CN110555372A - Data entry method, device, equipment and storage medium - Google Patents
Data entry method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN110555372A CN110555372A CN201910663209.9A CN201910663209A CN110555372A CN 110555372 A CN110555372 A CN 110555372A CN 201910663209 A CN201910663209 A CN 201910663209A CN 110555372 A CN110555372 A CN 110555372A
- Authority
- CN
- China
- Prior art keywords
- image
- text
- processed
- boundary
- data entry
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/414—Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
Abstract
The application relates to the field of data processing, and particularly discloses a data entry method, a device, equipment and a storage medium, wherein the data entry method comprises the following steps: acquiring an image to be processed corresponding to a text file, wherein the image to be processed comprises a text area; extracting a boundary box of the text area in the image to be processed to obtain a boundary image, wherein the boundary image is the image to be processed in the boundary box; inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image; inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information; and storing the text information and the keywords corresponding to the text information into a target database to finish data entry. And further improve the efficiency and accuracy of data entry.
Description
Technical Field
The present application relates to the field of data processing, and in particular, to a data entry method, apparatus, device, and storage medium.
Background
most of traditional data storage methods utilize paper documents to store data, but the paper documents are not only easy to lose, but also easy to damage due to accidents such as water and fire, and the paper documents are also low in searching efficiency when being searched.
In order to improve the security of data storage, the existing data storage method is generally to manually input data on paper documents into a data management system. However, the existing manual entry mode is not only low in entry efficiency, but also easy to cause errors during entry, and low in entry accuracy.
Therefore, how to improve the efficiency and accuracy of data entry becomes an urgent problem to be solved.
disclosure of Invention
The application provides a data entry method, a data entry device, data entry equipment and a storage medium, and provides important reference for screening fraud.
In a first aspect, the present application provides a data entry method, the method comprising:
Acquiring an image to be processed corresponding to a text file, wherein the image to be processed comprises a text area;
extracting a boundary box of the text area in the image to be processed to obtain a boundary image, wherein the boundary image is the image to be processed in the boundary box;
Inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image;
inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information;
and storing the text information and the keywords corresponding to the text information into a target database to finish data entry.
in a second aspect, the present application also provides a data entry device, the device comprising:
The image processing device comprises a to-be-processed image module, a to-be-processed image module and a text processing module, wherein the to-be-processed image module is used for acquiring a to-be-processed image corresponding to a text file, and the to-be-processed image comprises a text area;
A boundary frame extraction module, configured to extract a boundary frame of the text region in the image to be processed to obtain a boundary image, where the boundary image is the image to be processed in the boundary frame;
The text recognition module is used for inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image;
The keyword extraction module is used for inputting the text information into a pre-trained keyword extraction model for keyword extraction so as to obtain keywords corresponding to the text information;
And the data entry module is used for storing the text information and the keywords corresponding to the text information into a target database so as to complete data entry.
In a third aspect, the present application further provides a computer device comprising a memory and a processor; the memory is used for storing a computer program; the processor is configured to execute the computer program and to implement the data entry method as described above when the computer program is executed.
In a fourth aspect, the present application also provides a computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to implement the data entry method as described above.
the application discloses a data entry method, a device, equipment and a storage medium, wherein a to-be-processed image corresponding to a text file is obtained, then a boundary box of a text area in the to-be-processed image is extracted to obtain a boundary image, the boundary image is input into an image text recognition model to perform text recognition, text information is output, then the text information is input into a keyword extraction model to perform keyword extraction, so that keywords are obtained, and finally the text information and the keywords corresponding to the text information are stored in a target database to finish data entry. Compared with a mode that information is input into the content of a paper file manually, the scheme processes the image to be processed corresponding to the text file, so that the input of text information is realized, and the efficiency and the accuracy of data input are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart diagram of a training method of an image text recognition model provided by an embodiment of the application;
FIG. 2 is a schematic flow chart of a data entry method provided by an embodiment of the present application;
FIG. 3 is a schematic flow chart diagram of sub-steps provided by an embodiment of the present application;
FIG. 4 is a schematic flow chart of keyword extraction for text information according to an embodiment of the present disclosure;
FIG. 5 is a schematic flow chart diagram of a sub-step of keyword extraction on text information provided by an embodiment of the present application;
FIG. 6 is a flow chart illustrating steps of another data entry method provided by an embodiment of the present application;
FIG. 7 is a schematic flow chart diagram of sub-steps provided by an embodiment of the present application;
FIG. 8 is a schematic block diagram of a model training apparatus provided in an embodiment of the present application;
FIG. 9 is a schematic block diagram of a data entry device that is also provided by an embodiment of the present application;
FIG. 10 is a schematic block diagram of yet another data entry device provided by an embodiment of the present application;
Fig. 11 is a schematic block diagram of a computer device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The flow diagrams depicted in the figures are merely illustrative and do not necessarily include all of the elements and operations/steps, nor do they necessarily have to be performed in the order depicted. For example, some operations/steps may be decomposed, combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
It is to be understood that the terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
The embodiment of the application provides a data entry method and device, computer equipment and a storage medium. The data entry method can be applied to a terminal or a server to improve the accuracy and efficiency of data entry.
some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.
Referring to fig. 1, fig. 1 is a schematic flow chart of a training method of an image text recognition model according to an embodiment of the present application. The image text recognition model is obtained by model training based on a convolutional neural network, and can be obtained by training with other networks.
It should be noted that, in this embodiment, google lenet is used to perform model training to obtain the target recognition model, but of course, other networks may also be used, such as a Deep learning algorithm using one or more combinations of a Convolutional Neural Network (CNN), a Deep Residual Network (dressnet), a Long Short-Term Memory Network (LSTM), and the like. The following description will be given by taking google lenet as an example.
As shown in FIG. 1, the training method of the image text recognition model is used for training the image text recognition model to be applied to a data entry method. The training method of the image text recognition model comprises a step S101 and a step S102.
and S101, acquiring a text image sample.
Wherein the text image sample is an image including a text region. The content of the text area can be contract content, can also be identity card content, and can also be other text content. The following description will be made in detail taking the content of the text area as the contract content as an example.
In this embodiment, the text image samples may be different types of contract images such as a sales contract image, a transportation contract image, a technical contract image, and the like, and the contract images constitute the text image samples for training the image text recognition model. And a plurality of different contract images are used as samples, so that the identification accuracy of the image text identification model is improved.
and S102, based on a convolutional neural network, performing model training according to the text image sample to obtain an image text recognition model, and taking the image text recognition model as a preset image text recognition model.
Specifically, model training is performed through google lenet by using the constructed sample data, specifically, directional propagation training can be adopted, features are extracted from input sample data by using a convolution layer and a pooling layer of google lenet, a complete connection layer is used as a classifier, and the output of the classifier is probability values of different images and texts.
Initializing all filters and parameters/weights with random values; the convolutional neural network takes the training sample data as input, and finally obtains the output probability of each class through the forward propagation steps (convolution, ReLU activation and pooling operation for forward propagation in the fully connected layer).
And using the partial image in the sample data as calibration data (ground route), enabling a convolutional neural network to output the output probability of each text after learning the semantic information of the picture by utilizing the prepared sample data through large-scale iterative training, and reducing the loss function (loss) as much as possible in model training by using the output probability and the defined loss function (loss) of the calibration data (ground route) to ensure the accuracy of the model so as to finish the model training.
since the data entry method can be applied to a terminal or a server, the trained model needs to be saved in the terminal or the server. The terminal can be an electronic device such as a mobile phone, a tablet computer, a notebook computer, a desktop computer, a personal digital assistant and a wearable device; the servers may be independent servers or server clusters.
If the method is applied to the terminal, in order to ensure the normal operation of the terminal and quickly identify and detect the text information of the image, the image text identification model obtained by training needs to be compressed, and the compressed model is stored in the terminal.
The compression processing specifically comprises pruning processing, quantization processing, Huffman coding processing and the like on the image text recognition model so as to reduce the size of the image text recognition model and further facilitate storage in a terminal with smaller capacity.
According to the training method provided by the embodiment, the accuracy of text recognition in the image can be improved by acquiring various text image samples, then performing model training according to the text image samples based on the convolutional neural network to obtain the image text model, and applying the obtained image text model as the pre-trained image text model to the data entry method.
referring to fig. 2, fig. 2 is a schematic flowchart of a data entry method according to an embodiment of the present application. The data entry method is used for identifying and entering data, and improves entry efficiency and accuracy. The following description will take an example in which the data entry method is applied to a server.
as shown in fig. 2, the data entry method specifically includes: step S201 to step S205.
s201, acquiring the image to be processed corresponding to the text file.
specifically, the image to be processed includes a text region and a non-text region. In a specific implementation process, the image to be processed corresponding to the text file may be acquired by an image acquisition device such as a camera. After the user collects the image to be processed through the image collecting device, the server acquires the image to be processed collected by the image collecting device so as to perform subsequent operation on the image to be processed.
S202, extracting a boundary box of the text area in the image to be processed to obtain a boundary image.
And the boundary image is an image to be processed in the boundary frame. The method comprises the steps of extracting a boundary frame of a text area to obtain a boundary image, and then performing text recognition on the extracted boundary image, so that the calculated amount during contract image recognition is reduced, and the contract image recognition efficiency is improved.
In one embodiment, as shown in fig. 3, in order to improve the accuracy of the boundary box extraction, the boundary box of the text region in the image to be processed is extracted to obtain a boundary image, which specifically includes sub-steps S202a and S202 b.
S202a, identifying the text area in the image to be processed to obtain the size information and the position information of the text area.
The identification of the text area in the image to be processed refers to automatic analysis of texts, table information and position relations in the image to be processed. The position information of the text region can be acquired by identifying the text region in the image to be processed, and the size information of the text region can be judged according to the position information of the text region.
specifically, firstly, a text area in an image to be processed is identified to obtain area coordinates of a text in the text area, wherein the area coordinates refer to pixel position coordinates of the text area on the image to be processed; and calculating an outsourcing area according to the area coordinates of the text in the text area so as to obtain the size information of the text area, wherein the outsourcing area is the minimum area comprising the text area.
S202b, determining a boundary box of the text area according to the size information and the position information, and taking the image to be processed in the boundary box as a boundary image.
In a specific implementation process, a boundary box of the text region can be determined according to the size information and the position information of the text region, then an image to be processed in the boundary box is extracted based on the boundary box, and the extracted image to be processed is used as a boundary image.
s203, inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image.
Specifically, the boundary image is input to a pre-trained image text recognition model for text recognition, and the image text recognition model outputs text information included in the boundary image.
And S204, inputting the text information into a pre-trained keyword extraction model for keyword extraction so as to obtain keywords corresponding to the text information.
Specifically, the text information recognized from the boundary image is input into a pre-trained keyword extraction model for keyword extraction, so that corresponding keywords are obtained.
For example: the text information identified from the boundary image is: the validity period of the contract is from 28 days 6 and 28 months in 2017 to 28 days 6 and 28 months in 2019. Extracting the keywords through a keyword extraction model to obtain the corresponding keywords as follows: the effective date is 2017, 6 and 28, and the end date is 2018, 6 and 28.
In some embodiments, as shown in fig. 4, the extracting of the keywords from the text information specifically includes substeps S204a to S204 c.
S204a, performing word segmentation on the text information and obtaining word segmentation results.
Wherein the word segmentation result comprises at least one word segmentation. Specifically, the maximum matching algorithm may be adopted to perform word segmentation on the text information to obtain each word segmentation included in the text information, and a set of each word segmentation included in the text information is used as a word segmentation result.
The maximum matching algorithm is to take the longest word in the dictionary as the first scanning string based on the dictionary and scan in the dictionary. For example: the longest word in the dictionary is 7 Chinese characters in total, and the maximum matching initial word number is 7 Chinese characters. Then decreasing the number word by word and searching in the corresponding dictionary. It will be appreciated that in other embodiments, other methods may be used to segment the textual information, such as using the Viterbi (Viterbi) algorithm.
S204b, respectively inputting at least one word segmentation into a pre-trained keyword extraction model to obtain the importance weight corresponding to each word segmentation.
specifically, the word vector corresponding to each participle is input into the keyword extraction model, and the importance weight of each participle is output by the keyword extraction model, wherein the greater the importance weight is, the stronger the importance of the participle is.
The keyword extraction model can be obtained by selecting a preset number of standard contract samples on a network to perform model training on the deep learning model. For example, 1000 standard contract samples are selected for model training. Specifically, the keyword extraction model comprises a bidirectional long-time memory (BLSTM) algorithm model, a Max Pooling (Max Pooling) algorithm model and a Softmax algorithm model which are sequentially connected in series.
S204c, selecting the participle corresponding to the importance weight as the key word of the text message according to the importance weight.
The keywords can be keyword samples such as contract signing date, effective date, ending date, timeliness and the like. The keywords corresponding to the text information are extracted through the keyword extraction model, so that when the text information is inquired, each word in the text information does not need to be traversed, the condition that the inquiry data amount is too large is avoided, and the inquiry efficiency is improved.
In some embodiments, referring to fig. 5, step S204c includes steps S204c1 and S204c 2.
S204c1, sorting the participles according to the importance weight to obtain a sorting result.
S204c2, selecting participles as keywords of the text information based on the sorting result.
specifically, according to the importance weight, the segmentation words in the segmentation result are sorted in a descending order, and a preset number of segmentation words are selected as the keywords of the text information. For example, the word with the top three importance weights may be selected as the keyword of the ranking result.
for example: the contract date of a certain insurance contract is 2018, 4 and 10 months, the effective date is 2018, 4 and 20 months, the end date is 2019, 4 and 10 months, and the aging is one year. Wherein, the importance weight of the contract date is 0.28, the importance weight of the effective date is 0.22, the importance weight of the end date is 0.42, and the importance weight of the aging is 0.08. The ordering result is as follows: the expiration date is 2019, 4 and 10 months, the signing date is 2018, 4 and 10 months, the effective date is 2018, 4 and 20 months, and the aging is one year. The keywords are the expiration date of 2019, 4 and 10 months, the signing date of 2018, 4 and 10 months, and the effective date of 2018, 4 and 20 months.
S205, storing the text information and the keywords corresponding to the text information into a target database to complete data entry.
Specifically, after the keywords of the text information are obtained, the text information and the keywords are stored in the target database, so that the entry of the text information is completed, namely the entry of the contract related information is completed, and the entry efficiency and the entry accuracy are improved.
according to the data entry method provided by the embodiment, the image to be processed including the text area is obtained; then extracting a boundary frame of the text area in the image to be processed to obtain a boundary image; inputting the boundary image into a pre-trained image text recognition model for text recognition, thereby outputting text information; inputting the text information into a keyword extraction model for keyword extraction, thereby obtaining keywords; and finally, storing the text information and the keywords corresponding to the text information into a target database to finish data entry. The text information is input by processing the image to be processed corresponding to the text file, and the efficiency and accuracy of data input are improved.
Referring to fig. 6, fig. 6 is a schematic flow chart illustrating steps of another data entry method according to an embodiment of the present application. The data entry method can improve the accuracy and efficiency of data entry and improve the efficiency of data query.
as shown in fig. 6, the data entry method specifically includes: step S301 to step S308.
S301, acquiring the image to be processed corresponding to the text file.
Specifically, the image to be processed includes a text region and a non-text region. In a specific implementation process, the image to be processed corresponding to the text file may be acquired by an image acquisition device such as a camera. After the user collects the image to be processed through the image collecting device, the server acquires the image to be processed collected by the image collecting device so as to perform subsequent operation on the image to be processed.
S302, extracting a boundary box of the text area in the image to be processed to obtain a boundary image.
And the boundary image is an image to be processed in the boundary frame. The method comprises the steps of extracting a boundary frame of a text area to obtain a boundary image, and then performing text recognition on the extracted boundary image, so that the calculated amount during contract image recognition is reduced, and the contract image recognition efficiency is improved.
Specifically, as shown in fig. 7, to avoid interference of noise, background image, etc. in the contract picture to be recognized, and improve picture recognition accuracy, the boundary box of the text region in the image to be processed is extracted to obtain a boundary image, including substeps S302a to S302 c.
S302a, performing image smoothing processing and wavelet filtering processing on the image to be processed in the bounding box to obtain a denoised image.
Specifically, the image smoothing process and the wavelet filtering process can remove noise points of the boundary image, thereby causing less blurring of the boundary image.
the image smoothing process may adopt a neighborhood averaging method. The neighborhood averaging method is to assign the average value of one pixel and all the pixels in its neighborhood to the corresponding pixel in the output image to reach the aim of smoothing. Of course, in other embodiments, other methods of image balancing processing, such as median filtering, may be used.
S302b, performing direction correction processing on the denoised image to obtain a corrected image.
Since there may be multiple angles of rotation of the received contract, it is necessary to rotate the contract to the correct orientation for further operations. And carrying out direction correction processing on the denoised image so as to enable the contract to rotate in the correct direction, thereby obtaining a corrected image.
Specifically, the image compression normal position network can be adopted to rotate the denoised image to complete the direction correction of the denoised image, so that the contract text in the denoised image is in the correct direction to obtain the corrected image. The image compression normal position network is obtained by training through a machine learning method and has an image rotation function.
S302c, performing background removing processing on the corrected image to obtain a background-removed image as a boundary image.
Specifically, the corrected image includes a contract image and a background image, and the interference of the background image in the corrected image can be removed through background removal processing.
And S303, inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image.
Specifically, the boundary image is input to a pre-trained image text recognition model for text recognition, so that text information included in the boundary image is output.
S304, inputting the text information into a pre-trained text classification model for class recognition, so as to output a classification class corresponding to the text information.
for example, if the text information is an insurance contract, the text classification model may identify that the classification category of the insurance contract includes information such as an insurance target and insurance risk. The insurance mark comprises: personal insurance or property insurance. Insurance security risks include: life insurance, personal accidental injury insurance or health insurance, etc. The insurance target and the insurance security risk of the recognized text message are personal insurance and life insurance, respectively.
S305, inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information.
Specifically, the text information recognized from the boundary image is input into a pre-trained keyword extraction model for keyword extraction, so that corresponding keywords are obtained.
It should be noted that the present embodiment does not limit the execution sequence between step S304 and step S305.
s306, storing the text information and the keywords corresponding to the text information into a sub-database corresponding to a target database according to the classification category so as to complete data entry.
Specifically, each classification category corresponds to one sub-database, and the set of sub-databases constitutes a target database. After the classification category corresponding to the text information is obtained, a sub-database corresponding to the classification category in the target database can be determined, and the text information and the keyword information corresponding to the text information are stored in the sub-database, so that the efficient entry of the same information is completed, the entry accuracy is high, and the labor cost is reduced.
S307, receiving a data query request.
Wherein the data query request includes query keywords, and the query keywords may include classification category, effective date, and the like. Specifically, the query request may be a text with the query request converted from voice information input by the user through a question and talk mode, or a data query request directly sent to the server by the user.
S308, matching the keywords in the target database according to the query keywords to obtain target text information corresponding to the query keywords.
Specifically, after a data query request is received, matching is performed in the target database according to query keywords in the data query request. If the query key words are matched with the key words in the target database consistently, the text information corresponding to the matched key words is taken as the target text information to be output, and therefore query of the target contract is completed efficiently.
When the query keywords comprise classification categories, a sub-database corresponding to the classification categories is selected from the target database according to the classification categories, and then keyword matching is performed in the sub-database according to other query keywords, so that the query efficiency is improved.
The data entry method provided by the embodiment comprises the steps of obtaining an image to be processed comprising a text area; then extracting a boundary frame of the text area in the image to be processed to obtain a boundary image; performing text recognition on the boundary image, thereby outputting text information; performing category identification on the text information to obtain a classification category of the text information; extracting keywords from the text information to obtain keywords; and finally, storing the text information and the keywords corresponding to the text information into a target database according to the classification category so as to finish data entry. And then receiving a data query request, and matching keywords in the target database according to the data query request so as to obtain text information. The text information is input by processing the image to be processed corresponding to the text file, and the efficiency and accuracy of data input are improved. The text information is classified, the orderliness in data entry is improved, and the query efficiency is improved when data query is performed.
Referring to fig. 8, fig. 8 is a schematic block diagram of a model training apparatus according to an embodiment of the present application, which may be configured in a server for executing the aforementioned training method of the image-text recognition model.
As shown in fig. 8, the model training apparatus 400 includes: a sample acquisition module 401 and a model training module 402.
a sample obtaining module 401, configured to obtain a text image sample, where the text image sample is an image including a text region.
and the model training module 402 is configured to perform model training according to the text image sample based on a convolutional neural network to obtain an image text recognition model, and use the image text recognition model as a preset image text recognition model.
Referring to fig. 9, fig. 9 is a schematic block diagram of a data entry device for executing the foregoing data entry method according to an embodiment of the present application. Wherein, the data entry device can be configured in a server or a terminal.
The server may be an independent server or a server cluster. The terminal can be an electronic device such as a mobile phone, a tablet computer, a notebook computer, a desktop computer, a personal digital assistant and a wearable device.
As shown in fig. 9, the data entry device 500 includes: a pending image module 501, a bounding box extraction module 502, a text recognition module 503, a keyword extraction module 504, and a data entry module 505.
The image to be processed module 501 obtains an image to be processed corresponding to the text file, where the image to be processed includes a text region.
A bounding box extracting module 502, configured to extract a bounding box of the text region in the image to be processed to obtain a bounding image, where the bounding image is the image to be processed in the bounding box.
In one embodiment, the bounding box extraction module 502 includes a region identification submodule 5021 and a bounding box determination submodule 5022.
The region identification submodule 5021 is configured to identify a text region in the image to be processed, so as to obtain size information and position information of the text region. The bounding box determining sub-module 5022 is configured to determine a bounding box of the text region according to the size information and the position information, and use the image to be processed in the bounding box as a bounding image.
And a text recognition module 503, configured to input the boundary image into a pre-trained image text recognition model for text recognition, so as to output text information corresponding to the boundary image.
And a keyword extraction module 504, configured to input the text information into a pre-trained keyword extraction model to perform keyword extraction, so as to obtain a keyword corresponding to the text information.
In some embodiments, the keyword extraction module 504 includes a segmentation result sub-module 5041, a weight acquisition sub-module 5042, and a keyword determination sub-module 5043.
The word segmentation result sub-module 5041 is configured to perform word segmentation on the text information and obtain a word segmentation result. The weight obtaining sub-module 5042 is configured to input at least one of the segmented words into a pre-trained keyword extraction model, so as to obtain an importance weight corresponding to each of the segmented words. And the keyword determining submodule 5043 is configured to select the participle corresponding to the importance weight as the keyword of the text message according to the importance weight.
In some embodiments, the keyword determination sub-module 5043 is specifically configured to: sorting each participle according to the importance weight to obtain a sorting result; and selecting the participles as the keywords of the text information based on the sequencing result.
and a data entry module 505, configured to store the text information and the keyword corresponding to the text information in a target database, so as to complete data entry.
referring to fig. 10, fig. 10 is a schematic block diagram of another data entry device for executing the data entry method according to the embodiment of the present application. Wherein the data entry device may be configured in a server.
As shown in fig. 10, the data entry device 600 includes: a pending image module 601, a bounding box extraction module 602, a text recognition module 603, a category determination module 604, a keyword extraction module 605, a data entry module 606, a request receiving module 607, and a query matching module 608.
The image to be processed module 601 obtains an image to be processed corresponding to the text file, where the image to be processed includes a text region.
A bounding box extracting module 602, configured to extract a bounding box of the text region in the image to be processed to obtain a bounding image, where the bounding image is the image to be processed in the bounding box.
in one embodiment, the bounding box extraction module 602 includes a denoised image sub-module 6021, a rectified image sub-module 6022, and a background removed image sub-module 6023.
The denoised image sub-module 6021 is configured to perform image smoothing processing and wavelet filtering processing on the image to be processed in the bounding box to obtain a denoised image. And the corrected image submodule 6022 is configured to perform direction correction processing on the denoised image to obtain a corrected image. A background image removing submodule 6023 configured to perform background removing processing on the rectified image to obtain a background removed image as a boundary image.
The text recognition module 603 is configured to input the boundary image into a pre-trained image text recognition model for text recognition, so as to output text information corresponding to the boundary image.
The category determining module 604 is configured to input the text information into a pre-trained text classification model for category identification, so as to output a classification category corresponding to the text information.
And a keyword extraction module 605, configured to input the text information into a pre-trained keyword extraction model to perform keyword extraction, so as to obtain a keyword corresponding to the text information.
And a data entry module 606, configured to store the text information and the keyword corresponding to the text information in a sub-database corresponding to a target database according to the classification category, so as to complete data entry.
The request receiving module 607 is configured to receive a data query request, where the data query request includes a query keyword.
And a query matching module 608, configured to match the keywords in the target database according to the query keyword to obtain target text information corresponding to the query keyword.
it should be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the data entry apparatus and each module described above may refer to corresponding processes in the foregoing data entry method embodiment, and are not described herein again.
The data entry means described above may be implemented in the form of a computer program which can be run on a computer device as shown in figure 11.
Referring to fig. 11, fig. 11 is a schematic block diagram of a computer device according to an embodiment of the present disclosure. The computer device may be a server or a terminal.
referring to fig. 11, the computer device includes a processor, a memory, and a network interface connected through a system bus, wherein the memory may include a nonvolatile storage medium and an internal memory.
The non-volatile storage medium may store an operating system and a computer program. The computer program comprises program instructions which, when executed, cause a processor to perform any of the data entry methods.
The processor is used for providing calculation and control capability and supporting the operation of the whole computer equipment.
the internal memory provides an environment for the execution of a computer program on a non-volatile storage medium, which when executed by the processor, causes the processor to perform any of a variety of data entry methods.
The network interface is used for network communication, such as sending assigned tasks and the like. Those skilled in the art will appreciate that the architecture shown in fig. 11 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
it should be understood that the Processor may be a Central Processing Unit (CPU), and the Processor may be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of:
acquiring an image to be processed corresponding to a text file, wherein the image to be processed comprises a text area; extracting a boundary box of the text area in the image to be processed to obtain a boundary image, wherein the boundary image is the image to be processed in the boundary box; inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image; inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information; and storing the text information and the keywords corresponding to the text information into a target database to finish data entry.
in one embodiment, the processor, when implementing the extracting the bounding box of the text region within the image to be processed to obtain a bounding image, is configured to implement:
Identifying a text region in the image to be processed to acquire size information and position information of the text region; and determining a boundary box of the text area according to the size information and the position information, and taking the image to be processed in the boundary box as a boundary image.
in another embodiment, the processor, when implementing the extracting the bounding box of the text region within the image to be processed to obtain a bounding image, is configured to implement:
Performing image smoothing processing and wavelet filtering processing on the image to be processed in the boundary frame to obtain a de-noised image; carrying out direction correction processing on the denoised image to obtain a corrected image; and performing background removal processing on the corrected image to obtain a background-removed image as a boundary image.
In some embodiments, when implementing the inputting of the text information into a pre-trained keyword extraction model for keyword extraction to obtain a keyword corresponding to the text information, the processor is configured to implement:
Performing word segmentation on the text information and obtaining word segmentation results, wherein the word segmentation results comprise at least one word segmentation; respectively inputting at least one word segmentation into a pre-trained keyword extraction model to obtain the importance weight corresponding to each word segmentation; and selecting the participles corresponding to the importance weights as the keywords of the text information according to the importance weights.
In one embodiment, when the processor selects the participle corresponding to the importance weight as the keyword of the text message according to the importance weight, the processor is configured to:
Sorting each participle according to the importance weight to obtain a sorting result; and selecting the participles as the keywords of the text information based on the sequencing result.
wherein in another embodiment the processor is adapted to run a computer program stored in the memory to implement the steps of:
Acquiring an image to be processed corresponding to a text file, wherein the image to be processed comprises a text area; extracting a boundary box of the text area in the image to be processed to obtain a boundary image, wherein the boundary image is the image to be processed in the boundary box; inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image; inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information; storing the text information and the keywords corresponding to the text information into a target database to complete data entry; receiving a data query request, wherein the data query request comprises query keywords; and matching the keywords in the target database according to the query keywords to obtain target text information corresponding to the query keywords.
Wherein in another embodiment the processor is adapted to run a computer program stored in the memory to implement the steps of:
Acquiring a text image sample, wherein the text image sample is an image comprising a text area; based on a convolutional neural network, performing model training according to the text image sample to obtain an image text recognition model, and taking the image text recognition model as a preset image text recognition model.
The embodiment of the application also provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, the computer program comprises program instructions, and the processor executes the program instructions to realize any data entry method provided by the embodiment of the application.
The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the computer device.
While the invention has been described with reference to specific embodiments, the scope of the invention is not limited thereto, and those skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A data entry method, comprising:
acquiring an image to be processed corresponding to a text file, wherein the image to be processed comprises a text area;
extracting a boundary box of the text area in the image to be processed to obtain a boundary image, wherein the boundary image is the image to be processed in the boundary box;
Inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image;
Inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information;
And storing the text information and the keywords corresponding to the text information into a target database to finish data entry.
2. A data entry method as claimed in claim 1, wherein said extracting a bounding box of said text region within said image to be processed to obtain a bounding image comprises:
Identifying a text region in the image to be processed to acquire size information and position information of the text region;
And determining a boundary box of the text area according to the size information and the position information, and taking the image to be processed in the boundary box as a boundary image.
3. A data entry method as claimed in claim 1, wherein said extracting a bounding box of said text region within said image to be processed to obtain a bounding image comprises:
Performing image smoothing processing and wavelet filtering processing on the image to be processed in the boundary frame to obtain a de-noised image;
Carrying out direction correction processing on the denoised image to obtain a corrected image;
and performing background removal processing on the corrected image to obtain a background-removed image as a boundary image.
4. The data entry method of claim 1, wherein inputting the text information into a pre-trained keyword extraction model for keyword extraction to obtain keywords corresponding to the text information comprises:
performing word segmentation on the text information and obtaining word segmentation results, wherein the word segmentation results comprise at least one word segmentation;
respectively inputting at least one word segmentation into a pre-trained keyword extraction model to obtain the importance weight corresponding to each word segmentation;
And selecting the participles corresponding to the importance weights as the keywords of the text information according to the importance weights.
5. The data entry method of claim 4, wherein the selecting the participle corresponding to the importance weight as the keyword of the text message according to the importance weight comprises:
Sorting each participle according to the importance weight to obtain a sorting result;
And selecting the participles as the keywords of the text information based on the sequencing result.
6. A data entry method as claimed in claim 1, further comprising:
Receiving a data query request, wherein the data query request comprises query keywords;
And matching the keywords in the target database according to the query keywords to obtain target text information corresponding to the query keywords.
7. A data entry method as claimed in claim 1, further comprising:
Acquiring a text image sample, wherein the text image sample is an image comprising a text area;
based on a convolutional neural network, performing model training according to the text image sample to obtain an image text recognition model, and taking the image text recognition model as a preset image text recognition model.
8. A data entry device, comprising:
The image processing device comprises a to-be-processed image module, a to-be-processed image module and a text processing module, wherein the to-be-processed image module is used for acquiring a to-be-processed image corresponding to a text file, and the to-be-processed image comprises a text area;
A boundary frame extraction module, configured to extract a boundary frame of the text region in the image to be processed to obtain a boundary image, where the boundary image is the image to be processed in the boundary frame;
The text recognition module is used for inputting the boundary image into a pre-trained image text recognition model for text recognition so as to output text information corresponding to the boundary image;
The keyword extraction module is used for inputting the text information into a pre-trained keyword extraction model for keyword extraction so as to obtain keywords corresponding to the text information;
And the data entry module is used for storing the text information and the keywords corresponding to the text information into a target database so as to complete data entry.
9. A computer device, wherein the computer device comprises a memory and a processor;
The memory is used for storing a computer program;
The processor for executing the computer program and implementing the data entry method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, causes the processor to implement a data entry method as claimed in any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910663209.9A CN110555372A (en) | 2019-07-22 | 2019-07-22 | Data entry method, device, equipment and storage medium |
| PCT/CN2019/122812 WO2021012570A1 (en) | 2019-07-22 | 2019-12-03 | Data entry method and device, apparatus, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910663209.9A CN110555372A (en) | 2019-07-22 | 2019-07-22 | Data entry method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110555372A true CN110555372A (en) | 2019-12-10 |
Family
ID=68735720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910663209.9A Pending CN110555372A (en) | 2019-07-22 | 2019-07-22 | Data entry method, device, equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110555372A (en) |
| WO (1) | WO2021012570A1 (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111401042A (en) * | 2020-03-26 | 2020-07-10 | 支付宝(杭州)信息技术有限公司 | Method and system for training text key content extraction model |
| CN111507850A (en) * | 2020-03-25 | 2020-08-07 | 上海商汤智能科技有限公司 | Authority guaranteeing method and related device and equipment |
| CN111507333A (en) * | 2020-04-21 | 2020-08-07 | 腾讯科技（深圳）有限公司 | Image correction method and device, electronic equipment and storage medium |
| CN111652229A (en) * | 2020-05-25 | 2020-09-11 | 泰康保险集团股份有限公司 | Information input method and device, electronic equipment and storage medium |
| CN111723816A (en) * | 2020-06-28 | 2020-09-29 | 北京联想软件有限公司 | Teaching note acquisition method and electronic equipment |
| CN111950353A (en) * | 2020-06-30 | 2020-11-17 | 深圳市雄帝科技股份有限公司 | Seal text recognition method and device and electronic equipment |
| CN112381038A (en) * | 2020-11-26 | 2021-02-19 | 中国船舶工业系统工程研究院 | Image-based text recognition method, system and medium |
| CN112560411A (en) * | 2020-12-21 | 2021-03-26 | 深圳供电局有限公司 | Intelligent personnel information input method and system |
| CN112906801A (en) * | 2021-02-26 | 2021-06-04 | 北京房江湖科技有限公司 | Image processing method, device and storage medium |
| CN113033832A (en) * | 2021-01-21 | 2021-06-25 | 深圳市轱辘车联数据技术有限公司 | Vapour repair data entry method and device, terminal equipment and readable storage medium |
| CN116757886A (en) * | 2023-08-16 | 2023-09-15 | 南京尘与土信息技术有限公司 | Data analysis method and analysis device |
| CN116894937A (en) * | 2023-06-25 | 2023-10-17 | 德联易控科技(北京)有限公司 | Method, system and electronic equipment for acquiring parameters of wheel aligner |
| CN112381038B (en) * | 2020-11-26 | 2024-04-19 | 中国船舶工业系统工程研究院 | Text recognition method, system and medium based on image |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112966583A (en) * | 2021-02-26 | 2021-06-15 | 深圳壹账通智能科技有限公司 | Image processing method, image processing device, computer equipment and storage medium |
| CN113361442B (en) * | 2021-06-21 | 2024-03-29 | 梅卡曼德（北京）机器人科技有限公司 | Image recognition method, device, electronic equipment and storage medium |
| CN113505588A (en) * | 2021-07-14 | 2021-10-15 | 京东科技控股股份有限公司 | Information extraction method and device and electronic equipment |
| CN113762160A (en) * | 2021-09-08 | 2021-12-07 | 深圳前海环融联易信息科技服务有限公司 | Date extraction method and device, computer equipment and storage medium |
| CN114187605B (en) * | 2021-12-13 | 2023-02-28 | 苏州方兴信息技术有限公司 | Data integration method and device and readable storage medium |
| CN115862036B (en) * | 2022-12-14 | 2024-02-23 | 北京瑞莱智慧科技有限公司 | Information interference model training method, information interference method, related device and medium |
| CN116522011B (en) * | 2023-05-16 | 2024-02-13 | 深圳九星互动科技有限公司 | Big data-based pushing method and pushing system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102317955A (en) * | 2009-04-20 | 2012-01-11 | 万涛国际有限公司 | Data managing method and system based on image |
| CN108664473A (en) * | 2018-05-11 | 2018-10-16 | 平安科技（深圳）有限公司 | Recognition methods, electronic device and the readable storage medium storing program for executing of text key message |
| CN108764226A (en) * | 2018-04-13 | 2018-11-06 | 顺丰科技有限公司 | Image text recognition methods, device, equipment and its storage medium |
| US20190065841A1 (en) * | 2017-08-25 | 2019-02-28 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and device for acquiring text data from trademark image, computer device and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10817656B2 (en) * | 2017-11-22 | 2020-10-27 | Adp, Llc | Methods and devices for enabling computers to automatically enter information into a unified database from heterogeneous documents |
| CN109492143A (en) * | 2018-09-21 | 2019-03-19 | 平安科技（深圳）有限公司 | Image processing method, device, computer equipment and storage medium |
| CN110032990A (en) * | 2019-04-23 | 2019-07-19 | 杭州智趣智能信息技术有限公司 | A kind of invoice text recognition method, system and associated component |
| CN109977956B (en) * | 2019-04-29 | 2022-11-18 | 腾讯科技（深圳）有限公司 | Image processing method and device, electronic equipment and storage medium |
-
2019
- 2019-07-22 CN CN201910663209.9A patent/CN110555372A/en active Pending
- 2019-12-03 WO PCT/CN2019/122812 patent/WO2021012570A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102317955A (en) * | 2009-04-20 | 2012-01-11 | 万涛国际有限公司 | Data managing method and system based on image |
| US20190065841A1 (en) * | 2017-08-25 | 2019-02-28 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and device for acquiring text data from trademark image, computer device and storage medium |
| CN108764226A (en) * | 2018-04-13 | 2018-11-06 | 顺丰科技有限公司 | Image text recognition methods, device, equipment and its storage medium |
| CN108664473A (en) * | 2018-05-11 | 2018-10-16 | 平安科技（深圳）有限公司 | Recognition methods, electronic device and the readable storage medium storing program for executing of text key message |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111507850A (en) * | 2020-03-25 | 2020-08-07 | 上海商汤智能科技有限公司 | Authority guaranteeing method and related device and equipment |
| CN111401042B (en) * | 2020-03-26 | 2023-04-14 | 支付宝(杭州)信息技术有限公司 | Method and system for training text key content extraction model |
| CN111401042A (en) * | 2020-03-26 | 2020-07-10 | 支付宝(杭州)信息技术有限公司 | Method and system for training text key content extraction model |
| CN111507333A (en) * | 2020-04-21 | 2020-08-07 | 腾讯科技（深圳）有限公司 | Image correction method and device, electronic equipment and storage medium |
| CN111507333B (en) * | 2020-04-21 | 2023-09-15 | 腾讯科技（深圳）有限公司 | Image correction method and device, electronic equipment and storage medium |
| CN111652229A (en) * | 2020-05-25 | 2020-09-11 | 泰康保险集团股份有限公司 | Information input method and device, electronic equipment and storage medium |
| CN111652229B (en) * | 2020-05-25 | 2023-09-12 | 泰康保险集团股份有限公司 | Information input method and device, electronic equipment and storage medium |
| CN111723816B (en) * | 2020-06-28 | 2023-10-27 | 北京联想软件有限公司 | Acquisition method of teaching notes and electronic equipment |
| CN111723816A (en) * | 2020-06-28 | 2020-09-29 | 北京联想软件有限公司 | Teaching note acquisition method and electronic equipment |
| CN111950353A (en) * | 2020-06-30 | 2020-11-17 | 深圳市雄帝科技股份有限公司 | Seal text recognition method and device and electronic equipment |
| CN111950353B (en) * | 2020-06-30 | 2024-04-19 | 深圳市雄帝科技股份有限公司 | Seal text recognition method and device and electronic equipment |
| CN112381038B (en) * | 2020-11-26 | 2024-04-19 | 中国船舶工业系统工程研究院 | Text recognition method, system and medium based on image |
| CN112381038A (en) * | 2020-11-26 | 2021-02-19 | 中国船舶工业系统工程研究院 | Image-based text recognition method, system and medium |
| CN112560411A (en) * | 2020-12-21 | 2021-03-26 | 深圳供电局有限公司 | Intelligent personnel information input method and system |
| CN113033832A (en) * | 2021-01-21 | 2021-06-25 | 深圳市轱辘车联数据技术有限公司 | Vapour repair data entry method and device, terminal equipment and readable storage medium |
| CN113033832B (en) * | 2021-01-21 | 2024-04-26 | 深圳市轱辘车联数据技术有限公司 | Method and device for inputting automobile repair data, terminal equipment and readable storage medium |
| CN112906801A (en) * | 2021-02-26 | 2021-06-04 | 北京房江湖科技有限公司 | Image processing method, device and storage medium |
| CN116894937A (en) * | 2023-06-25 | 2023-10-17 | 德联易控科技(北京)有限公司 | Method, system and electronic equipment for acquiring parameters of wheel aligner |
| CN116894937B (en) * | 2023-06-25 | 2024-02-06 | 德联易控科技(北京)有限公司 | Method, system and electronic equipment for acquiring parameters of wheel aligner |
| CN116757886B (en) * | 2023-08-16 | 2023-11-28 | 南京尘与土信息技术有限公司 | Data analysis method and analysis device |
| CN116757886A (en) * | 2023-08-16 | 2023-09-15 | 南京尘与土信息技术有限公司 | Data analysis method and analysis device |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021012570A1 (en) | 2021-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110555372A (en) | Data entry method, device, equipment and storage medium | |
| CN108804512B (en) | Text classification model generation device and method and computer readable storage medium | |
| CN108664996B (en) | Ancient character recognition method and system based on deep learning | |
| CN109871446B (en) | Refusing method in intention recognition, electronic device and storage medium | |
| CN106547744B (en) | Image retrieval method and system | |
| US9256783B2 (en) | Systems and methods for tax data capture and use | |
| CN110033018B (en) | Graph similarity judging method and device and computer readable storage medium | |
| CN108491866B (en) | Pornographic picture identification method, electronic device and readable storage medium | |
| CN111460250B (en) | Image data cleaning method, image data cleaning device, image data cleaning medium, and electronic apparatus | |
| CN110110577B (en) | Method and device for identifying dish name, storage medium and electronic device | |
| CN108734159B (en) | Method and system for detecting sensitive information in image | |
| CN110503143B (en) | Threshold selection method, device, storage medium and device based on intention recognition | |
| CN111178147B (en) | Screen crushing and grading method, device, equipment and computer readable storage medium | |
| CN114780746A (en) | Knowledge graph-based document retrieval method and related equipment thereof | |
| CN107992872B (en) | Method for carrying out text recognition on picture and mobile terminal | |
| JP5480008B2 (en) | Summary manga image generation apparatus, program and method for generating manga content summary | |
| CN112818946A (en) | Training of age identification model, age identification method and device and electronic equipment | |
| CN111898378A (en) | Industry classification method and device for government and enterprise clients, electronic equipment and storage medium | |
| CN111078871A (en) | Method and system for automatically classifying contracts based on artificial intelligence | |
| CN114463673B (en) | Material recommendation method, device, equipment and storage medium | |
| CN113836297B (en) | Training method and device for text emotion analysis model | |
| CN112084979B (en) | Food ingredient identification method, device, equipment and storage medium | |
| CN115063826A (en) | Mobile terminal driver license identification method and system based on deep learning | |
| CN114743048A (en) | Method and device for detecting abnormal straw picture | |
| CN114283429A (en) | Material work order data processing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |