Image recognition-based book sorting method, device, medium and equipment

Abstract

The present application relates to the technical field of book logistics sorting, in particular to a book sorting method, device, medium and equipment based on image recognition, obtaining a to-be-sorted book order, the to-be-sorted book order including an order number and book information; the book information including a book code; configuring a seeding frame according to the order number; obtaining a seeding frame number; collecting a to-be-sorted book image, identifying the to-be-sorted book image through a convolutional neural network model algorithm to obtain a target book code; if the target book code is consistent with the book code; obtaining the seeding frame number, and conveying the to-be-sorted book corresponding to the target book code to the seeding area where the seeding frame is located through a belt conveyor; when the to-be-sorted book reaches the seeding area, opening the seeding frame door, and sorting the book into the seeding frame, which can solve the problem of low book sorting efficiency in the prior art, and improve the sorting efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of book logistics sorting technology, and in particular to a book sorting method, apparatus, medium and equipment based on image recognition. Background Technology

[0002] The sorting system is a crucial part of warehouse operations, especially in book warehouses, where its operation significantly impacts management and distribution efficiency.

[0003] In book warehousing operations management, book sorting and outbound processing is typically completed at the warehouse's distribution center. For orders awaiting shipment, staff print out the orders and then match each book with the corresponding item in the warehouse, sorting them into the appropriate packages. This process heavily relies on manual visual identification of books, verification of order information, and manual allocation of books to their respective orders. This method is inefficient and prone to errors, leading to missed or incorrect shipments. Furthermore, existing technologies also use scanners to identify book barcodes for sorting, but this requires precise book placement and precise barcode angles, making the operation inconvenient and cumbersome.

[0004] Therefore, finding a suitable book sorting method is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] In view of the technical problems of low efficiency and error in book sorting in the existing book warehouse, the present invention provides a book sorting method, device, medium and equipment based on image recognition, which can improve the efficiency and accuracy of book sorting in the book storage system.

[0006] Firstly, this application provides an image recognition-based book sorting method, applied to a warehouse book sorting system, comprising:

[0007] Obtain the book order to be sorted, wherein the book order includes an order number and book information; the book information includes a book code;

[0008] Configure the seeding frame according to the order number; obtain the seeding frame number;

[0009] Images of books to be sorted are collected, and the images are identified using a convolutional neural network model algorithm to obtain the target book code.

[0010] If the target book code is the same as the book code;

[0011] Then, obtain the seeding frame number, and transport the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor;

[0012] When the books to be sorted arrive at the sowing area, the sowing frame cover door is opened.

[0013] Furthermore, prior to the step of obtaining the book order to be sorted, the following steps are included:

[0014] Retrieve all pending book orders, wherein the pending book orders include order numbers and book information, wherein the book information includes book codes and book classification numbers;

[0015] Filter out the order numbers that have the same book category number among different book orders to be processed, and obtain the order number set;

[0016] Initiate a general picking request based on the book orders to be picked corresponding to the order number set; and obtain the general picking request information;

[0017] The total picking request information summarizes the book codes and quantities included in the book orders pending processing in the order number set;

[0018] The pending book orders in the overall picking request are treated as pending book orders for sorting.

[0019] Furthermore, the step of acquiring images of books to be sorted, identifying the images of books to be sorted using a convolutional neural network model algorithm, and obtaining the target book code includes:

[0020] Obtain images of books to be sorted, including book cover images; preprocess the images of books to be sorted to obtain preprocessed book images.

[0021] Feature extraction is performed on the preprocessed book image to obtain a convolutional layer;

[0022] The output of the convolutional layer is downsampled to obtain a pooling layer;

[0023] The convolutional layer and the pooling layer are connected to form a fully connected layer. The book code corresponding to the book image to be sorted is predicted and output based on the fully connected layer to obtain the target book code.

[0024] Further, the step of obtaining the seeding frame number and conveying the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor includes:

[0025] Obtain the seeding frame number to determine the location of the seeding area containing the seeding frame number;

[0026] The location of the acquisition component is obtained, the acquisition component including a camera, the camera being used to acquire images of the books to be sorted;

[0027] Plan the transport path for the books to be sorted based on the location of the sowing area and the collection location;

[0028] According to the conveying path, the books to be sorted corresponding to the target book code are conveyed to the sowing area where the sowing frame is located via a belt conveyor.

[0029] Furthermore, the step of opening the seeding frame cover when the books to be sorted arrive at the seeding area includes:

[0030] The books to be sorted are transported according to the transport path;

[0031] The detection device installed in the sowing area monitors whether the books to be sorted are transported to the sowing area. The detection device includes at least one of weight detection, visual detection, distance detection and sensor detection.

[0032] When the book to be sorted is detected to have arrived at the sowing area, an opening command is sent to open the sowing frame cover door of the sowing wall, and the sowing frame cover door is opened in response to the opening command.

[0033] Further, the step of obtaining the seeding frame number and determining the location of the seeding area where the seeding frame number is located includes:

[0034] The seeding wall number is obtained based on the seeding frame number, and the seeding wall is provided with multiple seeding frames;

[0035] The location of the sowing area is obtained based on the sowing frame number and the sowing wall number.

[0036] Furthermore, it also includes:

[0037] The algorithm identifies the images of books to be sorted based on a convolutional neural network model, outputs the target book code predicted from the images of books to be sorted, and outputs the probability value corresponding to the target book code predicted from the images of books to be sorted. If the probability value is lower than a preset value, it is determined that the book to be sorted corresponding to the image of the book to be sorted does not belong to the warehouse books.

[0038] Secondly, this application provides a book sorting device based on image recognition, comprising:

[0039] The acquisition module acquires book orders to be sorted, which include order numbers and book information; the book information includes book codes.

[0040] The configuration module configures the seeding frame according to the order number; and obtains the seeding frame number.

[0041] The acquisition module acquires images of books to be sorted, identifies the images of books to be sorted using a convolutional neural network model algorithm, and obtains the target book code.

[0042] The judgment module determines if the target book code matches the book code.

[0043] The conveying module then obtains the seeding frame number and conveys the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor.

[0044] The sowing module opens the sowing frame cover when the books to be sorted arrive at the sowing area.

[0045] Thirdly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the image recognition-based book sorting method as described in any of the first aspects.

[0046] Fourthly, this application provides a computer device including a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the image recognition-based book sorting method as described in any of the first aspects.

[0047] The book sorting method, apparatus, medium, and equipment based on image recognition provided by the present invention have the following advantages: First, an order for books to be sorted is obtained, including an order number and book information; the book information includes a book code. Second, a sorting frame is configured according to the order number. Third, a sorting frame number is obtained. Fourth, the order is bound to a sorting frame on the sorting wall. Fifth, images of the books to be sorted are collected, allowing for flexible placement of the books. The images of the books to be sorted are identified using a convolutional neural network model algorithm to obtain the target book code. If the target book code matches the book code, the sorting frame number is obtained. Sixth, the books corresponding to the target book code are transported to the sorting area where the sorting frame is located via a belt conveyor. Seventh, books are automatically transported within the area through automated sorting. When the books to be sorted arrive at the sorting area, the sorting frame cover is opened, enabling quick and accurate sorting of the books into the designated sorting frame, thus improving sorting efficiency and accuracy. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating a book sorting method based on image recognition in one embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram of a book sorting method based on image recognition in one embodiment of the present invention;

[0051] Figure 3 This is a schematic diagram of a book sorting device based on image recognition in one embodiment of the present invention;

[0052] Figure 4 This is a structural block diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0053] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0054] It should be noted that when a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly set on the other component; when a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to the other component.

[0055] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" or "several" means two or more, unless otherwise explicitly specified.

[0056] It should be noted that the structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which this application can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size should still fall within the scope of the technical content disclosed in this application, provided that they do not affect the effects and purposes that this application can produce.

[0057] In book warehousing operations management, book sorting and outbound processing is typically completed at the warehouse's distribution center. For orders awaiting shipment, staff print out the orders and then match them one by one with the books in the warehouse, sorting them into the corresponding packages according to the order. This process heavily relies on manual visual identification of books, verification of order information, and manual allocation of books to their respective orders. This method is inefficient and prone to errors, leading to missed or incorrect shipments. Furthermore, existing technologies also use scanners to identify book barcodes for sorting, but this requires precise book placement and barcode angle adjustments, making the operation inconvenient and cumbersome. Additionally, in some book warehouses, certain books...

[0058] For example, in a medium-sized book warehouse, the number of books leaving the warehouse daily ranges from 5,000 to 10,000. Manual sorting is labor-intensive, and prolonged operation can lead to a decrease in the accuracy of book sorting and high maintenance costs. Therefore, finding a suitable book sorting method is a technical problem that urgently needs to be solved by those skilled in the art.

[0059] In existing book warehouses, there are two methods for book sorting. The first is to manually identify the book type by eye, then check the order information and manually assign the book to which order it belongs. The second is to scan the barcode of the book, identify its type, and then use a drive to assign the book to which order it belongs. The first manual sorting method is inefficient, and the probability of error is greatly affected by the worker's supervisor's attitude. The second automated barcode scanning sorting method is faster and more accurate than the first manual method. However, in some book warehouses, some books have invalid or blank barcodes, or different types of books use the same barcode. In these cases, barcode scanning for sorting is ineffective. This poses a significant challenge to book sorting.

[0060] This application integrates the order system, data acquisition and recognition system, belt conveyor diversion system and sorting wall system in book warehouse operations into an image recognition-based book sorting system, and provides an image recognition-based book sorting method to improve the efficiency and accuracy of book sorting.

[0061] In one embodiment, see Figure 1 and Figure 2 As shown, this application provides a book sorting method based on image recognition, applied to a warehouse book sorting system, including:

[0062] Step 201: Obtain the book order to be sorted, wherein the book order includes an order number and book information; the book information includes a book code;

[0063] Specifically, the book order to be sorted includes an order number, with each order having a unique order number for easy tracking and management. The order also includes book information, including book codes and quantities. For example, in order number DDBH111, there are 10 books with book code TSBM111 and 20 books with book code TSBM112. The book information also includes book titles and publishers, further improving sorting accuracy.

[0064] The system can simultaneously acquire multiple book orders to be sorted. Sorting personnel pick books from multiple orders in the warehouse in the same batch and transport the books picked in the same batch to the sorting area. The sorting area is equipped with a belt conveyor and a data acquisition and identification device, which includes an industrial camera.

[0065] Step 202: Configure the seeding frame according to the order number; obtain the seeding frame number;

[0066] Specifically, a sorting frame is configured for each book order to be sorted, thus binding the book order to the sorting frame, improving sorting accuracy and preventing the wrong books from being sent and / or missing books from the book order to be sorted.

[0067] Step 203: Acquire images of books to be sorted, identify the images of books to be sorted using a convolutional neural network model algorithm, and obtain the target book code;

[0068] Specifically, images of the books to be sorted are captured by a data acquisition and recognition device, which includes an industrial camera.

[0069] Step 204: If the target book code matches the book code;

[0070] Specifically, if the target book code predicted by the convolutional neural network model algorithm matches the book code in the book order to be sorted, then the identified book to be sorted corresponds to the book in the book order to be sorted.

[0071] The process involves placing books picked in the same batch into the image acquisition and recognition area, and then placing the books to be sorted into the image acquisition and recognition area to capture images of the books to be sorted. The image acquisition and recognition area can be set on the belt pulley of the conveyor belt.

[0072] Step 205: Obtain the seeding frame number, and transport the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor;

[0073] Specifically, it can automatically transport the books to be sorted to the corresponding sowing area of ​​the sowing frame.

[0074] Step 206: When the books to be sorted arrive at the sowing area, open the sowing frame cover door;

[0075] Specifically, after opening the seeding frame cover, the sorting personnel will drop the books to be sorted from the seeding area into the seeding frame.

[0076] In one embodiment, prior to the step of obtaining the book order to be sorted, the following is included:

[0077] Step 101: Obtain all pending book orders, which include order numbers and book information, including book codes and book classification numbers;

[0078] Specifically, books with the same book code also have the same book classification number; books with the same book classification number do not necessarily have the same book code; for example, the book classification number of the book with book code TSBM111 and the book classification number of the book with book code TSBM112 are both book classification number TSFLH 001.

[0079] Step 102: Filter out the order numbers that have the same book category number among different book orders to be processed, and obtain the order number set;

[0080] Specifically, books with the same category number are placed in the same area. To facilitate picking books from multiple orders, the order numbers with the same category number among different orders are filtered out. For example, both order number DDBH113 and order number DDBH114 have books with category number TSFLH 001 to be picked. By filtering out the order numbers with the same category number among different orders, they are grouped into a set of order numbers. The picking staff then picks books from the warehouse in the same batch according to the set of order numbers, thus improving the efficiency of the picking system.

[0081] Of course, the selection criteria for the order number set can include not only book classification numbers, but also book codes, or the actual storage location of the books in the warehouse.

[0082] Step 103: Initiate a general picking request based on the book orders to be picked according to the order number set; and obtain the general picking request information;

[0083] Step 104: The total picking request information summarizes the book codes and quantities contained in the book orders pending processing in the order number set;

[0084] Specifically, sorting personnel pick books from the warehouse based on the book codes and quantities in the order orders, thereby improving the picking efficiency of the same batch of book orders.

[0085] Step 105: Treat the pending book orders in the overall picking request as pending book orders for sorting.

[0086] Specifically, after picking books with order numbers, sorting personnel transport the books to the sorting area. In the sorting area, data acquisition and identification devices and belt conveyors are used to automatically sort the books with order numbers, thereby improving sorting efficiency and accuracy.

[0087] In one embodiment, the step of acquiring images of books to be sorted, identifying the images of books to be sorted using a convolutional neural network model algorithm, and obtaining the target book code includes:

[0088] Step 2041: Obtain images of books to be sorted, including book cover images. Preprocess the images of books to be sorted to obtain preprocessed book images.

[0089] Step 2042: Extract features from the preprocessed book image to obtain a convolutional layer;

[0090] Step 2043: Downsample the output of the convolutional layer to obtain a pooling layer;

[0091] Step 2044: Connect the convolutional layer and the pooling layer to obtain a fully connected layer, and predict and output the book code corresponding to the book image to be sorted based on the fully connected layer to obtain the target book code.

[0092] In this embodiment, automatic sorting is specifically performed through image recognition, supporting the identification and sorting of any type of book in the warehouse. The book can be placed at any angle with its cover facing the recognition device. The recognition speed is fast. Simply place the book in the recognition area, and the convolutional neural network model algorithm will identify the image of the book to be sorted, thereby obtaining the target book code of the image of the book to be sorted. If the target book code matches the book code in the order to be sorted, the book will be automatically transported to the sorting area corresponding to the order to be sorted via a belt conveyor.

[0093] In one embodiment, the step of obtaining the seeding frame number and conveying the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor includes:

[0094] Step 2051: Obtain the seeding frame number and determine the location of the seeding area where the seeding frame number is located;

[0095] Step 2052: Obtain the location of the acquisition component, wherein the acquisition component includes a camera, and the camera is used to acquire images of the books to be sorted;

[0096] Step 2053: Plan the transport path for the books to be sorted based on the location of the sowing area and the collection location;

[0097] Step 2054: According to the conveying path, the book to be sorted corresponding to the target book code is conveyed to the sowing area where the sowing frame is located via a belt conveyor.

[0098] In this embodiment, specifically, sorting efficiency is improved by planning the transport path of the books to be sorted. Several seeding frames in close proximity can correspond to the same seeding area.

[0099] It should be noted that after installing the camera, you need to configure the camera's focal length, exposure, and other shooting parameters to ensure that the captured images are clear.

[0100] In one embodiment, the step of opening the seeding frame cover door when the books to be sorted arrive at the seeding area includes:

[0101] Step 2061: Convey the books to be sorted according to the conveying path;

[0102] Step 2062: Monitor whether the books to be sorted are transported to the sowing area by a detection device set in the sowing area. The detection device includes at least one of weight detection, visual detection, distance detection and sensor detection.

[0103] Step 2063: When the book to be sorted is detected to have arrived at the sowing area, an opening command is sent to open the sowing frame cover door of the sowing wall, and the sowing frame cover door is opened in response to the opening command.

[0104] In this embodiment, specifically, when the books to be sorted arrive at the corresponding sowing area, the sowing box cover automatically opens, making it convenient for sorting personnel to put the books into the sowing box. It also makes it easier for sorting personnel to distinguish which sowing box to put the books into when managing several sowing boxes at the same time, thus improving sowing efficiency.

[0105] In one embodiment, the step of obtaining the seeding frame number and determining the location of the seeding area where the seeding frame number is located includes:

[0106] Step 20511: Obtain the seeding wall number according to the seeding frame number, wherein the seeding wall is provided with multiple seeding frames;

[0107] Step 20512: Obtain the location of the sowing area based on the sowing frame number and the sowing wall number.

[0108] In this embodiment, the seeding wall can be composed of vertical supports or cabinets, and the seeding wall is provided with multiple layers of grooves to place the seeding frames.

[0109] In one embodiment, the image recognition-based book sorting method of this application further includes:

[0110] The algorithm identifies the images of books to be sorted based on a convolutional neural network model, outputs the target book code predicted from the images of books to be sorted, and outputs the probability value corresponding to the target book code predicted from the images of books to be sorted. If the probability value is lower than a preset value, it is determined that the book to be sorted corresponding to the image of the book to be sorted does not belong to the warehouse books.

[0111] In this embodiment, specifically, when sorting books that are not in the warehouse, a prompt will be displayed indicating that the book does not belong to this warehouse.

[0112] In one embodiment, see Figure 3 As shown, this application provides a book sorting device based on image recognition, comprising:

[0113] The acquisition module 100 acquires the book order to be sorted, which includes the order number and book information;

[0114] The book information includes the book code;

[0115] Configuration module 200 configures the seeding frame according to the order number; and obtains the seeding frame number;

[0116] The acquisition module 300 acquires images of books to be sorted, identifies the images of books to be sorted through a convolutional neural network model algorithm, and obtains the target book code.

[0117] The judgment module 400 determines if the target book code matches the book code.

[0118] The conveying module 500 then obtains the seeding frame number and conveys the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor.

[0119] The sowing module 600 opens the sowing frame cover door when the books to be sorted arrive at the sowing area.

[0120] The book sorting device based on image recognition provided by the present invention has the following advantages: It acquires book orders to be sorted, including order numbers and book information; the book information includes book codes; it configures sorting frames according to the order numbers; it obtains sorting frame numbers; it binds each order to a sorting frame on the sorting wall; it collects images of books to be sorted, allowing for flexible book placement; it identifies the images of books to be sorted using a convolutional neural network model algorithm to obtain the target book code; if the target book code matches the book code; it obtains the sorting frame number; and it transports the book to be sorted corresponding to the target book code to the sorting area where the sorting frame is located via a belt conveyor; it automatically sorts books within the area and transports them automatically; when the books arrive at the sorting area, it opens the sorting frame cover, enabling quick and accurate sorting of books into the designated sorting frames, thus improving sorting efficiency and accuracy.

[0121] In one embodiment, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the image recognition-based book sorting method as described in any of the first aspects.

[0122] Specifically, perform the following steps:

[0123] Step 201: Obtain the book order to be sorted, wherein the book order includes an order number and book information; the book information includes a book code;

[0124] Step 202: Configure the seeding frame according to the order number; obtain the seeding frame number;

[0125] Step 203: Acquire images of books to be sorted, identify the images of books to be sorted using a convolutional neural network model algorithm, and obtain the target book code;

[0126] Specifically, images of the books to be sorted are captured by a data acquisition and recognition device, which includes an industrial camera.

[0127] Step 204: If the target book code matches the book code;

[0128] Step 205: Then, obtain the seeding frame number and transport the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor.

[0129] Step 206: When the books to be sorted arrive at the sowing area, open the sowing frame cover door.

[0130] In one embodiment, see Figure 4As shown, this application provides a computer device including a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the image recognition-based book sorting method as described in any of the first aspects.

[0131] Specifically, perform the following steps:

[0132] Step 201: Obtain the book order to be sorted, wherein the book order includes an order number and book information; the book information includes a book code;

[0133] Step 202: Configure the seeding frame according to the order number; obtain the seeding frame number;

[0134] Step 203: Acquire images of books to be sorted, identify the images of books to be sorted using a convolutional neural network model algorithm, and obtain the target book code;

[0135] Specifically, images of the books to be sorted are captured by a data acquisition and recognition device, which includes an industrial camera.

[0136] Step 204: If the target book code matches the book code;

[0137] Step 205: Then, obtain the seeding frame number and transport the book to be sorted corresponding to the target book code to the seeding area where the seeding frame is located via a belt conveyor.

[0138] Step 206: When the books to be sorted arrive at the sowing area, open the sowing frame cover door.

[0139] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0140] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0141] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A book sorting method based on image recognition, characterized by, The application is applied to a warehouse book sorting system, comprising: Obtaining a to-be-sorted book order, the to-be-sorted book order comprising an order number and book information; the book information comprising a book code; Configuring a sowing frame according to the order number; obtaining a sowing frame number; Collecting a to-be-sorted book image, identifying the to-be-sorted book image through a convolutional neural network model algorithm, and obtaining a target book code; If the target book code is consistent with the book code; Obtaining the sowing frame number, and conveying the to-be-sorted book corresponding to the target book code to a sowing area where the sowing frame is located through a belt conveyor; When the to-be-sorted book reaches the sowing area, opening a sowing frame door; Before the step of obtaining the to-be-sorted book order, comprising: Obtaining all to-be-operated book orders, the to-be-operated book orders comprising order numbers and book information, the book information comprising book codes and book classification numbers; Screening out order numbers with the same book classification numbers among different to-be-operated book orders, and obtaining an order number set; Initiating a general sorting request according to the to-be-operated book orders corresponding to the order number set; and obtaining general sorting request information; The general sorting request information summarizes the book codes and the number of books contained in the to-be-operated book orders in the order number set; Taking the to-be-operated book orders in the general sorting request as to-be-sorted book orders; The step of obtaining the sowing frame number and conveying the to-be-sorted book corresponding to the target book code to the sowing area where the sowing frame is located through the belt conveyor, comprising: Obtaining the sowing frame number, and obtaining a sowing area position where the sowing frame number is located; Obtaining a collection position of a collection component, the collection component comprising a camera, the camera being used for collecting the to-be-sorted book image; Planning a to-be-sorted book conveying path according to the sowing area position and the collection position; Conveying the to-be-sorted book corresponding to the target book code to the sowing area where the sowing frame is located through the belt conveyor according to the conveying path; The step of opening the sowing frame door when the to-be-sorted book reaches the sowing area, comprising: Conveying the to-be-sorted book according to the conveying path; Monitoring whether the to-be-sorted book is conveyed to the sowing area through a detection device arranged in the sowing area, the detection device comprising at least one of weight detection, visual detection, distance detection, and sensor detection; When it is detected that the to-be-sorted book reaches the sowing area, sending an opening instruction of the sowing frame door of the sowing wall, and opening the sowing frame door in response to the opening instruction; The step of obtaining the sowing frame number and obtaining the sowing area position where the sowing frame number is located, comprising: Obtaining a sowing wall serial number according to the sowing frame number, the sowing wall being provided with a plurality of sowing frames; Obtaining a sowing area position according to the sowing frame number and the sowing wall serial number.

2. The image recognition-based book sorting method according to claim 1, characterized by, The step of collecting a to-be-sorted book image, identifying the to-be-sorted book image through a convolutional neural network model algorithm, and obtaining a target book code, comprising: An image of a book to be sorted is acquired, the image of the book to be sorted including a book cover image, the image of the book to be sorted is preprocessed to obtain a preprocessed book image; Feature extraction is performed on the preprocessed book image to obtain a convolution layer; Down-sampling is performed on the output of the convolution layer to obtain a pooling layer; The convolution layer and the pooling layer are connected to obtain a fully connected layer, and a book code corresponding to the image of the book to be sorted is predicted based on the fully connected layer to obtain a target book code.

3. The image recognition-based book sorting method according to claim 1, characterized by, Further comprising: According to the convolutional neural network model algorithm, the image of the book to be sorted is identified, and the target book code predicted based on the image of the book to be sorted is output, and the probability value corresponding to the target book code predicted based on the image of the book to be sorted is output, if the probability value is lower than the preset value, it is determined that the book to be sorted corresponding to the image of the book to be sorted does not belong to the warehouse book.

4. A book sorting apparatus based on image recognition, characterized by, The device comprises: An acquisition module acquires a book order to be sorted, the book order to be sorted including an order number and book information; the book information including a book code; A configuration module configures a sowing frame according to the order number to obtain a sowing frame number; A collection module collects an image of a book to be sorted, and identifies the image of the book to be sorted through a convolutional neural network model algorithm to obtain a target book code; A judgment module, if the target book code is consistent with the book code; A conveying module, then acquires the sowing frame number, and conveys the book to be sorted corresponding to the target book code to a sowing area where the sowing frame is located through a belt conveyor; A sowing module, when the book to be sorted reaches the sowing area, opens the sowing frame door.

5. A computer readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to make the processor execute the steps of the image recognition-based book sorting method according to any one of claims 1 to 3.

6. Computer device comprising a memory and a processor, characterized in that The memory stores a computer program, and the computer program is executed by the processor to make the processor execute the steps of the image recognition-based book sorting method according to any one of claims 1 to 3.

Images