INTELLIGENT SHELF MANAGEMENT SYSTEM AND OPERATING METHOD

Abstract

An intelligent shelf management system and operating method are revealed. The system includes an operating and display device by which users input their demands of goods; an information module electrically connected to the operating and display device and comprising a database of goods and an analysis module which receives and analyzes the demands of goods input and finds out product information matched after comparison with information in the database; a content management system electrically connected to the information module and receiving the product information from the analysis module; and a point-of-sale system having a plurality pieces of related sales information and electrically connected to the content management system or the information module. The content management system compares the product information with the plurality pieces of related sales information to collect and send the related sales information matched to the operating and display device.

Claims

1. An intelligent shelf management system comprising: an operating and display device by which a user inputs a demand for goods; an information module which is electrically connected to the operating and display device and comprises an analysis module and a database of goods; wherein the database of goods includes a plurality pieces of product information therein while the analysis module receives and analyzes the demands of goods and then compares the demand of goods with the plurality pieces of the product information in the database of goods to find out at least one piece of the product information matched the demand for goods; a content management system which is electrically connected to the information module to receive the at least one piece of the product information provided by the analysis module; and a point-of-sale system which is electrically connected to the information module or the content management system which including a plurality pieces of related sales information each of which includes a plurality pieces of position information of goods and stock information of goods; wherein the content management system compares the at least one piece of the product information with the plurality pieces of the sales related information to get at least one piece of the related sales information matched the at least one piece of the product information and then deliver the piece of the related sales information matched to the operating and display device.

2. The system as claimed in claim 1, wherein the information module establishes an association rule of the demand of goods by deep learning.

3. The system as claimed in claim 1, wherein the operating and display device includes a touch interface, a voice operated device, and a display interface.

4. The system as claimed in claim 1, wherein the intelligent shelf management system further includes a camera recognition module which is electrically connected to the point-of-sale system or the information module and comprises at least one image capture device and an image recognition module; wherein the image capture device is provided with a number and used to shoot at least one piece of the goods on a shelf to get a piece of image information and then send the piece of the image information to the image recognition module for recognition of appearance and position of the goods; thus a recognition result is obtained and delivered to the information module together with the number of the image capture device which captures the piece of the image information, the information module integrates the relevant information numbered in the database of goods and the price and promotion information in the point-of-sale information system for generating position information of the piece of the goods.

5. The system as claimed in claim 4, wherein the image recognition module sets up a recognition rule by deep learning.

6. The system as claimed in claim 4, wherein a product tag is disposed on the shelf and located around at least one piece of goods.

7. The system as claimed in claim 6, wherein the product tag is a touch display device for showing information related to the goods.

8. A method of operating an intelligent shelf management system comprising the steps of: Step 1: inputting a demand of goods on an operating and display device and sending the demand of goods to an information module; wherein the information module is electrically connected to the operating and display device and comprises an analysis module and a database of goods; Step 2: receiving and analyzing the demand of goods and compare the demand of goods with a plurality pieces of product information in the database of goods by the analysis module of the information module to find out at least one piece of the product information which matches the demand of goods; Step 3: delivering the at least one piece of the product information to a content management system which is electrically connected to not only the information module but also a point-of-sale system which includes a plurality pieces of related sales information; the point-of-sale system is also electrically connected to the information module; and Step 4: comparing the at least one piece of the product information with the plurality pieces of the related sales information of the point-of-sale system by the content management system to get at least one piece of the related sales information which matches the at least one piece of the product information and then send the piece of the related sales information to the content management system then deliver to the operating and display device.

9. The method of operating an intelligent shelf management system as claimed in claim 8, wherein a camera recognition module is used to obtain position information of at least one piece of goods and deliver the position information of the goods to the information module, the content management system, and the operating and display device in turn, wherein the information module compares the position information of goods with the point-of-sale system through commodity coding, and then transmits position information of goods and sales information to the content management system and the operating and display device; wherein the camera recognition module comprises at least one image capture device and an image recognition module; the image capture device shoots the piece of the goods on a shelf to get a piece of image information and send the piece of the image information to the image recognition module for recognition of appearance and position of the goods; thus a recognition result is obtained and delivered to the information module together with a number of the image capture device which captures the piece of the image information to be integrated in the information module for generating the position information of the piece of goods.

10. The method of operating an intelligent shelf management system as claimed in claim 9, wherein the information module establishes a relational analysis rule of the demand of goods by deep learning while the image recognition module sets up a recognition rule by deep learning.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The structure and the technical means adopted by the present invention to achieve the above and other objects can be best understood by referring to the following detailed description of the preferred embodiments and the accompanying drawings, wherein

[0017] FIG. 1 is a schematic drawing showing structure of the present invention;

[0018] FIG. 2 is a schematic drawing showing structure of a camera recognition module of an embodiment according to the present invention;

[0019] FIG. 3 is a schematic drawing showing an embodiment in use according to the present invention;

[0020] FIG. 4 is another schematic drawing showing an embodiment in use according to the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

[0021] Refer to FIG. 1 and FIG. 3, an intelligent shelf management system based on the present invention mainly includes an operating and display device 1, an information module 2, a content management system 3, and a point-of-sale system 4. The operating and display device 1 consists of a touch interface, a voice operated device, and a display interface. The operating and display device 1 which can also be a touch screen 11 or a display screen with a microphone is electrically connected to the information module 2 which comprises an analysis module 21 and a database of goods 22 for establishing relational analysis rules of goods demand based on deep learning approach. The content management system 3 is electrically connected to the information module 2 and the point-of-sale system 4 which includes a plurality pieces of related sales information such as position information of goods on shelves, price information of goods, amounts of goods in stock (stock information of goods), etc. As to the above relational analysis rules, dividing goods in the database of goods 22 into groups based on their characteristics at first. Taking a pharmacy store's system as an example. A demand-analysis model of ingredients or functions of all medicines or health food is trained by deep learning approach. The suitable ingredients of medicines or health food are quickly calculated and provided to consumers for reference after demands are input. Moreover, the deep learning approach integrates technologies of neural networks, decision trees, and random forests to develop the relational analysis rules of goods demand mentioned above.

[0022] Refer to FIG. 2 and FIG. 3, the present system further includes a camera recognition module 5 which is electrically connected to the point-of-sale system 4 and the information module 2 and comprises at least one image capture device 51 and an image recognition module 52. The image capture device 51 can be different types of cameras such as a wireless network internet protocol camera (IP cam) for surveillance.

[0023] A product tag 71 which can be a touch display device is placed on a shelf 6 and located around respective goods 7 for showing information related to the goods 7 such as prices, promotions, other available platforms for users to buy the goods 7, etc.

[0024] Refer to FIG. 1-4, pharmacies and drugstores are taken as an example in this embodiment. After entering into the store, a consumer U inputs his/her demand of goods (goods they are looking for) m1 by the operating and display device 1 such as the touch screen 11 or the voice operated device at first. For instance, the touch screen 11 in a pharmacy shows a schematic drawing of human organs or body parts. Once the consumer U is looking for medicines for nose allergy or stuffy nose, he/she clicks on a nose icon or says “nose” or “nasal”. Then the touch screen 11 displays symptoms of diseases associated with the nose and the demand for goods m1 is sent to the information module 2 after the consumer U selects the corresponding symptoms. The analysis module 21 of the information module 2 analyzes the demand of goods m1 received to get suitable product information m2 from the database of goods 22 after comparison and send the product information m2 to the content management system 3. In this embodiment, the product information m2 includes ingredients of medicines which correspond to nasal symptoms.

[0025] The analysis module 21 of the information module 2 also carries out relational analysis to get relational analysis rules between the demand of goods m1 and the product information m2. Taking the above pharmacy as an example. The ingredients of medicines associated with the symptoms are obtained after analysis of the symptoms input by the consumer U and then compared with information in the database of goods 22 to check whether the pharmacy has the medicines containing the related ingredients.

[0026] Refer to FIG. 2, the image capture device 51 of the camera recognition module 5 shoots the goods 7 on the shelf 6 to get image information and send the image information to the image recognition module 52 for recognition of appearance and positions of the goods 7 and learning positions of specific goods. This is a recognition result m5. The respective image capture devices 51 are labeled with different numbers or having different IP addresses. Both the recognition result m5 and the number of the image capture device 51 which sends image information are delivered to the database of goods 22 of the information module 2 then delivered to the content management system 3.

[0027] In a preferred embodiment, the image recognition module 52 sets up recognition rules based on deep learning. In order to establish the recognition rules, using images of the appearance of the goods 7 taken at different angles to set up a database is the first step. Step 2 is training an image recognition model by using deep learning approach. Thereby types and names of the goods 7 can be inferred by input of the images captured by the image capture device 51 into the image recognition model. The deep learning approach combines technologies of neural networks, pattern recognition, and digital image processing.

[0028] The present system uses the camera recognition module 5 to recognize the positions of the goods 7 on the shelf 6 automatically. Thus, even the positions of the goods 7 are changed, the camera recognition module 5 can still be used to recognize the positions of the goods 7 and deliver the recognition result m5 to the content management system 3, without input of the latest positions of the goods 7 again by labors. The possibility of input error caused by manual input is also lowered. At the same time, the content management system 3 also receives product information m2 from the database of goods 22 of the information module 2, or the content management system 3 also receives related sales information m3 from the point-of-sale system 4 while the sales related information m3 includes position information of goods m4.

[0029] The content management system 3 compares the product information m2 received with the plurality of the related sales information m3 received for finding out the related sales information m3 that matches the product information m2. Then the matched related sales information m3 is sent to the operating and display device 1 by the content management system 3 and provided to the consumer U so that the consumer U can move to the corresponding shelf 6 and take the goods 7 according to the related sales information m3 displayed.

[0030] The point-of-sale system 4 includes the stock information of goods so that the content management system 3 can also get current quantity of the goods 7 in stock while obtaining the matched goods 7 from the point-of-sale system 4. Once the quantity of the goods 7 in stock is insufficient, the content management system 3 reminds staff to replenish the stock.

[0031] Refer to FIG. 4, when the consumer U walks to the corresponding shelf 6, each kind of goods 7 on the shelf 6 is provided with the product tag 71 which can also be a touch display device. After being touched by consumers, the product tag 71 is activated and a message of activation is sent to the content management system 3 so that the content management system 3 recognizes the product tag 71 of which goods 7 is activated and then delivers a message related to the goods 7 corresponding to the product tag 71 to the point-of-sale system 4. Next the point-of-sale system 4 get the message related to the goods 7 across to an electronic commerce system. Thereby information related to the goods 7 such as price, promotions, other available platforms, etc. is displayed on the product tag 71 for the customer's (U) reference.

[0032] As shown in FIG. 3 and FIG. 4, the operating and display device 1, the information module 2, the content management system 3, the point-of-sale system 4, the image recognition module 52 of the camera recognition module 5, and the electronic commerce system can be arranged at one or a plurality of servers 8 which can be a physical server or cloud server.

[0033] In summary, the present invention has the following advantages compared with the techniques available now.

1. After input of the needs through the operating and display device of the present system, the customer can find out the goods he/she wants and recognize the positions of the goods quickly. Thus, the time it takes to search the goods in the store is significantly reduced. The store can be operated with fewer employees so that the operating cost is reduced.
2. Even the positions of the goods are changed, the position information of the goods in the point-of-sale system can be modified automatically in a real-time manner by the camera recognition module which recognizes the positions of the goods on the shelf automatically. Thus, there is no need to input the position information of the goods by labor and input error caused by manual operation can be avoided. In addition, the content management system can also set whether to modify the permissions to each store manager, once it is detected that the product is moved to the shelf that no longer corresponds, the store manager can confirm on the content management system whether he agrees to directly change the product information, or then send the clerk to adjust the product to the correct shelf.
3. The present system can be applied to various types of stores, especially those sell a plurality of products such as pharmacies, cosmetic stores, grocery stores, hardware stores, etc. and thus having a wide range of applications.

[0034] Additional advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details, and representative devices shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalent.