A warehouse goods intelligent inventory counting system and method

Abstract

The present invention discloses an intelligent inventory system and method for storage goods, and relates to the field of storage management technology. The system includes: a cargo shipping module for moving goods to a warehouse; an image acquisition module for acquiring three views of the goods entering the warehouse each time; a weighing module for measuring and recording the weighing data of the incoming pallet and the supported goods and the weighing data of the goods picked out on each pallet; a code reader for identifying the identification code information on the pallet; and a data processing module for taking inventory of the inventory goods on each pallet according to the weighing data of the incoming pallet and the supported goods, the weighing data of the goods picked out on each pallet, the current inventory data of the pallet recorded by the system, and the images and identification code information acquired by the camera. The present invention realizes accurate and rapid inventory of standardized goods such as books in a stereoscopic warehouse through a technical solution combining weighing and image recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of warehouse management, and in particular to an intelligent inventory counting system and method for warehoused goods. Background Art

[0002] The existing warehouse inventory technology, taking book inventory technology as an example, mainly includes: barcode-based inventory technology, wireless radio frequency identification RFID inventory technology, smart bookshelf inventory technology, and non-label visual recognition inventory technology. The above inventory technologies all have defects to varying degrees.

[0003] (1) Barcode-based inventory technology uses the barcode of each book as the unique identifier of the book. A handheld data collector (such as a barcode scanner) is used to scan and collect the barcode of the book. The collected data is transmitted to the computer system via a data cable or wirelessly, and compared with the data in the system database to obtain the book inventory results. Although this inventory technology has the advantages of simple operation and low cost, it often requires manual scanning one by one, which is inefficient and requires a lot of labor and time costs.

[0004] (2) Radio frequency identification (RFID) inventory technology: When books are produced, each book needs to be affixed with an RFID electronic tag, which stores the relevant information of the book. Using RFID readers and integrated devices for scanning and identification, books can be quickly counted in batches. Workers can scan directly in the warehouse using a handheld inspection instrument connected to a computer or data cable, or put the entire pallet into the scanning frame. The entire pallet of books can be counted without scanning each book. Although this inventory technology has the advantages of fast inventory speed, high accuracy, support for batch processing, and improved work efficiency, it is very expensive and requires professional RFID equipment and technical support. In the face of compactly stored books, the recognition accuracy rate is significantly reduced.

[0005] (3) Smart bookshelf inventory technology. Smart bookshelves are equipped with antennas on the bookshelf level, which can accurately read the bookshelf information on each layer. They can be set to count books automatically at regular or irregular times, or for the entire shelf or a single layer. Although this inventory technology enables real-time inventory and management of books, reduces manual intervention, and improves efficiency, it requires the library to be renovated and smart bookshelves to be installed, which is costly and not suitable for large-volume storage scenarios such as warehouses.

[0006] (4) Non-coding visual recognition inventory technology, based on advanced artificial intelligence visual algorithms, can identify and count books without any processing of the books. The books to be counted are photographed with a camera and the book information is identified through the algorithm. Although this inventory technology has high recognition accuracy, fast inventory speed, and strong environmental adaptability, it has high requirements on server performance, significantly increases costs, and has limited recognition effect on densely placed books.

[0007] The above-mentioned warehouse goods inventory counting technology has improved the inventory counting efficiency to a certain extent. However, when faced with high-density, large-scale storage commodities such as books, it still faces technical difficulties such as poor inventory counting accuracy and high costs. Summary of the invention

[0008] In view of the above-mentioned defects or deficiencies in the prior art, the present invention provides an intelligent inventory system and method for stored goods, which neither increases the cost of book products nor requires changing the existing storage layout. Through a technical solution combining weighing and image recognition, accurate and rapid inventory of standardized goods such as books in a stereoscopic warehouse is achieved.

[0009] In one aspect of the present invention, there is provided an intelligent inventory system for warehouse goods, comprising: a goods shipping module, an image acquisition module, a weighing module and a data processing module; the goods shipping module comprises a pallet and a pallet track, the pallet is used to load goods packages of a single category and standard size, each goods package comprises a fixed number of standard goods, an identification code is provided on the pallet, and a code reader for reading the identification code is provided on the pallet track; when the pallet moves along the pallet track to the position of the code reader, the code reader sends the read identification code information to the data processing module, the identification code information comprises one or more of the pallet ID, the goods category, the standard weight of the goods package, the quantity of goods corresponding to the goods package, the standard weight approved for the pallet, and the pallet in and out storage record; the image acquisition module comprises at least three modules arranged in the storage entrance area A camera is provided, wherein the camera is at least located in the front view direction, the top view direction and the side view direction of the supported goods, and is used to obtain at least the front view, the top view and the side view of the supported goods; the weighing module includes weighing sensors arranged in the warehouse entrance area and the picking area, which are used to measure and record the weighing data of the incoming pallets and the supported goods and the weighing data of the goods picked out on each pallet, and transmit them to the data processing module through the wireless network; the data processing module is used to associate and store the image obtained by the camera, the weighing data of each incoming pallet and the supported goods, the weighing data of the goods picked out on each pallet and the identification code information in a database; and the inventory of the inventory goods on each pallet is taken according to the weighing data of the incoming pallet and the supported goods, the weighing data of the goods picked out on each pallet, the current inventory data of the pallet recorded by the system, the image obtained by the camera and the identification code information.

[0010] Another aspect of the present invention further provides a method for intelligent inventory counting of stored goods based on the above-mentioned intelligent inventory counting system for stored goods, comprising the following steps:

[0011] The barcode reader identifies the identification code on the pallet to obtain identification code information, and determines whether the pallet is put into storage for the first time according to the identification code information;

[0012] If the pallet is being stored for the first time, the weight of the pallet and the cargo is measured by the weighing sensor in the storage entrance area, and at least three cameras take photos of the cargo through the same trigger signal to obtain at least the front view, side view and top view of the cargo, and the weight data, image data and pallet ID of the pallet and the cargo are associated and stored in the database; the data processing module calculates the absolute error value of the pallet and the cargo according to the following formula: and absolute error rate :

[0013]

[0014]

[0015] in, This is the weight of the pallet and the cargo when it is first put into storage. The standard weight approved for the pallet, The standard weight of each cargo package. is the number of cargo packages counted when the pallet is first put into storage; if the absolute error rate If the absolute error rate is less than the preset threshold, The corresponding absolute error value As the available error value; if the absolute error rate If the image is greater than or equal to the preset threshold, the front view, top view and side view of the pallet are retrieved, and the number of packages on the pallet is identified through the image recognition algorithm. , and the number of cargo packages identified by the image Calculate the absolute error rate again , if the absolute error rate calculated twice is and The same, then determine the absolute error value corresponding to the absolute error rate is the available error value and stores it in the database;

[0016] If the pallet is not put into storage for the first time, the weight of the pallet and the supported goods are measured by the weighing sensor in the storage entrance area, and at least three cameras take pictures of the supported goods through the same trigger signal to obtain at least the front view, side view and top view of the supported goods. The weight of the goods picked out of the pallet is measured by the weighing sensor in the picking area, and the weight data of the pallet and the supported goods, the weight data of the goods picked out of the pallet, the image data and the pallet ID are associated and stored in the database; the data processing module calculates the current inventory data of the goods on the pallet according to the following formula:

[0017]

[0018] in, The current inventory data of the goods on the calculated pallet; The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out after the pallet was last put into storage. The standard weight approved for the pallet, is the available error value of the pallet and the cargo supported. The standard weight of each cargo package is obtained; the inventory data of the cargo on the pallet recorded in the system database is compared with the calculated current inventory data of the cargo on the pallet. If there is inconsistency or the difference is greater than the preset threshold, the front view, top view and side view of the pallet when it was most recently put into storage are retrieved, and the front view, top view and side view of the supported cargo that is put into storage each time are subjected to denoising, image enhancement and image segmentation by using the image recognition algorithm, and the shape feature information of each cargo package in each view, the overall stacking shape feature information of the cargo, the number of layers of the cargo packages and the number of cargo packages in each layer are extracted; the overall stacking shape of the cargo is divided into a regular shape part and an irregular shape part according to the above feature information; the number of cargo packages in the regular shape part is calculated by the following formula:

[0019]

[0020] in, is the number of cargo packages in each layer in the length direction, is the number of cargo packages in each layer in the width direction, The number of layers of cargo packages; directly count the number of cargo packages with irregular shapes ; The number of packages of regular shaped goods and the number of packages of goods with irregular shapes The sum of the number of packages on the pallet ;

[0021] The number of packages on the pallet As the current inventory of the goods on the pallet , compare the inventory quantities calculated twice and If they are the same, the inventory is completed; if they are not the same, the current inventory data of the pallet calculated by the image recognition result is As the current inventory data of pallets after inventory verification and data update;

[0022] Recalculate the absolute error value of the pallet according to the following formula :

[0023] ;

[0024] ;

[0025] in To check the current inventory data of pallets after verification, The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out of the pallet after the most recent storage. The standard weight approved for the pallet, The standard weight of each package of goods on the pallet;

[0026] The absolute error value of the recalculated pallet As the new available error value for the pallet.

[0027] The intelligent inventory system and method for storage goods provided by the present invention realizes intelligent inventory of standard goods such as books in storage through a technical solution combining weighing and image recognition, that is, the current inventory is calculated using the weighing data of the incoming goods and the weighing data of the picked goods on each pallet, and then the calculated inventory is compared with the inventory recorded in the system. If there is a mismatch, the image of the incoming goods is called, and the number of goods on each pallet is automatically identified by image recognition, and then the number of all incoming goods is calculated, and then the current inventory is calculated based on the number of goods calculated by the image recognition result and the number of picked goods. The inventory result can be obtained by comparing the two calculated inventory quantities. The method of the present invention does not need to increase the storage cost, nor does it change the existing storage layout. By integrating storage equipment and intelligent software systems, it realizes accurate and rapid inventory of standard goods inventory, significantly reduces manual intervention, reduces the risk of human error, and improves the efficiency and accuracy of storage management. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0029] Figure 1 This is a logical composition diagram of a warehouse goods intelligent inventory counting system provided by an embodiment of the present application;

[0030] Figure 2 It is a structural diagram of a warehouse goods intelligent inventory counting system provided by an embodiment of the present application;

[0031] Figure 3 is a first flow chart of a method for intelligent inventory counting of stored goods provided by another embodiment of the present application;

[0032] Figure 4 This is the second flow chart of the intelligent inventory method for stored goods provided in another embodiment of the present application. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0035] It should be understood that although the terms first, second, third, etc. may be used to describe the acquisition modules in the embodiments of the present invention, the acquisition modules should not be limited to these terms. These terms are only used to distinguish the acquisition modules from each other.

[0036] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0037] It should be noted that the directional words such as "upper", "lower", "left", and "right" described in the embodiments of the present invention are described at the angles shown in the drawings and should not be understood as limiting the embodiments of the present invention. In addition, in the context, it should also be understood that when it is mentioned that an element is formed "on" or "under" another element, it can not only be formed directly "on" or "under" another element, but also be formed "on" or "under" another element indirectly through an intermediate element.

[0038] An embodiment of the present application provides an intelligent inventory system for stored goods, which realizes intelligent inventory of standard goods such as books in the warehouse through a technical solution combining weighing and image recognition. Standard goods in this embodiment refer to goods of the same category with fixed sizes. Several standard goods constitute a standard goods package, and each goods package has almost the same weight, size and quantity of goods. This embodiment is described by taking the inventory of stored books as an example, but the present invention is also applicable to the inventory of other similar standard goods.

[0039] See also Figure 1-2 The intelligent inventory counting system 100 for warehouse goods includes: a goods shipping module 101, an image acquisition module 102, a weighing module 103, a barcode reader 104 and a data processing module 105.

[0040] The cargo shipping module 101 includes a pallet 1011 and a pallet track 1012. The pallet 1011 is used to load cargo packages 1013 of a single category and standard size. The cargo packages 1013 can be various standard three-dimensional shapes, preferably rectangular or cube. Each cargo package 1013 includes a fixed number of books of a fixed category. An identification code 1014 is set on the pallet 1011, such as a barcode or a QR code. The identification code 1014 can be set on the side of the pallet to record information such as the pallet ID, cargo category, standard weight of the cargo package, the number of cargo corresponding to the cargo package, the standard weight of the pallet, and the pallet in and out of storage records.

[0041] The pallet track 1012 is used to provide a movement path for the pallet 1011, so that the pallet 1011 moves out of or into the warehouse along the pallet track 1012. Optionally, the pallet 1011 is moved by the operator pulling it, or the pallet 1011 is driven by a driving mechanism. Optionally, the pallet track 1012 can be a track line similar to a rail, a conveyor belt conveyor line, a powered roller line, or a non-powered roller line. A code reader 104 for reading the identification code 1014 is provided on the pallet track 1012. Preferably, the code reader 104 is provided in the vicinity of the warehouse entrance. More preferably, the code reader 104 is provided near the weighing module 103. In this way, while weighing the pallet 1011, the identification code information in the identification code 1014 can be read by the code reader 104. The identification code information includes information such as the pallet ID, the book category, the standard weight of the cargo package, the number of books corresponding to the cargo package, the standard weight of the pallet, and the pallet in and out of the warehouse. It should be pointed out that the pallet of the present embodiment is used as both a book shipping tool and a shelf in the warehouse, so the inventory of the stock books of the present invention is actually an inventory of the books or cargo packages (book packages) on each pallet in the warehouse.

[0042] The image acquisition module 102 includes: at least three cameras 1021 arranged in the storage entrance area. Taking the three cameras 1021 as an example, they are arranged in three directions, namely, the front view direction, the top view direction and the side view direction of the supported books, for acquiring the front view, the top view and the side view of the supported books. Furthermore, the three cameras 1021 are arranged on a fixing device 1022, which is preferably arranged in the storage entrance area. The fixing device 1022 may also be provided with an illumination light source 1023 for auxiliary shooting. Preferably, a weighing module 103 is provided below the fixing device 1022, and the tray 1011 and the tray track 1012 can pass through the bottom of the fixing device 1022. In this way, when the tray 1011 moves to the bottom of the fixing device 1022, the three cameras 1021 can take pictures of the supported books on the incoming tray 1011 through the same trigger signal, thereby obtaining the front view, top view and side view of the supported books at the same position. At this time, the tray 1011 is also located on the weighing module 103, and the weighing module 103 obtains the weighing data of the tray 1011 and the supported books. When the tray 1011 passes, the code reader 104 on the tray track 1012 near the weighing module 103 will also read the identification code information of the identification code 1014. The acquired image, weighing data and identification code information are sent to the data processing module 105 via the communication device.

[0043] Furthermore, this embodiment determines the placement positions and placement parameters of the three cameras according to the size of the warehouse entrance, the size of the pallet, and the size of the supported goods. Preferably, the side-view camera 1021 is mounted on the side of the pallet track 1012, 900 mm away from the side of the pallet track 1012, using a 12-megapixel industrial camera and a 6 mm lens, and the field of view on the side of the pallet track 1012 reaches 2100 mm * 1540 mm. Preferably, the overhead camera 1021 and the front camera 1021 are mounted on the top and front of the pallet track 1012, and a 12-megapixel camera is used to take a panoramic view of the books supported by the pallet, wherein the overhead camera 1021 reaches a field of view of 2100 mm * 1540 mm on the upper surface of the goods.

[0044] The weighing module 103 includes weighing sensors arranged in the warehouse entrance area and the picking area in the warehouse, which are used to measure and record the weighing data of the incoming pallet 1011 and the supported books and the weighing data of the goods picked out on each pallet, and transmit them to the data processing module 105 through the wireless network. Since the present embodiment is a three-dimensional warehouse, there are many warehouse entrances in the three-dimensional warehouse. Considering factors such as cost and installation difficulty, it is impossible to set weighing equipment at all warehouse entrances. Such practical factors make it impossible to directly calculate the current inventory data of the goods on the pallet by subtracting the weight of the outgoing books from the weight of the incoming books. Therefore, the present embodiment sets a high-precision weighing sensor in the picking area in the warehouse, and all the pallets and supported books that are pre-out of the warehouse in the warehouse are weighed here. The total weight of the goods packages 1013 picked out on the pallet is weighed each time, and the number of the picked goods packages 1013 is recorded, so that the number of books in the current inventory can be obtained by subsequent indirect calculation.

[0045] The data processing module 105 can be an industrial computer, a server or other computing device with data storage or database functions, on which the warehouse management system (WMS) software is run. The data processing module 105 is used to associate and store the images acquired by the camera 1021, the weighing data of each incoming pallet 1011 and the books supported, the weighing data of the goods picked out on each pallet, and the identification code information into a database, so that the database records the pallet ID of each pallet 1011 when it enters the warehouse, the weighing data of the books on the pallet, the category information of the books on the pallet, the standard weight information of the pallet, the three-view drawing of the pallet when it enters the warehouse, the standard weight information of the goods packaging, the picking quantity of such books, and other data, which is convenient for the subsequent calculation and inventory of the inventory of various types of books. More preferably, the warehouse management system fuses and associates the received data, outputs it through the communication protocol, and saves the goods image and related data through the FTP protocol.

[0046] Furthermore, the data processing module 105 periodically or according to the operator's instructions, uses the weighing data of the incoming goods and the weighing data of the picked-out goods on each pallet to calculate the current inventory of each pallet, and then compares the calculated inventory with the inventory recorded in the system. If there is a match, the incoming image of the pallet is called, and the number of cargo packages on each pallet is automatically identified by image recognition, and then the number of all incoming goods is calculated. The current inventory is calculated based on the number of goods calculated by the image recognition result and the number of picked-out goods, and the inventory result can be obtained by comparing the two calculated inventory quantities.

[0047] See also Figure 3-4 Another embodiment of the present application also provides a method for intelligent inventory counting of stored goods based on the above-mentioned intelligent inventory counting system 100 for stored goods. This method does not need to increase the storage cost and does not change the existing storage layout. By integrating storage equipment and intelligent software systems, it realizes accurate and rapid inventory counting of standard goods inventory, significantly reduces manual intervention, reduces the risk of human errors, and improves the efficiency and accuracy of warehouse management.

[0048] This embodiment still takes the inventory of warehouse books as an example for explanation. There is no absolute order relationship among the steps in this embodiment. The step sequence numbers are only used to distinguish the steps, not to limit the order of the steps. The intelligent inventory method for warehouse goods includes the following steps:

[0049] Step S101: Scan the code to identify whether the pallet is entering the warehouse for the first time

[0050] Specifically, when the pallet runs along the pallet track to under the fixing device near the entrance, the barcode reader on the pallet track recognizes the identification code on the pallet to obtain the identification code information, and determines whether the pallet is entering the warehouse for the first time based on the pallet entry and exit records in the identification code information.

[0051] Step S102: If the pallet is put into storage for the first time, the weighing data and image data of the pallet are obtained and the absolute error value of the pallet is calculated.

[0052] Specifically, the weighing sensor of the weighing module near the entrance measures the weighing data of the incoming pallet and the supported books, at least three cameras take pictures of the supported books through the same trigger signal to obtain at least the front view, side view and top view of the supported books on the pallet, and the code reader identifies the identification code to obtain the identification code information. The above-obtained data is transmitted to the data processing module through the communication device. The data processing module obtains the incoming weighing data, the image taken by the camera and the identification code information, and stores them in a database in association, so as to form the pallet, the category of the supported books, the image of the supported books, the weight of the supported books, the standard weight of the pallet, the standard weight of the cargo packaging and other related information in the database.

[0053] After long-term use, the outer packaging of pallets and cargo packages will be worn or attached with foreign objects, which will affect the data of each weighing. Since books are frequently put into and taken out of the warehouse, although the error amount each time is not large, the long-term accumulation of errors will also lead to errors in inventory calculation. Therefore, before checking the inventory books, the available error value must be calculated first. Specifically, the data processing module obtains the relevant data of the pallet and a certain category of supporting books (for example: magazines, collections of papers or other different categories of books) when they are first put into the warehouse from the database, and calculates the absolute error value of the pallet and the supporting books through the following formula: and absolute error rate :

[0054]

[0055]

[0056] in, This is the weight of the pallet and the cargo when it is first put into storage. The standard weight approved for the pallet, The standard weight of each cargo package. is the number of cargo packages counted when the pallet is first put into storage; if the absolute error rate If the absolute error rate is less than the preset threshold, The corresponding absolute error value As the available error value; if the absolute error rate If the image is greater than or equal to the preset threshold, the front view, top view and side view of the pallet are retrieved, and the number of packages on the pallet is identified through the image recognition algorithm. , and the number of cargo packages identified by the image Calculate the absolute error rate again , if the absolute error rate calculated twice is and The same, then determine the absolute error value corresponding to the absolute error rate The available error value is stored in the database.

[0057] Step S103: if the pallet is not put into storage for the first time, the image data of the pallet is obtained, the books picked out of the pallet are weighed, and then the inventory of the goods on the pallet is counted.

[0058] Specifically, if the pallet is not entering the warehouse for the first time, the weight of the incoming pallet and the supported goods is measured by a weighing sensor in the warehouse entrance area, at least three cameras take pictures of the supported goods through the same trigger signal to obtain at least a front view, a side view and a top view of the supported goods, and the weight of the goods picked off the pallet is measured by a weighing sensor in the picking area. The weight data of the incoming pallet and the supported goods, the weight data of the goods picked off the pallet, the image data and the pallet ID are associated and stored in the database.

[0059] The data processing module calculates the current inventory data of the goods on the pallet according to the following formula:

[0060]

[0061] in, The current inventory data of the goods on the calculated pallet; The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out after the pallet was last put into storage. The standard weight approved for the pallet, is the available error value of the pallet and the cargo supported. The standard weight of each cargo package is obtained; the inventory data of the cargo on the pallet recorded in the system database is compared with the calculated current inventory data of the cargo on the pallet. If there is inconsistency or the difference is greater than the preset threshold, the front view, top view and side view of the pallet when it was most recently put into storage are retrieved, and the front view, top view and side view of the supported cargo that is put into storage each time are subjected to denoising, image enhancement and image segmentation by using the image recognition algorithm, and the shape feature information of each cargo package in each view, the overall stacking shape feature information of the cargo, the number of layers of the cargo packages and the number of cargo packages in each layer are extracted; the overall stacking shape of the cargo is divided into a regular shape part and an irregular shape part according to the above feature information; the number of cargo packages in the regular shape part is calculated by the following formula:

[0062]

[0063] in, is the number of cargo packages in each layer in the length direction, is the number of cargo packages in each layer in the width direction, The number of layers of cargo packages; directly count the number of cargo packages with irregular shapes ; The number of packages of regular shaped goods and the number of packages of goods with irregular shapes The sum of the number of packages on the pallet ;

[0064] The number of packages on the pallet As the current inventory data of the goods on the pallet , compare the inventory quantities calculated twice and If they are the same, the inventory is completed; if they are not the same, the current inventory data of the pallet calculated by the image recognition result is The current inventory data of the pallets after inventory verification is updated.

[0065] Step S104: if the current inventory of the pallet calculated based on the weighing data is incorrect, the available error value of the pallet is recalculated.

[0066] Specifically, the current inventory data of the pallet is calculated through the image recognition results The most reliable one is used as the current inventory data of the pallet after the inventory check. If the current inventory data of the pallet after the inventory check is different from the current inventory data of the pallet calculated based on the weighing data or the difference exceeds the predetermined threshold, it means that the calculation result of the weighing data is wrong, which is most likely caused by the accumulated error in the calculation process, that is, the available error value of the pallet is inaccurate, and the reasons include but are not limited to the wear of the pallet, the increase of pallet attachments, etc., so the available error value of the pallet needs to be recalculated.

[0067] This step recalculates the absolute error value of the pallet according to the following formula :

[0068]

[0069] ;

[0070] in To check the current inventory data of pallets after verification, The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out of the pallet after the most recent storage. The standard weight approved for the pallet, The standard weight of each package of goods on the pallet;

[0071] The absolute error value of the recalculated pallet As the new available error value for the pallet, it will be used for subsequent inventory verification calculations.

[0072] It should be pointed out that one of the innovations of this embodiment is the introduction of the above-mentioned absolute error value and its dynamic adjustment mechanism for the absolute error value. This measure has played a vital role in the precise weighing and inventory process of the warehouse. Faced with the complex variable challenges commonly found in warehouse management, such as the inevitable weight differences between pallets, the weight changes caused by the replacement of book versions, and the weight fluctuations caused by the diversity of packaging materials, traditional technologies are often difficult to deal with effectively, and need to rely on cumbersome and error-prone algorithm rules for correction one by one. This embodiment cleverly takes the absolute error value into consideration and designs a set of absolute error adjustment mechanisms that are flexible and adaptable to actual scenarios. It can not only effectively integrate and uniformly handle the errors caused by the above-mentioned various uncontrollable variables, but also significantly improve the accuracy and efficiency of weighing and inventory counting. By dynamically adjusting the absolute error value and its related parameters, the system can automatically adapt to error fluctuations under different conditions, ensuring high-precision weight measurement and recording even in a changing environment.

[0073] The above description is only a preferred embodiment of the present invention. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but also should cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) to form a technical solution.

Claims

1. An intelligent inventory system for stored goods, characterized in that: include: Cargo shipping module, image acquisition module, weighing module and data processing module; The cargo shipping module includes a pallet and a pallet track. The pallet is used to load cargo packages of a single category and standard size. Each cargo package includes a fixed number of standard cargoes. An identification code is provided on the pallet. A code reader for reading the identification code is provided on the pallet track. When the pallet moves along the pallet track to the location of the code reader, the code reader sends the read identification code information to the data processing module. The identification code information includes one or more of the pallet ID, cargo category, standard weight of the cargo package, the number of cargoes corresponding to the cargo package, the approved standard weight of the pallet, and the pallet in and out storage record. The image acquisition module includes at least three cameras arranged in the warehouse entrance area, the cameras are at least located in the front view direction, the top view direction and the side view direction of the supported goods, and are used to obtain at least the front view, the top view and the side view of the supported goods; The weighing module includes weighing sensors arranged in the warehouse entrance area and the picking area, which are used to measure and record the weighing data of the warehouse pallets and the supported goods and the weighing data of the goods picked out on each pallet, and transmit them to the data processing module through the wireless network; The data processing module is used to: The images captured by the camera, the weighing data of each incoming pallet and the goods carried, the weighing data of the goods picked out on each pallet, and the identification code information are associated and stored in the database; Obtain the relevant data of each pallet and the cargo when it is first put into storage from the database, and calculate the absolute error value of each pallet and the cargo by the following formula and absolute error rate : in, This is the weight of the pallet and the cargo when it is first put into storage. The standard weight approved for the pallet, The standard weight of each cargo package. The number of cargo packages counted when the pallet is first put into storage; If the absolute error rate If the absolute error rate is less than the preset threshold, The corresponding absolute error value As the available error value; If the absolute error rate If the value is greater than or equal to the preset threshold, the front view, top view and side view of the cargo carried by the pallet when it is first put into storage are retrieved, and the number of cargo packages when the pallet is first put into storage is identified through the image recognition algorithm. , and the number of cargo packages identified by the image Calculate the absolute error rate again , if the absolute error rate calculated twice is and The same, then determine the absolute error value corresponding to the absolute error rate is the available error value; Get relevant data from the database and calculate the current inventory data of each pallet using the following formula: in, The current inventory data of the goods on each pallet is calculated; The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out after the pallet was last put into storage. The standard weight approved for the pallet, is the available error value of the pallet and the cargo supported. The standard weight of each cargo package; Compare the inventory data of each pallet recorded in the system database with the calculated current inventory data of each pallet. If there is any inconsistency or the difference is greater than the preset threshold, retrieve the front view, top view and side view of the pallet when it was last put into storage, and identify the number of packages on the pallet through the image recognition algorithm. , the number of packages on the pallet As the current inventory of the goods on the pallet ; Comparing the inventory levels of two calculations and Are they the same? If they are the same, the inventory is completed. If they are not the same, the current inventory data of the pallet calculated by the image recognition result is used as the verified current inventory data of the pallet and the data is updated; Recalculate the absolute error value of the pallet according to the following formula : ； ； in To check the current inventory data of pallets after verification, The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out of the pallet after the most recent storage. The standard weight approved for the pallet, The standard weight of each package of goods on the pallet; The absolute error value of the recalculated pallet As the new available error value for the pallet.

2. The intelligent inventory counting system for stored goods according to claim 1 is characterized in that: Also includes: A fixing device is arranged in the storage entrance area, the at least three cameras are arranged on the fixing device, the weighing module is arranged below the fixing device, and the pallet and the pallet track pass through below the fixing device.

3. The intelligent inventory counting system for stored goods according to claim 2 is characterized in that: The image acquisition module further includes a light source arranged on the fixing device.

4. The intelligent inventory counting system for stored goods according to claim 3 is characterized in that: The identification code is a barcode or a QR code.

5. The intelligent inventory counting system for stored goods according to claim 4 is characterized in that: The at least three cameras take photos of the supported goods on the incoming pallet through the same trigger signal.

6. The intelligent inventory counting system for stored goods according to claim 5 is characterized in that: The data processing module is further used for: The front view, top view and side view of the supported goods are de-noised, enhanced and segmented by an image recognition algorithm, and the shape feature information of the goods package, the overall stacking shape feature information of the goods, the number of layers of the goods package and the number of goods packages in each layer are extracted in each view; the overall stacking shape of the goods is segmented into a regular shape part and an irregular shape part according to the above feature information; The number of packages of regular-shaped goods is calculated by the following formula : in, is the number of cargo packages in each layer in the length direction, is the number of cargo packages in each layer in the width direction, The number of layers of cargo packaging; Directly count the number of irregular shaped cargo packages ; The number of packages of regular shaped goods and the number of packages of goods with irregular shapes The sum is the total number of packages on the pallet.

7. A method for intelligent inventory counting of stored goods based on the intelligent inventory counting system for stored goods according to any one of claims 1 to 5, characterized in that: The steps include: The barcode reader identifies the identification code on the pallet to obtain identification code information, and determines whether the pallet is put into storage for the first time according to the identification code information; If the pallet is being stored for the first time, the weight of the pallet and the cargo is measured by the weighing sensor in the storage entrance area, and at least three cameras take photos of the cargo through the same trigger signal to obtain at least the front view, side view and top view of the cargo, and the weight data, image data and pallet ID of the pallet and the cargo are associated and stored in the database; the data processing module calculates the absolute error value of the pallet and the cargo according to the following formula: and absolute error rate : in, This is the weight of the pallet and the cargo when it is first put into storage. The standard weight approved for the pallet, The standard weight of each cargo package. is the number of cargo packages counted when the pallet is first put into storage; if the absolute error rate If the absolute error rate is less than the preset threshold, The corresponding absolute error value As the available error value; if the absolute error rate If the image is greater than or equal to the preset threshold, the front view, top view and side view of the pallet are retrieved, and the number of packages on the pallet is identified through the image recognition algorithm. , and the number of cargo packages identified by the image Calculate the absolute error rate again , if the absolute error rate calculated twice is and The same, then determine the absolute error value corresponding to the absolute error rate is the available error value and stores it in the database; If the pallet is not put into storage for the first time, the weight of the pallet and the supported goods are measured by the weighing sensor in the storage entrance area, and at least three cameras take pictures of the supported goods through the same trigger signal to obtain at least the front view, side view and top view of the supported goods. The weight of the goods picked out of the pallet is measured by the weighing sensor in the picking area, and the weight data of the pallet and the supported goods, the weight data of the goods picked out of the pallet, the image data and the pallet ID are associated and stored in the database; the data processing module calculates the current inventory data of the goods on the pallet according to the following formula: in, The current inventory data of the goods on the calculated pallet; The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out after the pallet was last put into storage. The standard weight approved for the pallet, is the available error value of the pallet and the supported goods, The standard weight of each cargo package is obtained; the inventory data of the cargo on the pallet recorded in the system database is compared with the calculated current inventory data of the cargo on the pallet. If there is inconsistency or the difference is greater than the preset threshold, the front view, top view and side view of the pallet when it was most recently put into storage are retrieved, and the front view, top view and side view of the supported cargo that is put into storage each time are subjected to denoising, image enhancement and image segmentation by using the image recognition algorithm, and the shape feature information of each cargo package in each view, the overall stacking shape feature information of the cargo, the number of layers of the cargo packages and the number of cargo packages in each layer are extracted; the overall stacking shape of the cargo is divided into a regular shape part and an irregular shape part according to the above feature information; the number of cargo packages in the regular shape part is calculated by the following formula: in, is the number of cargo packages in each layer in the length direction, is the number of cargo packages in each layer in the width direction, The number of layers of cargo packages; directly count the number of cargo packages with irregular shapes ; The number of packages of regular shaped goods and the number of packages of goods with irregular shapes The sum of the number of packages on the pallet ; The number of packages on the pallet As the current inventory of the goods on the pallet , compare the inventory quantities calculated twice and If they are the same, the inventory is completed; if they are not the same, the current inventory data of the pallet calculated by the image recognition result is As the current inventory data of pallets after inventory verification and data update; Recalculate the absolute error value of the pallet according to the following formula : ； ； in To check the current inventory data of pallets after verification, The weight of the pallet and the cargo it carries when it was last put into storage. The weight of the goods picked out of the pallet after the most recent storage. The standard weight approved for the pallet, The standard weight of each package of goods on the pallet; The absolute error value of the recalculated pallet As the new available error value for the pallet.

Images