Inventory management device and inventory management method

Abstract

This inventory management device comprises: an analysis unit that recognizes an article on the basis of a video in which a user using the article and the article are captured, and generates information about the article; an estimation unit that estimates the satisfaction level of the user on the basis of the image of the user in the video and / or the user's voice data; a candidate determination unit that determines whether or not to replenish the article on the basis of the information about the article, and also determines, when a determination is made that the article is to be replenished, a candidate for the article to be replenished on the basis of the satisfaction level of the user; and a replenishment assistance unit that notifies the user of the candidate for the article to be replenished when a determination is made that the article is to be replenished.

Description

Inventory management device and inventory management method

[0001] The present invention relates to an inventory management device and an inventory management method.

[0002] Patent Document 1 discloses an inventory management device for checking whether an item is out of stock. In Patent Document 1, a marker printed with ink that has a high reflectivity for light in the infrared region is placed under the item (object to be managed). If the item is on a display shelf, a camera captures an image of the item. On the other hand, if the item is not on a display shelf, the camera captures an image of a marker that has been placed under the item in advance. The imaged marker is identified to determine whether the item is in stock.

[0003] Japanese Patent Application Publication No. 11-281754

[0004] The inventory management device disclosed in Patent Document 1 generates information about out-of-stock items based on the presence or absence of items on display shelves. Furthermore, out-of-stock items are replenished based solely on the information about out-of-stock items. This creates a problem in that it is not possible to prevent items that provide low user satisfaction from being replenished.

[0005] In view of the above circumstances, an object of the present invention is to provide an inventory management device and an inventory management method that can reduce the possibility of replenishing items that do not satisfy users.

[0006] One aspect of the present invention is an inventory management device that includes an analysis unit that recognizes an item based on a video of a user using the item and the item, and generates information about the item; an estimation unit that estimates the user's satisfaction level based on at least one of an image of the user in the video and voice data of the user; a candidate determination unit that determines whether to replenish the item based on the information about the item, and if it is determined that the item should be replenished, determines candidates for the item to be replenished based on the user's satisfaction level; and a replenishment support unit that, if it is determined that the item should be replenished, notifies the user of the candidates for the item to be replenished.

[0007] One aspect of the present invention is an inventory management method executed by an inventory management device, the inventory management method including the steps of: recognizing an item based on a video of a user using the item and the item, and generating information about the item; estimating the user's satisfaction level based on at least one of an image of the user in the video and voice data of the user; determining whether to replenish the item based on the information about the item, and if it is determined that the item should be replenished, determining candidate items to be replenished based on the user's satisfaction level; and if it is determined that the item should be replenished, notifying the user of the candidate items to be replenished.

[0008] The present invention can reduce the possibility of replenishing items that do not satisfy users.

[0009] It is a diagram showing an example of the configuration of an inventory management system in an embodiment. It is a diagram showing an example of inventory management data in an embodiment. It is a flowchart showing an example of the operation of an inventory management device in an embodiment. It is a diagram showing an example of the hardware configuration of an inventory management device in an embodiment.

[0010] An embodiment of the present invention will be described in detail with reference to the drawings. Fig. 1 is a diagram showing an example of the configuration of an inventory management system 1 (inventory replenishment support system) according to the embodiment. The inventory management system 1 includes an inventory management device 2, a communication line 3, one or more cameras 4, one or more microphones 5, and one or more sensors 6.

[0011] The inventory management device 2 includes an acquisition unit 21, a storage device 22, a memory 23, a learning device 24, an estimation device 25, and a replenishment support device 26. The learning device 24 includes a preprocessing unit 241, a learning unit 242, and an update unit 243. The estimation device 25 includes an analysis unit 251 and an estimation unit 252. The replenishment support device 26 includes a candidate determination unit 261 and a replenishment support unit 262.

[0012] The inventory management system 1 is a system that supports (assists) the inventory management of items (objects to be managed). For example, the inventory management system 1 supports the user in replenishing items. The location where the inventory management system 1 is installed is not limited to a specific location. In the following, the inventory management system 1 is installed in the user's residence as an example. Furthermore, the items (objects to be managed) are not limited to specific items. The items are, for example, fixtures.

[0013] The inventory management device 2 is a device that supports (assists) inventory management of goods. For example, the inventory management device 2 supports a user in replenishing goods. The inventory management device 2 is, for example, a personal computer.

[0014] The communication line 3 is not limited to a specific type of communication line. The communication line 3 may be a wireless communication line or a wired communication line. The communication line 3 is, for example, a local area network (LAN).

[0015] The camera 4 is installed at a predetermined position in the user's residence, facing in a predetermined direction. The camera 4 may be pre-installed in a predetermined device (e.g., a personal computer). The camera 4 captures a predetermined range (angle of view) within the residence from the position where the camera 4 is installed. As a result, the camera 4 generates a video (captured images) capturing, for example, items and a user within the residence. That is, the time-series still images (frames) constituting the video capture, for example, items within the residence and a user using the items. For example, if the item (managed object) within the residence is a plate, the user may be captured serving food on the plate. Alternatively, the user may be captured eating the food served on the plate.

[0016] The microphone 5 is installed at a predetermined position in the user's residence (for example, near the camera 4). The microphone 5 collects sound from around the position where the microphone 5 is installed. In this way, the microphone 5 generates voice data corresponding to, for example, the user's speech.

[0017] The sensor 6 is held by the user. The sensor 6 generates vital data of the user at a predetermined cycle. The vital data is, for example, pulse waveform data. The vital data may also be, for example, heart rate data.

[0018] The sensor 6 may be a positioning sensor. Alternatively, the sensor 6 may be a wireless tag. For example, the sensor 6 may measure the user's position within the residence at a predetermined period by triangulation using radio waves transmitted from wireless devices pre-installed in the residence. For example, the sensor 6 may measure the user's position outdoors at a predetermined period by using radio waves transmitted from an artificial satellite.

[0019] The acquisition unit 21 acquires video from the camera 4 via the communication line 3. The acquisition unit 21 acquires audio data from the microphone 5 via the communication line 3. The acquisition unit 21 acquires vital data from the sensor 6 via the communication line 3. The acquisition unit 21 may acquire user position information from the sensor 6 via the communication line 3. The acquisition unit 21 (communication unit) may communicate with an external device (not shown) via the communication line 3.

[0020] The method for identifying a user in a video (captured image) generated by the camera 4 is not limited to a specific identification method. For example, images of a plurality of users may be registered in advance in the storage device 22. The acquisition unit 21 may identify a user in the video generated by the camera 4 based on the degree of coincidence between the image of each user among the plurality of users registered in advance and the image of the user in the video generated by the camera 4.

[0021] For example, the acquisition unit 21 may acquire a wireless signal from a wireless tag held by the user, and may identify the user in the video based on identification information represented by the acquired wireless signal.

[0022] The storage device 22 stores a computer program, a learning model, a trained model, and training data. The learning model and the trained model have a neural network (not shown) for machine learning. The structure of the neural network is not limited to a specific structure. For example, the neural network has an input layer, an intermediate layer, and an output layer.

[0023] The storage device 22 stores the inventory control data. The storage device 22 may output the inventory control data to, for example, the estimation device 25.

[0024] 2 is a diagram illustrating an example of inventory management data according to an embodiment. The format of the inventory management data is not limited to a specific format. In FIG. 2, the format of the inventory management data is a table format (data table format) as an example.

[0025] The format of the inventory management data is not limited to a specific format. For example, the format of the inventory management data may be a file format, or a website format that can be viewed by a user terminal (not shown) accessing a server at a specific address. For example, the format of the inventory management data may be an email format that is periodically notified to the user.

[0026] When an item that has already been registered in the inventory management data is re-registered, only the information that needs to be updated is updated in the inventory management data. The inventory management data can also be updated manually by the user.

[0027] In the inventory management data illustrated in FIG. 2, for example, items (image recognition results), item information, and whether or not replenishment is required are associated with each other. The item information is not limited to specific information about the item. For example, the item information may include "remaining amount," "expiration date," "continuous use," and "satisfaction level."

[0028] "Continuous use" is information indicating whether the product is continuously used by the user, and is expressed as, for example, "YES" or "NO." "Continuous use" may also be expressed as a value (for example, frequency of use).

[0029] "Satisfaction level" is the degree to which a user is satisfied with an item (for example, a favorite level, an emotional index value, and a desire to replenish). "Need for replenishment" is the result of a determination as to whether or not an item needs to be replenished. "Need for replenishment" may be expressed using the level of desire for replenishment (desire level, replenishment determination value). Note that the inventory management data may be updated based on operational input (manual input) or voice input by the user.

[0030] 1 , we will continue to explain the inventory management device 2. When the inventory management device 2 is started up, a computer program is loaded from the storage device 22 into the memory 23. The loaded computer program is executed by each functional unit (processor) of the inventory management device 2.

[0031] In the learning stage, the learning device 24 executes a machine learning technique. The machine learning technique is not limited to a specific technique, but is, for example, supervised learning. The preprocessing unit 241 acquires training data from the storage device 22. The training data includes training data and correct answer labels (correct answer data).

[0032] The learning data is data prepared in advance and represents the state of the user. The data representing the state of the user is, for example, a facial image of the user for each facial expression. The data representing the state of the user may be, for example, voice data of the user for each tone of voice. The data representing the state of the user may be, for example, voice data of the user for each speech content.

[0033] The correct label is data (label) prepared in advance and represents the user's satisfaction with the product. The greater the user's satisfaction with the product, the greater the absolute value of the positive satisfaction value. The greater the user's dissatisfaction with the product, the greater the absolute value of the negative satisfaction value. The satisfaction value may also be normalized. For example, the satisfaction value may be expressed using multiple levels (e.g., five levels).

[0034] The user's satisfaction level is determined, for example, based on the user's vital data. For example, the higher the heart rate of a user using an item, the higher the user's satisfaction with that item (e.g., favorite level, emotion index value, and willingness to replenish). For example, the more positive utterances a user makes while using an item, the higher the user's satisfaction with that item. For example, the more negative utterances a user makes while using an item, the lower the user's satisfaction with that item.

[0035] The learning unit 242 acquires a training model from the storage device 22. The update unit 243 generates a trained model by applying a machine learning technique to the training model. Here, the update unit 243 inputs training data in a predetermined format (e.g., a predetermined size) to the training model. The update unit 243 acquires an estimation result (estimated value) of user satisfaction from the training model.

[0036] The update unit 243 compares the estimated result of the user satisfaction level with the correct label. An initial value is set for each parameter of the training model. The update unit 243 updates each parameter of the training model, for example, by backpropagation until a predetermined criterion is met. The predetermined criterion may be, for example, a criterion that must be met by the number of updates of each parameter, or a criterion that must be met by the difference between the estimated result and the correct label. In this way, the update unit 243 generates a trained model from the training model. The update unit 243 records the trained model in the storage device 22.

[0037] In the estimation stage, the estimation device 25 estimates the satisfaction level of a user who uses an item based on data representing the user's state. The analysis unit 251 acquires a video of the item and the user as one piece of data representing the user's state from the acquisition unit 21. The analysis unit 251 may also acquire user voice data from the acquisition unit 21 as one piece of data representing the user's state.

[0038] The analysis unit 251 recognizes an item using a predetermined analysis method based on a video captured of a user using the item. The predetermined analysis method is not limited to a specific analysis method. For example, the analysis unit 251 recognizes an item used by a user using a known image recognition technology (technology for recognizing items). The analysis unit 251 registers the item (image recognition result) in the inventory management data.

[0039] The analysis unit 251 generates information about items used by a user using a predetermined analysis method. The analysis unit 251 estimates the user's actions when using the item and the amount of the item used using the item using known motion analysis technology and image recognition technology. The analysis unit 251 estimates the remaining amount of the item by subtracting the amount of the item used (cumulative value) from the initial value of the remaining amount of the item. The analysis unit 251 updates the remaining amount of the item in the inventory management data.

[0040] The analysis unit 251 recognizes the expiration date (use-by date) printed on the item captured in the video by using known image recognition technology on the video. The analysis unit 251 may also recognize the expiration date printed on the container of the item captured in the video. The analysis unit 251 updates the expiration date of the item in the inventory management data. If the item used by the user does not have a specified expiration date, the analysis unit 251 may register an expiration date based on the end of a general usage period for the category to which the item belongs in the inventory management data. Furthermore, if the item is damaged, the analysis unit 251 may analyze the extent of damage to the item captured in the video by using known image recognition technology on the video. The analysis unit 251 updates the expiration date of the damaged item in the inventory management data based on the extent of the damage.

[0041] The analysis unit 251 estimates the frequency with which the user uses an item for each item by applying known motion analysis technology and image recognition technology to the video. The analysis unit 251 determines whether the item is being continuously used based on the estimated frequency with which the user uses the item. For example, the analysis unit 251 determines that the item is being continuously used if the frequency with which the user uses the item is equal to or greater than a threshold. If the analysis unit 251 determines that the item is being continuously used, it updates the "continuous use" column of the inventory management data for the item to "YES." If the analysis unit 251 determines that the item is not being continuously used, it updates the "continuous use" column of the inventory management data for the item to "NO." Furthermore, if the continuous use is expressed as a value, the analysis unit 251 may decrease the value of the continuous use of the item as the frequency with which the item is used decreases.

[0042] The estimation unit 252 (user request analysis unit) acquires the trained model from the storage device 22. The estimation unit 252 acquires videos of the item and the user from the analysis unit 251. The estimation unit 252 may acquire voice data of the user from the analysis unit 251.

[0043] The estimation unit 252 inputs estimation data in a predetermined format into the trained model. The estimation data is data representing the user's state at the estimation stage. The data representing the user's state at the estimation stage may be, for example, a facial image of the user. The data representing the user's state at the estimation stage may also be, for example, voice data of the user. That is, the estimation data includes videos captured of the item and the user at the estimation stage. The estimation data may also include voice data of the user at the estimation stage.

[0044] The estimation unit 252 acquires the estimation result (estimated value) of the user satisfaction level from the trained model, and registers the estimation result (estimated value) of the user satisfaction level for each item in the inventory management data.

[0045] The estimation unit 252 may estimate the relevance between different scenes using known scene understanding techniques. The estimation data may include not only a video captured by a user using an item, but also a video captured by a subsequent user. Furthermore, the estimation data may include not only audio data of a user using an item, but also audio data of subsequent users. For example, if the item is a seasoning, the estimation data may include not only video and audio data of a cooking scene using the seasoning, but also video and audio data of a subsequent eating scene.

[0046] The estimation unit 252 may perform a known face recognition process, a known region extraction process on time-series still images (frames) constituting a moving image, or a known speaker identification process on audio data.

[0047] The replenishment support device 26 is a device that supports the replenishment of items by a user, etc. After the estimation stage, the candidate determination unit 261 acquires the item recognition result, item information, and the estimation result (estimated value) of the user satisfaction level from the estimation unit 252 or the storage device 22 (inventory management data).

[0048] The candidate designator 261 determines, for each item, whether or not the item needs to be replenished, based on the item information. That is, the candidate designator 261 determines, for each item, the level of desire for replenishment (desirability) based on the item information. The method by which the candidate designator 261 determines whether or not the item needs to be replenished (desirability for replenishment) is not limited to a specific determination method. For example, the candidate designator 261 may determine that the lower the remaining amount of an item, the higher the desire for replenishment.

[0049] The candidate designator 261 registers the determination result of whether or not the item needs to be replenished in the inventory management data. The candidate designator 261 may also register the level of demand for replenishment in the inventory management data. The level of demand may be expressed using multiple levels (for example, five levels).

[0050] For an item whose remaining quantity decreases with use (e.g., seasoning), the candidate designator 261 determines whether or not to replenish the item (the level of desire) based on at least one of the remaining quantity, expiration date, and continued use (whether or not the item is being used continuously) in the inventory management data. Each of the remaining quantity, expiration date, and continued use may be weighted.

[0051] By setting the weight of "remaining amount" relatively large, the candidate designator 261 may estimate the timing when the remaining amount will be equal to or less than a predetermined amount based on the estimated result of the amount of use by the user per use (an example of the user's behavior pattern) and the remaining amount of the item. By setting the weight of "continuous use" relatively large, the candidate designator 261 does not replenish items that are not used continuously (items that are used infrequently by the user).

[0052] For items (e.g., plates) for which the remaining quantity does not decrease with use, the candidate designator 261 determines whether or not to replenish the item (the degree of desirability) based on at least one of the expiration date and continued use in the inventory management data. Each of the expiration date and continued use may be weighted. For example, even if an item has a large remaining quantity and is frequently used, the candidate designator 261 may set a low degree of desirability for replenishment for an item with low satisfaction.

[0053] This makes it possible to determine whether to replenish items based on the attributes of the items, and also to determine whether to replenish items based on the user's satisfaction level.

[0054] When it is determined that an item needs to be replenished (that an item needs to be replenished), the candidate determination unit 261 determines candidates for the item to be replenished based on the estimation result of the user's satisfaction level. The method by which the candidate determination unit 261 determines candidates for the item to be replenished is not limited to a specific method. For example, the candidate determination unit 261 may determine an item with a high level of satisfaction as a candidate for the item to be replenished. For example, when it is determined that an item with a high level of satisfaction needs to be replenished, the candidate determination unit 261 may determine the item with a high level of satisfaction or another item similar to the item with a high level of satisfaction (similar item) as a candidate for the item to be replenished.

[0055] Here, the candidate designator 261 determines whether the candidate item has a high level of satisfaction based on the comparison result between the estimated satisfaction level and a predetermined threshold value. The candidate designator 261 may adjust the predetermined threshold value according to the user's behavior (action).

[0056] When a candidate item is determined to be an item with low satisfaction, the candidate determination unit 261 may select a similar item other than the candidate item. For example, when a candidate item is determined to be an item with low satisfaction, the candidate determination unit 261 may determine another similar item with high satisfaction as a candidate item to be replenished.

[0057] The method for determining other similar items with high satisfaction as candidates is not limited to a specific determination method. For example, the candidate determination unit 261 may select other similar items that are popular in the same category as the item with low satisfaction as candidates for items to be replenished from a predetermined list of items. For example, the candidate determination unit 261 may analyze user preferences using a known preference analysis method. The candidate determination unit 261 may estimate ingredients, types, costs, etc. that suit the user's preferences. Based on the estimation results of ingredients, etc. that suit the user's preferences, the candidate determination unit 261 may select items with ingredients, etc. that suit the user's preferences as candidates for items to be replenished from a predetermined list of items.

[0058] The candidate designator 261 creates a list of items that are determined to need replenishment in the inventory management data (items with a high degree of demand for replenishment) as a purchase candidate list. The purchase candidate list may include, for each item, location information of a store where the item can be purchased.

[0059] The replenishment support unit 262 acquires information on the items to be replenished (e.g., a purchase candidate list) from the candidate determination unit 261. The replenishment support unit 262 supports (assists) the user in replenishing items based on the information on the items to be replenished.

[0060] The method by which the replenishment support unit 262 supports the replenishment of items is not limited to a specific support method. For example, the replenishment support unit 262 may acquire user location information from the sensor 6 via the communication line 3. When the distance between the user's outdoor location and the store location becomes less than a predetermined distance, the replenishment support unit 262 may transmit a purchase candidate list to a user terminal (not shown). In other words, the replenishment support unit 262 may notify the user, who is near a store where the item is available, of an item determined to need to be replenished, at an appropriate time.

[0061] When the cumulative number of items determined to need replenishment reaches a certain number, the replenishment support unit 262 may notify the user of multiple candidate items to be replenished. This can reduce the possibility that the user will be frequently notified of candidate items to be replenished.

[0062] When the inventory of an item that has been determined to require replenishment falls below a certain number, the replenishment support unit 262 may notify the user of candidate items to replenish before the inventory of the item runs out.

[0063] The replenishment support unit 262 may be linked in advance to a predetermined online shopping site. The replenishment support unit 262 may purchase an item determined to need replenishment from the linked online shopping site. The replenishment support unit 262 requests the online shopping site to deliver the purchased item to the user.

[0064] The replenishment support unit 262 may be linked in advance with stores within the user's range of activity. The replenishment support unit 262 may place an order with a store to reserve items determined to need replenishment at the store. When the distance between the user's outdoor location and the store location falls below a predetermined distance, the replenishment support unit 262 may transmit location information of the store to a user terminal (not shown).

[0065] When ordering other similar items that are highly satisfying, the replenishment support unit 262 may give the user a timing for ordering (purchasing) the items.

[0066] Next, an example of the operation of the inventory management device 2 will be described. FIG. 3 is a flowchart showing an example of the operation of the inventory management device 2 in an embodiment. The analysis unit 251 recognizes an item based on a video of a user using the item and the item (step S101). The analysis unit 251 generates information about the recognized item based on the video of a user using the item and the item (step S102). The estimation unit 252 estimates the user's satisfaction level based on at least one of an image of the user in the video and voice data of the user (step S103).

[0067] The candidates designator 261 determines whether to replenish the product based on the product information (step S104). If it is determined that the product should not be replenished (step S104: NO), the estimation device 25 returns the process to step S101.

[0068] If it is determined that the item should be replenished (step S104: YES), the candidate determination unit 261 determines candidates for the item to be replenished based on the user's satisfaction (step S105). The replenishment support unit 262 notifies the user of the candidates for the item to be replenished (step S106).

[0069] As described above, in the estimation device 25, the analysis unit 251 recognizes an item based on a video in which a user using the item and the item are captured. The analysis unit 251 generates information about the recognized item based on the video in which a user using the item and the item are captured. The estimation unit 252 estimates the user's satisfaction level based on at least one of an image of the user in the video and voice data of the user.

[0070] In the replenishment support device 26, the candidate determination unit 261 determines whether to replenish an item based on information about the item. If it is determined that an item should be replenished, the candidate determination unit 261 determines candidates for the item to be replenished based on the user's satisfaction level. If it is determined that a candidate item has low satisfaction, the candidate determination unit 261 may determine other similar items with high satisfaction levels as candidates for the item to be replenished. If it is determined that an item should be replenished, the replenishment support unit 262 notifies the user of the candidates for the item to be replenished.

[0071] This makes it possible to reduce the possibility of replenishing (ordering) items that do not satisfy the user.

[0072] By introducing image processing technology (image recognition technology) that recognizes items (objects), it is possible to manage inventory of items with fewer restrictions on the location, method, and type of items. Because items are image-recognized, there is no need to pre-register the image (appearance) of the item, and inventory management of items can be carried out efficiently.

[0073] Not only is the remaining amount of an item managed, but the necessity of replenishing the item is determined based on the estimated result of the user's satisfaction (preference) with respect to the item, so that optimal item management for the user is more efficiently performed. Furthermore, factors other than the image of the item (e.g., remaining amount and weight) (e.g., expiration date) are also used as criteria for determination, and inventory management (e.g., recommendation of items to be replenished) is more efficiently performed even for items with such factors that do not change.

[0074] Since the system suggests (supports) a method for replenishing items according to the user's behavior, it is possible to replenish items at the optimal timing that matches the user's behavior, and it is possible to prevent a decrease in the efficiency of inventory management due to forgetting to buy an item, etc.

[0075] In the above embodiment, inventory management in a home has been described as an example, but the inventory management device 2 can also be applied to the following applications.

[0076] <Inventory Management Applications in Offices> For example, the inventory management device 2 records the inventory status of items in inventory management data. This makes inventory registration more efficient and reduces the space required for storing inventory. It also makes it possible to reduce the possibility of replenishing (ordering) items that do not satisfy users.

[0077] <Inventory Management Applications in Retail Stores and Restaurants> For example, the inventory management device 2 manages food ingredient inventory based on customer satisfaction (preferences), thereby reducing food ingredient waste and providing food ingredients efficiently.

[0078] <Inventory Management Applications in Clothing Stores and Convenience Stores> For example, the inventory management device 2 adjusts the quantity and types of merchandise displayed in stores based on trends in customer satisfaction.

[0079] <Use for understanding remaining amount and status of installed items> For example, the inventory management device 2 manages the remaining amount of food installed in the aquarium and the status of the aquatic plants.

[0080] (Hardware Configuration) FIG. 4 is a diagram illustrating an example of the hardware configuration of the inventory management device 2 according to an embodiment. The inventory management device 2 is realized as software by a processor 101, such as a central processing unit (CPU), executing a program stored in a storage device 103 having a non-volatile storage medium (non-transitory storage medium) and a memory 102. The program may be recorded on a computer-readable storage medium. Examples of computer-readable storage media include portable media such as a flexible disk, a magneto-optical disk, a read-only memory (ROM), and a compact disc read-only memory (CD-ROM), and non-transitory storage media such as a hard disk or solid-state drive (SSD) built into a computer system. A communication unit 104 executes predetermined communication processing.

[0081] The inventory management device 2 may be realized using hardware including electronic circuits (electronic circuits or circuitry) using, for example, an LSI (Large Scale Integrated circuit), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0082] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention.

[0083] The present invention is applicable to an inventory management system (inventory replenishment support system).

[0084] 1...inventory management system, 2...inventory management device, 3...communication line, 4...camera, 5...microphone, 6...sensor, 21...acquisition unit, 22...storage device, 23...memory, 24...learning device, 25...estimation device, 26...replenishment support device, 101...processor, 102...memory, 103...storage device, 104...communication unit, 241...preprocessing unit, 242...learning unit, 243...update unit, 251...analysis unit, 252...estimation unit, 261...candidate determination unit, 262...replenishment support unit

Claims

1. An inventory management device comprising: an analysis unit that recognizes an item based on a video of a user using the item and the item, and generates information about the item; an estimation unit that estimates the user's satisfaction level based on at least one of an image of the user in the video and voice data of the user; a candidate determination unit that determines whether to replenish the item based on the information about the item, and if it is determined that the item should be replenished, determines candidates for the item to be replenished based on the user's satisfaction level; and a replenishment support unit that, if it is determined that the item should be replenished, notifies the user of the candidates for the item to be replenished.

2. The inventory management device of claim 1, wherein when the candidate determination unit determines that the candidate is an item with a low level of satisfaction, it determines other similar items with a high level of satisfaction as candidates for the item to be replenished.

3. An inventory management method executed by an inventory management device, comprising: a step of recognizing an item based on a video of a user using the item and the item, and generating information about the item; a step of estimating the user's satisfaction level based on at least one of an image of the user in the video and voice data of the user; a step of determining whether to replenish the item based on the information about the item, and if it is determined that the item should be replenished, determining candidate items to be replenished based on the user's satisfaction level; and a step of notifying the user of candidate items to be replenished if it is determined that the item should be replenished.

Images