CN114564941A - Method, device and equipment for judging option selection state and readable storage medium - Google Patents
Method, device and equipment for judging option selection state and readable storage medium Download PDFInfo
- Publication number
- CN114564941A CN114564941A CN202210215161.7A CN202210215161A CN114564941A CN 114564941 A CN114564941 A CN 114564941A CN 202210215161 A CN202210215161 A CN 202210215161A CN 114564941 A CN114564941 A CN 114564941A
- Authority
- CN
- China
- Prior art keywords
- information
- option
- question
- verification
- judged
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/226—Validation
Abstract
The invention provides a method, a device, equipment and a readable storage medium for judging an option selection state, wherein the method comprises the steps of obtaining picture information, wherein the picture information comprises information of the selection state to be judged; converting the picture information into text information, wherein the text information comprises at least one piece of first question information and option information of the selection state to be judged, which corresponds to the first question information; labeling the option information of the selection state to be judged in the text information by adopting a selected label or an unselected label to obtain label information; and sending the text information to the trained option state inference model to be converted into the label information and speculating to obtain first information, wherein the first information comprises the option information which is marked with the selected label and corresponds to each piece of first problem information.
Description
Technical Field
The invention relates to the field of intelligent document understanding, in particular to a method, a device and equipment for judging an option selection state and a readable storage medium.
Background
Even today with the rapid development of process automation, many scenarios inevitably apply to unstructured data such as paper documents, WORD documents, PDF documents, etc. In some questionnaire and application templates, to transform a question into a closed question, the template designer often presets options and provides check boxes for the user to select. Therefore, when converting unstructured data into structured data that can be understood by a computer program, analysis of check box selection states becomes an important issue.
However, since the habit of each person is different, different symbols are filled in the check boxes, and all the symbols cannot be listed, so that the selection state of the option corresponding to the check box cannot be judged, a technical route such as preset by jumping out of the template is required, and the judgment of the selection state of the option of the check box in the unfixed format is required.
Disclosure of Invention
The invention aims to provide a method, a device and equipment for judging an option selection state and a readable storage medium, so as to solve the problems.
In order to achieve the above object, the embodiments of the present application provide the following technical solutions:
in one aspect, an embodiment of the present application provides a method for determining an option selection state, where the method includes:
acquiring picture information, wherein the picture information comprises information of a selection state to be judged;
converting the picture information into text information, wherein the text information comprises at least one piece of first question information and option information of the selection state to be judged, which corresponds to the first question information;
labeling the option information of the selection state to be judged in the text information by adopting a selected label or an unselected label to obtain label information;
and sending the text information to the trained option state inference model to be converted into the label information and speculating to obtain first information, wherein the first information comprises the option information which is labeled with the selected label and corresponds to each piece of first question information.
Optionally, the converting the picture information into text information includes:
preprocessing the picture information to obtain a picture information set;
inputting a character detection model according to the picture information set to obtain a coordinate information set of the character slice, wherein the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice;
and sending the character slice coordinate information set to an identification model to obtain the text information corresponding to the character slice coordinate information set.
Optionally, after converting the picture information into text information, the method further includes:
detecting the table frame of the first question information, and finding out four vertexes of the table frame;
determining that the first question information belongs to the form box based on the four vertices of the form box;
presetting template information, wherein the template information comprises second question information set in different form frames and option information of the to-be-judged selection state corresponding to the second question information in sequence, positioning keywords in the first question information, matching the first question information with the template information, selecting the second question information with the highest matching degree with the keywords in the first question information from the template information, and obtaining the corresponding relation between the keywords in the first question information and the second question information of the template information.
Optionally, after obtaining the corresponding relationship between the keyword of the first question information and the second question information of the template information, the method further includes:
matching the keyword of the first question information corresponding to the first information with the second question information of the template information to obtain the second question information of the template information with the highest matching degree;
checking the first information and checking information, wherein the checking information comprises option information of the selected state to be judged corresponding to the second question information of the template information with the highest matching degree, and if the word number of the first information is the same as that of the checking information, judging to enter a first checking mode;
and if the word number of the first information is larger or smaller than the word number of the check information, judging to enter a second check mode.
Optionally, the verifying the first information and the verification information, and if the word number of the first information is the same as the word number of the verification information, determining to enter a first verification mode, including:
verifying the first information and the verification information, and outputting the first information as a selected option if the first information is the same as the verification information;
and verifying the first information and the verification information, and if the word number of the first information is different from that of the verification information, outputting the option information with the highest matching degree with the option information to be judged in the verification information as the selected option.
Optionally, the verifying the first information and the verification information, and if the number of words of the first information is greater than or less than the number of words of the verification information, determining to enter a second verification mode, including:
checking the first information and the check information, if the word number of the first information is larger than that of the check information, checking the check information in a sliding manner from left to right on the first information, and outputting option information to be judged in the check information with the highest matching degree with the first information as a selected option;
and verifying the first information and the verification information, if the number of words of the first information is smaller than that of the words in the verification information, verifying the first information in a sliding manner from left to right on the verification information, and outputting option information to be judged in the verification information with the highest matching degree with the verification information as a selected option.
In a second aspect, an embodiment of the present application provides an apparatus for determining an option selection state, where the apparatus includes: the device comprises an acquisition module, a conversion module, a marking module and a first determination module.
The acquisition module is used for acquiring picture information, and the picture information comprises information of a selection state to be judged;
the conversion module is used for converting the picture information into text information, and the text information comprises at least one piece of first question information and option information of the selection state to be judged, which corresponds to the first question information;
the marking module is used for marking the option information of the selection state to be judged in the text information by adopting a selected label or an unselected label to obtain label information;
the first determining module is configured to send the text information to the trained option state inference model to be converted into the label information, and infer to obtain first information, where the first information includes the option information labeled with the selected label and corresponding to each piece of the first question information.
Optionally, the conversion module includes:
the processing unit is used for preprocessing the picture information to obtain a picture information set;
the detection unit is used for inputting a character detection model according to the picture information set to obtain a coordinate information set of a character slice, and the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice;
and the recognition unit is used for inputting a character detection model according to the picture information set to obtain a coordinate information set of the character slice, and the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice.
Optionally, the apparatus further includes:
the detection module is used for detecting the table frame of the first question information and finding out four vertexes of the table frame;
a second determination module to determine that the first problem information belongs to the form box based on the four vertices of the form box;
the preset module is used for presetting template information, wherein the template information comprises second question information which is set in different form frames in sequence and option information of a to-be-judged selection state corresponding to the second question information, positioning keywords in the first question information, matching the first question information with the template information, selecting the second question information which has the highest matching degree with the keywords in the first question information from the template information, and obtaining the corresponding relation between the keywords in the first question information and the second question information of the template information.
Optionally, the apparatus further includes:
the matching module is used for matching the keyword of the first question information corresponding to the first information with the second question information of the template information to obtain the second question information of the template information with the highest matching degree;
the judging module is used for verifying the first information and verification information, the verification information comprises option information of the selected state to be judged, which corresponds to the second question information of the template information with the highest matching degree, and if the word number of the first information is the same as that of the verification information, the first information is judged to enter a first verifying unit;
and if the word number of the first information is larger or smaller than the word number of the check information, judging to enter a second check unit.
Optionally, the first verification unit includes:
the first determining unit is used for verifying the first information and the verification information, and outputting the first information as a selected option if the first information is the same as the verification information;
and the second determining unit is used for verifying the first information and the verification information, and outputting the option information with the highest matching degree with the option information to be judged in the verification information as the selected option if the word number of the first information is different from the word number of the verification information.
Optionally, the second verifying unit includes:
a third determining unit, configured to verify the first information and the verification information, and if the number of words of the first information is greater than the number of words in the verification information, perform sliding verification on the verification information from left to right on the first information, and output option information to be determined in the verification information with the highest matching degree with the first information as a selected option;
and the fourth determining unit is used for verifying the first information and the verification information, if the word number of the first information is smaller than that of the verification information, the first information is verified in a sliding mode from left to right on the verification information, and option information to be judged in the verification information with the highest matching degree with the verification information is output as a selected option.
In a third aspect, an embodiment of the present application provides an option selection state determination device, where the device includes a memory and a processor. The memory is used for storing a computer program; the processor is used for realizing the steps of the judging method of the option selection state when executing the computer program.
In a fourth aspect, an embodiment of the present application provides a readable storage medium, where a computer program is stored on the readable storage medium, and when the computer program is executed by a processor, the steps of the method for determining an option selection state are implemented.
The invention has the beneficial effects that:
the method aims at judging the option state of the picture or the WORD document with low picture definition and without a fixed preset check box template, can be suitable for the complex scene with low picture definition and without a fixed preset check box template by introducing an entity extraction technology, and provides a low-cost method compared with the method for improving the accuracy of judging the option state by an OCR engine.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a schematic flow chart of a method for determining an option selection status according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of an option selection state determination device according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an option selection state determining device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. 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 invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Example 1
As shown in fig. 1, the present embodiment provides a method for determining an option selection state, which includes step S1, step S2, step S6, and step S7.
Step S1, obtaining picture information, wherein the picture information comprises information of a selection state to be judged;
step S2, converting the picture information into text information, wherein the text information comprises at least one piece of first question information and option information of the selection state to be judged corresponding to the first question information;
step S6, labeling the option information of the selection state to be judged in the text information by adopting a selected label or an unselected label to obtain label information;
step S7, sending the text information to the trained option state inference model, converting the text information into the label information, and inferring to obtain first information, where the first information includes the option information labeled with the selected label corresponding to each first question information.
At present, the judgment of the option state in a document or a picture is mostly based on whether a check box is checked, but due to different personal habits, a preset check box may not contain a complete set of check boxes, some check boxes cannot be identified, so that the selection state of the option cannot be identified, meanwhile, if the object to be judged is the picture, some rare symbols may be generated through an OCR identification technology, or a blurred place of the picture is easily identified by mistake, and it is necessary to consume a large amount of labeling and research and development costs to optimize the small problems from an OCR level, however, in some scenes, the final result output is not influenced by the small OCR identification errors.
Therefore, the embodiment adopts the entity extraction technology, when the object to be judged is an image, the object to be judged is converted into text information through a series of processing, the option to be judged in the text information is labeled by using a label, when the object to be judged is a WORD text, the option to be judged in the text information of the WORD text is directly labeled by using the label, and the selection state of the option is judged by presuming the label information, so that errors caused by judgment through check boxes can be effectively reduced, the work efficiency is improved, the accuracy is improved, and when the object to be judged is an image, a low-cost method compared with an optimal OCR engine is provided.
According to the characteristics, the embodiment can realize the quick and efficient judgment of the selection state of the options in the picture or the WORD. An economical and efficient judgment method is provided for counting option selection state information, and the method can be widely used for processing feedback information of relevant business of customers and counting relevant proportions by banks.
In a specific embodiment of the present disclosure, the step S2 may further include a step S21, a step S22, and a step S23.
Step S21, preprocessing the picture information to obtain a picture information set;
step S22, inputting a character detection model according to the picture information set to obtain a coordinate information set of the character slice, wherein the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice;
and step S23, sending the character slice coordinate information set to an identification model to obtain the text information corresponding to the character slice coordinate information set.
In this embodiment, it is first determined that the type of the input data is a picture or a word document, if the input data is a picture, the uploaded picture is normalized, and the image is preprocessed by using technologies such as graphics and deep learning, including inclination angle correction, color chapter removal, and background filtering, so that the accuracy of OCR recognition on the picture can be improved, and if the input data is a word document, text information of the document is directly processed.
In a specific embodiment of the present disclosure, after the step S2, the method may further include a step S3, a step S4, and a step S5.
Step S3, detecting the form frame of the first question information, and finding out four vertexes of the form frame;
step S4, determining that the first question information belongs to the form frame based on the four vertexes of the form frame;
step S5, presetting template information, where the template information includes second question information set in different form frames in order and the option information of the to-be-determined selection state corresponding to the second question information, locating a keyword in the first question information, where the first question information matches the template information, and selecting the second question information having the highest matching degree with the keyword in the first question information from the template information, to obtain a correspondence between the keyword in the first question information and the second question information in the template information.
In the present embodiment, for example: every problem of opening the certificate application has the same problem to different customers, also has inequality problem, based on template information is predetermine according to certain order to the problem of the certificate application and the corresponding complete set of the option of waiting to judge, fixes a position through the keyword in the certificate application of waiting to judge to all form frames in the ergodic template information, will characters in the form frame with the keyword in the certificate application of waiting to judge matches, and the selected degree of matching with it is highest the keyword obtains the corresponding relation of keyword and content.
In a specific embodiment of the present disclosure, after the step S7, the method may further include a step S8 and a step S9.
Step S8, matching the keyword of the first question information corresponding to the first information with the second question information of the template information to obtain the second question information of the template information with the highest matching degree;
step S9, verifying the first information and verification information, where the verification information includes option information of the to-be-determined selection state corresponding to the second question information of the template information with the highest matching degree, and if the word count of the first information is the same as the word count of the verification information, determining to enter a first verification mode; and if the word number of the first information is larger or smaller than the word number of the check information, judging to enter a second check mode.
In this embodiment, the state of the option selected in the to-be-judged issuing application is inferred, the keyword of the first problem information in the to-be-judged issuing application is matched with the keyword of the second problem information in the template information, the position with the highest matching degree of each first problem information in the to-be-judged issuing application and the position with the highest matching degree of the second problem information in the template information are overlapped, and the to-be-judged option is checked by being placed in the same form frame, so that the accuracy of the to-be-judged option information is improved.
In a specific embodiment of the present disclosure, the step S91 further includes a step S911 and a step S912.
Step S911, the first information and the verification information are verified, and if the first information is the same as the verification information, the first information is output as a selected option;
step S912, verifying the first information and the verification information, and if the word number of the first information is different from the word number of the verification information, outputting the option information with the highest matching degree with the option information to be determined in the verification information as the selected option.
In a specific embodiment of the present disclosure, the step S92 may further include a step S921 and a step S922.
Step 921, checking the first information and the check information, if the number of words of the first information is greater than the number of words in the check information, checking the check information by sliding from left to right on the first information, and outputting option information to be judged in the check information with the highest matching degree with the first information as a selected option;
step S922, the first information and the check information are checked, if the word number of the first information is smaller than that of the check information, the first information is checked in a sliding mode from left to right on the check information, and option information to be judged in the check information with the highest matching degree with the check information is output to serve as a selected option.
In this embodiment, after the accuracy of the selection state information of the option to be determined is improved by verifying the option information to be determined and the verification information, a specific rule may be added according to a specific scenario to verify whether the inferred selection state of the option is correct, for example: in a known scene, the length of the option to be processed is certainly more than three words, if the length of a phrase is less than 3, but the label inferred by the prediction model is the selected label, the probability that the model conjectures wrongly is high, but the phrase is erased because the length of the option to be processed by adding a rule in advance is certainly more than three words; manual judgment rules can also be added: when the result deduced by the model conflicts with other known information, manual judgment can be carried out at the moment, different rules are added based on different scenes, and the accuracy of finally obtaining the option selection state can be effectively improved.
Example 2
As shown in fig. 2, the present embodiment provides a device for determining an option selection state, where the device includes an obtaining module 901, a converting module 902, a marking module 906, and a first determining module 907.
The obtaining module 901 is configured to obtain picture information, where the picture information includes information of a selection state to be determined;
the conversion module 902 is configured to convert the picture information into text information, where the text information includes at least one piece of first question information and option information of the selection state to be determined corresponding to the first question information;
the marking module 906 is configured to mark the option information of the selection state to be determined in the text information with a selected tag or an unselected tag to obtain tag information;
the first determining module 907 is configured to send the text information to the trained option state inference model to be converted into the tag information, and infer to obtain first information, where the first information includes the option information labeled with the selected tag and corresponding to each piece of the first question information.
The device of the embodiment can realize efficient and accurate option selection state statistics of the picture or the text, and simultaneously provides a low-cost method for optimizing an OCR engine when processing picture information.
In a specific embodiment of the present disclosure, the conversion module 902 includes a processing unit 9021, a detection unit 9022, and an identification unit 9023.
The processing unit 9021 is configured to perform preprocessing on the picture information to obtain a picture information set;
the detection unit 9022 is configured to input a character detection model according to the picture information set to obtain a coordinate information set of a character slice, where the character detection model is configured to search character coordinate information in the picture information and output coordinate information of the character slice;
the recognition unit 9023 is configured to input a text detection model according to the picture information set to obtain a coordinate information set of a text slice, where the text detection model is configured to search text coordinate information in the picture information and output coordinate information of the text slice.
In a specific embodiment of the present disclosure, the apparatus further includes a detection module 903, a second determination module 904, and a preset module 905.
The detecting module 903 is configured to detect the table frame of the first question information, and find four vertices of the table frame;
the second determining module 904, configured to determine that the first question information belongs to the form box based on the four vertices of the form box;
the preset module 905 is used for presetting template information, wherein the template information comprises second question information and option information, the second question information is set to be different in sequence, the option information is corresponding to the second question information and is used for judging the selection state, the keywords in the first question information are located, the first question information is matched with the template information, the second question information is selected from the template information, the matching degree of the keywords in the first question information is the highest, and the corresponding relation between the keywords in the first question information and the second question information is obtained.
In a specific embodiment of the present disclosure, the apparatus further includes a matching module 908 and a determining module 909.
The matching module 908 is configured to match the keyword of the first question information corresponding to the first information with the second question information of the template information, so as to obtain the second question information of the template information with the highest matching degree;
the determining module 909 is configured to perform verification on the first information and verification information, where the verification information includes option information of the to-be-determined selection state corresponding to the second question information of the template information with the highest matching degree, and if the number of words of the first information is the same as the number of words of the verification information, determine to enter a first verifying unit 9091;
and if the word number of the first information is larger than or smaller than the word number of the check information, judging to enter a second check unit 9092.
In a specific embodiment of the present disclosure, the first checking unit 9091 further includes a first determining unit 90911 and a second determining unit 90912.
The first determining unit 90911 verifies the first information with the verification information, and outputs the first information as a selected option if the first information is the same as the verification information;
the second determining unit 90912 verifies the first information and the verification information, and outputs the option information with the highest matching degree with the option information to be determined in the verification information as the selected option if the word number of the first information is different from the word number of the verification information.
In a specific embodiment of the present disclosure, the second checking unit 9092 further includes a third determining unit 90921 and a fourth determining unit 90922.
A third determining unit 90921, configured to verify the first information with the verification information, and if the number of words of the first information is greater than the number of words in the verification information, perform sliding verification on the verification information from left to right on the first information, and output option information to be determined in the verification information with the highest matching degree with the first information as a selected option;
a fourth determining unit 90922, configured to verify the first information with the verification information, and if the number of words of the first information is smaller than the number of words in the verification information, perform sliding verification on the verification information from left to right on the first information, and output option information to be determined in the verification information with the highest matching degree with the verification information as a selected option.
It should be noted that, regarding the apparatus in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.
Example 3
Corresponding to the above method embodiments, the embodiments of the present disclosure further provide an option selection state determination device, and the below-described option selection state determination device and the above-described option selection state determination method may be referred to in correspondence with each other.
Fig. 3 is a block diagram illustrating an option selection state determination device 800 according to an example embodiment. As shown in fig. 3, the option selection state determination device 800 may include: a processor 801, a memory 802. The option selection state determination device 800 may further include one or more of a multimedia component 803, an input/output (I/O) interface 804, and a communication component 805.
The processor 801 is configured to control the overall operation of the option selection state determination device 800, so as to complete all or part of the steps in the above-mentioned method for determining an option selection state. The memory 402 is used to store various types of data to support the operation of the option selection state determination device 800, such data may include, for example, instructions for any application or method operating on the option selection state determination device 800, as well as application-related data, such as contact data, messages sent or received, pictures, audio, video, and the like. The Memory 802 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk. The multimedia components 803 may include screen and audio components. Wherein the screen may be, for example, a touch screen and the audio component is used for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may further be stored in the memory 802 or transmitted through the communication component 805. The audio assembly also includes at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the determination device 800 of the option selection status and other devices. Wireless communication, such as Wi-Fi, bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them, so that the corresponding communication component 805 may include: Wi-Fi module, bluetooth module, NFC module.
In an exemplary embodiment, the option selection state determining Device 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components, and is used to perform the above option selection state determining method.
In another exemplary embodiment, a computer readable storage medium including program instructions, which when executed by a processor, implement the steps of the above method for determining an option selection state is also provided. For example, the computer readable storage medium may be the memory 802 including the program instructions, which are executable by the processor 801 of the option selection state determination device 800 to perform the above-described option selection state determination method.
Corresponding to the above method embodiment, the embodiment of the present disclosure further provides a readable storage medium, and a readable storage medium described below and the above method for determining the option selection status may be referred to correspondingly.
Example 4
A readable storage medium, on which a computer program is stored, the computer program, when being executed by a processor, implementing the steps of the method for determining an option selection state of the above method embodiment.
The readable storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various other readable storage media capable of storing program codes.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of the changes or substitutions within the technical scope of the present invention, and shall cover the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (14)
1. The method for judging the option selection state is characterized by comprising the following steps:
acquiring picture information, wherein the picture information comprises information of a selection state to be judged;
converting the picture information into text information, wherein the text information comprises at least one piece of first question information and option information of the selection state to be judged, which corresponds to the first question information;
labeling the option information of the selection state to be judged in the text information by adopting a selected label or an unselected label to obtain label information;
and sending the text information to the trained option state inference model to be converted into the label information and speculating to obtain first information, wherein the first information comprises the option information which is labeled with the selected label and corresponds to each piece of first question information.
2. The method for determining an option selection status according to claim 1, wherein the converting the picture information into text information includes:
preprocessing the picture information to obtain a picture information set;
inputting a character detection model according to the picture information set to obtain a coordinate information set of the character slice, wherein the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice;
and sending the character slice coordinate information set to an identification model to obtain the text information corresponding to the character slice coordinate information set.
3. The method for determining an option selection status according to claim 1, wherein after converting the picture information into text information, the method further comprises:
detecting the table frame of the first question information, and finding out four vertexes of the table frame;
determining that the first question information belongs to the form box based on the four vertices of the form box;
presetting template information, wherein the template information comprises second question information and option information of the to-be-judged selection state corresponding to the second question information which are set in different form frames in sequence, positioning a keyword in the first question information, matching the first question information with the template information, selecting the second question information which has the highest matching degree with the keyword in the first question information from the template information, and obtaining the corresponding relation between the keyword in the first question information and the second question information in the template information.
4. The method for determining an option selection status according to claim 3, wherein after obtaining the correspondence between the keyword of the first question information and the second question information of the template information, the method further comprises:
matching the keyword of the first question information corresponding to the first information with the second question information of the template information to obtain the second question information of the template information with the highest matching degree;
checking the first information and checking information, wherein the checking information comprises option information of the selected state to be judged corresponding to the second question information of the template information with the highest matching degree, and if the word number of the first information is the same as that of the checking information, judging to enter a first checking mode;
and if the word number of the first information is larger or smaller than the word number of the check information, judging to enter a second check mode.
5. The method for determining an option selection status according to claim 4, wherein the verifying the first information with the verification information, and if the word number of the first information is the same as the word number of the verification information, determining to enter the first verification mode includes:
verifying the first information and the verification information, and outputting the first information as a selected option if the first information is the same as the verification information;
and verifying the first information and the verification information, and if the word number of the first information is different from that of the verification information, outputting the option information with the highest matching degree with the option information to be judged in the verification information as the selected option.
6. The method as claimed in claim 4, wherein the verifying the first information with the verification information, and if the number of words of the first information is greater than or less than the number of words of the verification information, determining to enter the second verification mode comprises:
checking the first information and the check information, if the word number of the first information is larger than that of the check information, checking the check information in a sliding manner from left to right on the first information, and outputting option information to be judged in the check information with the highest matching degree with the first information as a selected option;
and checking the first information and the check information, if the word number of the first information is smaller than that of the check information, checking the first information in a sliding manner from left to right on the check information, and outputting the option information to be judged in the check information with the highest matching degree with the check information as the selected option.
7. The device for judging the option selection state is characterized by comprising:
the acquisition module is used for acquiring picture information, and the picture information comprises information of a selection state to be judged;
the conversion module is used for converting the picture information into text information, and the text information comprises at least one piece of first question information and option information of the selection state to be judged, which corresponds to the first question information;
the marking module is used for marking the option information of the selection state to be judged in the text information by adopting a selected label or an unselected label to obtain label information;
and the first determining module is used for sending the text information to the trained option state inference model to be converted into the label information and deducing to obtain first information, wherein the first information comprises the option information which is labeled with the selected label and corresponds to each piece of first question information.
8. The apparatus for determining an option selection status according to claim 7, wherein the conversion module comprises:
the processing unit is used for preprocessing the picture information to obtain a picture information set;
the detection unit is used for inputting a character detection model according to the picture information set to obtain a coordinate information set of a character slice, and the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice;
and the recognition unit is used for inputting a character detection model according to the picture information set to obtain a coordinate information set of the character slice, and the character detection model is used for searching character coordinate information in the picture information and outputting the coordinate information of the character slice.
9. The apparatus for determining an option selection status according to claim 7, further comprising:
the detection module is used for detecting the table frame of the first question information and finding out four vertexes of the table frame;
a second determination module to determine that the first problem information belongs to the form box based on the four vertices of the form box;
the preset module is used for presetting template information, wherein the template information comprises second question information which is set in different form frames in sequence and option information of a to-be-judged selection state corresponding to the second question information, positioning keywords in the first question information, matching the first question information with the template information, selecting the second question information which has the highest matching degree with the keywords in the first question information from the template information, and obtaining the corresponding relation between the keywords in the first question information and the second question information of the template information.
10. The apparatus for determining an option selection status according to claim 7, further comprising:
the matching module is used for matching the keyword of the first question information corresponding to the first information with the second question information of the template information to obtain the second question information of the template information with the highest matching degree;
the judging module is used for verifying the first information and verification information, the verification information comprises option information of the selected state to be judged, which corresponds to the second question information of the template information with the highest matching degree, and if the word number of the first information is the same as that of the verification information, the first information is judged to enter a first verifying unit;
and if the word number of the first information is larger or smaller than the word number of the check information, judging to enter a second check unit.
11. The apparatus for determining an option selection status according to claim 10, wherein the first checking unit includes:
the first determining unit is used for verifying the first information and the verification information, and outputting the first information as a selected option if the first information is the same as the verification information;
and the second determining unit is used for verifying the first information and the verification information, and outputting the option information with the highest matching degree with the option information to be judged in the verification information as the selected option if the word number of the first information is different from the word number of the verification information.
12. The apparatus for determining an option selection status according to claim 10, wherein the second checking unit includes:
a third determining unit, configured to verify the first information with the verification information, perform sliding verification on the verification information from left to right on the first information if the number of words of the first information is greater than the number of words in the verification information, and output option information to be determined in the verification information with the highest matching degree with the first information as a selected option;
and the fourth determining unit is used for verifying the first information and the verification information, if the word number of the first information is smaller than that of the verification information, the first information is verified in a sliding mode from left to right on the verification information, and option information to be judged in the verification information with the highest matching degree with the verification information is output as a selected option.
13. An option selection state judgment device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the method for determining the option selection status according to any one of claims 1 to 6 when executing the computer program.
14. A readable storage medium, characterized by: the readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the method for determining an option selection status according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210215161.7A CN114564941A (en) | 2022-03-04 | 2022-03-04 | Method, device and equipment for judging option selection state and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210215161.7A CN114564941A (en) | 2022-03-04 | 2022-03-04 | Method, device and equipment for judging option selection state and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114564941A true CN114564941A (en) | 2022-05-31 |
Family
ID=81717452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210215161.7A Pending CN114564941A (en) | 2022-03-04 | 2022-03-04 | Method, device and equipment for judging option selection state and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114564941A (en) |
-
2022
- 2022-03-04 CN CN202210215161.7A patent/CN114564941A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4829920B2 (en) | Form automatic embedding method and apparatus, graphical user interface apparatus | |
| CN113469067B (en) | Document analysis method, device, computer equipment and storage medium | |
| US10699112B1 (en) | Identification of key segments in document images | |
| CN108664471B (en) | Character recognition error correction method, device, equipment and computer readable storage medium | |
| CN110580308A (en) | information auditing method and device, electronic equipment and storage medium | |
| CN104750791A (en) | Image retrieval method and device | |
| US20230290169A1 (en) | Information Extraction Method and Apparatus for Text With Layout | |
| CN115935344A (en) | Abnormal equipment identification method and device and electronic equipment | |
| CN113450147B (en) | Product matching method, device, equipment and storage medium based on decision tree | |
| US11670067B2 (en) | Information processing apparatus and non-transitory computer readable medium | |
| CN111159354A (en) | Sensitive information detection method, device, equipment and system | |
| KR102282025B1 (en) | Method for automatically sorting documents and extracting characters by using computer | |
| CN111177387A (en) | User list information processing method, electronic device and computer readable storage medium | |
| CN114564941A (en) | Method, device and equipment for judging option selection state and readable storage medium | |
| CN115294593A (en) | Image information extraction method and device, computer equipment and storage medium | |
| CN115578736A (en) | Certificate information extraction method, device, storage medium and equipment | |
| CN116185812A (en) | Automatic testing method, device and medium for software system functions | |
| CN114359928A (en) | Electronic invoice identification method and device, computer equipment and storage medium | |
| CN114443834A (en) | Method and device for extracting license information and storage medium | |
| CN114065762A (en) | Text information processing method, device, medium and equipment | |
| CN110633457B (en) | Content replacement method and device, electronic equipment and readable storage medium | |
| CN109739981B (en) | PDF file type judgment method and character extraction method | |
| US10395081B2 (en) | Encoding document capture bounds with barcodes | |
| WO2023062799A1 (en) | Information processing system, manuscript type identification method, model generation method and program | |
| US20210342901A1 (en) | Systems and methods for machine-assisted document input |
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 |