Inventory management method and device based on image intelligent identification, equipment and medium

Abstract

The invention discloses an inventory management method, device and equipment based on intelligent image recognition and a medium, and the method comprises the steps: constructing a to-be-checked information base according to the current inventory statistical information, recognizing whether an input shot image is a copied image, segmenting the shot image if the input shot image is not the copied image, and recognizing to obtain image recognition information, checking whether the image identification information is consistent with certain article information in the information base to be checked, if the checking result is consistent, judging whether the shooting position information corresponding to the image identification information is the same as the article information, and if the shooting position information is the same as the article information, performing verification on the information base to be verified; and if not, converting the article information into an abnormal information base, and if the to-be-checked information base contains the article information which is not cancelled after verification, generating image input prompt information. According to the method, the input shot image can be identified and checked with the article information, so that the inventory can be quickly sorted, and the inventory management efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an inventory management method, device, equipment and medium based on image intelligent recognition. Background Art

[0002] In the field of supply chain management and inventory control, inventory counting is a vital task. Ensuring the accuracy and timeliness of inventory directly affects the operational efficiency of the enterprise. However, the traditional manual inventory counting method has many shortcomings. First, manual auditing is prone to errors, resulting in inaccurate inventory counting results, which in turn affects the decision-making of the enterprise. For example, employees may miss certain items or record incorrect quantities under high-intensity work conditions, which is particularly obvious when the inventory volume is large. Secondly, the traditional method is not only cumbersome to operate, but also time-consuming. It often takes hours or even days to complete a full inventory, which seriously affects the work efficiency of the enterprise. Furthermore, the traditional inventory counting method lacks systematic data archiving, which makes it impossible to effectively save the inventory counting results and difficult to conduct subsequent analysis and improvement. Therefore, the inventory management method in the prior art method has the problem of low management efficiency. Summary of the invention

[0003] The embodiments of the present invention provide an inventory management method, device, equipment and medium based on image intelligent recognition, aiming to solve the problem of low management efficiency existing in the inventory management method in the prior art.

[0004] In a first aspect, an embodiment of the present invention provides an inventory management method based on image intelligent recognition, wherein the method comprises:

[0005] Receive the inventory command input by the user and build the corresponding information database to be checked according to the current inventory statistics;

[0006] Identify whether the input captured image is a re-photographed image according to a preset re-photographed recognition model;

[0007] If the captured image is not a re-shot image, segmenting the captured image according to a preset image segmentation rule to obtain a corresponding segmented image;

[0008] Recognize the segmented image according to a preset image recognition model to obtain corresponding image recognition information;

[0009] According to a preset information verification strategy, the image recognition information and the corresponding segmented image are verified to see whether they are consistent with the information of a certain item in the information library to be verified, and a corresponding verification result is obtained;

[0010] If the verification result is consistent with a certain item information, determine whether the shooting position information corresponding to the image recognition information is the same as the verified item information to obtain a corresponding determination result;

[0011] If the judgment result is the same, the to-be-verified information database is cancelled according to the verification result;

[0012] If the judgment result is different, the item information that is consistent is transferred to the abnormal information database;

[0013] If the to-be-verified information database contains unverified item information, image input prompt information is generated to prompt the user to input the captured image again.

[0014] In a second aspect, an embodiment of the present invention further provides an inventory management device based on image intelligent recognition, wherein the device is used to execute the inventory management method based on image intelligent recognition as described in the first aspect above, and the device includes:

[0015] The to-be-checked information database generating unit is used to receive the inventory command input by the user and construct the corresponding to-be-checked information database according to the current inventory statistical information;

[0016] A re-shot image recognition unit, used to recognize whether the input captured image is a re-shot image according to a preset re-shot recognition model;

[0017] An image segmentation unit, configured to segment the captured image according to a preset image segmentation rule to obtain a corresponding segmented image if the captured image is not a reproduced image;

[0018] An image recognition information acquisition unit, used to recognize the segmented image according to a preset image recognition model to obtain corresponding image recognition information;

[0019] A verification result acquisition unit is used to verify whether the image recognition information and the corresponding segmented image are consistent with the information of a certain item in the information library to be verified according to a preset information verification strategy, and obtain a corresponding verification result;

[0020] a judgment result acquisition unit, configured to, if the verification result is consistent with a certain item information, determine whether the shooting position information corresponding to the image recognition information is the same as the verified item information, and obtain a corresponding judgment result;

[0021] A verification unit, configured to, if the judgment result is the same, verify the to-be-verified information base according to the verification result;

[0022] A transfer unit, configured to transfer the verified consistent item information to an abnormal information database if the judgment result is different;

[0023] The prompt information generating unit is used to generate image input prompt information to prompt the user to input the captured image again if the to-be-verified information library contains unverified item information.

[0024] In a third aspect, an embodiment of the present invention further provides a computer device, wherein the device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0025] Memory, used to store computer programs;

[0026] The processor is used to implement the steps of the inventory management method based on image intelligent recognition described in the first aspect when executing the program stored in the memory.

[0027] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the inventory management method based on image intelligent recognition as described in the first aspect above are implemented.

[0028] The embodiment of the present invention provides an inventory management method, device, equipment and medium based on image intelligent recognition, the method includes: constructing a to-be-verified information library according to current inventory statistical information, identifying whether the input photographed image is a re-photographed image, if it is not a re-photographed image, segmenting and identifying the photographed image to obtain image recognition information, verifying whether the image recognition information is consistent with a certain item information in the to-be-verified information library, if the verification result is consistent, judging whether the shooting location information corresponding to the image recognition information is the same as the item information, if they are the same, verifying the to-be-verified information library, if they are not the same, converting the item information to the abnormal information library, and generating image input prompt information if the to-be-verified information library contains unverified item information. The above inventory management method can identify the input photographed image and verify it with the item information, thereby realizing rapid sorting of the inventory, and at the same time, archiving the data in the verification process through the to-be-verified information library and the abnormal information library, greatly improving the inventory management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0030] Figure 1 A method flow chart of an inventory management method based on image intelligent recognition provided by an embodiment of the present invention;

[0031] Figure 2 A schematic block diagram of an inventory management device based on image intelligent recognition provided by an embodiment of the present invention;

[0032] Figure 3 It is a schematic block diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0035] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0036] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0037] See also Figure 1, as shown in the figure, the embodiment of the present invention application provides an inventory management method based on image intelligent recognition. The method is applied in a user terminal or a management server. The method is executed by an application installed in the user terminal or the management server. The user terminal can be a laptop, desktop computer, tablet computer or mobile phone and other devices; the management server is also a server side configured inside an enterprise or organization for managing commodity inventory. The management server can exchange information with the client. When the method is applied to the user terminal, the user can directly input data information in the terminal and manage commodity inventory locally through the user terminal; when the method is applied to the management server, the management server can provide services to the client, and the client sends data information to the management server. The management server executes the above method and manages commodity inventory on the management server, and feeds back the inventory verification results to the client. Figure 1 As shown, the method includes steps S110 to S180.

[0038] S110: receiving an inventory command input by a user, and constructing a corresponding information library to be checked according to current inventory statistical information.

[0039] After receiving the inventory command input by the user, a corresponding information library to be checked is constructed according to the current inventory statistics. The user can input the inventory command to the user terminal or the management server. When the user inputs the inventory command, the inventory consolidation process is entered. At this time, the user terminal can construct the corresponding information library to be checked based on the current inventory statistics stored locally. Generally speaking, a set of current inventory statistics corresponds to a warehouse, and the user terminal is usually used to manage the inventory of a local warehouse; the current inventory statistics is the information obtained by counting the current stored items. The current inventory statistics and the items stored in the warehouse may not completely match, so inventory verification is required. The user can also send an inventory command to the management server through the client. The management server obtains the current inventory statistics corresponding to the warehouse identifier according to the warehouse identifier in the inventory command and constructs the corresponding information library to be checked; the management server can simultaneously manage the inventory of multiple warehouses in different regions, so a warehouse that needs to be sorted can be specified through the inventory command for subsequent inventory sorting operations.

[0040] S120: Identify whether the input captured image is a re-shot image according to a preset re-shot identification model.

[0041] According to the preset remake recognition model, it is determined whether the input captured image is a remake image. Furthermore, the captured image can be collected and input, and the captured image has shooting location information. The shooting location information can be the positioning coordinate value for image shooting. The positioning coordinate value can be obtained in real time through the GPS chip, Beidou navigation chip and other positioning chips that come with the terminal device. During the specific implementation of this application, only real-time shooting and uploading of images are allowed to prevent remake phenomena.

[0042] If a photographed image is received, the photographed image can be identified by the copy recognition model to determine whether the photographed image is a copy image. Specifically, the photographed image can be convolved by the convolutional neural network in the copy recognition model to extract the image features of the photographed image, and the image features are input into the intelligent neural network in the copy recognition model to obtain the probability value of the photographed image being a copy image. If the probability value input by the intelligent neural network exceeds the preset probability threshold, the photographed image is identified as a copy image; if the probability value input by the intelligent neural network does not exceed the preset probability threshold, the photographed image is identified as not a copy image.

[0043] S130: If the captured image is not a reproduced image, segment the captured image according to a preset image segmentation rule to obtain a corresponding segmented image.

[0044] If the photographed image is not a re-photographed image, the photographed image is segmented according to a preset image segmentation rule to obtain a corresponding segmented image. If the photographed image is not a re-photographed image, the photographed image can be segmented according to the image segmentation rule to obtain a segmented image. The photographed image can contain the machine box serial code pattern to be identified, so the machine box serial code pattern can be located at the boundary to obtain the boundary vertices. Usually, the machine box serial code pattern is a rectangular pattern, and the corresponding boundary information can be determined by sequentially connecting the boundary vertices; the photographed image is segmented according to the boundary information obtained by locating, so as to obtain a segmented image corresponding to the machine box serial code pattern.

[0045] If the captured image is a duplicate image, an image error prompt message is generated to prompt the user that the captured image currently input is unqualified.

[0046] S140: Recognize the segmented image according to a preset image recognition model to obtain corresponding image recognition information.

[0047] The segmented image is identified according to a preset image recognition model to obtain corresponding image recognition information. Further, the segmented image obtained by the image recognition model can be identified to obtain corresponding string code characters as image recognition information.

[0048] In a specific embodiment, step S140 includes sub-steps: obtaining a recognition rule in the image recognition model corresponding to the barcode format of the segmented image; and recognizing the segmented image according to the recognition rule to obtain corresponding image recognition information.

[0049] Specifically, the recognition rules corresponding to the barcode format of the segmented image in the image recognition model may be obtained. The barcode format of the segmented image may be a QR code, a barcode or other format types. Different barcode formats correspond to different recognition rules.

[0050] The segmented image is identified according to the corresponding identification rules, thereby obtaining image identification information. Among them, the segmented image can be positioned according to the identification rules to determine the direction of adjusting the segmented image; for example, if the barcode format is a QR code, the segmented image needs to be rotated 90 degrees to the left, 90 degrees to the right, or 180 degrees to correspond to the standard direction in the identification rules, then the segmented image can be rotated according to the standard direction. Afterwards, the information in the rotated segmented image can be identified. For example, for a QR code image, the grayscale of the squares can be identified to obtain binary information of "0" or "1", and the binary information is further converted into a string code character, that is, the corresponding image identification information is obtained; for a barcode image, the bar lines can be identified and the sequentially arranged characters can be obtained, and the sequentially arranged characters can be combined to obtain the corresponding string code characters as image identification information.

[0051] S150, performing information verification on whether the image recognition information and the corresponding segmented image are consistent with information of an item in the to-be-verified information library according to a preset information verification strategy, and obtaining a corresponding verification result.

[0052] According to the preset information verification strategy, the image recognition information and the corresponding segmented image are checked for consistency with the information of a certain item in the information library to be verified, and the corresponding verification result is obtained. The image recognition information corresponds to the segmented image, and the image recognition information and the segmented image can be checked for consistency with the information of a certain item in the information library to be verified, wherein the information verification can be performed through the information verification strategy to obtain a verification result of whether they are consistent.

[0053] In a specific embodiment, step S150 includes the sub-steps of: checking whether the image recognition information is consistent with the serial code in any of the item information; if the image recognition information is consistent with the serial code in one of the item information, obtaining a consistent verification result; if the image recognition information is not consistent with the serial code in any of the item information, performing an image comparison between the segmented image and the item information in the information library to be verified according to a preset image comparison model to obtain a consistent verification result.

[0054] The information library to be checked contains multiple item information, each of which includes at least item name, item specification parameters, serial number, inventory location and other information, and each item uniquely corresponds to a set of serial numbers; specifically, it can be determined whether the image recognition information is consistent with the serial number of any item information. If it is consistent, it indicates that the currently acquired captured image can match a set of item information, and a consistent verification result is obtained. If the image recognition information does not match the serial number of any item information, the image recognition information may be inaccurate. The segmented image can be further compared with the item information in the information library to be checked according to the image comparison model, and whether the verification result is consistent can be determined according to the comparison result.

[0055] In a specific embodiment, the image comparison between the segmented image and the item information in the to-be-checked information library is performed according to a preset image comparison model, including: generating barcode graphics corresponding to the serial codes of each item information according to the barcode format corresponding to the segmented image; extracting segmented image features corresponding to the segmented image and barcode graphic features corresponding to the barcode graphics according to the convolutional neural network in the image comparison model; combining the segmented image features with each of the barcode graphic features and inputting them into the similarity analysis network in the image comparison model to obtain the similarity value corresponding to the segmented image features and each of the barcode graphic features; judging whether the barcode graphic feature with the highest similarity value is significant according to the significance judgment rule in the image comparison model, so as to determine whether the item information corresponding to the barcode graphic feature with the highest similarity value is consistent with the segmented image.

[0056] Specifically, a barcode graphic corresponding to the serial code of each item information in the information library to be verified can be generated according to the barcode format of the segmented image; that is, the serial code is converted into a corresponding barcode graphic based on the barcode format, and the serial code of each item information corresponds to a barcode graphic, and the size of the generated barcode graphic can be the same as the size of the segmented image.

[0057] According to the convolutional neural network in the image contrast model, the pixels contained in the segmented image are convoluted to obtain a corresponding set of output information as the segmented image features. The convolutional neural network is a neural network used to perform convolution operations on the pixel values ​​of the pixels in the image, and the convolutional neural network is configured with a convolution kernel. The segmented image features obtained can be a two-dimensional array, and the values ​​contained in the two-dimensional array are much smaller than the number of pixels contained in the segmented image. Based on the same processing process, the barcode graphic can be input into the convolutional neural network to extract the barcode graphic features recorded in the form of a two-dimensional array.

[0058] Furthermore, the obtained segmented image features can be combined with each barcode graphic feature respectively, and the number of combined features obtained after the combination is equal to the number of barcode graphic features, and each combined feature contains the segmented image features and a group of barcode graphic features. Each group of combined features is input into the similarity analysis network to obtain the corresponding similarity value. Among them, the similarity analysis network can be a neural network constructed based on artificial intelligence. The similarity analysis network includes input nodes, an intermediate layer and an output node. The input node is used to input the numerical value in the two-dimensional array, and each input node is used to input a corresponding numerical value. If the numerical values ​​in the segmented image features and the barcode graphic features are both N, the similarity analysis network needs to be configured with 2N input nodes to simultaneously input the numerical values ​​in a group of combined features. The middle layer in the similarity analysis network is configured between the input node and the output node. The middle layer is used to build the association relationship between the input node and the output node. Multiple middle nodes can be configured in the middle layer. The middle node is used to establish the association relationship between the upstream node (which can be the input node or the upstream middle node) and the downstream node (which can be the output node or the downstream middle node). Each middle node is configured with a different weight coefficient, and the weight coefficient is used to perform an association operation on the upstream input value. An output node can be configured in the similarity analysis network, and the output node is used to output a similarity value. The similarity value is used to reflect the similarity between the segmented image feature and the barcode graphic feature in a set of combined features. The larger the similarity value, the more similar the segmented image feature is to the barcode graphic feature. The similarity value between each barcode graphic feature and the segmented image feature can be obtained by the above method.

[0059] Furthermore, whether the barcode graphic feature with the highest similarity value is significant can be determined according to the significance judgment rules in the image comparison model. If the barcode graphic feature with the highest similarity value is relatively significant, it can be determined that the item information corresponding to the barcode graphic feature with the highest similarity value is consistent with the segmented image; if the barcode graphic feature with the highest similarity value is not significant, it is determined that the corresponding item information is inconsistent with the segmented image.

[0060] In a specific embodiment, judging whether the barcode graphic feature with the highest similarity value is significant according to the significance judgment rule in the image contrast model, so as to determine whether the item information corresponding to the barcode graphic feature with the highest similarity value is consistent with the segmented image, includes: determining that the barcode graphic corresponding to the barcode graphic feature with the highest similarity value is the target barcode graphic; calculating the significance coefficient corresponding to the standard deviation value of the similarity value with the non-target barcode graphic according to the significance coefficient calculation formula in the significance judgment rule; judging whether the overall standard deviation value of the similarity value of the barcode graphic feature is greater than the significance coefficient, and obtaining a judgment result of whether the barcode graphic feature with the highest similarity value is significant.

[0061] Specifically, it can be determined that the barcode graphic corresponding to the barcode graphic feature with the highest similarity value is the target barcode image, and the other barcode graphics are non-target barcode graphics. The standard deviation of the similarity values ​​of the non-target barcode graphics is further calculated, that is, the standard deviation of the similarity values ​​of all non-target barcode graphics is calculated. For example, after excluding the similarity value of the target barcode graphic, the standard deviation of the similarity values ​​of all other non-target barcode graphics is 3.712. The significant coefficient corresponding to the standard deviation of the similarity value is calculated according to the significant coefficient calculation formula. For example, the significant coefficient calculation formula can be shown as formula (1):

[0062] r z =√2×r p (1);

[0063] Among them, r z is the calculated significant coefficient, r p is the standard deviation of the similarity values.

[0064] It is further determined whether the overall standard deviation of the similarity values ​​of the barcode graphic features is greater than the significance coefficient obtained above. If the overall standard deviation is greater than the significance coefficient, it indicates that the barcode graphic feature with the highest similarity value is significant; if the overall standard deviation is not greater than the significance coefficient, it indicates that the barcode graphic feature with the highest similarity value is not significant.

[0065] S160: If the verification result is consistent with a certain item information, determine whether the shooting position information corresponding to the image recognition information is the same as the verified and consistent item information, and obtain a corresponding determination result.

[0066] If the verification result is consistent with a certain item information, the shooting position information corresponding to the image recognition information is judged to be the same as the verified item information, thereby obtaining a corresponding judgment result. If the verification result is consistent with a certain item information, the shooting position information corresponding to the image recognition information can be obtained, and it is judged whether the shooting position information is the same as the verified item information, thereby obtaining a corresponding judgment result.

[0067] In a specific embodiment, step S160 includes sub-steps: obtaining geographic information corresponding to the shooting location information; and verifying whether the geographic information is the same as the inventory location in the item information to obtain a judgment result of whether the shooting location information is the same as the item information.

[0068] Specifically, the shooting location information can be converted into corresponding geographic information according to the correspondence between the coordinate value and the location. For example, the positioning coordinate value "AA, BB" in the shooting location information (AA is the longitude coordinate value, BB is the latitude coordinate value) can be converted into the geographic information "Guangzhou-Tianhe" according to the correspondence between the coordinate value and the location. It is further verified whether the obtained geographic information is the same as the inventory location in the verified and consistent item information. If the geographic information is the same as the inventory location, the judgment result that the shooting location information is the same as the item information is obtained; if the geographic information is not the same as the inventory location, the judgment result that the shooting location information is not the same as the item information is obtained.

[0069] In a specific embodiment, after step S150, the method further includes: if the verification result is inconsistent, generating a prompt message of verification abnormality.

[0070] If the verification result is inconsistent, it means that the information library to be verified does not contain the item information corresponding to the captured image. At this time, a prompt message of verification abnormality can be generated to remind the user, and the user can verify the item information by himself.

[0071] S170: If the judgment result is the same, the to-be-checked information database is cancelled according to the verification result.

[0072] If the judgment result is the same, the to-be-checked information library is cancelled according to the verification result. If the judgment result is the same, the corresponding item information in the to-be-checked information library can be cancelled according to the verification result, that is, the item information corresponding to the verification result is deleted from the to-be-checked information library. At the same time, a verification mark can be added to the position corresponding to the item information in the current inventory statistics information. The verification mark includes information such as the verification time and the verification operator. By adding the verification mark, the verification result of the item information with the same judgment result can be recorded for easy subsequent viewing.

[0073] S180: If the judgment result is different, the item information that is verified to be consistent is transferred to the abnormal information library.

[0074] If the judgment result is different, the verified item information is transferred to the abnormal information library. If the judgment result is different, the verified item information can be transferred to the abnormal information library, that is, the item information is deleted from the information library to be verified, and the item information is added to the abnormal information library for storage, so as to continue to verify whether the inventory position of the item information is incorrect.

[0075] S190: If the to-be-verified information database contains unverified item information, generate image input prompt information to prompt the user to input the captured image again.

[0076] If the information to be verified contains unverified item information, image input prompt information is generated to prompt the user to input the captured image again. Further, after executing the above steps, it can be determined whether the information to be verified still contains unverified item information. If it does, image input prompt information can be generated. After receiving the image input prompt information, the user can re-photograph the box serial code image of other items in the warehouse, thereby obtaining the input captured image and cyclically executing the above verification process again (after receiving the input captured image, return to step S120). If the information to be verified does not contain unverified item information, it indicates that all item information in the information to be verified has completed preliminary verification, and there is no need to generate image input prompt information at this time.

[0077] In the inventory management method based on image intelligent recognition disclosed in the above embodiment, the method includes: constructing a to-be-verified information library according to the current inventory statistical information, identifying whether the input photographed image is a re-photographed image, if it is not a re-photographed image, segmenting and identifying the photographed image to obtain image recognition information, verifying whether the image recognition information is consistent with a certain item information in the to-be-verified information library, if the verification result is consistent, judging whether the shooting location information corresponding to the image recognition information is the same as the item information, if they are the same, verifying the to-be-verified information library, if they are not the same, converting the item information to the abnormal information library, and generating image input prompt information if the to-be-verified information library contains unverified item information. The above inventory management method can identify the input photographed image and verify it with the item information, thereby realizing rapid sorting of the inventory, and at the same time, archiving the data in the verification process through the to-be-verified information library and the abnormal information library, greatly improving the inventory management efficiency.

[0078] The embodiment of the present invention also provides an inventory management device based on image intelligent recognition, which can be configured in a user terminal or a management server, and is used to execute any embodiment of the aforementioned inventory management method based on image intelligent recognition. Figure 2 , Figure 2 A schematic block diagram of an inventory management device based on image intelligent recognition provided by an embodiment of the present invention.

[0079] like Figure 2 As shown, the inventory management device 100 based on image intelligent recognition includes a to-be-verified information library generation unit 110, a copied image recognition unit 120, an image segmentation unit 130, an image recognition information acquisition unit 140, a verification result acquisition unit 150, a judgment result acquisition unit 160, a cancellation unit 170, a transfer unit 180 and a prompt information generation unit 190.

[0080] The to-be-checked information library generating unit 110 is configured to receive an inventory instruction input by a user and construct a corresponding to-be-checked information library according to current inventory statistical information.

[0081] The re-shot image recognition unit 120 is used to recognize whether the input captured image is a re-shot image according to a preset re-shot recognition model.

[0082] The image segmentation unit 130 is configured to segment the captured image according to a preset image segmentation rule to obtain a corresponding segmented image if the captured image is not a reproduced image.

[0083] The image recognition information acquisition unit 140 is used to recognize the segmented image according to a preset image recognition model to obtain corresponding image recognition information.

[0084] The verification result acquisition unit 150 is used to verify whether the image recognition information and the corresponding segmented image are consistent with the information of a certain item in the information library to be verified according to a preset information verification strategy, and obtain a corresponding verification result.

[0085] The judgment result acquisition unit 160 is used to judge whether the shooting position information corresponding to the image recognition information is the same as the verified and consistent object information if the verification result is consistent with a certain object information, so as to obtain a corresponding judgment result.

[0086] The verification unit 1770 is used to verify the information base to be verified according to the verification result if the judgment result is the same.

[0087] The transfer unit 180 is used to transfer the verified consistent item information to the abnormal information library if the judgment result is different.

[0088] The prompt information generating unit 190 is used to generate image input prompt information to prompt the user to input the captured image again if the to-be-verified information library contains unverified item information.

[0089] The inventory management device based on image intelligent recognition provided in the embodiment of the present invention applies the above-mentioned inventory management method based on image intelligent recognition, builds a to-be-verified information library according to the current inventory statistical information, identifies whether the input photographed image is a re-photographed image, and if it is not a re-photographed image, segments and identifies the photographed image to obtain image recognition information, verifies whether the image recognition information is consistent with a certain item information in the to-be-verified information library, and if the verification result is consistent, determines whether the shooting location information corresponding to the image recognition information is the same as the item information, and if they are the same, the to-be-verified information library is verified, and if they are not the same, the item information is converted to the abnormal information library, and the to-be-verified information library contains unverified item information, and generates image input prompt information. The above-mentioned inventory management method can identify the input photographed image and verify it with the item information, thereby realizing rapid sorting of the inventory, and at the same time, the data in the verification process is archived through the to-be-verified information library and the abnormal information library, which greatly improves the inventory management efficiency.

[0090] The above-mentioned inventory management device based on image intelligent recognition can be implemented in the form of a computer program, and the computer program can be used in Figure 3 Runs on the computer device shown.

[0091] See also Figure 3 , Figure 3 1 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device may be a user terminal or a management server for executing an inventory management method based on image intelligent recognition to achieve commodity inventory management.

[0092] See also Figure 3 The computer device 500 includes a processor 502 , a memory and a communication interface 505 connected via a communication bus 501 , wherein the memory may include a storage medium 503 and an internal memory 504 .

[0093] The storage medium 503 may store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 may execute an inventory management method based on image intelligent recognition, wherein the storage medium 503 may be a volatile storage medium or a non-volatile storage medium.

[0094] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0095] The internal memory 504 provides an environment for the operation of the computer program 5032 in the storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the inventory management method based on image intelligent recognition.

[0096] The communication interface 505 is used for network communication, such as providing data information transmission, etc. Those skilled in the art will appreciate that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device 500 to which the solution of the present invention is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0097] The processor 502 is used to run the computer program 5032 stored in the memory to implement the corresponding functions in the above-mentioned inventory management method based on image intelligent recognition.

[0098] Those skilled in the art will understand that Figure 3 The embodiments of the computer device shown in the figure do not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structure and function of the memory and the processor are the same as those of the embodiment of the present invention. Figure 3 The embodiments shown are consistent and will not be described again here.

[0099] It should be understood that in the embodiment of the present invention, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0100] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps included in the above-mentioned inventory management method based on image intelligent recognition are implemented.

[0101] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment, devices and units can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. Those of ordinary skill in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0102] In the several embodiments provided by the present invention, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. Units with the same function may also be combined into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, or may be an electrical, mechanical or other form of connection.

[0103] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0104] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0105] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a computer-readable storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a magnetic disk or an optical disk.

[0106] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An inventory management method based on image intelligent recognition, characterized in that: The method comprises: Receive the inventory command input by the user and build the corresponding information database to be checked according to the current inventory statistics; Identify whether the input captured image is a re-photographed image according to a preset re-photographed recognition model; If the captured image is not a re-shot image, segmenting the captured image according to a preset image segmentation rule to obtain a corresponding segmented image; Recognize the segmented image according to a preset image recognition model to obtain corresponding image recognition information; According to a preset information verification strategy, the image recognition information and the corresponding segmented image are verified to see whether they are consistent with the information of a certain item in the information library to be verified, and a corresponding verification result is obtained; If the verification result is consistent with a certain item information, determine whether the shooting position information corresponding to the image recognition information is the same as the verified and consistent item information, and obtain a corresponding determination result; If the judgment result is the same, the to-be-verified information database is cancelled according to the verification result; If the judgment result is different, the item information that is consistent is transferred to the abnormal information database; If the to-be-verified information database contains unverified item information, image input prompt information is generated to prompt the user to input the captured image again.

2. The inventory management method based on image intelligent recognition according to claim 1 is characterized in that: The step of identifying the segmented image according to a preset image recognition model to obtain corresponding image recognition information includes: Obtaining a recognition rule in the image recognition model corresponding to the barcode format of the segmented image; The segmented image is identified according to the identification rule to obtain corresponding image identification information.

3. The inventory management method based on image intelligent recognition according to claim 1 is characterized in that: The information verification is performed based on the preset information verification strategy to determine whether the image recognition information and the corresponding segmented image are consistent with the information of a certain item in the to-be-verified information library, and a corresponding verification result is obtained, including: Whether the image recognition information is consistent with the serial code in any of the item information; If the image recognition information is consistent with a serial code in the item information, a consistent verification result is obtained; If the image recognition information is not consistent with any of the serial codes of the item information, an image comparison is performed between the segmented image and the item information in the to-be-checked information library according to a preset image comparison model to obtain a verification result of whether they are consistent.

4. The inventory management method based on image intelligent recognition according to claim 3 is characterized in that: The performing image comparison between the segmented image and the item information in the to-be-checked information library according to a preset image comparison model includes: Generate barcode graphics corresponding to the serial codes of each of the item information according to the barcode format corresponding to the segmented image; Extracting the segmented image features corresponding to the segmented image and the barcode graphic features corresponding to the barcode graphic respectively according to the convolutional neural network in the image contrast model; Combining the segmented image feature with each of the barcode graphic features respectively and inputting the combined features into a similarity analysis network in the image comparison model to obtain a similarity value corresponding to the segmented image feature and each of the barcode graphic features; Whether the barcode graphic feature with the highest similarity value is significant is determined according to the saliency judgment rule in the image comparison model, so as to determine whether the item information corresponding to the barcode graphic feature with the highest similarity value is consistent with the segmented image.

5. The inventory management method based on image intelligent recognition according to claim 4 is characterized in that: The determining whether the barcode graphic feature with the highest similarity value is significant according to the saliency determination rule in the image comparison model, so as to determine whether the item information corresponding to the barcode graphic feature with the highest similarity value is consistent with the segmented image, includes: Determine the barcode graphic corresponding to the barcode graphic feature with the highest similarity value as the target barcode graphic; Calculate the significance coefficient corresponding to the standard deviation value of the similarity value with the non-target barcode graphic according to the significance coefficient calculation formula in the significance judgment rule; It is determined whether the overall standard deviation of the similarity values ​​of the barcode graphic features is greater than the significant coefficient, and a determination result of whether the barcode graphic feature with the highest similarity value is significant is obtained.

6. The inventory management method based on image intelligent recognition according to claim 1 is characterized in that: The step of judging whether the shooting location information corresponding to the image recognition information is the same as the verified and consistent item information to obtain a corresponding judgment result includes: Acquire geographic information corresponding to the shooting location information; A check is performed to determine whether the geographic information is the same as the inventory location in the item information, so as to obtain a determination result of whether the shooting location information is the same as the item information.

7. The inventory management method based on image intelligent recognition according to claim 1 is characterized in that: After obtaining the corresponding verification result, the method further includes: If the verification result is inconsistent, a prompt message of verification abnormality is generated.

8. An inventory management device based on image intelligent recognition, characterized in that: The inventory management device based on image intelligent recognition is used to execute the inventory management method based on image intelligent recognition according to any one of claims 1 to 7, and the device includes: The to-be-checked information database generating unit is used to receive the inventory command input by the user and construct the corresponding to-be-checked information database according to the current inventory statistical information; A re-shot image recognition unit, used to recognize whether the input captured image is a re-shot image according to a preset re-shot recognition model; An image segmentation unit, configured to segment the captured image according to a preset image segmentation rule to obtain a corresponding segmented image if the captured image is not a reproduced image; An image recognition information acquisition unit, used to recognize the segmented image according to a preset image recognition model to obtain corresponding image recognition information; A verification result acquisition unit is used to verify whether the image recognition information and the corresponding segmented image are consistent with the information of a certain item in the information library to be verified according to a preset information verification strategy, and obtain a corresponding verification result; a judgment result acquisition unit, configured to, if the verification result is consistent with a certain item information, determine whether the shooting position information corresponding to the image recognition information is the same as the verified item information, and obtain a corresponding judgment result; A verification unit, configured to, if the judgment result is the same, verify the to-be-verified information base according to the verification result; A transfer unit, configured to transfer the verified consistent item information to an abnormal information database if the judgment result is different; The prompt information generating unit is used to generate image input prompt information to prompt the user to input the captured image again if the to-be-verified information library contains unverified item information.

9. A computer device, characterized in that: The device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the inventory management method based on image intelligent recognition described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the inventory management method based on image intelligent recognition as described in any one of claims 1 to 7 are implemented.

Images