TW201433992A - Graphical recognition inventory management and marketing system - Google Patents

Abstract

A graphical recognition inventory management and marketing system is disclosed. An image comprising a plurality of product elements is captured. Graphical recognition is utilized to detect, isolate, and identify each element by comparing characteristics of the element with element data stored in a database. When a match is found, the detected and isolated element data is stored with the matched database entry. Additional element data obtained during the graphical recognition process is added to element data stored in the database to expand the database entry. The current location of a user device is detected and product element data is displayed along with a map indicating the physical location of the product element and a route to the product element from the user's current location. A user can also capture an image and have product data and a map to the product displayed.

Description

Inventory management and marketing image recognition system

The present invention relates to an image recognition system, and more particularly to a system for utilizing image recognition technology to efficiently increase inventory management efficiency while at the same time achieving a better system for assisting customers to more easily shop in a physical store.

According to the demand of the thriving e-commerce shopping site, a large and diverse variety of goods must be stored in a large number of warehouses. Moreover, as consumer demand for shopping increases, the number of inventories in the warehouse and the size of each inventory warehouse need to grow and expand. Therefore, usually when the warehouse performs inventory inventory, the workers must spend a very large amount of time to perform the ordering action. Even when it is to be shipped, the workers also need to spend extra time searching for the goods purchased by the consumers, and using the manpower to remove the purchased items from the shelves of the warehouse for smooth shipment. It can be seen that the warehouse inventory/shipment management and inventory of the acquisitions are quite time-consuming and labor-intensive for the managers.

Moreover, since the warehouse usually includes more than one cargo rack, and each cargo rack has many different (but similar) merchandise items, the workers in the warehouse not only need to put the correct products one by one when carrying out the merchandise. Placed on the corresponding cargo rack, it is necessary to identify each specific single commodity among a large number of similar commodities to complete the correct ordering action.

Secondly, when inventory is in stock, the workers in the warehouse must also paste barcodes on each item one by one, and scan the barcode into the computer for documentation. Then, the scanned products are placed again. Back to the cargo rack. It can be seen that when these inventory actions are repeated, a very large amount of manpower is consumed, and the inventory efficiency is also extremely low.

In addition, when performing a warehouse inventory, the warehouse must briefly experience a period of freeze. Before the inventory is completed, the goods cannot flow (ie, are sold) during the closing period. Similarly, in a retail store in a local area, in order for an inventory to be counted, the store must be closed for a short period of time until the inventory is completed before the transaction can be reopened. It can be seen that in order to carry out inventory counting, these stores and warehouses must undergo a period of suspension of business. In this case, some of the business will be lost in the invisible, affecting the quality of service and reducing the interest of consumers.

Moreover, the traditional retail channel must attract customers to the physical store by issuing advertising flyers, and print preferential prices and promotional items on these flyers to thereby induce consumer consumption. purpose. However, it is worth noting that it is very inconvenient for consumers to carry these flyers to the store for each consumption, or to record which products are being promoted. Furthermore, when a consumer goes to the store to shop, it often takes a lot of time to find the location on which the product to be purchased is located on the shelf, which inevitably reduces the consumer's desire to shop.

Therefore, in view of the above, how to provide a system that is both efficient and economical, not only can effectively improve the inventory and merchandise management of the Xijian warehouse, but also enhance the desire of consumers to shop in the physical store, is familiar with this technology. One of the problems that domain players need to solve.

In order to solve the problems of the prior art, the main object of the present invention is to provide an image recognition system for inventory management and marketing, which utilizes graphic recognition technology to efficiently improve the efficiency of inventory management. . In addition, with this identification system, it is also possible to simultaneously assist the customer to more efficiently shop in the physical store and induce their shopping interest.

The image recognition system and the identification method thereof according to the present invention integrate image recognition technology and existing inventory management records to provide a system for inventory and marketing management with better inventory efficiency.

First of all, the user can regularly photograph the merchandise rack and the rest of the merchandise on the rack. Photo or dynamic video. Then, the digital image recognition technology is used to automatically separate the individual products in the captured image, thereby identifying and extracting the commodity object information of each commodity. Then, the server is connected to the database through a server to compare the identified product with the information of the object already stored in the database. In this way, the exact quantity of the item and the quantity of the stock can be updated to the latest status and stored in the database. The database then has the latest inventory count to quickly complete inventory management, product picking and order processing. In addition, the present invention automatically provides a map and uses the map to indicate the actual location of each item on the warehouse or shelf.

By using the image recognition system and the identification method thereof disclosed in the present invention, since the product does not need to be moved or scanned, the present invention has better inventory management efficiency, and can accurately and easily know all the commodities. The correct amount of inventory is beneficial to inventory and marketing management.

With the image recognition system and the identification method thereof disclosed by the present invention, the warehouse manager can display the product on the shelf at any time by using an electronic device that is carried with him, such as a tablet computer or a smart phone. The location, and using the screens of these electronic devices, displays a map indicating the actual location of each item in the warehouse on the shelf. By displaying the map thereby, the warehouse manager can easily find the corresponding product location at the time of shipment, thereby greatly reducing the time and labor cost required for the acquisition.

In addition, since the inventory of the warehouse is not inferior to the foreign language, when the foreign language is printed on the product, the inventory personnel often fail to understand the text and cause errors in the commodity inventory. In view of the above, the present invention proposes an optical character recognition (OCR) technology to identify text on a commodity, and accordingly provides a picture of the article of the commodity to solve the problem of the foreign language of the inventory personnel. Causes an error on the count.

Furthermore, the image recognition system for inventory management and marketing disclosed by the present invention can improve the satisfaction and loyalty of consumers when shopping. The consumer can take advantage of the image capturing device in the mobile device to take the desired item, and then, when the location of the consumer is successfully detected, the present invention can display the device on the mobile device. In order to purchase the item information of the commodity, in addition, the mobile device can display which position the object is placed in the store, and provide a path from the current position of the consumer to the location of the object. In view of this, when a consumer goes shopping in a physical store, not only can he/she be easily remembered What you want to buy, you can easily find the location of the product in the store, greatly improving the efficiency of shopping and consumer satisfaction.

In another aspect, the image recognition system for inventory management and marketing disclosed by the present invention detects the products that the consumer photographs for marketing, and can also detect the past shopping history of each consumer. Available shopping reference records.

The purpose, technical contents, features and effects achieved by the present invention will be more readily understood by the detailed description of the embodiments and the accompanying drawings.

100‧‧‧Photographing

105‧‧‧Blank area

110‧‧‧Commodities

110A, 110B, 110C, 110D, 110E, 110F‧‧‧ commodity items

115‧‧‧ text

115A‧‧‧ rotating text

115B‧‧‧ horizontal text

120‧‧‧Front pattern

150‧‧‧Commodity shelf

151‧‧‧ top shelf bottom

152‧‧‧ middle shelf top

153‧‧‧ middle shelf bottom

154‧‧‧ bottom shelf top

155‧‧‧high shelf height

156‧‧‧ bottom shelf height

157‧‧‧Shelf depth

160A, 160B, 160C‧‧‧ on the shelf position label

165‧‧‧Shelf high landmark

170‧‧‧Shelf wide landmark

180‧‧‧ top shelf

181‧‧‧shelf

182‧‧‧ bottom shelf

200‧‧‧Image capture device

210‧‧‧Wireless interface

220‧‧‧Server

230‧‧‧Database

300‧‧‧Image identification method

310, 320, 330, 331, 332, 333, 334, 340 ‧ ‧ steps

400‧‧‧Image identification method

405, 410, 415, 420, 425, 430, 435, 440, 445, 450 ‧ ‧ steps

500‧‧‧Image identification method

510, 520, 530, 540, 550 ‧ ‧ steps

600‧‧‧Image identification method

610, 620, 630, 640, 650‧ ‧ steps

700‧‧‧Image identification method

705, 710, 715, 720, 725, 730, 735, 740, 745‧‧ steps

1 is a schematic structural view of an image recognition system for inventory management and marketing according to an embodiment of the present invention.

Figure 2 is a schematic illustration of a product image taken in accordance with an embodiment of the present invention.

Figure 3 is a flow chart showing the steps of the image recognition method for inventory management and marketing according to an embodiment of the present invention.

4A to 4C are schematic views of a spacer object having a character according to an embodiment of the present invention.

Figure 5 is a side and front elevational view of a merchandise article in accordance with an embodiment of the present invention.

Figure 6 is a flow chart showing the steps of the object recognition program according to an embodiment of the present invention.

Figure 7 is a schematic view showing the structure of a merchandise rack having an on-position position label according to an embodiment of the present invention.

Figure 8 is a flow chart showing the steps of an image recognition method for inventory management and marketing according to an embodiment of the present invention.

Figure 9 is a schematic view showing the structure of a merchandise rack according to an embodiment of the present invention.

Figure 10 is a block diagram showing the structure of an image recognition system for inventory management and marketing according to an embodiment of the present invention.

Figure 11 is a flow chart showing the steps of an image recognition method for inventory management and marketing according to an embodiment of the present invention.

Figure 12 is a flow chart showing the steps of an image recognition method for inventory management and marketing according to another embodiment of the present invention.

Figure 13 is a flow chart showing the steps of an image recognition method for inventory management and marketing according to still another embodiment of the present invention.

The above description of the present invention is intended to be illustrative and illustrative of the spirit and principles of the invention, and to provide further explanation of the scope of the invention. The features, implementations, and utilities of the present invention are described in detail with reference to the preferred embodiments.

Please refer to FIG. 1, which is a schematic structural diagram of a graphic recognition inventory management and marketing system according to an embodiment of the present invention.

In general, warehouses, wholesalers, and retail stores have a variety of cargo racks for storing a variety of items to be sold. In order to maintain the exact total quantity of goods, warehouses, wholesalers, and retail stores must regularly perform inventory counting operations in order to accurately control the total number of goods.

Please refer to FIG. 1 , the image recognition system disclosed in the present invention includes a product shelf 150 , a plurality of products 110 stored on the product rack 150 , and an image capturing device ( An image capture device 200, a server 220, and a database 230.

Among them, the database 230 stores product element data about each item 110. For example, the commodity item information may include: a brand name of the product, a product name, a price, a cost, a current inventory level, and a size. (size), shape, area or volume, weight, logo, manufacturer, Supplier, color, design, graphic, image, text, total cost of current inventory, shelf label (shelf) A location identification, a product number, a barcode, and/or other identifying features that can be used to identify the item 110.

According to an embodiment of the present invention, the image capturing device 200 is configured to capture and record a still photo or a video of the article 110 stored on the product shelf 150. According to the embodiment of the present invention, first, the product 110 is first captured by the image capturing device 200, and the image of the product 110 is successively isolated and recognized to confirm The identity of the product 110. In detail, the image of the item 110 is compared with the information stored in the database 230. When the comparison is matched, the item information of the item 110 is updated and added to the account of the database 230 ( In the entry). For example, when the number of items is increased or the number of items is tracked, the position on which the item is placed is recognized, or additional item information is added, the information can be successfully added to the account of the database 230. Among them.

According to the embodiment of the present invention, the image capturing device 200 and the server 220 are connected by a wireless interface 220.

Thereby, the step of isolating and recognizing the image of the product 110 can be performed in the image capturing device 200 or the server 220.

Please refer to FIG. 2, which is a schematic diagram of a product image captured according to an embodiment of the present invention.

As shown in FIG. 2, the captured image 100 is captured by the image capturing device 200 and includes a plurality of product elements 110A-110F and a non-element area. ) 105.

Among them, in order to successfully identify each of the goods 110A~110F, these goods objects 110A~110F must first pass the isolation process. For example, by detecting the boundary between the product object 110A and the blank area 105, or detecting the edge of the product objects 110A and 110B, the product object 110A in the captured image 100 can be compared with other objects 110B-110F. Make a separation.

Please refer to FIG. 3, which is a flow chart of the steps of the image recognition method for inventory management and marketing according to an embodiment of the present invention.

As shown in FIG. 3, the image recognition method 300 includes steps 310-340 and begins at step 310. In step 310, the present invention first captures a product image by using the image capturing device 200, wherein the product image includes a plurality of commodity objects. Then, in step S320, each product object in the image can be used for each product object by, for example, edge detection, color change, or height change. Perform separate procedures for isolation. For example, by changing the boundary or color change between any two merchandise items, or any merchandise object and a blank area, and/or the height between the objects, the image can be detected by detecting these features. Each object in the object is isolated for the purpose.

Thereafter, step 330 confirms the identity of the object through graphical recognition. In detail, the object that is isolated will be compared with the information of the object stored in the database. For example, the present invention can utilize optical character recognition (OCR) to identify text on a commodity object, and then compare the text with the information of the object already stored in the database to Find out if the text matches the database account. In the present embodiment, the present invention adopts the feature of the physical appearance of the article of goods, and directly implements the method of comparing the database map.

When the comparison result is matched, step 340 is stored in the database together with the database account in which the isolated object matches the database. Therefore, with the flow chart of the steps disclosed in the present invention, the warehouse, the wholesaler, and the database of each retail store can effectively grasp the quantity of the existing commodity inventory. Alternatively, through this image recognition method, additional object information can be efficiently and quickly stored in the database, further expanding the size of the database and its purpose.

Continuing to refer to Figures 4A through 4C, which are schematic illustrations of a spacer article having text in accordance with an embodiment of the present invention.

According to an embodiment of the invention, in the image recognition process, the identification system can detect and identify the text of the object according to different orientations. For example, in Figure 4A, the text 115 on the item 110 is in a vertical direction; in Figure 4B, the item 115 is in a horizontal direction; in Figure 4C, The text 115 is rotated.

Therefore, according to the image recognition method and the image recognition system disclosed in the present invention, the system can effectively detect and discriminate regardless of the direction of the text in the object image. Identify and confirm the text on the object.

Continuing to refer to FIG. 5, which is a side view and a front view of a commercial article according to an embodiment of the present invention.

In general, in the image recognition process disclosed by the present invention, the present invention first collects a plurality of commodity object information, such as: color of an object, text on an object, or a graphic on an object (graphical) Object) and so on. Later, when an account matching the database is found, additional item information can be updated and added to the database to achieve the purpose of expanding the database size and accounts.

For example, as shown in FIG. 5, the isolated article 110 includes a rotated text 115A, an object shape, a size, and a back image thereof. According to Fig. 5, the rotated character 115A is matched to the horizontal character 115B stored in the database. Based on the results of the alignment, the existing database account can be updated to include the rotated text 115A, its front image, size, and front pattern 120. Thereby, the present invention can effectively update new or additional object information and add it to the database, thereby achieving the purpose of expanding the database size and its accounts.

Please refer to FIG. 6, which is a flow chart of the steps of the object recognition program according to an embodiment of the present invention.

As shown in FIG. 6, the object identification program 330 (ie, step 330 in FIG. 3) includes steps 331-334 and begins at step 331. First, in step 331, the present invention first detects and recognizes many features of the object, such as: text, color, graphics, fonts and patterns used by the text. and many more. Then, in step 332, the shape, size, and dimension of the object are further detected and recognized. Then, in step 333, the character on the product object is confirmed by optical character recognition (OCR). Finally, the identification results obtained in steps 331, 332, and 333 are compared with the object information already stored in the database in step 334 to find the result of the match.

According to an embodiment of the present invention, the result of the comparison must be compared with a preset threshold and greater than the predetermined threshold to determine whether the comparison is consistent with the database. Object information already stored.

Please refer to FIG. 7 , which is a stand-up position according to an embodiment of the present invention. Schematic diagram of the product shelf of the shelf location identification.

According to the technical features disclosed in the present invention, the product rack 150 shown in FIG. 7 is preferably used in the commodity shelf of the prior art, and can facilitate the storage and management of various commodities.

In detail, as shown in FIG. 7, the merchandise rack 150 includes a top shelf 180, a middle shelf 181, and a bottom shelf 182. The top shelf 180, the middle shelf 181, and/or the bottom shelf 182 are selectively configurable with one or more shelf position tags 160A, 160B, 160C. In accordance with an embodiment of the present invention, the on-stand position label 160A can be used to identify the location of the merchandise items 110A, 110B, 110C on the merchandise rack 150. Furthermore, the shelf position labels 160A, 160B, 160C can be further used to indicate that the merchandise rack 150 itself is located at a specific location in the warehouse.

For example, the shelf position labels 160A, 160B, 160C may include information such as the location of the merchandise rack 150 itself, whether the merchandiser rack 150 has a top shelf 180, the middle shelf 181, and/or the bottom shelf 182, and the merchandise itself. It is stored in which shelf on the shelf 150.

In general, the shelf position labels 160A, 160B, 160C can be recorded as numbers, letters, characters, colors, color gradients, measurements, barcodes. (barcode), quick response code (QR code), or other code that can be identified by image recognition technology.

In addition, as shown in FIG. 7, the on-position position tag 160C may further have at least two separate identification tags. Therefore, by using the separate identification tags on the shelf position label 160C, the present invention can not only identify the item information on the shelf 181 on the position label 160C, but also recognize the bottom shelf located under the position label 160C. Product information on 182.

Moreover, according to an embodiment of the present invention, the shelf 150 may further include a shelf width marker 170 and a shelf height marker 165. Among them, the shelf width landmark 170 is used to determine the width of the goods on the shelf, or the position of the goods on the shelf; the shelf height mark 165 is used to determine the height of the goods on the shelf, or the goods are located Which one is on the shelf. Therefore, through the design of the shelf width landmark 170 and the shelf height landmark 165, the present invention can quickly obtain various information about the commodity object.

Please refer to FIG. 8 , which is an inventory management according to an embodiment of the present invention. A flow chart of the steps of the image recognition method with marketing.

As shown in FIG. 8, the image recognition method 400 includes steps 405, 410, 415, 420, 425, 430, 435, 440, 445, 450, and begins at step 405. Initially, in step 405, the image capture device 200 first captures an image of the item on the shelf. Then, in step 410 and step 415, each item in the image is parsed and isolated. Among them, the captured image includes not only the image of the product object, but also the above-mentioned various position label on the shelf. These on-shelf position labels are identified in step 420. Thereafter, in step 425, each of the isolated items is identified by image recognition techniques and then, in step 430, linked to the corresponding on-shelf position label. Then, in step 435, the identified object features can be compared with the object information stored in the database, and after searching for the matching result, the product object and its corresponding shelf are The location tag is also stored in the account of the database.

According to an embodiment of the present invention, an expected quantity in which each commodity item can be stored is stored in the database. When a plurality of identifiable goods are placed on the product shelf, each successfully identified product object is updated and stored in the database after matching with the object information in the database, and the existing one is added. The quantity of inventory of goods. In step 440, the present invention compares the current current inventory to the expected quantity that can be stored in the database.

Next, in step 445, the present invention begins the execution of an inventory process. When the quantity of inventory in the database is updated to the latest inventory quantity, the system compares the latest inventory quantity with the expected quantity that can be stored in the database to provide sales histories of the goods item. . For example, in accordance with an embodiment of the present invention, from the inventory of the previous warehouse, it is known that the desired number of items ABC in the database that can be stored is 14. Then, if the number of the commodity objects ABC is 12 when the new inventory is executed, the inventory management system disclosed by the present invention can be automatically linked to the warning of theft or low stock.

Finally, in step 450, the present invention automatically creates a flow map based on the on-position location label of the identified object, and the process map is formed from the data account for each object. This process map shows the location of the merchandise items on the warehouse and on the merchandise shelf. Moreover, this process map can also guide the user from the current location to the merchandise object. The way.

Please refer to FIG. 9, which is a schematic structural view of a product shelf according to an embodiment of the present invention.

According to an embodiment of the present invention, if the size of the product rack 150 is known, the size of the article 110A, 110B, 110C is also determined according to the distance, width, and depth between the shelf and the shelf. Can be known. For example, if there is a shelf height 155 between the top shelf bottom 151 and the middle shelf top 152, the bottom shelf height 156, the shelf depth 157, and the shelf width between the middle shelf bottom 153 and the bottom shelf top 154 are known. Then, when the image capturing device 200 captures the image of the product, and the upper and lower shelves of the product are known, the size of the product can be estimated by the ratio of the relative shelf size. Therefore, the system disclosed in the present invention can automatically calculate the size of the product and further store the object information in the database. In addition, each of the on-shelf position labels can also be used to identify and calculate the size of the merchandise rack 150 itself.

Please refer to FIG. 10, which is a schematic structural diagram of an image recognition system for inventory management and marketing according to an embodiment of the present invention.

As shown in FIG. 10, according to the technical content disclosed in the present invention, the user, the consumer, or the manager of the warehouse can connect the commodity object 110 by the user device 200 of the image recognition management system. With database 230. The user device 200 and the database 230 are connected through a wireless communication connection 210. For example, the database 230 can display the location and the access path of the product object 110 on the user device 200. The user device 200 can extract the price or inventory quantity of the commodity object 110 from the database 230 to borrow This profit and help users to shop.

Please refer to FIG. 11 , which is an inventory management according to an embodiment of the present invention. A flow chart of the steps of the image recognition method with marketing.

In general, retail stores or dealers typically print a large number of marketing flyers, and display preferential prices and promotional items on these flyers in order to induce user consumption. However, it is worth noting that it is very inconvenient for the user to carry these flyers to the store for each consumption, or to record which products are being promoted.

In view of this, the inventory management and marketing image identification method disclosed by the present invention 500, as shown in FIG. 11, includes steps 510, 520, 530, 540, and 550. First, in step 510, the user can first capture a product image on the leaflet using a user device (eg, an image capture device). In general, the user can obtain the flyer in advance at home, and use the image capturing device at hand to capture the image of the product that the user wants to purchase. Thereafter, in step 520, the product item in the image is recognized, and the original price, discount price, brand, logo, product category of the product. ), or the store name can also be successfully identified.

Thereafter, in step 530, the location of the user is detected. For example, when the user is entering or already in the store, the inventory management and marketing image recognition system can be, for example, a global positioning satellite (GPS) or a wireless network (Wi-Fi). The location of the user device is detected by means of Bluetooth, radio frequency identification (RFID) or near field communication (NFC).

Then, in step 540, the system can capture and display the product object and its object information on the user device. Finally, in step 550, a flow map can be displayed on the user device, and the process map includes the exact location of the product item in the store and the product shelf, and the user from the current location. A convenient path to get this item.

Therefore, according to the inventory management and marketing image identification method disclosed in the present invention, the user can easily take the items on the leaflet that they want to purchase, and when they arrive at the store, they can learn more quickly. The exact location of the product increases the efficiency and convenience of shopping.

In addition, marketers of retailers or stores can also provide similar competitive or free goods and their location through this system to further stimulate users' additional consumer interest.

Please refer to FIG. 12, which is a flow chart of the steps of the image recognition method for inventory management and marketing according to another embodiment of the present invention.

As shown in FIG. 12, the inventory management and marketing image identification method 600 disclosed herein includes steps 610, 620, 630, 640, and 650. First, in step 610, the user can first capture a product image using a user device (eg, an image capture device). Thereafter, in step 620, the item of merchandise in the image is recognized. Thereafter, in step 630, the location of the user is detected. Then, in step 640, all locations where the merchandise item is sold are retrieved from the database, and each location where the merchandise item is sold is displayed on the user device. Finally, in step 650, the item information of the item can be displayed on the user device at each location.

For example, the user can take an image of the computer to be purchased in advance. After that, the system can compare the object information in the image, such as color, brand name, size, size, or model, with the information stored in the database to identify the computer. Then, when the system detects the location of the user, a map can be retrieved from the database and displayed on the user device, and all the places where the computer can be purchased are displayed on the map. In addition, the map also provides information on how to purchase all of the above computers from the user's current location. Further, the user device can further display the price of the computer sold by each location, the same type of competitive products, or the like, in order to facilitate the user to decide the final store to go to the store, or What is the final product to buy?

Please refer to FIG. 13, which is a flow chart of the steps of the image recognition method for inventory management and marketing according to an embodiment of the present invention.

As shown in FIG. 13, the inventory management and marketing image identification method 700 disclosed herein includes steps 705, 710, 715, 720, 725, 730, 735, 740, and 745. First, in step 705, the user can first capture a product image using a user device (eg, an image capture device). Thereafter, in step 710, the item of the item in the image and its object information are identified. Then, in step 715, the identified commodity objects and their object information are stored in the user's database. Thereafter, in step 720, a user wish list of the user is created based on the product items and stored in the database. In general, users can record the items they want to buy by compiling a list of interests, or open this list of interest to let other users know what they want to buy, and let others buy it as a gift. Use.

Thereafter, in step 725, the location of the user is detected. Then, in step 730, a map is displayed on the user device, and the map indicates the actual location of each item of merchandise on the list of interests. Then, in step 735, the item information of the commodity item at each location can be displayed on the user device. After that, in step 740, another map is displayed. On the user device, and the other map indicates the actual location of the competitive item for each item of merchandise on the list of interests. Finally, in step 745, the item information of the competing items at each location can be displayed on the user device.

According to an embodiment of the present invention, the user can decide whether to share his or her interest list with other users. In general, when one of the users chooses to share their interest list with other users, the other users' interest lists will receive a notification (ie, step 735 of Figure 13). Under this circumstance, the user can know at any time what the other user wants to purchase, and at the same time know the location of the nearest purchase by the map displayed on the user device. In this way, the present invention not only enhances the value of the social network and the satisfaction of giving each other gifts, but also eliminates the regional restrictions on shopping.

For example, if user A takes a picture of the product to be purchased, and the item has not yet been sold in the city where user A lives. At this time, if user A shares the product image on his own interest list and shares it with other users, if user B sells the product in user B, user B is watching After the user A's interest list, the user can purchase the product for the user A in the city where the user B lives, and mail it back to the city where the user A lives.

Therefore, in summary, the image recognition system for inventory management and marketing disclosed by the present invention and the identification method thereof can not only eliminate the regional limitation of the user's shopping, but also enhance the efficiency of international shopping and the social network. The value and the ability to extend the Internet to a commercial use effectively promotes economic development.

The embodiments described above are merely illustrative of the technical spirit and the features of the present invention, and the objects of the present invention can be understood by those skilled in the art, and the scope of the present invention cannot be limited thereto. That is, the equivalent variations or modifications made by the spirit of the present invention should still be included in the scope of the present invention.

Claims (20)

An image recognition system for inventory management and marketing, comprising: capturing an image having a plurality of products, and the products are disposed on a commodity shelf; analyzing the image; isolating the products from the image An object; identifying a position label on the shelf of the product shelf; identifying the object information of the object and its object feature; linking the object to the position label on the shelf; searching for the object in a database The data account matching the object information; updating the data account according to the object; and storing the object information of the object and the updated data account back to the database. The image recognition system for inventory management and marketing according to item 1 of the patent application scope further includes the use of optical character recognition (OCR) to identify the text on the object. The image recognition system for inventory management and marketing according to item 1 of the patent application scope further includes generating a map according to the position label on the shelf, the map indicating the actual position of the object. The image recognition system for inventory management and marketing according to item 1 of the patent application scope, wherein the position label on the shelf can be recorded as a number, a letter, a character, a color, a color. Color gradient, size (measurement), bar code, quick response code (QR code), or other code that can be recognized by image recognition technology, and the on-position location tag can be used to identify the product. The position of the shelf, any particular shelf in the shelf, and the location of the particular shelf on the shelf. The image recognition system for inventory management and marketing according to item 1 of the patent application scope includes: a shelf width landmark, which determines the width and position of the object on the shelf; and a shelf height landmark, which is determined The height of the item on the shelf and the item on which the item is placed. The image recognition system for inventory management and marketing according to item 1 of the patent application scope further includes linking other items in the image to the data account. The image recognition system for inventory management and marketing according to item 1 of the patent application scope, wherein the step of identifying the object information of the object and the object feature thereof further comprises: detecting the text, color, pattern, and text of the object. Using the font and style; detecting the shape, size and size of the object; using optical character recognition to identify the text on the object; and comparing the identification result with the information stored in the database To find the result of the match. An image recognition system for inventory management and marketing according to item 1 of the patent application scope, wherein the article information includes: a brand name, a product name, a selling price, and a cost of the article Price, existing stock level, size, shape, area or volume, weight, label, manufacturer, supplier, color, design, pattern, image, text, total cost of existing stock, product shelf label, item number, product barcode And/or other identifying features that can be used to identify the object. The image recognition system for inventory management and marketing according to claim 1 of the patent application scope includes: detecting a current location of a user device; and capturing and displaying the object and its object information according to the database; And capturing and displaying a map on the user device according to the database, the map showing the actual location of the object, and the map indicating the current location of the user device to the actual location of the object path. The image recognition system for inventory management and marketing according to claim 1 of the patent application scope includes: capturing a product image on a leaflet by using a user device; identifying one of the product images to be purchased; detecting the user The device is in a current location in the store; displaying information on the user device that matches the item to be purchased; and displaying a map on the user device, the map indicating that the item to be purchased is located in the store The actual location, and the map indicates the path from the current location of the user device to the actual location of the item to be purchased. The image recognition system for inventory management and marketing according to the scope of claim 1 further includes: capturing a user product image by using a user device; identifying one of the user product images to purchase the product; detecting the use The current location of the device; Displaying, on the user device, object information that matches the item to be purchased; and displaying a map on the user device, the map showing a plurality of locations where the item to be purchased is sold, and the map is instructed A plurality of lines from the current location of the user device to the locations where the item to be purchased is sold. The image recognition system for inventory management and marketing according to claim 1 of the patent application scope includes: capturing a user product image by using a user device; detecting product information in the user product image; from the product information Identifying a product to be purchased; storing the product to be purchased and its product information in a user account of the database; and linking the product to be purchased and the product information in the user account to the database And an object information matching the same; generating a user interest list in the database, the user interest list includes a plurality of the products to be purchased in the user account; detecting a current location of the user device; Displaying a map on the user device, the map displays the locations of the user's interest list for the items to be purchased; and displaying the product information of the desired product for each location on the user device. The image recognition system for inventory management and marketing according to claim 12 of the patent application scope further includes displaying, on the user device, a plurality of positions from the current location of the user device to the locations where the goods to be purchased are sold. path of. The image recognition system for inventory management and marketing according to item 12 of the patent application scope includes: Displaying another map on the user device, and the other map indicates the location of each of the competing items of the item to be purchased on the user interest list; and displaying the content of each position on the user device Product information of the product. The image recognition system for inventory management and marketing according to item 12 of the patent application scope further includes sharing the user interest list with other users to promote interaction and communication between other users and the user interest list. An image recognition system for inventory management and marketing, comprising: capturing an image having a plurality of products, and the products are disposed on a commodity shelf; analyzing the image; isolating the products from the image An object; identifying a position label on the shelf of the product shelf; identifying the object information of the object and its object feature; linking the object to the position label on the shelf; searching for the object in a database The data account matching the object information; updating the data account according to the object; storing the object information of the object and the updated data account back to the database; generating a map according to the location label on the shelf, The map indicates the actual location of the object; a user device is used to capture a product image on a leaflet; one of the product images is identified to be purchased; and the user device is detected in a current location in the store; Displaying, on the user device, information about the object that matches the item to be purchased; A map is displayed on the user device, the map showing the actual location of the item to be purchased in the store, and the map indicating the path from the current location of the user device to the actual location of the item to be purchased. The image recognition system for inventory management and marketing according to claim 16 of the patent application scope further includes: displaying, on the user device, complementary element data sold together with the product to be purchased; A map is displayed on the user device, and the map displays the actual location of the complementary product information. The image recognition system for inventory management and marketing according to claim 16 of the patent application scope further includes: displaying, on the user device, competing element data competing with the product to be purchased; and at the user device A map is displayed on the map, and the map displays the actual location of the competitive merchandise information. An image recognition system for inventory management and marketing, comprising: capturing an image having a plurality of products, and the products are disposed on a commodity shelf; analyzing the image; isolating the products from the image An object; identifying a position label on the shelf of the product shelf; identifying the object information of the object and its object feature; linking the object to the position label on the shelf; searching for the object in a database The information account with the object information matching; Updating the data account according to the object; storing the object information of the object and the updated data account back to the database; generating a map according to the location tag on the shelf, the map indicating the actual location of the object Using a user device to capture a user product image; identifying one of the user product images to purchase the product; detecting the current location of the user device; displaying the user device with the desired product to be purchased And displaying a map on the user device, the map showing a plurality of locations where the item to be purchased is sold, and the map indicates a plurality of lines from the current location of the user device to the There is a route to sell the location where the item is to be purchased. The image recognition system for inventory management and marketing according to claim 19 of the patent application scope includes: displaying, on the user device, competitive product information and a map competing with the product to be purchased, wherein the map shows the competition The actual location of the merchandise; and displaying on the user device complementary merchandise information and a map sold with the merchandise to be purchased, wherein the map displays the actual location of the complementary merchandise.