WO2020258500A1 - 光学字符识别辅助方法、装置、计算机设备及存储介质 - Google Patents

光学字符识别辅助方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2020258500A1
WO2020258500A1 PCT/CN2019/103231 CN2019103231W WO2020258500A1 WO 2020258500 A1 WO2020258500 A1 WO 2020258500A1 CN 2019103231 W CN2019103231 W CN 2019103231W WO 2020258500 A1 WO2020258500 A1 WO 2020258500A1
Authority
WO
WIPO (PCT)
Prior art keywords
result
target
recognition
credibility
recognition result
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.)
Ceased
Application number
PCT/CN2019/103231
Other languages
English (en)
French (fr)
Inventor
郭俊雄
王健宗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2020258500A1 publication Critical patent/WO2020258500A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/217Validation; Performance evaluation; Active pattern learning techniques
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/40Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10Character recognition
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Definitions

  • This application relates to the field of artificial intelligence technology, and in particular to an optical character recognition auxiliary method, device, computer equipment and storage medium.
  • OCR Optical Character Recognition, Optical Character Recognition
  • the embodiments of the present application provide an optical character recognition auxiliary method, device, computer equipment, and storage medium to solve the problem of inaccurate OCR recognition that affects user work efficiency.
  • An auxiliary method for optical character recognition including:
  • the target result is sent to the operating user as an optical character recognition result.
  • An optical character recognition auxiliary device including:
  • the first obtaining module is configured to obtain a target picture to be recognized sent by an operating user, wherein the target picture includes character information;
  • a recognition module configured to use an optical character recognition tool to perform character recognition on the character information in the target picture to obtain the recognition result and the credibility corresponding to the recognition result;
  • the first comparison module is used to compare the reliability with a preset threshold
  • the first comparison result module is configured to determine the recognition result corresponding to the reliability as a target result if the reliability is greater than or equal to a preset threshold;
  • a second comparison result module configured to preprocess the recognition result corresponding to the credibility if the credibility is less than a preset threshold to determine the target result
  • the sending module is configured to send the target result as an optical character recognition result to the operating user.
  • a computer device including a memory, a processor, and computer-readable instructions stored in the memory and capable of running on the processor, and the processor implements the above-mentioned optical character recognition aid when the computer-readable instructions are executed Method steps.
  • a non-volatile computer-readable storage medium stores computer-readable instructions, and the computer-readable instructions implement any of the foregoing when executed by a processor The steps of the optical character recognition auxiliary method.
  • FIG. 1 is a flowchart of an optical character recognition auxiliary method provided by an embodiment of the present application
  • step S5 is a flowchart of step S5 in the optical character recognition auxiliary method provided by the embodiment of the present application.
  • step S51 in the optical character recognition auxiliary method provided by the embodiment of the present application.
  • step S513 is a flowchart of step S513 in the optical character recognition auxiliary method provided by the embodiment of the present application.
  • FIG. 5 is a flowchart of determining a target result in the case where the first target result, the second target result, and the recognition result are all different in the optical character recognition auxiliary method provided by the embodiment of the present application;
  • FIG. 6 is a flowchart of updating the OCR recognition library in the auxiliary method for optical character recognition provided by an embodiment of the present application
  • FIG. 7 is a schematic diagram of an optical character recognition auxiliary device provided by an embodiment of the present application.
  • Fig. 8 is a basic structural block diagram of a computer device provided by an embodiment of the application.
  • optical character recognition auxiliary method provided in this application is applied to the server, and the server can be implemented by an independent server or a server cluster composed of multiple servers.
  • an auxiliary method for optical character recognition is provided, which includes the following steps:
  • S1 Acquire a target picture to be recognized sent by the operating user, where the target picture includes character information.
  • the target picture to be recognized refers to a picture that includes character information and needs to be recognized by an OCR tool.
  • the target picture when the operating user sends the target picture to be recognized in the client, the target picture will be automatically saved in the preset picture library.
  • the preset picture library refers to a database dedicated to storing target pictures to be identified.
  • the target picture is obtained from the preset picture library, the target picture is deleted from the preset picture library.
  • S2 Use an optical character recognition tool to perform character recognition on the character information in the target image, and obtain the recognition result and the reliability corresponding to the recognition result.
  • an optical character recognition tool refers to a tool that specifically uses a character recognition method to translate the shape of characters in a picture into computer text.
  • Credibility refers to the degree of trust set by the OCR tool for the recognized recognition results. The higher the credibility, the higher the accuracy of OCR recognition.
  • the recognition result output by the OCR tool after recognition and the credibility corresponding to the recognition result are obtained.
  • the recognition result obtained is "ABCD", and the reliability corresponding to the recognition result is 80%.
  • the credibility is compared with a preset threshold.
  • the credibility is compared with a preset threshold, and if the comparison result is that the credibility is greater than or equal to the preset threshold, the recognition result corresponding to the credibility is determined as the target result.
  • the preset threshold may be 80%, or it may be set according to the actual needs of the user, and there is no limitation here.
  • recognition result A its corresponding credibility is 80%
  • the preset threshold is 80%
  • the credibility is compared with the preset threshold. Since the credibility and the preset threshold are both 80%, the The recognition result A is determined as the target result.
  • the credibility is compared with the preset threshold, and if the comparison result is that the credibility is less than the preset threshold, it means that the accuracy of the recognition result corresponding to the credibility does not reach the user setting.
  • the recognition result is preprocessed.
  • the specific process of preprocessing can include: modify the recognition result to obtain the reference result, and then send the reference result and the target image to be recognized to the first target user and the second target user for Check, obtain the first target result and the second target result fed back after the first target user and the second target user are checked, and perform matching processing on the first target result, the second target result and the recognition result to obtain the same matching
  • the result is the target result.
  • S6 Send the target result as the optical character recognition result to the operating user.
  • the target result is used as the OCR recognition result and sent to the operating user in a preset manner, where the preset manner may specifically be in the form of feedback on the client, for example, directly feedback the OCR recognition result to the client.
  • the preset manner may specifically be in the form of feedback on the client, for example, directly feedback the OCR recognition result to the client.
  • the display interface it can also be set according to the actual needs of the user.
  • the recognition result and the credibility corresponding to the recognition result are obtained, and the credibility is compared with the preset threshold, and the credibility greater than or equal to the preset threshold is obtained.
  • the recognition result corresponding to the degree of recognition is determined as the target result, and the recognition result corresponding to the credibility less than the preset threshold is preprocessed to obtain the preprocessed target result.
  • the target result is sent to the operating user as the OCR recognition result.
  • step S5 that is, if the credibility is less than the preset threshold, preprocessing the recognition result corresponding to the credibility to determine the target result includes the following steps:
  • the recognition result corresponding to the credibility is modified according to a preset modification plan, where the preset modification plan refers to a plan preset by the user .
  • the specific process may include: first generating a random number according to the recognition result, and then selecting a modification scheme corresponding to the random number to modify the recognition result to obtain the modified reference result.
  • random number refers to a positive integer greater than 0.
  • S52 Send the reference result and the target picture to the first target user and the second target user for verification.
  • the first target user and the second target user refer to legitimate users who specifically verify the character information in the target picture and the reference result.
  • the reference result and target picture obtained by step S51 are respectively sent to the first target user and the second target user for verification processing in a preset manner.
  • the recognition results are directly sent to the first target user and the second target user for verification, the first target user and the second target user may subjectively believe that the accuracy of the OCR tool recognition is high, leading to failure
  • the recognition results will be carefully checked, and the recognition results will be directly used for feedback. Therefore, in order to avoid this situation, it is necessary to process the OCR recognition result into a reference result, and then send the reference result to the first target user and the second target user for verification.
  • S53 Acquire the first target result fed back by the first target user, and the second target result fed back by the second target user.
  • the reference result is analyzed according to the character information in the target picture, and it is determined whether the reference result is the same as the character information in the target picture. If they are the same, the reference can be directly referenced. The result is regarded as the first target result. If it is not the same, the recognition result can be re-formulated according to the analysis situation, and the recognition result is regarded as the first target result.
  • the second target user receives the reference result and target picture according to the The second target result is obtained when the first target user obtains the first target result.
  • the first target user and the second target user feed back the first target result and the second target result to the server through the client and respectively.
  • the server receives the feedback information from the first target user and the second target user, it will directly Acquire the first target result fed back by the first target user and the second target result fed back by the second target user.
  • the first target result, the second target result, and the recognition result are matched respectively, that is, the first target result is matched with the second target result, the first target result is matched with the recognition result, and the second target result is matched with the recognition result. Make a match.
  • the first target result is the same as the second target result, or the first target result is the same as the recognition result, or the first target result is the same as the recognition result.
  • the second target result is the same as the recognition result, or the first target result is the same as the second target result and the recognition result, it means that at least two of the first target result, the second target result, and the recognition result are the same, and Determine the same result as the target result.
  • the first target result is A
  • the second target result is B
  • the recognition result is A
  • the first target result is obtained by matching the first target result A, the second target result B and the recognition result A respectively A is the same as the recognition result A, which means that at least two of the first target result, the second target result, and the recognition result are the same, and the same result A is determined as the target result.
  • the recognition result when the credibility of the recognition result is less than the preset threshold, the recognition result is modified to obtain the reference result, and the reference result is sent to the first target user and the second target user for verification to obtain The first target result and the second target result fed back by the first target user and the second target user, and finally the target result is obtained by matching the first target result, the second target result and the recognition result, so as to achieve the corresponding OCR recognition result
  • the recognition result is further verified by manual to avoid the inaccurate OCR recognition result, and further ensure that the operator can obtain accurate recognition results and improve the work efficiency of the operator.
  • step S51 that is, if the credibility is less than the preset threshold, the recognition result corresponding to the credibility is modified according to the preset modification plan, and the reference result includes the following steps :
  • the recognition result is traversed, and the number of characters n containing characters in the recognition result is obtained through traversal.
  • the recognition result is "Tianxia Taiping”
  • the recognition result includes 4 characters: " ⁇ ", “ ⁇ ”, “ ⁇ ” and " ⁇ ”.
  • S512 Generate a random numbers according to the number of characters n, where a and the random number are both positive integers, and a ⁇ [1,2], the value range of the random number is [1, n].
  • the random number refers to the character position corresponding to the character in the recognition result, and the value range of the random number is a positive integer greater than 0. For example, if the recognition result is "QWE", the random number corresponding to the "Q" character is 1, the random number corresponding to the "W” character is 2, and the random number corresponding to the "E” character is 3.
  • a random number is generated according to the number n of characters obtained in step S511, and both a and the random number are positive integers, the value range of a is 1 or 2, and the value range of the random number is within the number of characters n For example, if the number of characters is 4, the value range of the random number is between 1 and 4.
  • the value of the random number can be 1 or 2 or 3 or 4; if a is 2, it is generated If there are two random numbers, the values of the two random numbers can be 1 and 2, or 1 and 3, or 1 and 4, etc.
  • S513 Select a modification plan corresponding to the number a of the random number to modify the recognition result to obtain a reference result, where the modification plan includes but is not limited to one or more of the following processes: exchanging X and Y, X replaces Y, removes X, and inserts X.
  • X and Y are characters corresponding to different random numbers.
  • exchanging X and Y refers to the exchange of different characters corresponding to different random numbers, and this modification scheme is used as the exchange scheme; replacing X with Y means replacing the character corresponding to the random number with another random number The character corresponding to the number, the modification scheme is used as the replacement scheme; the elimination of X means the deletion of the character corresponding to the random number, and the modification scheme as the elimination scheme; the insertion of X means between the character corresponding to the random number and the next character Insert the character corresponding to the random number at the position of, and use the modified scheme as the insertion scheme.
  • the number a of the random number generated in step S512 is matched with the identification number in the preset scheme library, and when the number a of the matched random number is the same as the identification number in the preset scheme library , Select the modification plan corresponding to the identification number to modify the recognition result, and use the modified result as the reference result.
  • the preset scheme library is specifically used to store modification schemes and the identification number corresponding to each modification scheme, for example, the identification number corresponding to the exchange scheme is 2.
  • the corresponding identification numbers are 2, 2, 1, and 1, respectively. If the number a of random numbers is 1, Then the number 1 is matched with the identification number in the preset scheme library, and the identification number 1 is the same as the number 1, so the elimination scheme is selected at random to modify the recognition result "ASDF", if the modified result is "ADF” ", the "ADF” is determined as the reference result.
  • a random number is generated according to the number n of characters corresponding to the recognition result, and a corresponding modification scheme is selected according to the number a of the random number to modify the recognition result.
  • step S513 that is, selecting a modification scheme corresponding to the number a of random numbers to modify the recognition result, and obtaining the reference result includes the following steps:
  • the number a of random numbers is directly compared with a preset value.
  • the preset value can be 1 or 2, and the specific value range can be set according to the actual needs of the user.
  • the value of the preset value in this proposal is set to 1.
  • step S5131 the number a of random numbers is compared with a preset value, and if the comparison result is that the number a of random numbers is greater than the preset value, an exchange scheme or an alternative scheme is randomly selected, and if an exchange scheme is selected, then according to Each random number determines the position corresponding to the character in the recognition result, and randomly selects the characters between the positions corresponding to the two random numbers to exchange, and obtains the reference result after the character exchange; if the replacement scheme is selected, according to each random number Determine the position corresponding to the character in the recognition result, randomly determine the replacement position and the replaced position, and replace the character corresponding to the replaced position with the character corresponding to the replaced position to obtain the reference result after the character replacement.
  • the recognition result is ABCD
  • the exchange is selected According to the random number 1 and 2, the position between the character A and the character B is exchanged, and the reference result obtained is BACD; if the replacement scheme is selected, the randomly determined replacement position is 1 and the replaced position is 2 , After replacing the character B corresponding to the replacement position 2 with the character A corresponding to the replacement position 1, the reference result obtained is AACD.
  • step S5131 the number a of random numbers is compared with a preset value, and if the comparison result is that the number a of random numbers is less than or equal to the preset value, a rejection scheme or an insertion scheme is randomly selected, and if a rejection scheme is selected, then Determine the position corresponding to the character in the recognition result according to the random number, and delete the character corresponding to the position to obtain the reference result after the character deletion processing; if the insertion scheme is selected, the position corresponding to the character in the recognition result is determined according to the random number, The character corresponding to the random number in the recognition result is inserted between the position and the position next to the position, and the reference result after the character insertion is obtained.
  • the recognition result is FREE
  • the random number is 2, that is, the random number 2 corresponds to the character in the recognition result as R.
  • the elimination scheme is selected, the random number 2 is corresponding to the random number 2.
  • the reference result is FEE; if the insertion scheme is selected, the corresponding character R in the recognition result is determined according to the random number 2, and the character corresponding to the next position of the character R is E After inserting a random number 2 between the character R and the character E corresponding to the character R in the recognition result, the reference result obtained is FRREE.
  • the corresponding modification scheme is selected to modify the recognition result, and the reference result is obtained, so as to realize the adaptability of the recognition result according to the number a of random numbers. Modification to avoid the situation that the first target user and the second target user will subjectively believe that the OCR recognition accuracy is high during the subsequent verification, and directly use the recognition result for feedback, thereby ensuring the subsequent first target user and second target user The effectiveness of feedback results.
  • the auxiliary method for optical character recognition further includes the following steps:
  • step S54 if the first target result, the second target result, and the recognition result are different from each other, the value of the number a of random numbers is updated to another value in the value range of a, Since the value range of a is 1 or 2, if the value of a in step S512 is 1, then the value of a is updated to 2; if the value of a in step S512 is 2, the value of a is updated to 1. ; And based on the updated a value, return to step S51 to continue execution until step S54 is executed.
  • the same result is determined as the target result.
  • the audit user refers to a user who specifically performs manual identification processing on the acquired target picture in the case of a matching failure.
  • the target result is the target result.
  • step S51 is returned to continue execution until step S54 is executed.
  • the target result is determined; if the matching fails, the target image and the failed result are sent to the audit user for audit processing, and the result feedback from the audit user is received as the target result. In this way, the matching can be performed again in the case of a matching failure, avoiding errors in the manual intervention process, thereby ensuring the accuracy of the matching, and further ensuring the accuracy of the subsequent feedback of the target result to the operating user.
  • the optical character recognition auxiliary method further includes the following steps:
  • the credibility of the target result is set, and the credibility is set to be greater than or equal to the preset in step S3 Threshold, and establish the mapping relationship between the target result and the target picture.
  • mapping relationship obtained in step S8 is updated to the recognition library of the optical character recognition tool.
  • the optical character recognition tool After the recognition library of the optical character recognition tool is updated, if the optical character recognition tool recognizes the target image, it will directly output the updated target result that has a mapping relationship with the target image and the reliability of the target result .
  • the recognition of the optical character recognition tool can be improved
  • the recognition range and accuracy of the library thereby improving the accuracy of OCR recognition.
  • an optical character recognition auxiliary device is provided, and the optical character recognition auxiliary device corresponds one-to-one with the optical character recognition auxiliary method in the foregoing embodiment.
  • the optical character recognition auxiliary device includes a first acquisition module 71, a recognition module 72, a first comparison module 73, a first comparison result module 74, a second comparison result module 75, and a sending module 76.
  • the detailed description of each functional module is as follows:
  • the first obtaining module 71 is configured to obtain a target picture to be recognized sent by the operating user, where the target picture includes character information;
  • the recognition module 72 is configured to use an optical character recognition tool to perform character recognition on the character information in the target picture, to obtain the recognition result and the reliability corresponding to the recognition result;
  • the first comparison module 73 is configured to compare the credibility with a preset threshold
  • the first comparison result module 74 is configured to determine the recognition result corresponding to the credibility as the target result if the credibility is greater than or equal to the preset threshold;
  • the second comparison result module 75 is configured to preprocess the recognition result corresponding to the credibility if the credibility is less than the preset threshold to determine the target result;
  • the sending module 76 is configured to send the target result as an optical character recognition result to the operating user.
  • the second comparison result module 75 includes:
  • the modification sub-module is used to modify the recognition result corresponding to the credibility according to the preset modification plan if the credibility is less than the preset threshold to obtain the reference result;
  • the verification sub-module is used to send the reference result and the target picture to the first target user and the second target user for verification;
  • the second obtaining submodule is used to obtain the first target result fed back by the first target user, and the second target result fed back by the second target user;
  • the matching sub-module is used to match the first target result, the second target result and the recognition result
  • the same matching sub-module is used to determine the same result as the target result if at least two of the first target result, the second target result and the recognition result are the same.
  • modified sub-module includes:
  • the third acquiring unit is configured to acquire the number n of characters corresponding to the recognition result if the credibility is less than the preset threshold, where n is a positive integer;
  • the random number generating unit is used to generate a random numbers according to the number of characters n, where a and the random number are both positive integers, and a ⁇ [1,2], the value range of the random number is [1, n ];
  • the reference result obtaining unit is used to select the modification plan corresponding to the number a of the random number to modify the recognition result to obtain the reference result, where the modification plan includes but is not limited to one or more of the following processing: Swap with Y, replace X with Y, remove X and insert X, X and Y are characters corresponding to different random numbers.
  • the reference result obtaining unit includes:
  • the second comparison subunit is used to compare the number a of random numbers with a preset value
  • the third comparison result subunit is used to modify the recognition result by randomly adopting a modification scheme of exchanging X with Y or replacing X with Y if the number a of random numbers is greater than the preset value, to obtain a reference result;
  • the fourth comparison result subunit is used to modify the recognition result by randomly adopting a modification scheme of excluding X or inserting X if the number a of random numbers is less than or equal to the preset value, to obtain a reference result.
  • optical character recognition auxiliary device further includes:
  • the value update module is used to update the value of the number a of random numbers to another value in the value range of a if the first target result, the second target result, and the recognition result are different from each other, and the The recognition result corresponding to the reliability is modified according to the preset modification plan, and the step of obtaining the reference result continues to be executed until the first target result, the second target result and the recognition result are matched;
  • the target result determination module is used to determine the target result if the matching is successful
  • the audit module is used to send the target picture and the result of the failed matching to the audit user for audit processing if the matching fails, and receive the result feedback from the audit user as the target result.
  • optical character recognition auxiliary device further includes:
  • An add module is used to set the credibility of the target result to be greater than or equal to a preset threshold when the target result is determined from the first target result, the second target result and the recognition result, and to establish the relationship between the target result and the target picture Mapping relations.
  • the recognition library update module is used to update the mapping relationship to the recognition library of the optical character recognition tool.
  • FIG. 8 is a block diagram of the basic structure of the computer device 90 in an embodiment of the present application.
  • the computer device 90 includes a memory 91, a processor 92, and a network interface 93 that are communicatively connected to each other through a system bus. It should be pointed out that FIG. 8 only shows a computer device 90 with components 91-93, but it should be understood that it is not required to implement all the illustrated components, and more or fewer components may be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculation and/or information processing in accordance with pre-set or stored instructions.
  • Its hardware includes but is not limited to microprocessors, dedicated Integrated Circuit (Application Specific Integrated Circuit, ASIC), Programmable Gate Array (Field-Programmable Gate Array, FPGA), Digital Processor (Digital Signal Processor, DSP), embedded devices, etc.
  • ASIC Application Specific Integrated Circuit
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • DSP Digital Processor
  • the computer device may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
  • the computer device can interact with the user through a keyboard, a mouse, a remote control, a touch panel, or a voice control device.
  • the memory 91 includes at least one type of readable storage medium, the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static memory Random access memory (SRAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disks, optical disks, etc.
  • the memory 91 may be an internal storage unit of the computer device 90, such as a hard disk or memory of the computer device 90.
  • the memory 91 may also be an external storage device of the computer device 90, such as a plug-in hard disk equipped on the computer device 90, a smart memory card (Smart Media Card, SMC), and a secure digital (Secure Digital, SD) card, Flash Card, etc.
  • the memory 91 may also include both the internal storage unit of the computer device 90 and its external storage device.
  • the memory 91 is generally used to store the operating system and various application software installed in the computer device 90, for example, the program code of the optical character recognition auxiliary method.
  • the memory 91 can also be used to temporarily store various types of data that have been output or will be output.
  • the processor 92 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments.
  • the processor 92 is generally used to control the overall operation of the computer device 90.
  • the processor 92 is configured to run the program code or process data stored in the memory 91, for example, run the program code of the optical character recognition auxiliary method.
  • the network interface 93 may include a wireless network interface or a wired network interface, and the network interface 93 is generally used to establish a communication connection between the computer device 90 and other electronic devices.
  • This application also provides another implementation manner, that is, to provide a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores an OCR identification information entry program, the OCR identification
  • the information entry program may be executed by at least one processor, so that the at least one processor executes the steps of any one of the aforementioned optical character recognition auxiliary methods.
  • the method of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. ⁇
  • the technical solution of this application essentially or the part that contributes to the existing technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, The optical disc) includes several instructions to enable a computer device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in the various embodiments of the present application.
  • a computer device which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Character Discrimination (AREA)
  • Character Input (AREA)

Abstract

一种光学字符识别辅助方法、装置、计算机设备及存储介质,涉及人工智能技术领域,所述光学字符识别辅助方法包括:获取操作用户发送的待识别的目标图片(S1);使用光学字符识别工具对目标图片中的字符信息进行字符识别,得到识别结果及识别结果对应的可信度(S2);将可信度与预设阈值进行比较(S3);若可信度大于等于预设阈值,则将可信度对应的识别结果确定为目标结果(S4);若可信度小于预设阈值,则对可信度对应的识别结果进行预处理,以确定目标结果(S5);将目标结果作为光学字符识别结果发送给操作用户(S6)。上述技术方案实现对光学字符识别结果进行辅助校验,进一步提高光学字符识别的准确性,保证操作用户的工作效率。

Description

光学字符识别辅助方法、装置、计算机设备及存储介质
【交叉引用】
本申请以2019年6月26日提交的申请号为2019105583199,名称为“光学字符识别辅助方法、装置、计算机设备及存储介质”的中国发明专利申请为基础,并要求其优先权。
【技术领域】
本申请涉及人工智能技术领域,尤其涉及一种光学字符识别辅助方法、装置、计算机设备及存储介质。
【背景技术】
OCR(Optical Character Recognition,光学字符识别)是指通过AI算法对图像中的文字进行识别。但由于目前的算法处理能力有限,往往识别出来的结果准确性无法达到100%或者达到使用者的需求,发明人意识到,如果将OCR识别结果原封不动的作为参考答案发放给用户进行校验,会存在识别结果不准确的情况,从而影响用户的工作效率。
【发明内容】
本申请实施例提供一种光学字符识别辅助方法、装置、计算机设备及存储介质,以解决OCR识别不准确,影响用户工作效率的问题。
一种光学字符识别辅助方法,包括:
获取操作用户发送的待识别的目标图片,其中,所述目标图片包括字符信息;
使用光学字符识别工具对所述目标图片中的字符信息进行字符识别,得到识别结果及所述识别结果对应的可信度;
将所述可信度与预设阈值进行比较;
若所述可信度大于等于预设阈值,则将所述可信度对应的所述识别结果确定为目标结果;
若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果;
将所述目标结果作为光学字符识别结果发送给所述操作用户。
一种光学字符识别辅助装置,包括:
第一获取模块,用于获取操作用户发送的待识别的目标图片,其中,所述目标图片包括字符信息;
识别模块,用于使用光学字符识别工具对所述目标图片中的字符信息进行字符识别,得到识别结果及所述识别结果对应的可信度;
第一比较模块,用于将所述可信度与预设阈值进行比较;
第一比较结果模块,用于若所述可信度大于等于预设阈值,则将所述可信度对应的所述识别结果确定为目标结果;
第二比较结果模块,用于若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果;
发送模块,用于将所述目标结果作为光学字符识别结果发送给所述操作用户。
一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现上述光学字符识别辅助方法的步骤。
一种非易失性的计算机可读存储介质,所述非易失性的计算机可读存储介质存储有计算机可读指令,所述计算机可读指令被一种处理器执行时实现上述任一种光学字符识别辅助方法的步骤。
本申请的一个或多个实施例的细节在下面的附图和描述中提出,本申请的其他特征和优点将从说明书、附图以及权利要求变得明显。
【附图说明】
为了更清楚地说明本申请实施例的技术方案,下面将对本申请实施例的描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的光学字符识别辅助方法的流程图;
图2是本申请实施例提供的光学字符识别辅助方法中步骤S5的流程图;
图3是本申请实施例提供的光学字符识别辅助方法中步骤S51的流程图;
图4是本申请实施例提供的光学字符识别辅助方法中步骤S513的流程图;
图5是本申请实施例提供的光学字符识别辅助方法中在第一目标结果、第二目标结果和识别结果都不相同的情况下确定目标结果的流程图;
图6是本申请实施例提供的光学字符识别辅助方法中更新OCR识别库的流程图;
图7是本申请实施例提供的光学字符识别辅助装置的示意图;
图8是本申请实施例提供的计算机设备的基本机构框图。
【具体实施方式】
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请提供的光学字符识别辅助方法应用于服务端,服务端具体可以用独立的服务器或者多个服务器组成的服务器集群实现。在一实施例中,如图1所示,提供一种光学字符识别辅助方法,包括如下步骤:
S1:获取操作用户发送的待识别的目标图片,其中,目标图片包括字符信息。
在本申请实施例中,待识别的目标图片是指包括字符信息,且需要通过OCR工具帮助识别的图片。
具体地,当操作用户在客户端中发送待识别的目标图片时,目标图片将自动保存到预设图片库中,通过对预设图片库进行检测,若检测到预设图片库中存在待识别的目标图片,则对该目标图片进行获取。其中,预设图片库是指专门用于存储待识别的目标图片的数据库。
需要说明的是,当从预设图片库中获取到目标图片后,将该目标图片从预设图片库中进行删除处理。
S2:使用光学字符识别工具对目标图片中的字符信息进行字符识别,得到识别结果及识别结果对应的可信度。
在本申请实施例中,光学字符识别工具是指专门利用字符识别方法将图片中的字符形状翻译成计算机文字的工具。可信度是指OCR工具针对识别出来的识别结果设定的信赖程度,可信度越高,表示OCR识别的准确率越高。
具体地,通过直接使用OCR工具识别目标图片中的字符信息,得到OCR工具识别后输出的识别结果以及该识别结果对应的可信度。
例如,通过利用OCR识别工具对目标图片中的字符信息进行识别后,得到的识别结果为“ABCD”,且该识别结果对应的可信度为80%。
S3:将可信度与预设阈值进行比较。
具体地,根据步骤S2中得到识别结果对应的可信度,将该可信度与预设阈值进行比较。
S4:若可信度大于等于预设阈值,则将可信度对应的识别结果确定为目标结果。
具体地,根据步骤S3中将该可信度与预设阈值进行比较,若比较结果为可信度大于等于预设阈值,则将该可信度对应的识别结果确定为目标结果。其中,预设阈值具体可以是80%,也可以根据用户的实际需求进行设置,此处不做限制。
例如,存在识别结果A,其对应的可信度为80%,预设阈值为80%,将可信度与预设阈值进行比较,由于可信度与预设阈值都为80%,故将该识别结果A确定为目标结果。
S5:若可信度小于预设阈值,则对可信度对应的识别结果进行预处理,以确定目标结果。
具体地,据步骤S3中将该可信度与预设阈值进行比较,若比较结果为可信度小于预设阈值,则表示该可信度对应的识别结果的准确性达不到用户设定的标准,对识别结果进行预处理,预处理的具体过程可以包括:对识别结果进行修改处理得到参考结果,再将参考结果及待识别的目标图片发送给第一目标用户和第二目标用户进行校验,获取第一目标用户和第二目标用户校验后反馈的第一目标结果和第二目标结果,并对第一目标结果、第二目标结果和识别结果进行匹配处理,获取匹配相同的结果作为目标结果。
S6:将目标结果作为光学字符识别结果发送给操作用户。
具体地,将目标结果作为OCR识别结果,并按照预设的方式发送给操作用户,其中,预设的方式具体可以是以反馈在客户端的形式,例如,直接将OCR识别结果反馈在客户端指定的显示界面中,也可以根据用户的实际需求进行设定。
本实施例中,通过利用OCR工具对待识别的目标图片进行识别,得到识别结果及识别结果对应的可信度,并将可信度与预设阈值进行比较,将大于等于预设阈值的可信度对应的识别结果确定为目标结果,将小于预设阈值的可信度对应的识别结果进行预处理,得到预处理后的目标结果,最后再把目标结果作为OCR识别结果发送给操作用户,从而实现对OCR识别出来的识别结果进行辅助校验,针对可信度不高的识别结果进行预处理,能够有效弥补OCR算法识别的不足,进一步提高OCR识别的准确性,保证操作用户的工作效率。
在一实施例中,如图2所示,步骤S5中,即若可信度小于预设阈值,则对可信度对应的识别结果进行预处理,以确定目标结果包括如下步骤:
S51:若可信度小于预设阈值,则对可信度对应的识别结果按预设修改方案进行修改处理,得到参考结果。
在本申请实施例中,若可信度小于预设阈值,则对该可信度对应的识别结果按照预设修改方案进行修改处理,其中,预设修改方案是指用户预先设定好的方案,其具体过程可以包括:先根据识别结果生成随机数,再选取随机数对应的修改方案对识别结果进行修改,得到修改后的参考结果。
需要说明的是,随机数是指大于0正整数。
S52:将参考结果及目标图片分别发送给第一目标用户和第二目标用户进行校验。
在本申请实施例中,第一目标用户和第二目标用户是指专门对目标图片中的字符信息与参考结果进行校验的合法用户。通过将步骤S51得到的参考结果及目标图片按照预设的方式分别发送给第一目标用户和第二目标用户进行校验处理。
需要说明的是,如果直接将识别结果发送给第一目标用户和第二目标用户进行校验,第一目标用户和第二目标用户可能会凭主观意识认为OCR工具识别的准确率高,导致不会对识别结果进行认真校验,直接利用识别结果进行反馈。因此,为了避免这种情况,需要对OCR识别结果进行处理成参考结果,再将参考结果发送给第一目标用户和第二目标用户进行校验。
S53:获取第一目标用户反馈的第一目标结果,第二目标用户反馈的第二目标结果。
具体地,当第一目标用户接收到参考结果和目标图片之后,根据目标图片中的字符信息对参考结果进行分析,判断参考结果是否与目标图片中的字符信息相同,若相同则可直接将参考结果作为第一目标结果,若不相同则可根据分析情况重新制定识别结果,并将该识别结果作为第一目标结果,同理,第二目标用户在接收到参考结果和目标图片之后,根据与第一目标用户得到第一目标结果的情况得到第二目标结果。
第一目标用户和第二目标用户通过客户端及分别将第一目标结果和第二目标结果反馈给服务端,服务端在接收到第一目标用户和第二目标用户的反馈信息时,将直接获取第一目标用户反馈的第一目标结果和第二目标用户反馈的第二目标结果。
S54:对第一目标结果、第二目标结果和识别结果进行匹配。
具体地,将第一目标结果、第二目标结果和识别结果分别进行匹配,即将第一目标结果与第二目标结果进行匹配,第一目标结果与识别结果进行匹配,第二目标结果与识别结果进行匹配。
S55:若第一目标结果、第二目标结果和识别结果之中至少有两个相同,则将相同的结果确定为目标结果。
具体地,根据步骤S54中的匹配过程,若第一目标结果、第二目标结果和识别结果之中存在第一目标结果与第二目标结果相同,或者第一目标结果与识别结果相同,或者第二目标结果与识别结果相同,或者第一目标结果与第二目标结果和识别结果都相同,则表示第一目标结果、第二目标结果和识别结果之中至少有两个结果是相同的,并将相同的结果确定为目标结果。
例如,若第一目标结果为A,第二目标结果为B,识别结果为A,通过对第一目标结果A、第二目标结果B与识别结果A之间分别进行匹配,得到第一目标结果A与识别结果A相同,则表示第一目标结果、第二目标结果和识别结果之中至少有两个结果是相同的,并将相同的结果A确定为目标结果。
本实施例中,在识别结果的可信度小于预设阈值的情况下,对识别结果进行修改处理得到参考结果,并将参考结果发送给第一目标用户和第二目标用户进行校验,获取第一目标用户和第二目标用户反馈的第一目标结果和第二目标结果,最后利用第一目标结果、第二目标结果和识别结果进行匹配的方式得到目标结果,从而实现在OCR识别结果对应的可信度小于预设阈值的情况下,结合人工对识别结果做进一步校验,避免OCR识别结果不准确的情况,进一步保证操作用户能够获取准确的识别结果,提供操作用户工作效率。
在一实施例中,如图3所示,步骤S51中,即若可信度小于预设阈值,则对可信度对应的识别结果按预设修改方案进行修改处理,得到参考结果包括如下步骤:
S511:若可信度小于预设阈值,则获取识别结果对应的字符个数n,其中,n为正整数。
具体地,若可信度小于预设阈值,则对识别结果进行遍历,通过遍历得到识别结果中包含字符的字符个数n。
例如,存在识别结果为“天下太平”,该识别结果存在“天”、“下”、“太”和“平”4个字符。通过对该识别结果进行遍历,得到该识别结果包含字符的字符个数n为4。
S512:根据字符个数n,生成a个随机数,其中,a和随机数均为正整数,且a∈[1,2]、随机数的取值范围为[1,n]。
在本申请实施例中,随机数是指识别结果中的字符对应的字符位置,且随机数的取值范围为大于0的正整数。例如,若识别结果为“QWE”,则“Q”字符对 应的随机数为1,“W”字符对应的随机数为2,“E”字符对应的随机数为3。
具体地,根据步骤S511得到的字符个数n,生成a个随机数,且a和随机数都为正整数,a的取值范围为1或者2,随机数的取值范围在字符个数n的范围内,例如,字符个数为4,则随机数的取值范围为1到4之间。
需要说明的是,当a的取值范围大于1时,随机数的取值不可重复。
例如,存在识别结果对应的字符个数n为4,若a为1,即生成1个随机数,则随机数的取值具体可以为1或者2或者3或者4;若a为2,即生成2个随机数,则2个随机数的取值具体可以为1和2,或者1和3,或者1和4等等。
S513:选取随机数的数量a对应的修改方案,以对识别结果进行修改处理,得到参考结果,其中,修改方案包括但不限于以下处理中的一种或者多种:将X与Y交换、将X替换Y、剔除X和插入X,X和Y为不同的随机数对应的字符。
在本申请实施例中,将X与Y交换是指对不同随机数对应的不同字符进行交换,将该修改方案作为交换方案;将X替换Y是指将随机数对应的字符替换成另个随机数对应的字符,将该修改方案作为替换方案;剔除X是指对随机数对应的字符进行删除,将该修改方案作为剔除方案;插入X是指在随机数对应的字符与下一字符之间的位置插入随机数对应的字符,将该修改方案作为插入方案。
具体地,根据步骤S512中生成的随机数的数量a,将该数量a与预设方案库中的标识数字进行匹配,当匹配到随机数的数量a与预设方案库中的标识数字相同时,选取该标识数字对应的修改方案对识别结果进行修改处理,并将修改之后的结果作为参考结果。其中,预设方案库是指专门用于存储修改方案以及每种修改方案对应的标识数字,例如,交换方案对应的标识数字为2。
需要说明的是,若预设方案库中存在的多种修改方案对应的标识数字相同时,则在根据标识数字选取修改方案时,将随机选取其中的修改方案进行修改处理。
例如,存在识别结果为“ASDF”,预设方案库中存在交换方案、替换方案、剔除方案和插入方案分别对应的标识数字为2、2、1和1,若随机数的数量a为1,则将数量1与预设方案库中的标识数字进行匹配,得到标识数字1与数量1相同,故随机选取剔除方案对识别结果“ASDF”进行修改处理,若修改处理后得到的结果为“ADF”,则将该“ADF”确定为参考结果。
本实施例中,在可信度小于预设阈值的情况下根据识别结果对应的字符个数n生成a个随机数,并根据随机数的数量a选取对应的修改方案对识别结果进行修改处理,从而实现将识别结果修改成参考结果,避免后续第一目标用户和第二目标用户在进行校验时会凭主观意识认为OCR识别准确率高,直接利用识别结果进行反馈的情况,进而保证后续第一目标用户和第二目标用户反馈结果的有效性。
在一实施例中,如图4所示,步骤S513中,即选取随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果包括如下步骤:
S5131:将随机数的数量a与预设数值进行比较。
具体地,直接将随机数的数量a与预设数值进行比较,其中,预设数值具 体可以是1,也可以是2,其具体的取值范围可以根据用户的实际需求进行设置,此处不做限制,优选的,本提案对预设数值的取值设置为1。
S5132:若随机数的数量a大于预设数值,则随机采用将X与Y交换或将X替换Y的修改方案对识别结果进行修改处理,得到参考结果。
具体地,根据步骤S5131中将随机数的数量a与预设数值进行比较,若比较结果为随机数的数量a大于预设数值,随机选取交换方案或者替换方案,若选取到交换方案,则根据每个随机数确定识别结果中字符对应的位置,并随机选取2个随机数对应的位置之间的字符进行交换,得到字符交换后的参考结果;若选取到替换方案,则根据每个随机数确定识别结果中字符对应的位置,随机确定替换位置与被替换位置,并将被替换位置对应的字符替换成替换位置对应的字符,得到字符替换后的参考结果。
例如,若预设数值为1,识别结果为ABCD,存在两位随机数1和2,即随机数1对应识别结果的字符为A,随机数2对应识别结果的字符为B,若选取到交换方案,则根据随机数1和2,将字符A和字符B之间的位置进行交换之后,得到的参考结果为BACD;若选取到替换方案,随机确定的替换位置为1,被替换位置为2,将被替换位置2对应的字符B替换成替换位置1对应的字符A之后,得到的参考结果为AACD。
S5133:若随机数的数量a小于等于预设数值,则随机采用剔除X或插入X的修改方案对识别结果进行修改处理,得到参考结果。
具体地,根据步骤S5131中将随机数的数量a与预设数值进行比较,若比较结果为随机数的数量a小于等于预设数值,随机选取剔除方案或者插入方案,若选取到剔除方案,则根据随机数确定识别结果中字符对应的位置,将该位置对应的字符进行删除处理,得到字符删除处理后的参考结果;若选取到插入方案,则根据随机数确定识别结果中字符对应的位置,在该位置与该位置的下一位置之间插入随机数在识别结果中对应的字符,得到字符插入后的参考结果。
例如,若预设数值为1,识别结果为FREE,存在随机数为2,即随机数2对应识别结果中的字符为R,若选取到剔除方案,则根据随机数2,将随机数2对应识别结果中的字符R进行删除处理后,得到的参考结果为FEE;若选取到插入方案,则根据随机数2确定识别结果中对应的字符R,以及字符R的下一位置对应的字符为E,在字符R和字符E之间插入随机数2对应识别结果中的字符R后,得到的参考结果为FRREE。
本实施例中,通过将随机数的数量a与预设数值进行比较,选取对应的修改方案对识别结果进行修改处理,得到参考结果,从而实现根据随机数的数量a能够对识别结果进行适应性修改,避免后续第一目标用户和第二目标用户在进行校验时会凭主观意识认为OCR识别准确率高,直接利用识别结果进行反馈的情况,进而保证后续第一目标用户和第二目标用户反馈结果的有效性。
在一实施例中,如图5所示,步骤S54之后,该光学字符识别辅助方法还包括如下步骤:
S71:若第一目标结果、第二目标结果和识别结果互不相同,则将随机数的数量a的值更新为a的取值范围中的另一个取值,并对可信度对应的识别结果按 预设修改方案进行修改处理,得到参考结果的步骤继续执行,直到对第一目标结果、第二目标结果和识别结果进行匹配为止。
具体地,根据步骤S54的匹配情况,若第一目标结果、第二目标结果和识别结果互不相同,则将随机数的数量a的值更新为a的取值范围中的另一个取值,由于a的取值范围为1或者2,若步骤S512中a的取值为1,则将a的值更新为2;若步骤S512中a的取值为2,则将a的值更新为1;并基于更新后的a值,重返步骤S51继续执行,直到执行到步骤S54为止。
S72:若匹配成功,则确定目标结果。
具体地,若第一目标结果、第二目标结果和识别结果之中至少有两个相同,则表示匹配成功,并将相同的结果确定为目标结果。
S73:若匹配失败,则将目标图片和匹配失败的结果发送给审核用户进行审核处理,并接收审核用户反馈的结果作为目标结果。
在本申请实施例中,审核用户是指专门在匹配失败的情况下,针对获取到的目标图片进行人工识别处理的用户。
具体地,若第一目标结果、第二目标结果和识别结果互不相同,则表示匹配失败,将目标图片和匹配失败的结果按照预设的方式发送给审核用户进行审核处理,并获取审核用户反馈的结果作为目标结果。
本实施例中,在第一目标结果、第二目标结果和识别结果都不相同的情况下,对随机数的数量a的值进行更新,并重返步骤S51继续执行,直到执行到步骤S54为止,若匹配成功则确定目标结果,若匹配失败则将目标图片和失败结果发送给审核用户进行审核处理,接收审核用户反馈的结果作为目标结果。从而实现在出现匹配失败的情况下能够重新进行匹配,避免人工干预过程中出现失误的情况,从而确保匹配的准确性,进一步保证后续将目标结果反馈给操作用户的准确性。
在一实施例中,如图6所示,步骤S55之后,该光学字符识别辅助方法还包括如下步骤:
S8:当从第一目标结果、第二目标结果和识别结果之中确定出目标结果时,设置目标结果的可信度大于等于预设阈值,并建立目标结果与目标图片之间的映射关系。
具体地,当从第一目标结果、第二目标结果和识别结果之中确定出目标结果时,对该目标结果的可信度进行设置,将可信度设置成大于等于步骤S3中的预设阈值,并建立目标结果与目标图片之间的映射关系。
S9:将映射关系更新到光学字符识别工具的识别库中。
具体地,将步骤S8得到的映射关系更新到光学字符识别工具的识别库中。
需要说明的是,光学字符识别工具的识别库经过更新之后,若光学字符识别工具识别到目标图片,则直接输出经过更新后与目标图片存在映射关系的目标结果以及该目标结果对应的可信度。
本实施例中,通过设置目标结果的可信度,然后建立目标结果与目标图片之间的映射关系,最后将映射关系更新到光学字符识别工具的识别库中,能够提高光学字符识别工具的识别库的识别范围以及识别的准确性,进而提高OCR 识别的准确性。
应理解,上述实施例中各步骤的序号的大小并不意味着执行顺序的先后,各过程的执行顺序应以其功能和内在逻辑确定,而不应对本申请实施例的实施过程构成任何限定。
在一实施例中,提供一种光学字符识别辅助装置,该光学字符识别辅助装置与上述实施例中光学字符识别辅助方法一一对应。如图7所示,该光学字符识别辅助装置包括第一获取模块71、识别模块72、第一比较模块73、第一比较结果模块74、第二比较结果模块75和发送模块76。各功能模块详细说明如下:
第一获取模块71,用于获取操作用户发送的待识别的目标图片,其中,目标图片包括字符信息;
识别模块72,用于使用光学字符识别工具对目标图片中的字符信息进行字符识别,得到识别结果及识别结果对应的可信度;
第一比较模块73,用于将可信度与预设阈值进行比较;
第一比较结果模块74,用于若可信度大于等于预设阈值,则将可信度对应的识别结果确定为目标结果;
第二比较结果模块75,用于若可信度小于预设阈值,则对可信度对应的识别结果进行预处理,以确定目标结果;
发送模块76,用于将目标结果作为光学字符识别结果发送给操作用户。
进一步地,第二比较结果模块75包括:
修改子模块,用于若可信度小于预设阈值,则对可信度对应的识别结果按预设修改方案进行修改处理,得到参考结果;
校验子模块,用于将参考结果及目标图片分别发送给第一目标用户和第二目标用户进行校验;
第二获取子模块,用于获取第一目标用户反馈的第一目标结果,第二目标用户反馈的第二目标结果;
匹配子模块,用于对第一目标结果、第二目标结果和识别结果进行匹配;
匹配相同子模块,用于若第一目标结果、第二目标结果和识别结果之中至少有两个相同,则将相同的结果确定为目标结果。
进一步地,修改子模块包括:
第三获取单元,用于若可信度小于预设阈值,则获取识别结果对应的字符个数n,其中,n为正整数;
随机数生成单元,用于根据字符个数n,生成a个随机数,其中,a和随机数均为正整数,且a∈[1,2]、随机数的取值范围为[1,n];
参考结果获取单元,用于选取随机数的数量a对应的修改方案,以对识别结果进行修改处理,得到参考结果,其中,修改方案包括但不限于以下处理中的一种或者多种:将X与Y交换、将X替换Y、剔除X和插入X,X和Y为不同的随机数对应的字符。
进一步地,参考结果获取单元包括:
第二比较子单元,用于将随机数的数量a与预设数值进行比较;
第三比较结果子单元,用于若随机数的数量a大于预设数值,则随机采用 将X与Y交换或将X替换Y的修改方案对识别结果进行修改处理,得到参考结果;
第四比较结果子单元,用于若随机数的数量a小于等于预设数值,则随机采用剔除X或插入X的修改方案对识别结果进行修改处理,得到参考结果。
进一步地,该光学字符识别辅助装置还包括:
取值更新模块,用于若第一目标结果、第二目标结果和识别结果互不相同,则将随机数的数量a的值更新为a的取值范围中的另一个取值,并对可信度对应的识别结果按预设修改方案进行修改处理,得到参考结果的步骤继续执行,直到对第一目标结果、第二目标结果和识别结果进行匹配为止;
目标结果确定模块,用于若匹配成功,则确定目标结果;
审核模块,用于若匹配失败,则将目标图片和匹配失败的结果发送给审核用户进行审核处理,并接收审核用户反馈的结果作为目标结果。
进一步地,该光学字符识别辅助装置还包括:
添加模块,用于当从第一目标结果、第二目标结果和识别结果之中确定出目标结果时,设置目标结果的可信度大于等于预设阈值,并建立目标结果与目标图片之间的映射关系。
识别库更新模块,用于将映射关系更新到光学字符识别工具的识别库中。
本申请的一些实施例公开了计算机设备。具体请参阅图8,为本申请的一实施例中计算机设备90基本结构框图。
如图8中所示意的,所述计算机设备90包括通过系统总线相互通信连接存储器91、处理器92、网络接口93。需要指出的是,图8中仅示出了具有组件91-93的计算机设备90,但是应理解的是,并不要求实施所有示出的组件,可以替代的实施更多或者更少的组件。其中,本技术领域技术人员可以理解,这里的计算机设备是一种能够按照事先设定或存储的指令,自动进行数值计算和/或信息处理的设备,其硬件包括但不限于微处理器、专用集成电路(Application Specific Integrated Circuit,ASIC)、可编程门阵列(Field-Programmable Gate Array,FPGA)、数字处理器(Digital Signal Processor,DSP)、嵌入式设备等。
所述计算机设备可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述计算机设备可以与用户通过键盘、鼠标、遥控器、触摸板或声控设备等方式进行人机交互。
所述存储器91至少包括一种类型的可读存储介质,所述可读存储介质包括闪存、硬盘、多媒体卡、卡型存储器(例如,SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘等。在一些实施例中,所述存储器91可以是所述计算机设备90的内部存储单元,例如该计算机设备90的硬盘或内存。在另一些实施例中,所述存储器91也可以是所述计算机设备90的外部存储设备,例如该计算机设备90上配备的插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)等。当然,所述存储器91还可以既包括所述计算机设备90的内部存储单元也包括其外部存储设备。本实施例中,所述 存储器91通常用于存储安装于所述计算机设备90的操作系统和各类应用软件,例如所述光学字符识别辅助方法的程序代码等。此外,所述存储器91还可以用于暂时地存储已经输出或者将要输出的各类数据。
所述处理器92在一些实施例中可以是中央处理器(Central Processing Unit,CPU)、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器92通常用于控制所述计算机设备90的总体操作。本实施例中,所述处理器92用于运行所述存储器91中存储的程序代码或者处理数据,例如运行所述光学字符识别辅助方法的程序代码。
所述网络接口93可包括无线网络接口或有线网络接口,该网络接口93通常用于在所述计算机设备90与其他电子设备之间建立通信连接。
本申请还提供了另一种实施方式,即提供一种非易失性的计算机可读存储介质,所述非易失性的计算机可读存储介质存储有OCR识别信息录入程序,所述OCR识别信息录入程序可被至少一个处理器执行,以使所述至少一个处理器执行上述任意一种光学字符识别辅助方法的步骤。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台计算机设备(可以是手机,计算机,服务器,空调器,或者网络设备等)执行本申请各个实施例所述的方法。
最后应说明的是,显然以上所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例,附图中给出了本申请的较佳实施例,但并不限制本申请的专利范围。本申请可以以许多不同的形式来实现,相反地,提供这些实施例的目的是使对本申请的公开内容的理解更加透彻全面。尽管参照前述实施例对本申请进行了详细的说明,对于本领域的技术人员来而言,其依然可以对前述各具体实施方式所记载的技术方案进行修改,或者对其中部分技术特征进行等效替换。凡是利用本申请说明书及附图内容所做的等效结构,直接或间接运用在其他相关的技术领域,均同理在本申请专利保护范围之内。

Claims (20)

  1. 一种光学字符识别辅助方法,其特征在于,所述光学字符识别辅助方法包括:
    获取操作用户发送的待识别的目标图片,其中,所述目标图片包括字符信息;
    使用光学字符识别工具对所述目标图片中的字符信息进行字符识别,得到识别结果及所述识别结果对应的可信度;
    将所述可信度与预设阈值进行比较;
    若所述可信度大于等于预设阈值,则将所述可信度对应的所述识别结果确定为目标结果;
    若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果;
    将所述目标结果作为光学字符识别结果发送给所述操作用户。
  2. 如权利要求1所述的光学字符识别辅助方法,其特征在于,所述若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果的步骤包括:
    若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果;
    将所述参考结果及所述目标图片分别发送给第一目标用户和第二目标用户进行校验;
    获取所述第一目标用户反馈的第一目标结果,所述第二目标用户反馈的第二目标结果;
    对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配;
    若所述第一目标结果、所述第二目标结果和所述识别结果之中至少有两个相同,则将相同的结果确定为所述目标结果。
  3. 如权利要求2所述的光学字符识别辅助方法,其特征在于,所述若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果的步骤包括:
    若所述可信度小于预设阈值,则获取所述识别结果对应的字符个数n,其中,n为正整数;
    根据所述字符个数n,生成a个随机数,其中,所述a和所述随机数均为正整数,且a∈[1,2]、所述随机数的取值范围为[1,n];
    选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果,其中,所述修改方案包括但不限于以下处理中的一种或者多种:将X与Y交换、将X替换Y、剔除X和插入X,X和Y为不同的随机数对应的字符。
  4. 如权利要求3所述的光学字符识别辅助方法,其特征在于,所述选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参 考结果的步骤包括:
    将所述随机数的数量a与预设数值进行比较;
    若所述随机数的数量a大于预设数值,则随机采用将X与Y交换或将X替换Y的修改方案对所述识别结果进行修改处理,得到所述参考结果;
    若所述随机数的数量a小于等于预设数值,则随机采用剔除X或插入X的修改方案对所述识别结果进行修改处理,得到所述参考结果。
  5. 如权利要求2所述的光学字符识别辅助方法,其特征在于,所述对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配的步骤之后,所述光学字符识别辅助方法还包括:
    若所述第一目标结果、所述第二目标结果和所述识别结果互不相同,则将所述随机数的数量a的值更新为a的取值范围中的另一个取值,并对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果的步骤继续执行,直到对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配为止;
    若匹配成功,则确定所述目标结果;
    若匹配失败,则将所述目标图片和匹配失败的结果发送给审核用户进行审核处理,并接收所述审核用户反馈的结果作为所述目标结果。
  6. 如权利要求2所述的光学字符识别辅助方法,其特征在于,所述若所述第一目标结果、所述第二目标结果和所述识别结果之中至少有两个相同,则将相同的结果确定为所述目标结果的步骤之后,所述光学字符识别辅助方法还包括:
    当从所述第一目标结果、所述第二目标结果和所述识别结果之中确定出所述目标结果时,设置所述目标结果的可信度大于等于所述预设阈值,并建立所述目标结果与所述目标图片之间的映射关系;
    将所述映射关系更新到所述光学字符识别工具的识别库中。
  7. 一种光学字符识别辅助装置,其特征在于,所述光学字符识别辅助装置包括:
    第一获取模块,用于获取操作用户发送的待识别的目标图片,其中,所述目标图片包括字符信息;
    识别模块,用于使用光学字符识别工具对所述目标图片中的字符信息进行字符识别,得到识别结果及所述识别结果对应的可信度;
    第一比较模块,用于将所述可信度与预设阈值进行比较;
    第一比较结果模块,用于若所述可信度大于等于预设阈值,则将所述可信度对应的所述识别结果确定为目标结果;
    第二比较结果模块,用于若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果;
    发送模块,用于将所述目标结果作为光学字符识别结果发送给所述操作用户。
  8. 如权利要求7所述的光学字符识别辅助装置,其特征在于,所述第二比较结果模块包括:
    修改子模块,用于若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果;
    校验子模块,用于将所述参考结果及所述目标图片分别发送给第一目标用户和第二目标用户进行校验;
    第二获取子模块,用于获取所述第一目标用户反馈的第一目标结果,所述第二目标用户反馈的第二目标结果;
    匹配子模块,用于对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配;
    匹配相同子模块,用于若所述第一目标结果、所述第二目标结果和所述识别结果之中至少有两个相同,则将相同的结果确定为所述目标结果。
  9. 如权利要求8所述的光学字符识别辅助装置,其特征在于,所述修改子模块包括:
    第三获取单元,用于若所述可信度小于预设阈值,则获取所述识别结果对应的字符个数n,其中,n为正整数;
    随机数生成单元,用于根据所述字符个数n,生成a个随机数,其中,所述a和所述随机数均为正整数,且a∈[1,2]、所述随机数的取值范围为[1,n];
    参考结果获取单元,用于选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果,其中,所述修改方案包括但不限于以下处理中的一种或者多种:将X与Y交换、将X替换Y、剔除X和插入X,X和Y为不同的随机数对应的字符。
  10. 如权利要求9所述的光学字符识别辅助装置,其特征在于,所述参考结果获取单元包括:
    随机数比较子单元,用于将所述随机数的数量a与预设数值进行比较;
    第三比较结果子单元,用于若所述随机数的数量a大于预设数值,则随机采用将X与Y交换或将X替换Y的修改方案对所述识别结果进行修改处理,得到所述参考结果;
    第四比较结果子单元,用于若所述随机数的数量a小于等于预设数值,则随机采用剔除X或插入X的修改方案对所述识别结果进行修改处理,得到所述参考结果。
  11. 一种计算机设备,包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机可读指令,其特征在于,所述处理器执行所述计算机可读指令时实现如下步骤:
    获取操作用户发送的待识别的目标图片,其中,所述目标图片包括字符信息;
    使用光学字符识别工具对所述目标图片中的字符信息进行字符识别,得到识别结果及所述识别结果对应的可信度;
    将所述可信度与预设阈值进行比较;
    若所述可信度大于等于预设阈值,则将所述可信度对应的所述识别结果确定为目标结果;
    若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预 处理,以确定目标结果;
    将所述目标结果作为光学字符识别结果发送给所述操作用户。
  12. 如权利要求11所述的计算机设备,其特征在于,所述若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果的步骤包括:
    若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果;
    将所述参考结果及所述目标图片分别发送给第一目标用户和第二目标用户进行校验;
    获取所述第一目标用户反馈的第一目标结果,所述第二目标用户反馈的第二目标结果;
    对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配;
    若所述第一目标结果、所述第二目标结果和所述识别结果之中至少有两个相同,则将相同的结果确定为所述目标结果。
  13. 如权利要求12所述的计算机设备,其特征在于,所述若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果的步骤包括:
    若所述可信度小于预设阈值,则获取所述识别结果对应的字符个数n,其中,n为正整数;
    根据所述字符个数n,生成a个随机数,其中,所述a和所述随机数均为正整数,且a∈[1,2]、所述随机数的取值范围为[1,n];
    选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果,其中,所述修改方案包括但不限于以下处理中的一种或者多种:将X与Y交换、将X替换Y、剔除X和插入X,X和Y为不同的随机数对应的字符。
  14. 如权利要求13所述的计算机设备,其特征在于,所述选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果的步骤包括:
    将所述随机数的数量a与预设数值进行比较;
    若所述随机数的数量a大于预设数值,则随机采用将X与Y交换或将X替换Y的修改方案对所述识别结果进行修改处理,得到所述参考结果;
    若所述随机数的数量a小于等于预设数值,则随机采用剔除X或插入X的修改方案对所述识别结果进行修改处理,得到所述参考结果。
  15. 如权利要求12所述的计算机设备,其特征在于,所述对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配的步骤之后,所述处理器执行所述计算机可读指令时还包括实现如下步骤:
    若所述第一目标结果、所述第二目标结果和所述识别结果互不相同,则将所述随机数的数量a的值更新为a的取值范围中的另一个取值,并对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果的步骤继续执行,直到对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配 为止;
    若匹配成功,则确定所述目标结果;
    若匹配失败,则将所述目标图片和匹配失败的结果发送给审核用户进行审核处理,并接收所述审核用户反馈的结果作为所述目标结果。
  16. 一种非易失性的计算机可读存储介质,所述非易失性的计算机可读存储介质存储有计算机可读指令,其特征在于,所述计算机可读指令被一种处理器执行时,使得所述一种处理器执行如下步骤:
    获取操作用户发送的待识别的目标图片,其中,所述目标图片包括字符信息;
    使用光学字符识别工具对所述目标图片中的字符信息进行字符识别,得到识别结果及所述识别结果对应的可信度;
    将所述可信度与预设阈值进行比较;
    若所述可信度大于等于预设阈值,则将所述可信度对应的所述识别结果确定为目标结果;
    若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果;
    将所述目标结果作为光学字符识别结果发送给所述操作用户。
  17. 如权利要求16所述的非易失性的计算机可读存储介质,其特征在于,所述若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果进行预处理,以确定目标结果的步骤包括:
    若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果;
    将所述参考结果及所述目标图片分别发送给第一目标用户和第二目标用户进行校验;
    获取所述第一目标用户反馈的第一目标结果,所述第二目标用户反馈的第二目标结果;
    对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配;
    若所述第一目标结果、所述第二目标结果和所述识别结果之中至少有两个相同,则将相同的结果确定为所述目标结果。
  18. 如权利要求17所述的非易失性的计算机可读存储介质,其特征在于,所述若所述可信度小于预设阈值,则对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果的步骤包括:
    若所述可信度小于预设阈值,则获取所述识别结果对应的字符个数n,其中,n为正整数;
    根据所述字符个数n,生成a个随机数,其中,所述a和所述随机数均为正整数,且a∈[1,2]、所述随机数的取值范围为[1,n];
    选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果,其中,所述修改方案包括但不限于以下处理中的一种或者多种:将X与Y交换、将X替换Y、剔除X和插入X,X和Y为不同的随机数对应的字符。
  19. 如权利要求18所述的非易失性的计算机可读存储介质,其特征在于,所述选取所述随机数的数量a对应的修改方案,以对所述识别结果进行修改处理,得到参考结果的步骤包括:
    将所述随机数的数量a与预设数值进行比较;
    若所述随机数的数量a大于预设数值,则随机采用将X与Y交换或将X替换Y的修改方案对所述识别结果进行修改处理,得到所述参考结果;
    若所述随机数的数量a小于等于预设数值,则随机采用剔除X或插入X的修改方案对所述识别结果进行修改处理,得到所述参考结果。
  20. 如权利要求17所述的非易失性的计算机可读存储介质,其特征在于,所述对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配的步骤之后,所述计算机可读指令被一种处理器执行时,使得所述一种处理器还执行如下步骤:
    若所述第一目标结果、所述第二目标结果和所述识别结果互不相同,则将所述随机数的数量a的值更新为a的取值范围中的另一个取值,并对所述可信度对应的所述识别结果按预设修改方案进行修改处理,得到参考结果的步骤继续执行,直到对所述第一目标结果、所述第二目标结果和所述识别结果进行匹配为止;
    若匹配成功,则确定所述目标结果;
    若匹配失败,则将所述目标图片和匹配失败的结果发送给审核用户进行审核处理,并接收所述审核用户反馈的结果作为所述目标结果。
PCT/CN2019/103231 2019-06-26 2019-08-29 光学字符识别辅助方法、装置、计算机设备及存储介质 Ceased WO2020258500A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910558319.9A CN110458184B (zh) 2019-06-26 2019-06-26 光学字符识别辅助方法、装置、计算机设备及存储介质
CN201910558319.9 2019-06-26

Publications (1)

Publication Number Publication Date
WO2020258500A1 true WO2020258500A1 (zh) 2020-12-30

Family

ID=68481069

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/103231 Ceased WO2020258500A1 (zh) 2019-06-26 2019-08-29 光学字符识别辅助方法、装置、计算机设备及存储介质

Country Status (2)

Country Link
CN (1) CN110458184B (zh)
WO (1) WO2020258500A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114637474A (zh) * 2022-04-07 2022-06-17 苏州浪潮智能科技有限公司 数据块处理的方法、装置、电子设备

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111078569B (zh) * 2019-12-20 2022-01-18 腾讯科技(深圳)有限公司 光学字符识别应用的测试方法、装置及存储介质
CN112287936B (zh) * 2020-09-24 2024-10-15 深圳市智影医疗科技有限公司 光学字符识别测试方法、装置、可读存储介质及终端设备
CN115147914A (zh) * 2021-03-29 2022-10-04 中移(上海)信息通信科技有限公司 一种行为检测方法、装置、电子设备和可读存储介质
CN113177551B (zh) * 2021-05-21 2024-08-16 中国工商银行股份有限公司 报表处理方法及装置
CN114693815A (zh) * 2022-04-02 2022-07-01 北京字跳网络技术有限公司 信息处理方法、装置、终端和存储介质
CN121564310A (zh) * 2022-04-25 2026-02-24 追觅创新科技(苏州)有限公司 地面材质的识别方法及装置、存储介质、电子装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160110759A1 (en) * 2014-10-16 2016-04-21 Verizon Patent And Licensing Inc. Presenting smart billboard content based on optical pattern recognition of vehicle information
CN106845478A (zh) * 2016-12-30 2017-06-13 同观科技(深圳)有限公司 一种字符置信度的二次车牌识别方法及装置
CN109840519A (zh) * 2019-01-25 2019-06-04 青岛盈智科技有限公司 一种自适应的智能单据识别录入装置及其使用方法
CN109918984A (zh) * 2018-12-15 2019-06-21 深圳壹账通智能科技有限公司 保险单号码识别方法、装置、电子设备及存储介质

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4056745B2 (ja) * 2002-01-04 2008-03-05 株式会社リコー 文字認識方法および文字認識装置
CN109685100B (zh) * 2018-11-12 2024-05-10 平安科技(深圳)有限公司 字符识别方法、服务器及计算机可读存储介质
CN109933973B (zh) * 2019-01-24 2024-01-19 平安科技(深圳)有限公司 密码校验方法、装置、计算机设备及存储介质

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160110759A1 (en) * 2014-10-16 2016-04-21 Verizon Patent And Licensing Inc. Presenting smart billboard content based on optical pattern recognition of vehicle information
CN106845478A (zh) * 2016-12-30 2017-06-13 同观科技(深圳)有限公司 一种字符置信度的二次车牌识别方法及装置
CN109918984A (zh) * 2018-12-15 2019-06-21 深圳壹账通智能科技有限公司 保险单号码识别方法、装置、电子设备及存储介质
CN109840519A (zh) * 2019-01-25 2019-06-04 青岛盈智科技有限公司 一种自适应的智能单据识别录入装置及其使用方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114637474A (zh) * 2022-04-07 2022-06-17 苏州浪潮智能科技有限公司 数据块处理的方法、装置、电子设备

Also Published As

Publication number Publication date
CN110458184B (zh) 2023-06-30
CN110458184A (zh) 2019-11-15

Similar Documents

Publication Publication Date Title
WO2020258500A1 (zh) 光学字符识别辅助方法、装置、计算机设备及存储介质
WO2020151319A1 (zh) 密码校验方法、装置、计算机设备及存储介质
WO2021012382A1 (zh) 配置聊天机器人的方法、装置、计算机设备和存储介质
WO2021147219A1 (zh) 基于图像的文本识别方法、装置、电子设备及存储介质
CN108009928B (zh) 电子保单加签方法、装置、计算机设备及存储介质
WO2020155740A1 (zh) 信息查询方法、装置、计算机设备及存储介质
CN111680480A (zh) 基于模板的作业批改方法、装置、计算机设备及存储介质
CN110008997A (zh) 图像纹理相似度识别方法、装置及计算机可读存储介质
WO2021189911A1 (zh) 基于视频流的目标物位置检测方法、装置、设备及介质
WO2020232883A1 (zh) 脚本缺陷扫描方法、装置、计算机设备和存储介质
CN108366052A (zh) 验证短信的处理方法及系统
CN110019873B (zh) 人脸数据处理方法、装置及设备
WO2019085338A1 (zh) 电子装置、基于图像的年龄分类方法、系统及存储介质
CN112286780B (zh) 识别算法的测试方法、装置、设备及存储介质
CN114218110A (zh) 财务数据的对账测试方法、装置、计算机设备及存储介质
WO2021139443A1 (zh) 数据访问控制方法及装置、数据访问设备及系统
JP2013105373A (ja) データ取得装置、方法及びプログラム
CN113936286B (zh) 图像文本识别方法、装置、计算机设备及存储介质
WO2019179028A1 (zh) 电子装置、基于动态图片的用户验证方法及存储介质
CN114218191A (zh) 系统功能迁移方法、装置、计算机设备及存储介质
CN115630663A (zh) 一种二维码识别方法、装置及电子设备
CN107430533B (zh) 一种数据校验方法和装置
CN115359272A (zh) 理赔检测方法、装置、设备及存储介质
CN111191006B (zh) 图例间连接关系的确定方法、装置和电子系统
WO2021151274A1 (zh) 图像文档处理方法、装置、电子设备及计算机可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19935379

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19935379

Country of ref document: EP

Kind code of ref document: A1