Shelving allocation data generation system, shelving allocation data generation method, and program

Abstract

To provide a shelf allocation data generation system capable of generating shelf allocation data in consideration of stockout on the basis of an image including a commodity shelf on which commodities are displayed.SOLUTION: A planogram data generation system includes image acquisition means for acquiring a first image including a product shelf on which products are displayed, specification means for specifying an out-of-stock region of the product shelf included in the first image, and generation means for determining an image in which an amount of the specified out-of-stock region satisfies a predetermined condition among a plurality of the first images as a second image and generating planogram data of the product shelf based on the second image.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a shelf planogram data generating device, a shelf planogram data generating system, a shelf planogram data generating method, and a storage medium. [Background technology]

[0002] In retail stores, shelf layout data, which is a digital representation of where each product should be displayed, is one piece of information required for store management. One technology for generating shelf layout data involves taking pictures of the shelves on which products are displayed and analyzing the images.

[0003] For example, Patent Document 1 discloses a technology that identifies each product displayed in a sales area from a digital image of the sales area, and based on the identification results, places a product master containing product information in the corresponding position on the fixture model, thereby reproducing shelf layout. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-187482 Summary of the Invention [Problem to be solved by the invention]

[0005] When generating planogram data by performing image recognition on an image including product shelves, it is desirable to generate the planogram data taking into account out-of-stock items.

[0006] Therefore, one object of the present invention is to provide a shelf layout data generation device, shelf layout data generation system, shelf layout data generation method, and storage medium that are capable of generating shelf layout data that takes stockouts into account based on an image including product shelves on which products are displayed. [Means for solving the problem]

[0007] One aspect of the shelf layout data generation device of the present invention comprises an image acquisition means for acquiring a first image including a product shelf on which products are displayed, an identification means for identifying a stockout area of ​​the product shelf included in the first image, and a generation means for determining a second image from a plurality of the first images based on the stockout area, and generating shelf layout data for the product shelf based on the second image.

[0008] One aspect of the shelf layout data generation system of the present invention comprises an image acquisition means for acquiring a first image including a product shelf on which products are displayed, an identification means for identifying a stockout area of ​​the product shelf included in the first image, and a generation means for determining a second image from a plurality of the first images based on the stockout area, and generating shelf layout data for the product shelf based on the second image.

[0009] One aspect of the shelf layout data generation method of the present invention acquires a first image including a product shelf on which products are displayed, identifies a stockout area of ​​the product shelf included in the first image, determines a second image from a plurality of first images based on the stockout area, and generates shelf layout data for the product shelf based on the second image.

[0010] One aspect of a computer-readable storage medium storing a program of the present invention causes a computer to acquire a first image including a product shelf on which products are displayed, identify a stockout area of ​​the product shelf included in the first image, determine a second image from a plurality of first images based on the stockout area, and generate shelf layout data for the product shelf based on the second image. [Effects of the Invention]

[0011] According to the present invention, it is possible to improve the accuracy of shelf planogram data generated by image recognition. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a shelf planogram data generating device according to first and second embodiments. [Figure 2]4 is a flowchart showing an example of the operation of the shelf planogram data generating device according to the first embodiment. [Figure 3] FIG. 10 is a block diagram showing an example of the configuration of a shelf planogram data generating device according to a third embodiment. [Figure 4] An example of a product database. [Figure 5] 10 is a flowchart showing an example of the operation of the shelf planogram data generating device according to the third embodiment. [Figure 6] FIG. 10 is a block diagram showing an example of the configuration of a shelf planogram data generating device according to a fourth embodiment. [Figure 7] 10 is a flowchart showing an example of the operation of the shelf planogram data generating device according to the fourth embodiment. [Figure 8] FIG. 2 is a block diagram illustrating an example of the hardware configuration of an information processing device that can constitute each shelf planogram data generation device in each embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] [First embodiment] FIG. 1 shows an example of the configuration of a shelf planogram data generating device 100 according to a first embodiment of the present invention.

[0014] The shelf planogram data generating device 100 includes an image acquiring unit 110, an identifying unit 120, and a generating unit .

[0015] The image acquisition means 110 acquires a first image including a product shelf for which planogram data is to be generated. Products are displayed on the product shelf.

[0016] The identification means 120 identifies a stockout area of ​​the product shelf included in the first image. A stockout area is an area where no products are displayed.

[0017] The generating means 130 generates shelf planogram data for the product shelf based on the second image, where the second image is an image determined from the plurality of first images based on the out-of-stock area.

[0018] 1, each block included in the shelf plan data generation device 100 shows a functional unit configuration. Therefore, each block included in the shelf plan data generation device 100 shown in Fig. 1 may be implemented in a single device, or may be implemented separately in multiple devices and configured as a shelf plan data generation system.

[0019] The operation of the shelf planogram data generation device 100 according to the first embodiment will be described with reference to the drawings. Fig. 2 is a flowchart showing a shelf planogram data generation method used in the shelf planogram data generation device 100 according to the first embodiment. The shelf planogram data generation device 100 operates according to this flowchart.

[0020] The image acquisition means 110 acquires a first image including a product shelf for which planogram data is to be generated (step S10), and the identification means 120 identifies a stockout area of ​​the product shelf included in the first image (step S11). Then, a second image is determined from the plurality of first images based on the stockout area (step S12). However, the determination of the second image may be performed by the identification means 120 or the generation means 130. Alternatively, the second image may be determined by a determination means (not shown). Thereafter, the generation means 130 generates planogram data of the product shelf based on the second image (step S13), and the planogram data generation device ends the processing.

[0021] The shelf planogram data generation device according to the first embodiment generates shelf planogram data for a product shelf from an image determined based on a stockout area, thereby making it possible to generate shelf planogram data that takes stockouts into consideration.

[0022] [Second embodiment] In the second embodiment, the shelf planogram data generating device 100 of the present invention will be described in more detail. Hereinafter, the same configurations and operations as those in the first embodiment will be assigned the same reference numerals, and explanations of overlapping parts will be omitted.

[0023] An example of the configuration of a shelf planogram data generating device 100 according to the second embodiment of the present invention is the same as that shown in FIG.

[0024] The image acquisition means 110 acquires a first image including a product shelf for which planogram data is to be generated. Examples of images acquired by the image acquisition means 110 include, but are not limited to, images taken by a terminal carried by a store clerk or customer, images taken by a robot patrolling in the store, and images taken by a camera in the store.

[0025] The image acquisition means 110 may acquire the image directly from the terminal that captured the image, or may acquire the image via a network, etc. Alternatively, the image may be acquired from a storage device such as a cloud.

[0026] The image acquisition means 110 may acquire an image each time an image is taken, or may acquire a plurality of images at once.

[0027] The identification means 120 processes the first image to identify an area where no products are displayed (out-of-stock area). Furthermore, the identification means 120 may identify an area where products are displayed (product area). The identification means 120 may, for example, use existing image recognition technology to recognize objects in the first image and identify an area where the object cannot be recognized as an out-of-stock area. In this case, the identification means 120 is only required to recognize the presence or absence of an object, and does not need to recognize which product the recognized object is. Furthermore, the identification means 120 may, for example, store the background of a product shelf and identify an area where the background cannot be recognized as a product area and an area where the background can be recognized as an out-of-stock area.

[0028] Furthermore, the identification means 120 may identify how many types of products are displayed in the identified out-of-stock area. For example, the identification means 120 may recognize shelf tags in addition to the out-of-stock area. In this case, it is possible to identify how many types of products are displayed in the identified out-of-stock area based on how many shelf tags are attached to the out-of-stock area. As an example, if three shelf tags are attached to one identified out-of-stock area, it is possible to identify that three types of products are displayed in that out-of-stock area.

[0029] As another example, the identification unit 120 may identify how many types of products will be displayed in the identified out-of-stock area based on the display status around the out-of-stock area. In this case, the identification unit 120 recognizes the products surrounding the identified out-of-stock area and can identify how many types of products will be displayed in the identified out-of-stock area based on the display width of the surrounding products. Specifically, if the display width of the surrounding products is constant, the number of types of products displayed in the identified out-of-stock area can be identified by dividing the width of the identified out-of-stock area by the display width of the surrounding products. As an example, consider a case where the identified out-of-stock area is 90 cm and the display width of the surrounding products is constant at 30 cm. In this case, by dividing the width of the identified out-of-stock area (90 cm) by the display width of the surrounding products (30 cm), it can be identified that three types of products will be displayed in the identified out-of-stock area. Furthermore, if the display width of the surrounding products is not constant, the number of types of products displayed in the identified out-of-stock area can be identified by dividing the width of the identified out-of-stock area by the average display width of the surrounding products. For example, consider a case where the identified out-of-stock area is 80 cm and the display widths of the surrounding products are 15 cm, 17 cm, 23 cm, and 25 cm. In this case, by dividing the width of the identified out-of-stock area (80 cm) by the average display width of the surrounding products (20 cm), it can be determined that four types of products are displayed in the identified out-of-stock area.

[0030] Alternatively, the identification means 120 may identify how many types of products are displayed in the identified out-of-stock area based on the display width of the identified surrounding products. Specifically, the number of types of products displayed in the identified out-of-stock area can be identified by dividing the width of the identified out-of-stock area by the display width of the identified surrounding products. Here, the identified surrounding product is one surrounding product that is in a predetermined positional relationship with the identified out-of-stock area. The identified surrounding product may be, for example, a product adjacent to the identified out-of-stock area or a product above or below the identified out-of-stock area, but is not limited to these.

[0031] The generation means 130 generates product shelf planogram data based on the second image. The generation means 130 recognizes products included in the second image using, for example, an existing image recognition technology, and generates the product shelf planogram data.

[0032] Here, the second image is an image determined from the plurality of first images based on the missing-item areas. The determination of the second image may be performed by the identification means 120 or the generation means 130. Alternatively, the second image may be determined by a determination means (not shown). For example, the second image may be determined as an image among the plurality of first images in which the missing-item areas are less than a predetermined threshold. Specifically, the missing-item areas may be determined to be less than the predetermined threshold when the area of ​​the missing-item areas is less than a predetermined threshold or when the number of pixels in the missing-item areas is less than a predetermined threshold. Alternatively, the missing-item areas may be determined to be less than the predetermined threshold when the number of missing-item areas is less than a predetermined threshold. The number of missing-item areas may be determined as the number of types of products displayed in the missing-item areas. The number of types of products displayed in the missing-item areas can be determined using the method described above. Furthermore, if there are multiple images in which the missing-item areas are less than the predetermined threshold, the image captured most recently among the images may be determined as the second image.

[0033] As another example of a method for determining the second image, the image with the fewest missing parts among the plurality of first images may be selected as the second image. Specifically, the image with the smallest missing part area among the plurality of first images, or the image with the smallest number of pixels in the missing part area, may be selected as the second image. Also, the image with the fewest missing parts among the plurality of first images may be selected as the second image. However, these are merely examples, and the method for determining the second image is not limited to these examples.

[0034] The operation of the shelf planogram data generating device 100 according to the second embodiment of the present invention is the same as that shown in FIG.

[0035] The shelf planogram data generating device according to the second embodiment generates shelf planogram data for a product shelf from an image determined based on a stockout area, thereby making it possible to generate shelf planogram data that takes stockouts into consideration.

[0036] Furthermore, by selecting an image among the multiple first images with fewer out-of-stock areas than a predetermined threshold as the second image, it is possible to generate planogram data using images with fewer out-of-stock areas. This improves the accuracy of the generated planogram data. Furthermore, because the out-of-stock areas included in the second image are always below the predetermined threshold, it is possible to maintain a high level of accuracy in the generated planogram data.

[0037] Furthermore, by selecting the image with the fewest out-of-stock areas from among the multiple first images as the second image, it is possible to generate planogram data using images with few out-of-stock areas. This improves the accuracy of the generated planogram data. Furthermore, by using the image with the fewest out-of-stock areas from among the first images, it is possible to generate planogram data with the highest accuracy.

[0038] [Third embodiment] The shelf planogram data generation device 200 of the third embodiment differs from the shelf planogram data generation device 100 of the second embodiment in that it includes estimation means 140. Hereinafter, the same configurations and operations as those of the second embodiment will be assigned the same reference numerals, and explanations of overlapping parts will be omitted.

[0039] 3 shows an example of the configuration of a shelf planogram data generation device 200 according to the third embodiment of the present invention. The shelf planogram data generation device 200 includes an image acquisition unit 110, an identification unit 120, a generation unit 130, and an estimation unit 140.

[0040] The estimation means 140 identifies candidate out-of-stock items by comparing the displayed items included in the second image with the available items that are handled in the store. Specifically, the estimation means 140 compares the displayed items with the available items, and identifies items that are included in the available items but not in the displayed items as candidate out-of-stock items.

[0041] The displayed products are products displayed on the product shelves included in the second image. The displayed products are recognized, for example, using existing image recognition technology. The recognition of the displayed products may be performed by the identification means 120 or the estimation means 140. Alternatively, the recognition of the displayed products may be performed by a recognition means (not shown).

[0042] The merchandise items are those sold in the store. Information about the merchandise items is stored in a storage means (not shown) as a merchandise item database. An example of the merchandise item database 300 is shown in FIG. 4. The merchandise item database 300 includes product names, product IDs (Identifications), display areas, sales quantities, sizes, weights, prices, and sales periods. The information included in the merchandise item database 300 shown in FIG. 4 is an example and is not limited to this. For example, the merchandise item database 300 may further store merchandise categories.

[0043] The product name is the name of each product. The product ID is identification information assigned to each product so that it can be identified. The product ID can be any character string, number sequence, or combination of letters and numbers.

[0044] The display area indicates the area where each product is displayed. The display area may be a product shelf ID indicating the product shelf where the product is displayed. Alternatively, the display area may be a designation indicating a sales area for each product category, such as a "drinks area" or a "snacks area."

[0045] The sales figures indicate the number of sales of each product. The sales figures stored in the product database may be the number of sales on a given day or the cumulative number of sales over a given period.

[0046] The size indicates the size of each product. The size includes at least one of the width, height, and depth of the product. Alternatively, the front area of ​​the product when it is displayed may be stored as the size. Alternatively, the volume of the product may be stored as the size.

[0047] The weight indicates the weight of each product. The price indicates the selling price of each product.

[0048] The sales period indicates a period during which each product is sold at a store. The sales period may be stored as either the start time or the end time of the product, or may be stored as the period from the start time to the end time of the product.

[0049] The estimation means 140 may compare the items with all the available items contained in the available item database 300, or may compare the items with a portion of the available items in the available item database 300. For example, among the items contained in the available item database 300, items whose display area is in the area included in the second image may be compared with the displayed items included in the second image. Specifically, the displayed items included in the second image may be compared with the available items whose display area is in the area included in the second image, and items that are included in the available items but not in the displayed items may be identified as candidate out-of-stock items.

[0050] Furthermore, the estimation means 140 may compare products in the same category as the products included in the second image, among the products included in the handled product database 300, with the displayed products included in the second image. Specifically, the estimation means 140 may compare the displayed products included in the second image with the handled products in the same category as the category of the displayed products included in the second image, and identify products that are included in the handled products but not included in the displayed products as candidate out-of-stock products.

[0051] Furthermore, the estimation means 140 may compare products included in the handled product database 300 that are within the handling period at the time of comparison with the displayed products included in the second image. Specifically, the estimation means 140 may compare the displayed products included in the second image with the handled products that are within the handling period at the time of comparison, and identify products that are included in the handled products but not included in the displayed products as candidate out-of-stock products. Additionally, the estimation means 140 may compare products included in the handled product database 300 that have a price range that matches the price range of the products displayed in the area included in the second image with the displayed products included in the second image. Specifically, the estimation means 140 may compare the displayed products included in the second image with the handled products that have a price range that matches the price range of the products displayed in the area included in the second image, and identify products that are included in the handled products but not included in the displayed products as candidate out-of-stock products.

[0052] Then, the estimation means 140 estimates out-of-stock items that are displayed in out-of-stock areas from among the out-of-stock item candidates. For example, if there is one out-of-stock item candidate, the out-of-stock item candidate is estimated to be the out-of-stock item. Also, if the number of out-of-stock areas matches the number of out-of-stock item candidates, the out-of-stock item candidate may be estimated to be the out-of-stock item. Furthermore, if there are multiple out-of-stock item candidates, the out-of-stock item may be estimated using the method described below.

[0053] Furthermore, if a candidate out-of-stock item cannot be identified, it is highly likely that the out-of-stock area is part of the display area of ​​an item displayed adjacent to the out-of-stock area. Therefore, if a candidate out-of-stock item cannot be identified, the estimation means 140 may estimate an item displayed adjacent to the out-of-stock area as an out-of-stock item for which the out-of-stock area is to be displayed. Alternatively, the estimation means 140 may estimate an item displayed above or below the out-of-stock area as an out-of-stock item for which the out-of-stock area is to be displayed. Specifically, the displayed items may be compared with the available items, and if there is no item that is included in the available items but not included in the displayed items, the item displayed adjacent to the out-of-stock area may be estimated as an out-of-stock item for which the out-of-stock area is to be displayed.

[0054] The generating means 130 generates shelf planogram data for the product shelves based on the second image and the out-of-stock products estimated by the estimating means 140.

[0055] The operation of the shelf planogram data generating device according to the third embodiment will be described with reference to the drawings. Fig. 5 is a flowchart showing an example of the operation of the shelf planogram data generating device according to the third embodiment.

[0056] Steps S10 to S12 are the same as those in Fig. 2, and therefore will not be described further. The estimation means 140 identifies candidate out-of-stock items by comparing the displayed items included in the second image with the available items (step S31). Then, from among the candidate out-of-stock items, it estimates out-of-stock items whose display location will be the out-of-stock area (step S32). Thereafter, the generation means 130 generates shelf planogram data for the product shelf based on the second image and the out-of-stock items estimated by the estimation means 140 (step S33).

[0057] The shelf planogram data generating device according to the third embodiment generates shelf planogram data for a product shelf from an image determined based on a stockout area, thereby making it possible to generate shelf planogram data that takes stockouts into consideration.

[0058] Furthermore, out-of-stock items are estimated, and shelf planogram data for the product shelves is generated based on the images and the estimated out-of-stock items. This makes it possible to generate shelf planogram data for out-of-stock areas as well, making it possible to generate shelf planogram data that takes out-of-stock items into consideration. Furthermore, even when out-of-stock areas exist, shelf planogram data can be generated with high accuracy.

[0059] [Fourth embodiment] The shelf planogram data generation device 400 of the fourth embodiment differs from the shelf planogram data generation device 200 of the third embodiment in that it includes a product information acquisition means 150. Hereinafter, the same configurations and operations as those of the third embodiment will be assigned the same reference numerals, and explanations of overlapping parts will be omitted.

[0060] 6 shows an example of the configuration of a shelf planogram data generation device 400 according to the fourth embodiment of the present invention. The shelf planogram data generation device 400 includes an image acquisition unit 110, an identification unit 120, a generation unit 130, an estimation unit 140, and a product information acquisition unit 150.

[0061] The estimation means 140 identifies candidate out-of-stock items by comparing the displayed items included in the second image with the available items handled in the store. The candidate out-of-stock items may be identified using the method described in the third embodiment. Furthermore, items displayed adjacent to the out-of-stock area may also be added to the candidate out-of-stock items. Furthermore, items displayed above or below the out-of-stock area may also be added to the candidate out-of-stock items. Then, the product information acquisition means 150 acquires product information including at least one of the sales quantity, size, weight, and price of the candidate out-of-stock items. For example, the product information acquisition means 150 acquires product information from the available product database 300. Alternatively, the product information may be acquired from a POS (Point of Sales) terminal or a store computer.

[0062] The estimation means 140 estimates out-of-stock products that will be displayed in the out-of-stock area from among the out-of-stock product candidates based on the product information acquired by the product information acquisition means 150.

[0063] A specific example of the estimation of out-of-stock items by the estimation means 140 will be described.

[0064] <Example 1 of missing item estimation> The product information acquisition means 150 acquires the sales volume of candidate out-of-stock products as product information. The product information acquisition means 150 may further acquire the sales volume of all products available, or may acquire the sales volume of some of the products available. If the sales volume is high, there is a possibility that the product has been removed from the shelves, and therefore the possibility of being out of stock increases. Therefore, the estimation means 140 estimates candidate out-of-stock products with high sales volumes as out-of-stock products.

[0065] As an example, consider a case where there is one out-of-stock area and two out-of-stock product candidates (out-of-stock product candidate A1, out-of-stock product candidate B1). However, it is assumed that one type of product is displayed in this out-of-stock area. Also, it is assumed that the sales numbers of each out-of-stock product candidate acquired by the product information acquisition means 150 are as follows: Out-of-stock product candidate A1: 15 units Out-of-stock item candidate B1: 7 units At this time, the estimation means 140 estimates the out-of-stock product candidate A1 as the out-of-stock product.

[0066] In addition, when there is one out-of-stock area and three or more out-of-stock product candidates, the estimation means 140 estimates the out-of-stock product candidate with the highest sales volume as the out-of-stock product. Furthermore, when there are multiple out-of-stock areas, the out-of-stock product candidates may be estimated in descending order of sales volume.

[0067] <Example 2 of estimating missing items> A specific example will be described for the case where there are two or more out-of-stock areas. In this case, the estimation means 140 estimates the out-of-stock product based on the positional relationship of the out-of-stock areas.

[0068] As an example, a case will be described in which the product information acquisition means 150 acquires the weight of candidate out-of-stock products as product information. The product information acquisition means 150 may further acquire the weight of all available products, or may acquire the weight of some of the available products. Heavier products are more likely to be placed on the lower shelves of a product shelf. Therefore, the estimation means 140 estimates out-of-stock products based on the relative positions of the out-of-stock areas and the weight of candidate out-of-stock products.

[0069] As a more specific example, consider a case where there are two out-of-stock areas (out-of-stock area X1, out-of-stock area Y1) and two out-of-stock product candidates (out-of-stock product candidate A2, out-of-stock product candidate B2). However, assume that each out-of-stock area displays one type of product. Assume that out-of-stock area X1 is located above out-of-stock area Y1. Also, assume that the weights of each out-of-stock product candidate acquired by product information acquisition means 150 are as follows: Out-of-stock product candidate A2: 1kg Out-of-stock item candidate B2: 0.5kg At this time, the estimation means 140 estimates the out-of-stock product that will be displayed in the out-of-stock area Y1 as the out-of-stock product candidate A2, and the out-of-stock product that will be displayed in the out-of-stock area X1 as the out-of-stock product candidate B2.

[0070] The estimation means 140 may also estimate out-of-stock products in descending order of weight so that the out-of-stock product candidates correspond to out-of-stock areas closer to the bottom shelves of the product shelves.

[0071] As another example, a case will be described in which the product information acquisition means 150 acquires the sales numbers of candidate out-of-stock products as product information. The product information acquisition means 150 may further acquire the sales numbers of all available products, or may acquire the sales numbers of some available products. The sales numbers may vary depending on the display position of the products. As an example, products displayed at a height that allows customers to easily pick them up may have higher sales numbers. Therefore, the estimation means 140 estimates the out-of-stock products based on the positional relationship of the out-of-stock areas and the sales numbers of candidate out-of-stock products.

[0072] As a more specific example, consider a case where there are two out-of-stock areas (out-of-stock area X2, out-of-stock area Y2) and two out-of-stock product candidates (out-of-stock product candidate A3, out-of-stock product candidate B3). However, assume that each out-of-stock area displays one type of product. Out-of-stock area X2 is an area where sales volume is high. Areas where sales volume is high may be set in advance. Also, assume that the sales volume of each out-of-stock product candidate acquired by product information acquisition means 150 is as follows: Missing product candidate A3: 15 units Out-of-stock item candidate B3: 7 units At this time, the estimation means 140 estimates the out-of-stock product to be displayed in the out-of-stock area X2 as the out-of-stock product candidate A3, and the out-of-stock product to be displayed in the out-of-stock area Y2 as the out-of-stock product candidate B3.

[0073] <Example 3 of estimating missing items> A specific example will be described in which the estimation means 140 estimates out-of-stock items based on the size of the out-of-stock area.

[0074] As an example, a case will be described in which the product information acquisition means 150 acquires the size of candidate out-of-stock products as product information. The product information acquisition means 150 may further acquire the sizes of all available products, or may acquire the sizes of only some of the available products. Large products cannot be displayed in a small out-of-stock area. Therefore, the estimation means 140 estimates the out-of-stock products based on the size of the out-of-stock area and the size of the candidate out-of-stock products.

[0075] As a more specific example, consider a case where there is one out-of-stock area and two out-of-stock product candidates (out-of-stock product candidate A4, out-of-stock product candidate B4). However, assume that one type of product is displayed in this out-of-stock area, and that the area of ​​this out-of-stock area is 10 cm. Also, assume that the size of each out-of-stock product candidate acquired by product information acquisition means 150 is as follows: Missing product candidate A4:15cm Missing product candidate B4:5cm At this time, the estimation means 140 estimates the out-of-stock product candidate B4 as the out-of-stock product.

[0076] Furthermore, when there are multiple out-of-stock areas, the estimation means 140 may estimate out-of-stock items in descending order of size of out-of-stock item candidates so that they correspond to larger out-of-stock areas.

[0077] As another example, a case will be described in which the product information acquisition means 150 acquires the sales numbers of candidate out-of-stock products as product information. The product information acquisition means 150 may further acquire the sales numbers of all available products, or may acquire the sales numbers of some available products. The size of the display area may change depending on the sales numbers of the products. As an example, products that sell well may be displayed in a large area. Therefore, the estimation means 140 estimates the out-of-stock products based on the size of the out-of-stock area and the sales numbers of candidate out-of-stock products.

[0078] As a more specific example, consider a case where there are two out-of-stock areas (out-of-stock area X3, out-of-stock area Y3) and two out-of-stock product candidates (out-of-stock product candidate A5, out-of-stock product candidate B5). However, assume that each out-of-stock area displays one type of product. Assume that out-of-stock area X3 is a larger out-of-stock area than out-of-stock area Y3. Also, assume that the sales figures for each out-of-stock product candidate acquired by product information acquisition means 150 are as follows: Out-of-stock product candidate A5: 15 units Out-of-stock item candidate B5: 7 units At this time, the estimation means 140 estimates the out-of-stock product to be displayed in the out-of-stock area X3 as out-of-stock product candidate A5, and the out-of-stock product to be displayed in the out-of-stock area Y3 as out-of-stock product candidate B5.

[0079] While specific examples have been described above for estimating out-of-stock items by the estimation means 140, specific examples 1 to 3 are merely examples and are not intended to be limiting. The above specific examples may also be combined.

[0080] An example of combining the above-mentioned specific examples will be described. For example, consider a case where there are N out-of-stock areas and M out-of-stock item candidates. Here, N and M are natural numbers, and N is a number smaller than M. First, the estimation means 140 identifies the N out-of-stock item candidates in descending order of sales volume based on the sales volume of each out-of-stock item candidate acquired by the product information acquisition means 150. Then, based on the positional relationship of the out-of-stock areas and / or the size of the out-of-stock areas, the out-of-stock items are estimated from the N out-of-stock item candidates. In this way, by combining the above-mentioned specific examples, it is possible to accurately estimate the out-of-stock items that will be displayed in each out-of-stock area.

[0081] Next, another example combining the above specific examples will be described. First, the estimation means 140 identifies out-of-stock product candidates that are smaller than the size of the out-of-stock area based on the size of each out-of-stock product candidate acquired by the product information acquisition means 150. Then, based on the sales volume of the out-of-stock product candidates and / or the positional relationship of the out-of-stock area, the estimation means 140 estimates out-of-stock products from the out-of-stock product candidates that are smaller than the size of the out-of-stock area. Even in this example, it is possible to accurately estimate out-of-stock products that will be displayed in each out-of-stock area.

[0082] Furthermore, the product information acquired by the product information acquisition means 150 may be product information for all available products, or product information for some available products. For example, product information for available products whose display area is the area included in the second image may be acquired, or product information for available products in the same category as the product category included in the second image may be acquired. Also, product information for available products whose sales period includes the time when out-of-stock products are estimated may be acquired. Also, product information for available products whose prices match the price range of the products displayed in the area included in the second image may be acquired. In this way, when product information for some available products is acquired, less processing is required than when product information for all available products is acquired.

[0083] The operation of the shelf planogram data generating device according to the fourth embodiment will be described with reference to the drawings. Fig. 7 is a flowchart showing an example of the operation of the shelf planogram data generating device according to the fourth embodiment.

[0084] Steps S10 to S12 are the same as those in FIG. 2, and steps S31 and S33 are the same as those in FIG. 5, so their explanations are omitted. After step S31, product information acquisition means 150 acquires product information of out-of-stock product candidates (step S41). Then, based on the product information, out-of-stock products that will be displayed in the out-of-stock area are estimated from the out-of-stock product candidates (step S42). Then, the process proceeds to step S33.

[0085] The shelf planogram data generating device according to the fourth embodiment generates shelf planogram data for a product shelf from an image determined based on a stockout area, thereby making it possible to generate shelf planogram data that takes stockouts into consideration.

[0086] Furthermore, out-of-stock items are estimated, and shelf planogram data for the product shelves is generated based on the images and the estimated out-of-stock items. This makes it possible to generate shelf planogram data for out-of-stock areas as well, making it possible to generate shelf planogram data that takes out-of-stock items into consideration. Furthermore, even when out-of-stock areas exist, shelf planogram data can be generated with high accuracy.

[0087] Furthermore, product information including at least one of the sales volume, size, weight, and price of candidate out-of-stock items is acquired, and the out-of-stock items are estimated based on the product information. This makes it possible to generate planogram data that takes out-of-stocks into account. This also improves the accuracy of product estimation of out-of-stock areas, making it possible to generate planogram data with high accuracy.

[0088] Furthermore, among the candidate out-of-stock items, the item with the highest sales volume is estimated as the out-of-stock item. This makes it possible to generate planogram data that takes out-of-stocks into account. This also improves the accuracy of product estimation of out-of-stock areas, allowing for more accurate planogram data to be generated.

[0089] Furthermore, when there are multiple out-of-stock areas, the out-of-stock items are estimated based on the relative positions of the out-of-stock areas. This makes it possible to generate planogram data that takes out-of-stock items into account. Even when there are multiple out-of-stock areas, the accuracy of product estimation for the out-of-stock areas is improved, making it possible to generate planogram data with high accuracy.

[0090] Furthermore, among the candidate out-of-stock items, the item with the heaviest weight is estimated as the out-of-stock item in the out-of-stock area closest to the bottom shelf. This makes it possible to generate planogram data that takes out-of-stock items into account. This also improves the accuracy of estimating items in out-of-stock areas, allowing for more accurate planogram data to be generated.

[0091] Furthermore, out-of-stock items are estimated based on the size of the out-of-stock area. This makes it possible to generate planogram data that takes stockouts into account. Even when there are multiple out-of-stock areas, the accuracy of product estimation for the out-of-stock area is improved, making it possible to generate planogram data with high accuracy.

[0092] Furthermore, products whose size is equal to or smaller than the size of the out-of-stock area are estimated to be out-of-stock items. This allows for the generation of planogram data that takes out-of-stock items into consideration. This also improves the accuracy of product estimation of out-of-stock areas, allowing for the generation of planogram data with high accuracy.

[0093] In each embodiment of the present disclosure, each shelf planogram data generation device is realized by a combination of hardware and a program. Fig. 8 is a block diagram illustrating an example of the hardware configuration of an information processing device 1000 that can constitute each shelf planogram data generation device in each embodiment.

[0094] The information processing device 1000 includes a processor 1001, a memory 1002, a network interface 1003, an input / output interface 1004, and a storage device 1005, and the components of the information processing device 1000 are communicably connected via a bus 1006.

[0095] The processor 1001 is realized by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.

[0096] The memory 1002 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0097] The network interface 1003 is an interface for connecting to a network, which may be a LAN (Local Area Network) or a WAN (Wide Area Network).

[0098] The input / output interface 1004 is an interface for connecting to various input / output devices.

[0099] The storage device 1005 is an auxiliary storage device realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a memory card, a ROM (Read Only Memory), etc. The storage device 1005 may store a program for realizing the function of each of the shelf planogram data generation devices in each embodiment.

[0100] The processor 1001 loads a program stored in a storage device 1005 into a memory 1002 and executes it to realize the functions of each shelf planogram generation device. The program may also be supplied from a network via a network interface 1003. Alternatively, the program may be stored in advance in a storage medium (not shown) and supplied by reading the program.

[0101] Furthermore, this program can display the processing results, including intermediate states, at each stage as necessary on a display device, or can communicate with the outside via the network interface 1003. Furthermore, this program can be recorded on a computer-readable (non-transitive) recording medium.

[0102] The present disclosure is not limited to the above-described embodiments, and various modifications are possible. Embodiments obtained by appropriately combining the configurations, operations, and processes disclosed in different embodiments are also included in the technical scope of the present disclosure.

[0103] The present disclosure is not limited to the above-described embodiments, and various aspects that can be understood by a person skilled in the art can be applied to the present disclosure within the scope of the present disclosure.

[0104] [Appendix 1] image acquisition means for acquiring a first image including a product shelf on which products are displayed; An identification means for identifying a shortage area of ​​the product shelf included in the first image; a generating means for determining a second image from the plurality of first images based on the out-of-stock area, and generating shelf planogram data for the product shelf based on the second image; A shelf layout data generating device comprising: [Appendix 2] The second image is an image among the plurality of first images in which the number of missing areas is less than a threshold value. 2. The shelf planogram data generating device according to claim 1. [Appendix 3] The second image is an image having the fewest missing areas among the plurality of first images. 2. The shelf planogram data generating device according to claim 1. [Appendix 4] The method further comprises: an estimation means for, when there is a shortage area in which the product is not displayed on the product shelf included in the second image, identifying a shortage product candidate by comparing the displayed products included in the second image with the products handled in the store, and estimating a shortage product from the shortage product candidate that will be displayed in the shortage area; the generating means generates planogram data for the product shelf based on the second image and the estimated out-of-stock product. 4. The shelf planogram data generating device according to claim 2 or 3. [Appendix 5] The system further includes a product information acquisition unit for acquiring product information including at least one of the sales quantity, size, weight, and price of the out-of-stock product candidate. The estimation means further estimates the out-of-stock items based on the item information. 5. The shelf planogram data generating device according to claim 4. [Appendix 6] the product information acquisition means acquires the sales volume as the product information, The estimation means estimates the product with the highest sales volume among the candidate out-of-stock products as the out-of-stock product. 6. The shelf planogram data generating device according to claim 5. [Appendix 7] When there are a plurality of out-of-stock areas on the product shelf included in the second image, The estimation means further estimates the out-of-stock product based on a positional relationship of the out-of-stock area. 7. The shelf planogram data generating device according to claim 5 or 6. [Appendix 8] the product information acquisition means acquires the weight as the product information, the estimation means estimates the product with the largest weight among the out-of-stock product candidates as the out-of-stock product in the out-of-stock area closest to a lower shelf of the product shelf; 8. A shelf plan data generating device according to any one of appendices 5 to 7. [Appendix 9] The estimation means further estimates the out-of-stock product based on the size of the out-of-stock area. 9. A shelf plan data generating device according to any one of appendices 5 to 8. [Appendix 10] the product information acquisition means acquires the size as the product information, The estimation means estimates a product whose size is equal to or smaller than the area of ​​the out-of-stock area as the out-of-stock product. 10. The shelf planogram data generating device according to any one of appendices 5 to 9. [Appendix 11] image acquisition means for acquiring a first image including a product shelf on which products are displayed; An identification means for identifying a shortage area of ​​the product shelf included in the first image; a generating means for determining a second image from the plurality of first images based on the out-of-stock area, and generating shelf planogram data for the product shelf based on the second image; A shelf allocation data generation system comprising: [Appendix 12] acquiring a first image including a product shelf on which products are displayed; Identifying a shortage area of ​​the product shelf included in the first image; determining a second image from the plurality of first images based on the out-of-stock area, and generating shelf planogram data for the product shelf based on the second image; How to generate shelf planogram data. [Appendix 13] acquiring a first image including a product shelf on which products are displayed; Identifying a shortage area of ​​the product shelf included in the first image; determining a second image from the plurality of first images based on the out-of-stock area, and generating shelf planogram data for the product shelf based on the second image; A computer-readable storage medium that stores a program that causes a computer to execute the following: [Explanation of symbols]

[0105] 100, 200, 400 shelf layout data generator 110 Image acquisition means 120 Specific means 130 Generation means 140 Estimation means 150 Product information acquisition means 300 product database 1000 Information Processing Device 1001 processor 1002 memory 1003 Network Interface 1004 Input / Output Interface 1005 Storage Devices 1006 Bus

Claims

1. an image acquisition means for acquiring a first image including a product shelf on which products are displayed; an identification means for identifying a shortage area of ​​the product shelf included in the first image; a generating means for determining, as a second image, an image among the plurality of first images in which the amount of the identified out-of-stock area satisfies a predetermined condition, and generating shelf planogram data for the product shelf based on the second image; A shelf allocation data generation system comprising:

2. the predetermined condition is a condition for determining, as the second image, an image in which the amount of missing areas is less than a threshold value or an image in which the amount of missing areas is the smallest. The shelf planogram data generation system according to claim 1 .

3. The amount of the missing area is any one of the area, the number of pixels, and the number of missing areas.

3. The shelf planogram data generating system according to claim 1 or 2.

4. The method further comprises: an estimation means for, when there is a shortage area in which the product is not displayed on the product shelf included in the second image, identifying a shortage product candidate by comparing the displayed products included in the second image with the products handled in the store, and estimating a shortage product from the shortage product candidate that will be displayed in the shortage area; the generating means generates shelf planogram data for the product shelf based on the second image and the estimated out-of-stock product.

3. The shelf planogram data generating system according to claim 1 or 2.

5. The system further includes a product information acquisition unit for acquiring product information including at least one of the sales quantity, size, weight, and price of the out-of-stock product candidate. The estimation means further estimates the out-of-stock items based on the item information. The shelf planogram data generating system according to claim 4.

6. the product information acquisition means acquires the weight as the product information, the estimation means estimates the product with the largest weight among the out-of-stock product candidates as the out-of-stock product in the out-of-stock area closest to a lower shelf of the product shelf; The shelf planogram data generating system according to claim 5.

7. The estimation means further estimates the out-of-stock product based on the size of the out-of-stock area. The shelf planogram data generating system according to claim 5.

8. the product information acquisition means acquires the size as the product information, The estimation means estimates a product whose size is equal to or smaller than the area of ​​the out-of-stock area as the out-of-stock product. The shelf planogram data generating system according to claim 5.

9. The information processing device acquiring a first image including a product shelf on which products are displayed; Identifying a shortage area of ​​the product shelf included in the first image; determining, as a second image, an image in which the amount of the identified out-of-stock area satisfies a predetermined condition among the plurality of first images, and generating shelf planogram data for the product shelf based on the second image; How to generate shelf planogram data.

10. acquiring a first image including a product shelf on which products are displayed; Identifying a shortage area of ​​the product shelf included in the first image; determining, as a second image, an image in which the amount of the identified out-of-stock area satisfies a predetermined condition among the plurality of first images, and generating shelf planogram data for the product shelf based on the second image; A program that makes a computer do something.

Images