Article information identification processing method and device, storage medium and program product

Abstract

The invention relates to an article information identification processing method and device, a storage medium and a program product. The article information identification processing method comprises the following steps: acquiring article information pictures from a plurality of acquisition points; for each acquisition point, determining an acquisition point type of the acquisition point; determining a picture layout format template corresponding to each acquisition point according to a corresponding relationship between the acquisition point type and the picture layout format template; and for the article information picture acquired at each acquisition point, according to an information identification mode of the picture layout format template corresponding to the acquisition point, the acquisition information of each article in the article information picture is identified, and the acquisition information comprises an article name and article parameter information. According to the method and the device, different picture layout format templates and information identification modes can be determined according to different acquisition points, so that article information pictures acquired in different channels can be conveniently and accurately identified.

Description

Technical Field

[0001] The present disclosure relates to the technical field of item information identification, and in particular to an item information identification processing method and device, a storage medium, and a program product. Background Art

[0002] The market circulation price of goods reflects the actual situation of the market and is a direct reflection of the supply and demand relationship of goods. Since the market circulation price will fluctuate with the changes in market conditions and has a strong dynamic nature, if an effective commodity price analysis mechanism is to be established for decision-making reference, the commodity circulation price must be collected extensively and frequently. Summary of the invention

[0003] The inventor found through research that the price collection channels of related art items are also special. Generally, they are obtained from designated physical retail stores, wholesalers, distributors, and even notice boards in vegetable markets, and the results are mainly obtained in the form of pictures in the form of cooperation. Therefore, the partners are in a relatively passive position in the management of the form and quality of item circulation prices. In the related art, the price pictures of items collected from different channels have different table formats, complex printed fonts and handwritten numbers, and are difficult to identify. There is a lack of automated processing procedures to improve collection efficiency and quality.

[0004] In view of at least one of the above technical problems, the present disclosure provides a method and device for object information identification and processing, a storage medium and a program product, which can determine different image layout format templates and information identification methods according to different collection points, so as to conveniently and accurately identify object information images collected through different channels.

[0005] According to one aspect of the present disclosure, a method for identifying and processing object information is provided, comprising:

[0006] Get item information images from multiple collection points;

[0007] For each collection point, determine the collection point type of the collection point;

[0008] According to the correspondence between the collection point type and the picture layout format template, determine the picture layout format template corresponding to each collection point;

[0009] For the item information picture acquired at each collection point, the collection information of each item in the item information picture is identified according to the information identification method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information.

[0010] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0011] Determine the item standard name table corresponding to each collection point according to the correspondence between the collection point type and the item standard name table, wherein the item standard name table is a mapping table between the item name at the collection point and the item standard name;

[0012] According to the item name of the collection point identified in the item information picture, the standard item name corresponding to the item name of the collection point is determined from the table of standard item names corresponding to the collection point.

[0013] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0014] The recognition result of each object recognized in the object information picture is stored, wherein the recognition result includes the collection time, collection point information, standard object name and object parameter information.

[0015] In some embodiments of the present disclosure, determining the picture layout format template corresponding to each collection point according to the correspondence between the collection point type and the picture layout format template includes: determining the picture layout format template corresponding to the collection point according to the item information picture obtained by each collection point and the correspondence between the collection point type and the picture layout format template.

[0016] In some embodiments of the present disclosure, the collection information includes collection time, item name and item parameter information.

[0017] In some embodiments of the present disclosure, the information identification method of the image layout template corresponding to the collection point to identify the collection time, the name of the item and the item parameter information in the item information image includes:

[0018] When the image layout format template of the item information image is the first mode or the second mode, the acquisition information coordinates in the item information image are fixed, wherein the first mode and the second mode are both fixed electronic table formats, and the item name and item parameter information of the same item in the first mode are respectively in adjacent cells in the table, and the item name and item parameter information of the same item in the second mode are in the same cell in the table;

[0019] The collection information corresponding to each collection information coordinate is identified, wherein the collection information includes collection time, item name and item parameter information.

[0020] In some embodiments of the present disclosure, the information identification method of the image layout template corresponding to the collection point to identify the collection time, the name of the item and the item parameter information in the item information image includes:

[0021] When the picture layout format template of the item information picture is the third mode, the item information picture is cut and partially enlarged, wherein in the third mode: the item name is in a fixed electronic spreadsheet format, the item parameter information is filled in manually, and the third mode picture is in a fixed scan format or photographed form;

[0022] Perform affine transformation on the cut image and the local magnification results to achieve deformation correction;

[0023] The corrected result is subjected to text recognition to obtain the collection information of each item, wherein the collection information includes the collection time, the item name and the item parameter information.

[0024] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0025] Pre-establish the correspondence between the collection point type and the image layout format template;

[0026] Establish the correspondence between the collection point type and the standard name table of items in advance.

[0027] In some embodiments of the present disclosure, determining the standard item name corresponding to the item name at the collection point from a table of standard item names corresponding to the collection point according to the item name at the collection point identified in the item information picture includes:

[0028] Identify online spreadsheet results from the item information image;

[0029] Filtering text content belonging to the names of items at the collection point from the online electronic spreadsheet results;

[0030] The text contents of the names of the items at the collection points are input into the generative large language model one by one, and the standard item names corresponding to the names of the items at the collection points are generated by comparing with the generative large language model through the prompt word engineering.

[0031] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0032] Generate a mapping result between the item name at the collection point and the item standard name;

[0033] The mapping results are displayed online in a table format for easy checking and modification.

[0034] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0035] When the name of the collection point item identified in the item information image cannot be used to determine the standard item name corresponding to the collection point item name from the standard item name table corresponding to the collection point, it is determined that the collection point item name recognition is abnormal, and the collection abnormality information corresponding to the abnormal collection point item name is stored separately.

[0036] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0037] Determine whether the item parameter information is within a predetermined item parameter range, wherein the predetermined item parameter range is greater than a predetermined item parameter lower limit and less than a predetermined item parameter upper limit;

[0038] In the case where the item parameter information is not within the predetermined item parameter range, the item parameter information is determined to be abnormal, and the collection abnormality information corresponding to the abnormal item parameter information is stored separately.

[0039] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0040] Sending the collection abnormality information to the staff terminal for staff confirmation;

[0041] Receive abnormal information feedback returned by the staff terminal;

[0042] The collected abnormal information is processed accordingly according to the abnormal information feedback.

[0043] In some embodiments of the present disclosure, the corresponding processing of the collected abnormal information according to the abnormal information feedback includes at least one of the following steps:

[0044] If it is confirmed in the abnormal information feedback that the abnormal item parameter information is correct, the collection abnormality information corresponding to the abnormal item parameter information is stored together with the normal collection information;

[0045] In the case where the abnormal item parameter information is corrected in the abnormal information feedback, the corrected acquisition abnormal information is stored together with the normal acquisition information;

[0046] In the case where the name of the item at the abnormal collection point is maintained as the standard item name in the abnormal information feedback, the maintained collection abnormality information is stored together with the normal collection information;

[0047] The collection exception information is deleted according to the deletion instruction in the exception information feedback.

[0048] In some embodiments of the present disclosure, the object information identification and processing method further includes:

[0049] Receive item parameter information query request;

[0050] Determine the target item parameter information based on the collection time, collection point information and standard item name in the item parameter information query request;

[0051] Feedback the target item parameter information to the user terminal.

[0052] According to another aspect of the present disclosure, there is provided an object information identification and processing device, comprising:

[0053] An information picture acquisition module is configured to acquire item information pictures from multiple collection points;

[0054] A collection point type determination module is configured to determine the collection point type of each collection point;

[0055] The layout format determination module is configured to determine the image layout format template corresponding to each acquisition point according to the correspondence between the acquisition point type and the image layout format template;

[0056] The collection information recognition module is configured to identify the collection information of each item in the item information picture obtained at each collection point according to the information recognition method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information

[0057] According to another aspect of the present disclosure, there is provided an object information identification and processing device, comprising:

[0058] a memory configured to store instructions; and

[0059] The processor is configured to execute the instructions so that the item information identification and processing device implements the item information identification and processing method as described in any of the above embodiments.

[0060] According to another aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the object information identification and processing method as described in any of the above embodiments is implemented.

[0061] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the object information identification and processing method as described in any of the above embodiments is implemented.

[0062] The present disclosure can determine different picture layout format templates and information recognition methods according to different collection points, so that item information pictures collected through different channels can be easily and accurately identified. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0064] Figure 1 Schematic diagram of some embodiments of the object information identification and processing method disclosed in the present invention.

[0065] Figure 2 Schematic diagrams of other embodiments of the object information identification and processing method disclosed in the present invention.

[0066] Figure 3 Schematic diagrams of some further embodiments of the object information identification and processing method disclosed in the present invention.

[0067] Figure 4 This is a schematic diagram of the image recognition effect collected in the first mode in some embodiments of the present disclosure.

[0068] Figure 5 This is a schematic diagram of the image recognition effect collected in the second mode in some embodiments of the present disclosure.

[0069] Figure 6 Schematic diagrams of some further embodiments of the object information identification and processing method disclosed in the present invention.

[0070] Figure 7 This is a schematic diagram of the picture cutting effect in the third mode in some embodiments of the present disclosure.

[0071] Figure 8 It is a schematic diagram of affine transformation in the third mode in some embodiments of the present disclosure.

[0072] Fig. 9 It is a schematic diagram of the structure of some embodiments of the object information identification and processing device disclosed in the present invention.

[0073] Fig.10 Schematic diagram of the structure of other embodiments of the object information identification and processing device disclosed in the present invention. DETAILED DESCRIPTION

[0074] The technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present disclosure and its application or use. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0075] Unless specifically stated otherwise, the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0076] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0077] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0078] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0079] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0080] Figure 1 Schematic diagram of some embodiments of the object information identification and processing method disclosed in the present invention. Figure 1 The embodiment can be executed by the object information identification processing device disclosed in the present invention. Figure 1 As shown, Figure 1 The method of the embodiment may include at least one of steps 11 to 14.

[0081] Step 11, obtaining item information pictures from multiple collection points.

[0082] In some embodiments of the present disclosure, the item may be a commodity.

[0083] In some embodiments of the present disclosure, the item information may include information such as the price of the item.

[0084] Step 12: for each collection point, determine the collection point type of the collection point.

[0085] Step 13: Determine the picture layout format template corresponding to each collection point according to the correspondence between the collection point type and the picture layout format template.

[0086] Step 14, for the item information picture obtained by each collection point, the collection information of each item in the item information picture is identified according to the information identification method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information.

[0087] In some embodiments of the present disclosure, the item parameter information may be item price information.

[0088] Figure 2 Schematic diagrams of other embodiments of the object information identification and processing method disclosed in the present invention. Figure 2 The embodiment can be executed by the item information identification processing device disclosed in the present invention. Figure 1 Steps 11 to 14 of the embodiment are respectively Figure 2 Step 1, step 3, and steps 6-7 of the embodiment are the same or similar. Figure 2 As shown, Figure 2 The method of the embodiment may include at least one of steps 1 to 10.

[0089] Step 1: Obtain item information images from multiple collection points.

[0090] Step 2: Classify channels.

[0091] In some embodiments of the present disclosure, step 1 may include: continuing to classify the collection points.

[0092] In some embodiments of the present disclosure, step 1 may include: pre-establishing a correspondence between the collection point type and the image layout format template; pre-establishing a correspondence between the collection point type and the item standard name table.

[0093] In some embodiments of the present disclosure, step 1 may include: the number of collection points is limited, and in order to improve the accuracy of the collection results, channel classification is performed, and item name and picture layout format templates are constructed for different collection points.

[0094] Step 3: for each collection point, determine the collection point type of the collection point.

[0095] Step 4: Map the standard item names.

[0096] In some embodiments of the present disclosure, step 4 may include at least one of steps 41 to 43 .

[0097] Step 41, according to the correspondence between the collection point type and the item standard name table, determine the item standard name table corresponding to each collection point, wherein the item standard name table is a mapping table between the collection point item name and the item standard name.

[0098] Step 42, according to the item name of the collection point identified in the item information picture, determine the standard item name corresponding to the item name of the collection point from the table of standard item names corresponding to the collection point.

[0099] In some embodiments of the present disclosure, step 42 may include: identifying an online spreadsheet result from the item information picture; filtering out text content belonging to the item name at the collection point from the online spreadsheet result; inputting the text content belonging to the item name at the collection point into a generative large language model one by one, and generating a standard item name corresponding to the item name at the collection point by using a generative large language model through a prompt word project.

[0100] Step 43, generating a mapping result between the item name at the collection point and the item standard name; and displaying the mapping result online in a table form for easy checking and modification.

[0101] The inventor discovered through research that different channels for collecting item prices usually have many personalized naming methods for item names and abbreviations, and the rules are difficult to list exhaustively.

[0102] In some embodiments of the present disclosure, step 42 and step 43 may include: to improve the standardization level of item names, first establish a mapping table from the item names of each channel to the standard items; introduce PP-OCRv4 to quickly obtain the recognition online spreadsheet results of the collected pictures based on layout recognition, filter out the text content belonging to the item names, input them into the generative large language model one by one, and through the prompt word project, let the generative large language model generate a mapping result of (standard item name: collection point item name) by comparison, and display the results online in a table form.

[0103] In some embodiments of the present disclosure, step 43 may include: outputting the mapping results of the collection point item name and the item standard name so that the mapping results of the standard item name of the collection point (standard item name: collection point item name) can be manually checked and modified, wherein the output format is a final mapping result table of (collection point: standard item name: collection point item name).

[0104] Step 5: determine whether the item parameter information is within a predetermined item parameter range, wherein the predetermined item parameter range is greater than a predetermined item parameter lower limit and less than a predetermined item parameter upper limit.

[0105] In some embodiments of the present disclosure, step 5 may include: when the item parameter information is not within a predetermined item parameter range, determining that the item parameter information is abnormal, and separately storing the collection abnormality information corresponding to the abnormal item parameter information.

[0106] In some embodiments of the present disclosure, step 5 may include: performing maintenance within a predetermined parameter range of the item.

[0107] In some embodiments of the present disclosure, step 5 may include: collecting and maintaining normal price ranges of items.

[0108] In some embodiments of the present disclosure, step 5 may include: each collection point sets a normal price range for the collected items, the default lower limit of the range is: the lowest price of the collection point in the past N times × (1-the decline rate); the default upper limit of the range is: the highest price of the collection point in the past N times × (1+the increase rate), and the final value of the price range is manually adjusted and confirmed.

[0109] In some embodiments of the present disclosure, N is a natural number greater than 1.

[0110] In some embodiments of the present disclosure, N is 10.

[0111] Step 6: Determine the image layout format template corresponding to each collection point according to the correspondence between the collection point type and the image layout format template.

[0112] In some embodiments of the present disclosure, step 6 may include: loading a table recognition template.

[0113] In some embodiments of the present disclosure, step 6 may include: determining the image layout format template corresponding to the collection point according to the item information image obtained by each collection point and the correspondence between the collection point type and the image layout format template.

[0114] In some embodiments of the present disclosure, step 6 may include: constructing a pattern table recognition template.

[0115] Step 7: for the item information picture obtained by each collection point, the collection information of each item in the item information picture is identified according to the information identification method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information.

[0116] In some embodiments of the present disclosure, the collection information includes collection time, item name and item parameter information.

[0117] Figure 3 Schematic diagrams of some further embodiments of the object information identification and processing method disclosed in the present invention. Figure 3 The embodiment can be executed by the object information identification processing device disclosed in the present invention. Figure 3 As shown, Figure 3 The method of the embodiment (for example Figure 2 Step 7) of the embodiment may include at least one of steps 71 to 78 .

[0118] Step 71, obtain item information picture.

[0119] In some embodiments of the present disclosure, the item information picture may be an item price information picture.

[0120] Step 72, perform layout recognition of the item information picture.

[0121] Step 73, fix the coordinates of the collected information.

[0122] In some embodiments of the present disclosure, step 73 may include: when the picture layout format template of the item information picture is the first mode or the second mode, fixing the acquisition information coordinates in the item information picture.

[0123] Step 74, loading the table recognition mode.

[0124] In some embodiments of the present disclosure, the first mode and the second mode are both fixed spreadsheet formats, and the item name and item parameter information of the same item in the first mode are respectively located in adjacent cells in the table, and the item name and item parameter information of the same item in the second mode are located in the same cell in the table.

[0125] Step 75, extracting the collected information.

[0126] In some embodiments of the present disclosure, step 75 may include: identifying the collection information corresponding to each collection information coordinate, wherein the collection information includes collection time, item name, and item parameter information.

[0127] Step 76, processing the collection abnormal information.

[0128] Step 77, obtaining the collection information result.

[0129] Step 78, data storage of the collected information results.

[0130] Figure 4 Schematic diagram of the image collection recognition effect in the first mode in some embodiments of the present disclosure. The price collection image in the first mode is exported in a fixed spreadsheet format, and the image size, format, and text layout remain unchanged. The font used is a standard printing font. Figure 4As shown in the figure, in the first mode, the name and parameter information of the same item are located in adjacent cells in the table. Based on the layout recognition of PP-OCRv4, the recognition coordinates (upper left corner, lower right corner) of each cell in the image are obtained, and after manual fine-tuning, a fixed recognition template is formed. In this template, the acquisition time coordinate is fixed (such as Figure 4 In the embodiment, the collection time is specifically 2024.7.24). Each pair of item name:collection price coordinates is also fixed. Figure 4 As shown, the adjacent cells give the name of product 1 and the price of product 1, ..., the name of product N and the price of product N. In this mode, the recognition template finally outputs: the acquisition time coordinates (upper left corner, lower right corner), each pair of item name: item name coordinates of the acquisition price (upper left corner, lower right corner), and the corresponding acquisition price coordinates (upper left corner, lower right corner).

[0131] Figure 5 The following is a schematic diagram of the recognition effect of the collected pictures in the second mode in some embodiments of the present disclosure. The collected pictures in the second mode are exported in a fixed spreadsheet format, the picture size and format are neat and unchanged, and the font used is a standard printing font, but the layout of the text content is not completely aligned, and the name of the item in the cell may not be fully displayed. Figure 5 As shown in FIG. 1 , the item name and item parameter information of the same item in the second mode are in the same cell in the table. Figure 5 As shown in the figure, the name and price of product 1 are given in the same cell. Based on the layout recognition of PP-OCRv4, the recognition coordinates (upper left corner, lower right corner) of each cell in the image are obtained. After manual fine-tuning, a fixed recognition template is formed. In this template, the acquisition time coordinate is fixed (such as Figure 5 In the embodiment, the collection time is specifically July 25), but each pair of (item name: collection price) coordinates in the cell is not fixed.

[0132] Figure 6 Schematic diagrams of some further embodiments of the object information identification and processing method disclosed in the present invention. Figure 6 The embodiment can be executed by the object information identification processing device disclosed in the present invention. Figure 6 As shown, Figure 6 The method of the embodiment (for example Figure 2 Step 7) of the embodiment may include at least one of steps 71 to 72, steps 703 to 704, and steps 75 to 78. Figure 3 Steps 71 to 72, and steps 75 to 78 of the embodiment are respectively Figure 7 Steps 71 to 72 and steps 75 to 78 of the embodiment are the same or similar.

[0133] Step 703, when the picture layout format template of the item information picture is in the third mode, the item information picture is cut and partially enlarged, wherein in the third mode: the item name is in a fixed electronic spreadsheet format, the item parameter information is filled in manually, and the third mode picture is in a fixed scan format or photographed form.

[0134] Step 704, performing affine transformation on the cut image and the local magnification result to achieve deformation correction.

[0135] Figure 7 The following is a schematic diagram of the image cutting effect in the third mode in some embodiments of the present disclosure. In the collected images in the third mode, the names of the items are exported and printed in a fixed spreadsheet format, but the prices of the items are filled in manually. If it is a fixed scan format, the image size remains unchanged; but it may also be a manual photo shoot, in which case the image size and format will change to a certain extent. In the third mode, the acquired image is first cut and partially enlarged, and the cut result is deformed and corrected by affine transformation, and then the corrected result is sent to PP-OCRv4 for text recognition to obtain the collected image information. Figure 8 It is a schematic diagram of affine transformation in the third mode in some embodiments of the present disclosure.

[0136] Step 75, performing text recognition on the corrected result to obtain the collection information of each item, wherein the collection information includes the collection time, item name and item parameter information.

[0137] Step 8: Perform exception handling on the collected exception information.

[0138] In some embodiments of the present disclosure, step 8 may include at least one of steps 81 to 82 .

[0139] Step 81, handle the name exception.

[0140] In some embodiments of the present disclosure, step 81 may include: at least one step from step 811 to step 813 .

[0141] Step 811, when the name of the collection point item identified in the item information picture cannot be used to determine the standard item name corresponding to the collection point item name from the standard item name table corresponding to the collection point, it is determined that the collection point item name recognition is abnormal, and the collection abnormality information corresponding to the abnormal collection point item name is stored separately.

[0142] In some embodiments of the present disclosure, step 811 may include: three recognition modes: a first mode, a second mode, and a third mode. If the OCR (Optical Character Recognition) recognition result cannot be mapped to the standard item name mapping table of the collection point, it is determined that the item name recognition is abnormal, and the corresponding collection information is stored separately.

[0143] Step 812: Send the collection exception information to the staff terminal so that the staff can manually confirm the collection exception information.

[0144] Step 813, receiving the abnormal information feedback returned by the staff terminal; and performing corresponding processing on the collected abnormal information according to the abnormal information feedback.

[0145] In some embodiments of the present disclosure, step 813 may include at least one of the following steps: when the name of the item at the abnormal collection point is maintained as the standard item name in the abnormal information feedback, the maintained collection exception information is stored together with the normal collection information (other normal results); and according to the deletion indication in the abnormal information feedback, the collection exception information is deleted.

[0146] Step 82, processing the numerical anomaly.

[0147] In some embodiments of the present disclosure, the numerical anomaly is an abnormal item parameter or an abnormal item acquisition price.

[0148] In some embodiments of the present disclosure, step 81 may include: at least one of steps 821 to 823 .

[0149] Step 821, determining whether the item parameter information is within a predetermined item parameter range, wherein the predetermined item parameter range is greater than a predetermined item parameter lower limit and less than a predetermined item parameter upper limit; if the item parameter information is not within the predetermined item parameter range, determining that the item parameter information is abnormal, and storing the collection abnormality information corresponding to the abnormal item parameter information separately;

[0150] Step 822: Send the collection exception information to the staff terminal so that the staff can manually confirm the collection exception information.

[0151] Step 823, receiving the abnormal information feedback returned by the staff terminal; and performing corresponding processing on the collected abnormal information according to the abnormal information feedback.

[0152] In some embodiments of the present disclosure, step 823 may include at least one of steps 8231 to 8233.

[0153] Step 8231, when it is confirmed in the abnormal information feedback that the abnormal item parameter information is correct, the collection abnormality information corresponding to the abnormal item parameter information is stored together with the normal collection information (other correct results).

[0154] In some embodiments of the present disclosure, step 8231 may include: confirming that the collection price is correct, and storing the collection result together with other normal results.

[0155] Step 8232, when the abnormal item parameter information is corrected in the abnormal information feedback, the corrected acquisition abnormality information is stored together with the normal acquisition information (other correct results).

[0156] In some embodiments of the present disclosure, step 8232 may include: correcting the collection price information, and storing the maintained collection results together with other normal results.

[0157] Step 8233: Delete the collection exception information according to the deletion instruction in the exception information feedback.

[0158] Step 9, storing the recognition result of each object recognized in the object information picture, wherein the recognition result includes the collection time, collection point information, standard object name and object parameter information.

[0159] In some embodiments of the present disclosure, step 9 may include: storing the identified collection results in a relational database, a vector database, or a CSV (Comma-Separated Values) format file, and the stored information includes: collection time, collection point, standard name of the item, and price.

[0160] Step 10: retrieve and process the collected information.

[0161] In some embodiments of the present disclosure, step 10 may include: receiving an item parameter information query request; determining the target item parameter information based on the collection time, collection point information and standard item name in the item parameter information query request; and feeding back the target item parameter information to the user terminal.

[0162] In some embodiments of the present disclosure, step 10 may include: performing item price query and answering based on a generative big language model, wherein the generative big language retrieves date, collection point, and item name information from a vector database and feeds back price query results.

[0163] In some embodiments of the present disclosure, step 10 may include: based on a relational database or a CSV file, collecting price trends of items and comparing prices at collection points for reference in decision analysis.

[0164] The above-mentioned embodiments of the present disclosure propose an automatic processing flow and device for collecting price pictures of items that combines a generative large language model, PP-OCRv4's layout recognition and text recognition technology, and vector database retrieval, thereby effectively improving the processing efficiency and analysis quality of item price information.

[0165] The above embodiments of the present disclosure collect market circulation prices, and the usage method of the above embodiments of the present disclosure is mainly based on OCR technology and a generative large model.

[0166] The above-mentioned embodiments of the present disclosure collect channel prices, and the channel party has already conducted certain data aggregation in advance. Therefore, the price information collected in the above-mentioned embodiments of the present disclosure is dense, the scenarios are complex, and the solution is difficult.

[0167] The channel prices collected by the above-mentioned embodiments of the present disclosure cannot be obtained in batches from the Internet, and the density of the identified price information is large. The above-mentioned embodiments of the present disclosure also need to use layout recognition and other methods for processing.

[0168] Fig. 9 Schematic diagram of the structure of some embodiments of the object information identification and processing device disclosed in the present invention. Fig. 9 As shown, the object information identification and processing device disclosed in the present invention may include an information picture acquisition module 91, a collection point type determination module 92, a layout format determination module 93 and a collection information identification module 94.

[0169] The information picture acquisition module 91 is configured to acquire item information pictures from multiple collection points.

[0170] The collection point type determination module 92 is configured to determine the collection point type of each collection point.

[0171] The layout format determination module 93 is configured to determine the picture layout format template corresponding to each acquisition point according to the correspondence between the acquisition point type and the picture layout format template.

[0172] In some embodiments of the present disclosure, the layout format determination module 93 may be configured to determine the image layout format template corresponding to the collection point based on the item information image acquired at each collection point and the correspondence between the collection point type and the image layout format template.

[0173] The collection information recognition module 94 is configured to identify the collection information of each item in the item information picture obtained at each collection point according to the information recognition method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information

[0174] In some embodiments of the present disclosure, the collection information includes collection time, item name and item parameter information.

[0175] In some embodiments of the present disclosure, the collection information identification module 94 can be configured to fix the collection information coordinates in the item information picture obtained at each collection point when the picture layout format template of the item information picture is the first mode or the second mode, wherein the first mode and the second mode are both fixed spreadsheet formats, the item name and item parameter information of the same item in the first mode are respectively located in adjacent cells in the table, and the item name and item parameter information of the same item in the second mode are located in the same cell in the table; identify the collection information corresponding to each collection information coordinate, wherein the collection information includes the collection time, the item name and the item parameter information.

[0176] In some embodiments of the present disclosure, the collection information identification module 94 may also be configured to cut and partially enlarge the item information picture obtained at each collection point when the picture layout format template of the item information picture is the third mode, wherein in the third mode: the item name is in a fixed spreadsheet format, the item parameter information is filled in manually, and the third mode picture is in a fixed scan format or photographed form; perform affine transformation on the results after cutting and partial enlargement to achieve deformation correction; perform text recognition on the corrected results to obtain the collection information of each item, wherein the collection information includes the collection time, item name and item parameter information.

[0177] In some embodiments of the present disclosure, the item information identification and processing device of the present disclosure may also be configured to determine the item standard name table corresponding to each collection point according to the correspondence between the collection point type and the item standard name table, wherein the item standard name table is a mapping table of collection point item names and item standard names; and determine the standard item name corresponding to the collection point item name from the item standard name table corresponding to the collection point according to the collection point item name identified in the item information picture.

[0178] In some embodiments of the present disclosure, in some embodiments of the present disclosure, the object information identification and processing device of the present disclosure, when determining the standard object name corresponding to the collection point object name from the object standard name table corresponding to the collection point according to the collection point object name identified in the object information picture, can be configured to identify an online spreadsheet result from the object information picture; filter out text content belonging to the collection point object name from the online spreadsheet result; input the text content belonging to the collection point object name into the generative large language model one by one, and generate the standard object name corresponding to the collection point object name by comparison using the generative large language model through prompt word engineering.

[0179] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may also be configured to store the identification results of each object identified in the object information picture, wherein the identification results include collection time, collection point information, standard object name and object parameter information.

[0180] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may also be configured to pre-establish a correspondence between the collection point type and the image layout format template; and pre-establish a correspondence between the collection point type and the object standard name table.

[0181] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may also be configured to generate a mapping result between the object name at the collection point and the object standard name; and display the mapping result online in a table form for easy verification and modification.

[0182] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may also be configured to determine that the object name recognition of the collection point is abnormal when the object name of the collection point identified in the object information picture cannot determine the standard object name corresponding to the object name of the collection point from the object standard name table corresponding to the collection point, and store the collection abnormality information corresponding to the abnormal collection point object name separately.

[0183] In some embodiments of the present disclosure, the item information identification and processing device of the present disclosure may also be configured to determine whether the item parameter information is within a predetermined item parameter range, wherein the predetermined item parameter range is greater than a predetermined item parameter lower limit and less than a predetermined item parameter upper limit; when the item parameter information is not within the predetermined item parameter range, the item parameter information is determined to be abnormal, and the collection abnormality information corresponding to the abnormal item parameter information is stored separately.

[0184] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may also be configured to send the collection exception information to a staff terminal for confirmation by the staff; receive exception information feedback returned by the staff terminal; and perform corresponding processing on the collection exception information according to the exception information feedback.

[0185] In some embodiments of the present disclosure, when the object information identification and processing device of the present disclosure processes the collection abnormality information accordingly according to the abnormal information feedback, it can be configured to perform at least one of the following operations: when it is confirmed in the abnormal information feedback that the abnormal object parameter information is correct, the collection abnormality information corresponding to the abnormal object parameter information is stored together with the normal collection information; when the abnormal object parameter information is corrected in the abnormal information feedback, the corrected collection abnormality information is stored together with the normal collection information; when the abnormal collection point object name is maintained as the standard object name in the abnormal information feedback, the maintained collection abnormality information is stored together with the normal collection information; and the collection abnormality information is deleted according to the deletion indication in the abnormal information feedback.

[0186] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may also be configured to receive an object parameter information query request; determine the target object parameter information based on the collection time, collection point information and standard object name in the object parameter information query request; and feed back the target object parameter information to the user terminal.

[0187] In some embodiments of the present disclosure, the object information identification and processing device of the present disclosure may be configured to implement the object information identification and processing method as described in any of the above embodiments.

[0188] Fig.10 Schematic diagram of the structure of some other embodiments of the object information identification and processing device disclosed in the present invention. Fig.10 As shown, the object information identification and processing device disclosed in the present invention may include a memory 101 and a processor 102 .

[0189] The memory 101 is used to store instructions. The processor 102 is coupled to the memory 101. The processor 102 is configured to execute the object information identification processing method involved in the above embodiment based on the instructions stored in the memory.

[0190] like Fig.10 As shown, the item information identification processing device also includes a communication interface 103 for information exchange with other devices. At the same time, the item information identification processing device also includes a bus 104, through which the processor 102, the communication interface 103, and the memory 101 communicate with each other.

[0191] The memory 101 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory. The memory 101 may also be a memory array. The memory 101 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules.

[0192] In addition, the processor 102 may be a central processing unit (CPU), or may be an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present disclosure.

[0193] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the object information identification and processing method as described in any of the above embodiments is implemented.

[0194] According to another aspect of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the object information identification and processing method as described in any of the above embodiments is implemented.

[0195] The computer-readable storage medium of the present disclosure may be implemented as a non-transitory computer-readable storage medium.

[0196] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, devices, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0197] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0198] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0199] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0200] The object information identification and processing device, information picture acquisition module, collection point type determination module, layout format determination module and collection information identification module described above can be implemented as a general-purpose processor, programmable logic controller, digital signal processor (DSP), application-specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component or any appropriate combination thereof for performing the functions described in the present disclosure.

[0201] Those skilled in the art will appreciate that all or part of the steps of the above-described embodiment method of the present disclosure may be accomplished by hardware, and the hardware may be implemented as a general-purpose processor, a programmable logic controller, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or any appropriate combination thereof for executing the method described in the present disclosure.

[0202] So far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0203] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or may be accomplished by instructing the relevant hardware through a program, and the program may be stored in a non-transitory computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0204] The description of the present disclosure is given for the purpose of illustration and description, and is not intended to be exhaustive or to limit the present disclosure to the disclosed form. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present disclosure, and to enable those of ordinary skill in the art to understand the present disclosure and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. A method for identifying and processing object information, comprising: Get item information images from multiple collection points; For each collection point, determine the collection point type of the collection point; According to the correspondence between the collection point type and the picture layout format template, determine the picture layout format template corresponding to each collection point; For the item information picture acquired at each collection point, the collection information of each item in the item information picture is identified according to the information identification method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information.

2. The object information identification and processing method according to claim 1, further comprising: Determine the item standard name table corresponding to each collection point according to the correspondence between the collection point type and the item standard name table, wherein the item standard name table is a mapping table between the item name at the collection point and the item standard name; According to the item name of the collection point identified in the item information picture, the standard item name corresponding to the item name of the collection point is determined from the table of standard item names corresponding to the collection point.

3. The object information identification and processing method according to claim 2, further comprising: The recognition result of each object recognized in the object information picture is stored, wherein the recognition result includes the collection time, collection point information, standard object name and object parameter information.

4. The method for identifying and processing item information according to any one of claims 1 to 3, wherein: Determining the picture layout format template corresponding to each collection point according to the correspondence between the collection point type and the picture layout format template includes: According to the item information pictures obtained at each collection point and the correspondence between the collection point type and the picture layout format template, the picture layout format template corresponding to the collection point is determined.

5. The method for identifying and processing item information according to any one of claims 1 to 3, wherein: The information identification method of the image layout template corresponding to the collection point to identify the collection time, the name of the item and the item parameter information in the item information image includes: When the image layout format template of the item information image is the first mode or the second mode, the acquisition information coordinates in the item information image are fixed, wherein the first mode and the second mode are both fixed electronic table formats, and the item name and item parameter information of the same item in the first mode are respectively in adjacent cells in the table, and the item name and item parameter information of the same item in the second mode are in the same cell in the table; The collection information corresponding to each collection information coordinate is identified, wherein the collection information includes collection time, item name and item parameter information.

6. The method for identifying and processing item information according to any one of claims 1 to 3, wherein: The information identification method of the image layout template corresponding to the collection point to identify the collection time, the name of the item and the item parameter information in the item information image includes: When the picture layout format template of the item information picture is the third mode, the item information picture is cut and partially enlarged, wherein in the third mode: the item name is in a fixed electronic spreadsheet format, the item parameter information is filled in manually, and the third mode picture is in a fixed scan format or photographed form; Perform affine transformation on the cut image and the local magnification results to achieve deformation correction; The corrected result is subjected to text recognition to obtain the collection information of each item, wherein the collection information includes the collection time, the item name and the item parameter information.

7. The object information identification and processing method according to claim 2 or 3, further comprising: Pre-establish the correspondence between the collection point type and the image layout format template; Establish the correspondence between the collection point type and the standard name table of items in advance.

8. The method for identifying and processing item information according to claim 2 or 3, wherein: The step of determining the standard item name corresponding to the item name at the collection point identified in the item information picture from the item standard name table corresponding to the collection point includes: Identify online spreadsheet results from the item information image; Filtering text content belonging to the names of items at the collection point from the online electronic spreadsheet results; The text contents of the names of the items at the collection points are input into the generative large language model one by one, and the standard item names corresponding to the names of the items at the collection points are generated by comparing with the generative large language model through the prompt word engineering.

9. The method for identifying and processing item information according to claim 8, further comprising: Generate a mapping result between the item name at the collection point and the item standard name; The mapping results are displayed online in a table format for easy checking and modification.

10. The method for identifying and processing object information according to claim 2 or 3, further comprising: When the name of the collection point item identified in the item information image cannot be used to determine the standard item name corresponding to the collection point item name from the standard item name table corresponding to the collection point, it is determined that the collection point item name recognition is abnormal, and the collection abnormality information corresponding to the abnormal collection point item name is stored separately.

11. The method for identifying and processing object information according to claim 10, further comprising: Determine whether the item parameter information is within a predetermined item parameter range, wherein the predetermined item parameter range is greater than a predetermined item parameter lower limit and less than a predetermined item parameter upper limit; In the case where the item parameter information is not within the predetermined item parameter range, the item parameter information is determined to be abnormal, and the collection abnormality information corresponding to the abnormal item parameter information is stored separately.

12. The method for identifying and processing object information according to claim 11, further comprising: Sending the collection abnormality information to the staff terminal for staff confirmation; Receive abnormal information feedback returned by the staff terminal; The collected abnormal information is processed accordingly according to the abnormal information feedback.

13. The method for identifying and processing item information according to claim 12, wherein: The corresponding processing of the collected abnormal information according to the abnormal information feedback comprises at least one of the following steps: If it is confirmed in the abnormal information feedback that the abnormal item parameter information is correct, the collection abnormality information corresponding to the abnormal item parameter information is stored together with the normal collection information; In the case where the abnormal item parameter information is corrected in the abnormal information feedback, the corrected acquisition abnormal information is stored together with the normal acquisition information; In the case where the name of the item at the abnormal collection point is maintained as the standard item name in the abnormal information feedback, the maintained collection abnormality information is stored together with the normal collection information; The collection exception information is deleted according to the deletion instruction in the exception information feedback.

14. The method for identifying and processing item information according to claim 3, further comprising: Receive item parameter information query request; Determine the target item parameter information based on the collection time, collection point information and standard item name in the item parameter information query request; Feedback the target item parameter information to the user terminal.

15. An object information identification and processing device, comprising: An information picture acquisition module is configured to acquire item information pictures from multiple collection points; A collection point type determination module is configured to determine the collection point type of each collection point; The layout format determination module is configured to determine the image layout format template corresponding to each acquisition point according to the correspondence between the acquisition point type and the image layout format template; The collection information recognition module is configured to identify the collection information of each item in the item information picture obtained at each collection point according to the information recognition method of the picture layout format template corresponding to the collection point, wherein the collection information includes the item name and item parameter information.

16. An object information identification and processing device, comprising: a memory configured to store instructions; and The processor is configured to execute the instruction so that the item information identification and processing device implements the item information identification and processing method as described in any one of claims 1 to 14.

17. A computer-readable storage medium, wherein: The computer-readable storage medium stores computer instructions, and when the instructions are executed by a processor, the object information identification and processing method according to any one of claims 1 to 14 is implemented.

18. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the object information identification and processing method according to any one of claims 1 to 14 is implemented.

Images