monitor image. Here, the privacy processing refers to partial mosaic processing, partial image replacement processing, low-resolution processing of the entire monitored person, color conversion processing, or a combination of these.

[0027] A distribution control unit 16-6 executes a process of distributing monitor images to each user randomly selected from among the users registered by the registration unit 16-1 according to a predetermined monitor condition. Specifically, the process executes a process of selecting and randomly distributing store images at intervals of, for example, 5 minutes, 7 minutes, and 15 minutes. Note that a configuration may be adopted in which a schedule is allocated so that multiple users can simultaneously monitor the same store or the same floor. In addition, the distribution control unit 16-6 can prevent personal information from being spread to registered users by concealing and distributing information about the store in which the network camera is installed.

[0028] This allows for the construction of a more accurate crime prevention monitoring system by simultaneously recognizing the combination of a user focusing on head movements and hand movements among the behavior of the person being monitored.

[0029] In addition, by implementing control that switches between 5-minute, 7-minute, and 15-minute intervals, it is possible to prevent the monitor from becoming too tense and overlooking the person being monitored, while also controlling the monitor to ensure that visual recognition ability does not decrease at all times.

[0030] 3 and 4 are block diagrams for explaining the usage of the shoplifting prevention system shown in FIG. 1, and correspond to an example of system operation in which the store BL1 is the object of monitoring.

[0031] When a registered user (registered user Y1) operates data terminal 2-1 and finds suspicious behavior (including hand gestures, body language, and movement into the camera's blind spot) on the display screen of privacy-processed security video information distributed from server device 1 on any floor of store BL1, which indicates shoplifting itself or preparatory behavior for shoplifting (frequent looking around or going back and forth along the same route (wandering)) (B1), the user presses a discovery button on a user interface provided by an application downloaded from server device 1, which sends an alert to server device 1 (B2). Next, server device 1 starts recording security video information from camera CA1-1 of store BL1 (B3).

[0032] Next, the server device 1 analyzes the security video information from the camera CA1-1 in the store BL1 using a behavior pattern analysis application to determine whether or not shoplifting has occurred (B4).

[0033] If the behavior pattern analysis application determines through analysis that there is a high probability of shoplifting, an alert is sent to the data terminal 3-1 operated by the security staff of the store BL1 (B5).

[0034] In response to this, the store confirms the alert sent from the server device 1, instructs the monitor stationed at the store BL1 or the floor manager, etc. to begin monitoring activities, and takes measures including protection and reporting to the authorities (B6).

[0035] The processing time during this period is assumed to be approximately one minute for the notification from the data terminal 2-1 and four minutes for the notification of this alert from the server device 1, but it is possible to shorten this time.

[0036] 5 is a block diagram for explaining a distribution process of security video information in the shoplifting prevention system according to the present embodiment. In this example, an example will be explained in which stores BL1 to BL6 are distributed to data terminals 2-1 to 2-N operated by registered users Y1 to Y5.

[0037] In FIG. 5, a communication unit 1-1 of a server device 1 receives security video information from three locations of cameras CA1-1 to CA1-3 in a store BL1 via channels A-1 to A-3.

[0038] Similarly, communication unit 1-1 of server device 1 receives security video information from three locations of cameras CA2-1 to CA2-3 in store BL2 via channels B-4 to B-6.

[0039] Similarly, communication unit 1-1 of server device 1 receives security video information from three locations, captured by cameras CA3-1 to CA3-3 in store BL3, via channels C-7 to C-9.

[0040] Similarly, communication unit 1-1 of server device 1 receives security video information from three locations, from cameras CA4-1 to CA4-3 in store BL4, via channels D-10 to D-12.

[0041] Similarly, communication unit 1-1 of server device 1 receives security video information from three locations of cameras CA5-1 to CA5-3 in store BL5 via channels E-13 to E-15.

[0042] Similarly, communication unit 1-1 of server device 1 receives security video information from three locations of cameras CA6-1 to CA6-3 in store BL6 via channels F-16 to F-18.

[0043] In contrast, the image processing unit 16-5 of the server device 1 is equipped with multiple filters FIL-1 to FIL-18 for multiple privacy processing, and the distribution control unit 16-6 distributes security video information processed by filters FIL-1, FIL-6, and FIL-10 to the registered user Y1.

[0044] Similarly, the image processing unit 16-5 of the server device 1 is equipped with multiple filters FIL-1 to FIL-18 for multiple privacy processing, and the distribution control unit 16-6 distributes security video information processed by filters FIL-3, FIL-4, and FIL-8 to registered user Y2.

[0045] Similarly, the image processing unit 16-5 of the server device 1 is equipped with multiple filters FIL-1 to FIL-18 for multiple privacy processing, and the distribution control unit 16-6 distributes security video information processed by filters FIL-2, FIL-9, FIL-13, and FIL-15 to registered user Y3.

[0046] Similarly, the image processing unit 16-5 of the server device 1 is equipped with multiple filters FIL-1 to FIL-18 for multiple privacy processing, and the distribution control unit 16-6 distributes security video information processed by filters FIL-5, FIL-11, FIL-14, and FIL-18 to registered user Y4.

[0047] Similarly, the image processing unit 16-5 of the server device 1 is equipped with multiple filters FIL-1 to FIL-18 for multiple privacy processing, and the distribution control unit 16-6 distributes security video information processed by filters FIL-7, FIL-12, FIL-16, and FIL-17 to registered user Y5.

[0048] In this way, the distribution control unit 16-6 provided in the server device 1 dynamically distributes security video information while applying filter processing such as blurring faces and lowering resolution to the applications installed by account registrants in the data terminals 2-1 to 2-N.

[0049] At this time, the distribution control unit 16-6 included in the server device 1 controls the distribution of security video information of the same floor of the same store by randomly allocating the security video information in chronological order.

[0050] Furthermore, the distribution control unit 16-6 included in the server device 1 may perform control so that security video information is randomly distributed in chronological order, and security video information of the same floor of the same store is distributed to registered users without duplication.

[0051] Furthermore, the distribution control unit 16-6 included in the server device 1 may control the allocation of security video information transmitted from each of the network cameras CA1-1 to CA1-6 to CAN-1 to CAN-6 of each of the stores BL1 to BLN so that the allocation corresponds to an average monitoring time.

[0052] 6 and 7 are flow charts showing an example of security video information processing in the shoplifting prevention system according to the present embodiment. Note that (1) to (19) indicate each step, and each step is realized by the CPU 13 executing a control program stored in the external memory 18.

[0053] The CPU 13 of the server device 1 determines whether the request received by the communication unit 1-1 via the network 4 is a user registration application (1), and if the CPU 13 determines that it is not a user registration application, the CPU 13 further determines whether the request received by the communication unit 1-1 via the network 4 is a store registration application (2), and if the CPU 13 determines that it is not a store application, terminates the processing.

[0054] On the other hand, if the CPU 13 determines in step (2) that the application is for a store, it registers the store name, camera ID (for each floor), account, and administrator name in the database 1-3 (3). Next, it registers an account for any of the data terminals 3-1 to 3-N used by the administrator in the database 1-3 (4), and proceeds to step (7) and onward.

[0055] On the other hand, if the CPU 13 determines in step (1) that the request is a user registration application, it registers in the database 1-3 the user name, account, and remuneration payee of the user who monitors the store video information that is distributed from the network camera CA1-1 of the store BL1, etc. to the server device 1 and privately processed (5). Next, the CPU 13 sets the monitoring time for monitoring the store BL1, etc., for each user in a management table provided in the database 1-3 (6).

[0056] Next, CPU 13 judges whether the current time is the monitoring start time (7), and if CPU 13 judges that it is not the monitoring start time, it waits for the monitoring start time. If CPU 13 judges that it is the monitoring start time, it judges whether it has received store video information from store BL1 or the like via network 4 (8), and repeats step (7) while it has not received store video information (NO). On the other hand, if CPU 13 judges that it has received store video information (YES) in step (7), when CPU 13 starts receiving store video information, CPU 13 starts image processing unit 16-5 and performs, for example, low-resolution processing (9) so that the monitored person is not clearly identifiable by registered users Y1 to YN.

[0057] Next, the CPU 13 activates the distribution control unit 16-6 and performs a process of randomly selecting a distribution destination (the account of any one of the registered users Y1 to YN) of the privately processed store video information (10).

[0058] Next, the CPU 13 distributes the image that has been subjected to image processing and further edited, such as size processing, to the selected registered user Y1 to YN via the network 4 (11).

[0059] It is assumed that the controller unit 1-2 checks the login status of the data terminals 2-1 to 2-N and recognizes a valid account as a delivery destination.

[0060] Next, the CPU 13 determines (12) whether it is time to switch the delivery destination based on the difference between the current time and the delivery time determined in step (7). If the CPU 13 determines that it is not time to switch the delivery destination, the process returns to step (11) and repeats the process.

[0061] On the other hand, if the CPU 13 determines in step (12) that it is time to switch the distribution destination, the CPU 13 starts the distribution control unit 16-6 and executes a process of randomly switching the distribution destination (13). As a result, the store video information displayed on the display unit of the data terminal 2-1 to 2-N of each registered user Y1 to YN is dynamically switched, and the registered users Y1 to YN start monitoring the store video information of the next target store BL1 to BLN (store name is not disclosed) with a fresh awareness.

[0062] Next, the CPU 13 determines (14) whether or not an alert (first alert) that is transmitted when the registered users Y1 to YN determine that there is a high possibility that they will be identified as engaging in suspicious behavior has been received from any of the data terminals 2-1 to 2-N via the network 4. Here, if the CPU 13 determines that the first alert has not been received, the process returns to step (11) and repeats the process.

[0063] On the other hand, if the CPU 13 determines in step (14) that the first alert has been received, it receives recorded data from, for example, the data terminals 2-1 to 2-N operated by the registered users Y1 to YN via the network 4 (15).

[0064] It is assumed that, at the time of user registration, the data terminals 2-1 to 2-N are configured such that a compressed image recording area is reserved in the memory resources of the data terminals 2-1 to 2-N by a monitoring application downloaded from the server device 1, making it possible to record video for one hour continuously.

[0065] Next, CPU 13 starts up certification unit 16-2 and plays back video information for a total of one hour, for example, 30 minutes before and after the timing of transmitting the first alert received from data terminals 2-1 to 2-N, and analyzes the pre-registered types of shoplifting and the played-back images to determine whether or not shoplifting can be recognized (16).

[0066] If the CPU 13 determines that shoplifting has occurred, it pays a reporting reward to the registered users Y1 to YN set in the database 1-3 (17). Next, it is determined whether the closing time of each store BL1 to BLN has arrived, that is, whether the store has closed (18). If it determines that the store has closed, the CPU 13 ends the process.

[0067] On the other hand, if the CPU 13 determines in step (18) that it is not the closing time, the process returns to step (8) and the process is repeated.

[0068] On the other hand, if the CPU 13 determines in step (16) that the act was not shoplifting, it registers the person as a person to be warned in the database 1-3 (19), and proceeds to step (17).

[0069] [Effects of the first embodiment] According to this embodiment, the burden on stores BL1 to BLN in the security surveillance system within stores BL1 to BLN and the data processing burden on the server device 1 are reduced, and with support from external monitors by multiple registered users Y1 to YN, it is possible to build a system that is cheaper and can significantly improve the rate of identifying criminals.

[0070] Second Embodiment In the above embodiment, an example was described in which the amount of remuneration paid to registered users Y1 to YN was fixed. However, for registered users Y1 to YN with a high identification rate (users with a high monitor ranking), a fixed remuneration amount plus a contribution bonus may be paid.

[0071] [Effects of the second embodiment] According to this embodiment, each of the registered users Y1 to YN can expect to receive a higher reward amount than the other registered users Y1 to YN by contributing more to the crime prevention activities of the stores BL1 to BLN.

[0072] Third Embodiment A configuration may be incorporated in which registered users Y1 to YN with high identification rates are registered in the database 1-3 as one team, so that the monitoring work within the stores BL1 to BLN can be entrusted to a dynamically organized set of registered users.

[0073] [Effects of the third embodiment] According to this embodiment, by providing a monitoring service with a high crime prevention effect, it is possible to conclude advantageous contracts with the stores BL1 to BLN. [Industrial Applicability]

[0074] In the above embodiment, a system is envisaged for preventing shoplifting within stores BL1 to BLN, but by applying this to surveillance work in individual homes and housing complexes in the nursing care business, it is also possible to significantly reduce the contract fees for nursing care monitoring systems provided by security companies.

[0075] The disclosure regarding the present invention described above can be summarized at least as follows.

[0076] (1) A shoplifting prevention system capable of communicating via a specified communication medium among a store terminal, a network camera installed in a store, a server device that acquires video information captured by the network camera, and a data terminal operated by a user that monitors the video information distributed by the server device, wherein the server device comprises a registration means that registers an account for identifying the data terminal in a database, a certification means that receives identified individual information reported from the data terminal and determines whether the identified individual is a confirmed individual, and an issuance means that issues an alert to the store terminal based on information regarding the confirmed individual certified by the certification means.

[0077] (2) The server device is characterized by having a settlement means for paying a reward to the user based on answer information regarding the confirmed person notified from the store terminal.

[0078] (3) The server device is characterized by comprising an image processing means for applying privacy processing to the video information captured by the network camera to generate a monitor image, and a distribution control means for distributing the monitor image to a user randomly selected from among the users registered by the registration means in accordance with predetermined monitor conditions.

[0079] (4) The distribution control means distributes information about the store in which the network camera is installed in a confidential manner.

[0080] (5) The data terminal is characterized in that it has an installation means for installing a monitoring application distributed from the server device, and the monitoring application has a function for allowing a user to set a monitoring time specified by the time to start and end monitoring.

[0081] (6) The issuing means is characterized in that it issues alerts to a plurality of report destinations according to a predetermined alert level.

[0082] (7) The settlement means is characterized in that it pays a variable remuneration amount based on the answer information about the confirmed person and the monitor ranking notified from the store terminal. [Explanation of symbols]

[0083] 1. Server device 2-1~2-N Data terminal 3-1~3-N Data terminal 16-1 Registration Section 16-2 Certification Department 16-3 Publishing Department 16-4 Payment Department 16-5 Image Processing Section 16-6 Distribution control section

Claims

1. A shoplifting prevention system capable of communicating with a store terminal, a network camera installed in a store, a server device that acquires video information captured by the network camera, and a data terminal operated by a user that monitors the video information distributed by the server device, via a predetermined communication medium, comprising: The server device includes: a registration means for registering an account for identifying the data terminal in a database; A certification means for receiving the specified person information reported from the data terminal and certifying whether the specified person is a confirmed person or not; an issuing means for issuing an alert to the store terminal based on information about the confirmed person certified by the certification means; A shoplifting prevention system comprising:

2. The server device includes:

2. The shoplifting prevention system according to claim 1, further comprising a settlement means for paying a reward to the user based on answer information regarding the confirmed individual notified from the store terminal.

3. The server device includes: an image processing means for performing privacy processing on the video information captured by the network camera to generate a monitor image; Among the users registered by the registration means, the a distribution control means for distributing a monitor image to a selected user; 2. The shoplifting prevention system according to claim 1, further comprising:

4. 4. The shoplifting prevention system according to claim 3, wherein the distribution control means distributes information about the store in which the network camera is installed in a confidential manner.

5. The data terminal comprises: an installation means for installing a monitoring application distributed from the server device; 2. The shoplifting prevention system according to claim 1, wherein the monitoring application has a function of setting a monitoring time specified by a time when the user starts monitoring and a time when the user ends monitoring.

6. 2. The shoplifting prevention system according to claim 1, wherein said issuing means issues an alert to a plurality of report destinations according to a predetermined alert level.

7. 3. The shoplifting prevention system according to claim 2, wherein the settlement means pays a variable remuneration amount based on the answer information on the confirmed shoplifter notified from the store terminal and the monitor ranking.

Images