JP4948343B2 - Image extraction apparatus and image extraction method - Google Patents

Description

  The present invention relates to an image extraction apparatus and an image extraction method.

  Conventionally, in a convenience store, a supermarket, or the like, a monitoring system is generally known in which a surveillance camera is installed as a measure against shoplifting, and an image including a shoplifter is captured by imaging the inside of the store. In such a monitoring system, since a large amount of images are accumulated for each monitoring camera, there is a problem that enormous confirmation time is required to confirm the shoplifter with the images after the fact.

  As a solution to such a problem, a monitoring system for detecting a shoplifter from an image has been disclosed (see Patent Document 1). In such a monitoring system, an action table for each person is created in advance, and a person suspected of shoplifting can be specified by detecting a person from an image based on actions recorded in the action table.

JP 200387777 A

  However, in the technique described in Patent Document 1, the algorithm for creating the behavior table and the algorithm for determining a specific person based on the behavior recorded in the behavior table are complicated, and it takes time for processing in the monitoring system. There was a problem that. In addition, even if the person is the same person, if the action recorded in the action table is not taken, there is a problem that the person to be noticed cannot be detected, and even if the person is captured, it cannot be confirmed on the image. .

  The present invention has been made in view of the above, and an object of the present invention is to provide an image extraction apparatus and an image extraction method capable of extracting an image to be confirmed without omission by simple processing.

  In order to solve the above-described problems and achieve the object, the invention according to claim 1 includes product identification information for identifying a product and region identification information for identifying a region where a person stays in order to view the product. Display information storage means for storing in association with each other, image information captured by the monitoring camera, imaging time of the image information, and monitoring camera identification information for identifying the monitoring camera that has captured the image information are stored in association with each other. Image information storage means and image information in which a person stays for a predetermined time or more is detected from the image information stored in the image information storage means, and a region in which a person stays is detected from the detected image information. Residence detection means for determining the corresponding area identification information, imaging time and monitoring camera identification information corresponding to the image information detected by the residence detection means, and the area identification information Based on retention information storage means for storing in association, inventory information, purchase information, and sales information, insufficient product specifying means for specifying the product identification information whose quantity is insufficient, and specification by the insufficient product specifying means Area acquisition means for acquiring the area identification information corresponding to the product identification information obtained from the display information storage means, the imaging time corresponding to the area identification information acquired by the area acquisition means, and the monitoring camera identification Information from the staying information storage means, and the image extraction means for extracting the image information corresponding to the acquired imaging time and the monitoring camera identification information from the image information storage means, and the image extraction means Display means for displaying the extracted image information.

  According to a second aspect of the present invention, in the image extracting device according to the first aspect, the image information captured by the monitoring camera, the image capturing time of the image information, and the monitoring of the monitoring camera capturing the image information. Image information receiving means for receiving camera identification information; and the image information storage means associates the image information received by the image information receiving means with the imaging time and the monitoring camera identification information. It memorizes.

  According to a third aspect of the present invention, in the image extraction apparatus according to the first or second aspect, the sales information stores the product identification information and a sales time indicating a time when the product is sold in association with each other. The sales time corresponding to the product identification information specified by the shortage product specifying means acquired from the storage means and the sales information storage means, and the imaging time at the region identification information corresponding to the product identification information And a possibility determination means for determining the possibility of shoplifting based on the above.

  According to a fourth aspect of the present invention, in the image extraction device according to the third aspect, the possibility determining means includes the product identification information specified by the insufficient product specifying means from the sales time and the imaging time. When it is determined that the corresponding product is not sold within a predetermined time, it is determined that the possibility of shoplifting is higher than when the product is sold within a predetermined time.

  According to a fifth aspect of the present invention, in the image extraction method executed by the image extraction device, the image extraction device identifies product identification information for identifying a product and a region where a person stays in order to view the product. Display information storage means for associating and storing region identification information to be performed, image information captured by the surveillance camera, imaging time of the image information, and surveillance camera identification information for identifying the surveillance camera that captured the image information Image information storage means for storing information in association with each other, and the stay detection means detects and detects image information in which a person stays for a predetermined time or more from the image information stored in the image information storage means. The stay detection step for determining the region identification information corresponding to the region where the person is staying from the image information thus obtained, and the stay information storage means are detected by the stay detection step. A staying information storing step for storing the imaging time corresponding to the recorded image information and the monitoring camera identification information and the area identification information in association with each other in the staying information storage means, and the shortage product specifying means include inventory information and purchase information. Based on the sales information, the shortage product specifying step for specifying the product identification information whose quantity is insufficient, and the region acquisition means corresponding to the product identification information specified by the shortage product specifying means An area acquisition step for acquiring information from the display information storage means; and an image extraction means for storing the imaging time and the monitoring camera identification information corresponding to the area identification information acquired by the area acquisition step. And acquiring the image information corresponding to the acquired imaging time and the monitoring camera identification information from the image information storage unit. An image extracting step of extracting from the display means, characterized in that and a display step of displaying the image information extracted by the image extracting unit.

  According to the present invention, image information in which a person staying in order to select a product and an area in which the person is staying are detected and detected from a large amount of image information captured by the surveillance camera. The image capture time, surveillance camera identification information, and area identification information corresponding to the image information are retained, and the image information corresponding to the missing product is extracted and displayed based on the retained information. There is an effect that it is possible to extract an image to be confirmed without any.

  Exemplary embodiments of an image extraction apparatus and an image extraction method according to the present invention are explained in detail below with reference to the accompanying drawings. The present invention is not limited to these embodiments.

(First embodiment)
A first embodiment will be described with reference to the accompanying drawings. First, a configuration example of an image extraction apparatus to which the present invention is applied will be described. FIG. 1 is a block diagram illustrating a configuration of an image extraction apparatus according to the first embodiment. The image extraction apparatus 100 is connected to a surveillance camera 200 installed in a security area via a wired or wireless communication line.

  The image extraction apparatus 100 according to the present embodiment includes an image information receiving unit 101, an image information storage unit 102, a stay detection unit 103, a stay information storage unit 104, an insufficient product specifying unit 105, and an image information extraction unit. 106, display unit 107, extracted image information storage unit 108, inventory information acquisition unit 109, purchase information acquisition unit 110, sales information acquisition unit 111, display information database 120, monitoring area database 130, inventory An information database 140, a purchase information database 150, and a sales information database 160 are provided.

  The image information receiving unit 101 receives the image information captured by the monitoring camera 200, the imaging time, and the monitoring camera ID that identifies the monitoring camera that captured the image information from each of the monitoring cameras 200 connected by communication lines. The imaging time may not be measured by each monitoring camera 200 but may be acquired from a timer unit (not shown) provided on the image extraction apparatus 100 side. Further, instead of the imaging time, a counter value synchronized with all the monitoring cameras 200 may be received. The image information storage unit 102 stores the image information received by the image information receiving unit 101, the imaging time, and the monitoring camera ID in association with each other.

  The stay detection unit 103 detects a person staying in one area from the image information for each monitoring camera 200 stored in the image information storage unit 102, and the image pickup time when the person starts staying and the person stays there. A region ID for identifying the determined imaging time, the region, and the surveillance camera ID are stored in the staying information storage unit 104. Here, the region is a place where a person's feet staying (that is, stopping their feet) are placed when selecting a product displayed on the shelf. In this way, by detecting staying, an image of a person who has stopped to purchase a product and an image of a person who has stopped in the vicinity of the product to capture a gap for the purpose of shoplifting, etc. are extracted. Is possible. Note that image information received by the image information receiving unit 101 may be used instead of the image information stored in the image information storage unit 102 as image information for detecting a person staying in the stay detection unit 103. . Moreover, as long as the algorithm used for detection of staying can detect the area | region where the person stayed from the image, you may utilize arbitrary algorithms.

  The stay information storage unit 104 stores the stay information detected by the stay detection unit 103. FIG. 2 is an explanatory diagram illustrating an example of a data configuration of the staying information storage unit. The staying information storage unit 104 stores an imaging time when a person starts staying, an area ID where the person stays, and a monitoring camera ID that captures the staying person in association with each other.

  The inventory information database 140 stores inventory information that is the inventory quantity for each product confirmed by inventory. FIG. 3 is an explanatory diagram showing an example of the data structure of the inventory information database. The inventory information database 140 stores the product ID and the quantity for each inventory date in association with each other.

  The purchase information database 150 stores purchase information that is a purchase quantity for each product. FIG. 4 is an explanatory diagram showing an example of the data structure of the purchase information database. The purchase information database 150 stores a product ID and a quantity for each purchase date in association with each other.

  The sales information database 160 stores sales information that is a sales quantity for each product. FIG. 5 is an explanatory diagram showing an example of the data structure of the sales information database. The sales information database 160 stores the product ID and the quantity for each sales date in association with each other.

  The shortage product specifying unit 105 determines the quantity from the inventory information of the predetermined period acquired from the inventory information database 140, the purchase information of the predetermined period acquired from the purchase information database 150, and the sales information of the predetermined period acquired from the sales information database 160. Identify product IDs that do not match.

  The display information database 120 stores an area where goods and people stay to select goods. FIG. 6 is an explanatory diagram showing an example of the data configuration of the display information database. The display information database 120 stores a product ID for identifying a product and a region ID for identifying a region in association with each other. As shown in FIG. 6, since the product ID “food A” and the region ID “E1” are stored in association with each other, when the person selects the product with the product ID “food A”, the region ID It stays in the area “E1”. Note that a plurality of area IDs associated with the product ID may be used as necessary, for example, when a product is displayed across a plurality of areas, or when displayed near a boundary between a plurality of areas.

  The monitoring area database 130 stores an area captured by the monitoring camera 200. FIG. 7 is an explanatory diagram showing an example of the data configuration of the monitoring area database. The monitoring area database 130 stores the monitoring camera ID and the area ID in association with each other. As shown in FIG. 7, for example, the monitoring camera ID “1” captures an image of the area ID “A1 to A4, B1 to B4”.

  The image information extraction unit 106 acquires the area ID corresponding to the product ID specified by the shortage product specifying unit 105 from the display information database 120, and stores the imaging time and the monitoring camera ID corresponding to the acquired area ID. 104. Further, the image information extraction unit 106 extracts image information corresponding to the acquired surveillance camera ID and imaging time from the image information stored in the image information storage unit 102. The image information extraction unit 106 constitutes an area acquisition unit and an image extraction unit according to the present invention.

  The display unit 107 displays the image information extracted by the image information extraction unit 106 on a display monitor. As the image information, the image information at the imaging time may be displayed one after another, or the retention time is held and the image information from the imaging time to the end of the retention time is displayed one after another. Also good. The extracted image information storage unit 108 stores the image information extracted by the image information extraction unit 106. The extracted image information storage unit 108 may be a RAM (Random Access Memory), a HDD (Hard Disk Drive), a CD (Compact Disk) drive device, or a USB (Universal Serial Bus) memory.

  The inventory information acquisition unit 109 acquires the quantity for each product ID confirmed by the inventory as inventory information, and stores the acquired inventory information in the inventory information database 140. The purchase information acquisition unit 110 acquires the quantity of each delivered product as purchase information, and stores the acquired purchase information in the purchase information database 150. The sales information acquisition unit 111 acquires the product ID and quantity corresponding to the sales product acquired from the POS (Point Of Sales) register as sales information, and stores the acquired sales information in the sales information database 160.

  The monitoring camera 200 captures an imaging region, and sends the captured image information, imaging time, and monitoring camera ID to the image extraction apparatus 100 via a communication line. Each of the monitoring cameras 200 captures an area associated with the monitoring camera ID stored in the monitoring area database 130.

  Next, the staying information generation process by the image extraction apparatus 100 configured as described above will be described. FIG. 8 is a flowchart illustrating a stay information generation processing procedure performed by the image information reception unit and the stay detection unit.

  First, the image information receiving unit 101 receives image information, an imaging time, and a monitoring camera ID from the monitoring camera 200 (step S801). The image information receiving unit 101 stores the received image information, imaging time, and surveillance camera ID in the image information storage unit 102 (step S802).

  The stay detection unit 103 detects a person staying from the image information (step S803). For example, image information in which a human body, in particular, a foot portion is stopped for a certain time in continuous image information is detected. Accordingly, it is possible to detect image information that is stopped when a person selects a product. The stay detection unit 103 acquires the imaging time and the monitoring camera ID corresponding to the detected image information from the image information storage unit 102 (step S804).

  Next, the stay detection unit 103 determines an area ID of an area where a person is stopped from the detected image information (step S805). For example, position information corresponding to a region in image information captured in an unattended state in advance by the monitoring camera 200 is registered, and it is determined in which region the position of a person's foot is in the detected image information. Thus, the area where the person is stopped is determined, and the area ID corresponding to the determined area is acquired.

  The stay detection unit 103 stores the imaging time, the area ID, and the monitoring camera ID in the stay information storage unit 104 (step S806). In this way, the staying information storage unit 104 stores staying information as an index for searching for image information when a person stays as shown in FIG.

  Next, image information extraction processing by the image extraction apparatus 100 will be described. FIG. 9 is a flowchart showing an image information extraction processing procedure performed by the shortage product specifying unit and the image information extracting unit.

  First, the shortage product specifying unit 105 acquires inventory information for a predetermined period from the inventory information database 140 (step S901). For example, in the inventory information database 140 shown in FIG. 3, the inventory quantity on August 1st 0:00 and the inventory quantity on August 2nd 0:00 are acquired. Next, the shortage product specifying unit 105 acquires purchase information purchased in a predetermined period from the purchase information database 150 (step S902). For example, as shown in FIG. 4, the product ID and quantity purchased between 00:00 on August 1 and 23:59 on August 1 are acquired.

  The shortage product specifying unit 105 acquires the sales information sold in the predetermined period from the sales information database 160 (step S903). For example, as shown in FIG. 5, the product ID and quantity sold between 00:00 on August 1 and 23:59 on August 1 are acquired.

  The missing product identification unit 105 identifies the missing product ID from the inventory information, the purchase information, and the sales information (step S904). For example, in the case of the stock information, purchase information, and sales information as described above, the product ID “miscellaneous goods A” has a stock quantity of “15” on August 1, 0:00 and a stock quantity of August 2, 0:00. Is “15”, the purchase quantity from August 1 00: 00 to 23:59 is “10”, and the sales quantity is “8”. Therefore, the unknown quantity is “2”, which is specified as a shortage product. The The product ID “Cosmetic A” has an inventory quantity of “12” on August 1, 0:00, an inventory quantity of “12” on August 2, 00: 00, and an August 1, 00: 00-23: Since the purchase quantity of 59 is “10” and the sales quantity is “8”, the unknown quantity is “2”, which is specified as an insufficient product.

  The image information extraction unit 106 acquires an area ID corresponding to the product ID identified as the shortage product from the display information database 120 (step S905). For example, as shown in FIG. 6, the area ID “C3” corresponding to the product ID “miscellaneous goods A” is acquired. Next, the image information extraction unit 106 acquires the imaging time and the monitoring camera ID corresponding to the region ID from the staying information storage unit 104 (step S906). For example, in the case shown in FIG. 2, the imaging time “08/01 16:05:05” and the monitoring camera ID “2” corresponding to the region ID “C3” are acquired.

  The image information extraction unit 106 acquires image information corresponding to the imaging time and the monitoring camera ID from the image information storage unit 102 (step S907). The display unit 107 displays the acquired image information on the display monitor (step S908). The image information extraction unit 106 outputs the acquired image information to the extracted image information storage unit 108 (step S909). FIG. 10 is an explanatory diagram illustrating an example of an index of image information stored in the extracted image information storage unit. As shown in FIG. 10, the image information corresponding to FIG. 10 is extracted from the staying information shown in FIG. 2, and the image information is stored in the extracted image information storage unit 108. Thereby, it is possible to confirm the image information repeatedly after the fact, and it is possible to submit the image information to the police for cooperation. The extracted image information storage unit 108 may hold an index as shown in FIG. 10 in addition to the image information, and may select the image information using the held index.

  In this way, from the enormous amount of image information captured by the surveillance camera, image information in which a person staying in order to select a product and an area in which the person is staying are detected, and the detected image By holding the imaging time corresponding to the information, the monitoring camera identification information, and the area identification information, it is possible to easily extract and display the image information corresponding to the missing product. As a result, it is possible to display without omission image information that is highly likely that a shoplifter has been imaged from a vast amount of image information, and the time for confirming the image information can be greatly shortened.

  Next, the relationship between the product shelf, the monitoring camera, and the area will be described. FIG. 11 is an explanatory diagram illustrating an example of a product shelf, a monitoring camera, and an area in a store. As shown in FIG. 11, surveillance cameras 200-1 to 200-7 are arranged in a store such as a convenience store. For example, the monitoring camera 200-2 images the area IDs “C1 to C4, D1 to D4”.

  Here, as described above, the image information captured by the monitoring camera 200-2 (monitoring camera ID = 2) is stored in the image information storage unit 102 together with the imaging time and the monitoring camera ID. Further, a person staying in the captured image information is detected, and an imaging time and a monitoring camera ID corresponding to the detected image information are stored in the retention information storage unit 104. On the other hand, the missing product ID “miscellaneous goods A” is identified from the inventory information database 140, the purchase information database 150, and the sales information database 160, and the area ID “C3” corresponding to the identified product ID “miscellaneous goods A” is displayed. Obtained from the information database 120.

  The imaging time and the monitoring camera ID corresponding to the area ID “C3” are acquired from the staying information storage unit 104, and the image information corresponding to the acquired imaging time “08/01 16:05:05” and the monitoring camera ID “2” is acquired. Is acquired from the image information storage unit 102, and the acquired image information is displayed. As a result, the person staying in the area ID “C3”, which is the area where the product ID “miscellaneous goods A” is displayed, can be displayed and confirmed without omission. Further, since unnecessary image information is not displayed, unnecessary confirmation time can be reduced.

  11 is merely an example, and the size and number of areas to be set may be arbitrarily selected, and adjacent areas may overlap.

(Second Embodiment)
A second embodiment will be described with reference to the accompanying drawings. With respect to the configuration example of the image extraction apparatus to which the present invention is applied, a part different from the first embodiment will be described. The other parts are the same as those in the first embodiment, so the description is omitted with reference to the above description. FIG. 12 is a block diagram illustrating a configuration of an image extraction apparatus according to the second embodiment.

  The image extraction apparatus 300 according to the present embodiment includes an image information receiving unit 101, an image information storage unit 102, a stay detection unit 103, a stay information storage unit 304, an insufficient product specifying unit 105, and an image information extraction unit 306. A display unit 107, an extracted image information storage unit 108, an inventory information acquisition unit 109, a purchase information acquisition unit 110, a sales information acquisition unit 111, a possibility determination unit 312, a display information database 120, and a monitor. An area database 130, an inventory information database 140, a purchase information database 150, and a sales information database 360 are provided.

  Here, the image information receiving unit 101, the image information storage unit 102, the stay detection unit 103, the shortage product specifying unit 105, the display unit 107, the extracted image information storage unit 108, the inventory information acquisition unit 109, The configuration and functions of the purchase information acquisition unit 110, the sales information acquisition unit 111, the display information database 120, the monitoring area database 130, the inventory information database 140, and the purchase information database 150 are the same as those in the first embodiment. Therefore, explanation is omitted.

  In addition to the data configuration in the first embodiment, the sales information database 360 stores a product ID and a sales time in association with each other. FIG. 13 is an explanatory diagram showing an example of the data configuration of the sales information database. As shown in FIG. 13, the sales time is stored corresponding to the product ID. In this way, by maintaining the time of sale for each product, information for determining whether there is a high possibility that a product lacking in inventory management has been purchased or shoplifting is likely to be obtained. Obtainable.

  The possibility determination unit 312 determines the possibility of shoplifting based on whether or not the product corresponding to the area ID in which the stay is detected is sold within a predetermined time, and stores the shoplifting possibility in the stay information storage unit 304.

  The staying information storage unit 304 stores the imaging time, the area ID, and the monitoring camera ID detected by the staying detection unit 103, and stores them with the possibility of shoplifting determined by the possibility determination unit 312 thereafter. FIG. 14 is an explanatory diagram illustrating an example of a data configuration of the staying information storage unit. The staying information storage unit 304 stores the imaging time, the area ID, the monitoring camera ID, and the possibility in association with each other.

  In addition to the function and configuration of the first embodiment, the image information extraction unit 306 extracts image information according to the possibility, for example, in descending order of possibility.

  Next, the possibility storing process by the image extracting apparatus 300 configured as described above will be described. FIG. 15 is a flowchart illustrating the possibility storage processing procedure performed by the possibility determination unit. The other processing is the same as the processing in the first embodiment described above, and a description thereof is omitted here.

  First, the possibility determination unit 312 acquires an imaging time and a region ID from the staying information storage unit 304 (step S1501). The possibility determination unit 312 acquires a product ID corresponding to the region ID from the display information database 120 (step S1502). The possibility determination unit 312 acquires the sales time corresponding to the product ID from the sales information database 360 (step S1503).

  Next, the possibility determination unit 312 determines whether or not the product has been sold within 5 minutes after staying in front of the product shelf from the imaging time and the sales time (step S1504). If it is determined that the product has been sold within 5 minutes (step S1504: Yes), “low” is stored in the possibility of the staying information storage unit 304 (step S1505). For example, when the imaging time is “08/01 19:20:55” as shown in FIG. 14 and the sales time is “08/01 19:22:45” as shown in FIG. Therefore, “low” is stored in the possibility as shown in FIG.

  When it is determined that the product is not sold within 5 minutes (step S1504: No), the possibility determination unit 312 stays in front of the product shelf from the imaging time and the sales time, and the product is sold within 10 minutes. It is determined whether or not it has been done (step S1506). If it is determined that the product has been sold within 10 minutes (step S1506: Yes), “medium” is stored in the possibility of the staying information storage unit 304 (step S1507). For example, when the imaging time is “08/01 19:48:05” as shown in FIG. 14 and the sales time is “08/01 19:54:00” as shown in FIG. 13, it exceeds 5 minutes. Therefore, “medium” is stored in the possibility as shown in FIG.

  If it is determined that the product has not been sold within 10 minutes (step S1506: No), “high” is stored in the possibility of the staying information storage unit 304 (step S1508). For example, as shown in FIG. 14, the imaging time is “08/01 19:55:40”, and the sales time is not stored within 10 minutes from the imaging time as shown in FIG. The possibility is stored as “high”.

  In this way, the imaging time of a person staying near a product and the sales time of the product are compared, and the possibility of shoplifting is determined based on the time from the imaging time to the sales time. If the product is sold immediately after imaging, the possibility of “low” is determined based on the assumption that the person who has purchased the product has been imaged. By extracting image information in consideration of the possibility of shoplifting in the image information extraction unit, it is possible to confirm from more likely image information, and therefore it is possible to shorten the confirmation time.

  Note that the time used for the determination here is not limited to 5 minutes or 10 minutes, and can be freely changed according to the size or type of business of the store and the customer base. Further, in the present embodiment, an example in which shoplifting possibility is divided into three stages is shown, but it may be divided into two stages or four or more stages.

  As described above, the present invention has been described using the first and second embodiments, but various changes or improvements can be added to the above-described embodiments. The configurations and functions described in the first embodiment and the second embodiment described above can be freely combined.

It is a block diagram which shows the structure of the image extraction apparatus concerning 1st Embodiment. It is explanatory drawing which shows an example of a data structure of a residence information storage part. It is explanatory drawing which shows an example of a data structure of an inventory information database. It is explanatory drawing which shows an example of a data structure of the purchase information database. It is explanatory drawing which shows an example of a data structure of a sales information database. It is explanatory drawing which shows an example of a data structure of a display information database. It is explanatory drawing which shows an example of a data structure of the monitoring area | region database. It is a flowchart which shows the stay information generation process procedure which an image information receiving part and a stay detection part perform. It is a flowchart which shows the image information extraction process procedure which an insufficient goods specific | specification part and an image information extraction part perform. It is explanatory drawing which shows an example of the index of the image information memorize | stored in an extracted image information storage part. It is explanatory drawing which shows an example of the goods shelf in a store, a surveillance camera, and an area | region. It is a block diagram which shows the structure of the image extraction apparatus concerning 2nd Embodiment. It is explanatory drawing which shows an example of a data structure of a sales information database. It is explanatory drawing which shows an example of a data structure of a residence information storage part. It is a flowchart which shows the possibility storage process procedure which a possibility determination part performs.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 300 Image extraction apparatus 101 Image information receiving part 102 Image information storage part 103 Stay detection part 104 304 Stay information storage part 105 Insufficient goods specific part 106 306 Image information extraction part 107 Display part 108 Extraction image information storage part 109 Inventory information acquisition part 110 Purchase Information Acquisition Unit 111 Sales Information Acquisition Unit 120 Display Information Database 130 Monitoring Area Database 140 Inventory Information Database 150 Purchase Information Database 160 360 Sales Information Database 200 Monitoring Camera 312 Possibility Determination Unit

Claims (5)

Display information storage means for storing product identification information for identifying a product and region identification information for identifying a region where a person stays to view the product in association with each other;
Image information storage means for storing the image information captured by the monitoring camera, the imaging time of the image information, and the monitoring camera identification information for identifying the monitoring camera that captured the image information;
The area identification information corresponding to the area where the person is staying from the detected image information is detected from the image information stored in the image information storage means. A stay detection means for determining
Retention information storage means for storing the imaging time corresponding to the image information detected by the stay detection means, the monitoring camera identification information, and the area identification information in association with each other;
Based on inventory information, purchase information, and sales information, a shortage product specifying unit that specifies the product identification information whose quantity is short,
Area acquisition means for acquiring the area identification information corresponding to the product identification information specified by the shortage commodity specifying means from the display information storage means;
The imaging time corresponding to the area identification information acquired by the area acquisition means and the monitoring camera identification information are acquired from the staying information storage means, and the acquired imaging time and the monitoring camera identification information are corresponded Image extracting means for extracting the image information from the image information storage means;
Display means for displaying the image information extracted by the image extraction means;
An image extracting apparatus comprising: Image information receiving means for receiving image information captured by the monitoring camera, an imaging time of the image information, and the monitoring camera identification information of the monitoring camera capturing the image information,
The image extraction according to claim 1, wherein the image information storage unit stores the image information received by the image information reception unit, the imaging time, and the monitoring camera identification information in association with each other. apparatus. Sales information storage means for storing the product identification information in association with the sales time indicating the time when the product was sold;
Based on the sales time corresponding to the product identification information specified by the shortage product specifying means and the imaging time in the region identification information corresponding to the product identification information acquired from the sales information storage means. The image extraction apparatus according to claim 1, further comprising a possibility determination unit that determines the possibility of shoplifting.   When the possibility determination means determines that the product corresponding to the product identification information specified by the shortage product specification means from the sales time and the imaging time is not sold within a predetermined time, a predetermined time The image extraction apparatus according to claim 3, wherein it is determined that the possibility of shoplifting is higher than when sold within. In the image extraction method executed by the image extraction device,
The image extraction device includes:
Display information storage means for storing product identification information for identifying a product and region identification information for identifying a region where a person stays to view the product in association with each other;
Image information storage means for storing the image information captured by the monitoring camera, the imaging time of the image information, and the monitoring camera identification information for identifying the monitoring camera capturing the image information in association with each other;
The stay detection means detects image information in which a person stays for a predetermined time or more from the image information stored in the image information storage means, and corresponds to an area in which a person stays from the detected image information A stay detection step for determining the region identification information to be performed;
A stay information storage step in which the stay information storage means stores the image capturing time and the monitoring camera identification information corresponding to the image information detected in the stay detection step in association with the area identification information in the stay information storage means. When,
A deficient product specifying means for specifying the product identification information whose quantity is insufficient based on inventory information, purchase information and sales information;
An area acquisition step in which an area acquisition means acquires the area identification information corresponding to the product identification information specified by the shortage product specification means from the display information storage means;
An image extraction unit acquires the imaging time and the monitoring camera identification information corresponding to the region identification information acquired by the region acquisition step from the staying information storage unit, and the acquired imaging time and the monitoring camera An image extraction step of extracting the image information corresponding to the identification information from the image information storage means;
A display step of displaying the image information extracted by the image extraction unit;
An image extraction method characterized by comprising:

Images