Verification method and device of handwritten transcribed text, computer equipment and storage medium

Abstract

The invention relates to the technical field of image processing and financial science and technology, and particularly discloses a verification method and device for a handwritten transcribed text, computer equipment and a storage medium. According to the method, the target template corresponding to the to-be-verified transcribing text is determined through the table area of the to-be-verified transcribing text, and then the verification result is obtained based on the character sequence of the target template and the character sequence of the to-be-verified transcribing text, so that intelligent verification is realized, manual verification is not needed, manual false detection is avoided, and the verification efficiency and accuracy are improved. And secondly, through target template matching, different verification character sequences can be matched for the to-be-verified transcribing texts in different scenes, so that accurate verification is realized, the situation that a universal character recognition scheme wrongly recognizes the to-be-verified transcribing texts in different scenes is avoided, and the verification accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of image processing and financial technology, and in particular to a method, device, computer equipment and storage medium for verifying handwritten transcripts. Background Art

[0002] With the development of the economy and the improvement of people's living standards, people's awareness of buying insurance has been continuously improved. Insurance, as a protection product, brings risk protection to customers and reduces losses. In the process of purchasing insurance, customers are required to read the relevant terms and confirm the relevant terms. Sometimes it involves the need for customers to hand-write a text such as "I have read the insurance terms". In the underwriting process, a very important part is to review whether the customer has correctly copied the corresponding sentences as required to ensure that the customer is fully aware of the insurance regulations. Currently, in insurance scenarios, manual verification is usually used for such transcription scenarios. There are cases of false detection and low efficiency of manual verification, resulting in low verification efficiency and accuracy of handwritten transcripts. Therefore, how to improve the verification efficiency and accuracy of handwritten transcripts has become an urgent problem to be solved. Summary of the invention

[0003] The present application provides a method, apparatus, computer device and storage medium for verifying handwritten transcripts, so as to improve the verification efficiency and accuracy of handwritten transcripts.

[0004] In a first aspect, the present application provides a method for verifying a handwritten transcript, the method comprising:

[0005] Obtaining a handwritten transcription video of a customer, recognizing the handwritten transcription video, and obtaining a transcription text to be verified and a table area corresponding to the transcription text to be verified;

[0006] Matching is performed in a preset transcription template library based on the table area to obtain a target template corresponding to the transcription text to be verified, and a character sequence of the target template is obtained as a first character sequence;

[0007] Recognize the to-be-verified transcribed text based on the target template to obtain a character sequence of the to-be-verified transcribed text as a second character sequence;

[0008] Based on the first character sequence and the second character sequence, a verification result of the transcribed text to be verified is obtained.

[0009] In a second aspect, the present application further provides a device for verifying a handwritten transcript, the device comprising:

[0010] A handwriting transcription video recognition module is used to obtain a customer's handwriting transcription video, recognize the handwriting transcription video, and obtain a transcription text to be verified and a table area corresponding to the transcription text to be verified;

[0011] A first text sequence obtaining module, configured to perform matching in a preset transcription template library based on the table area, obtain a target template corresponding to the transcription text to be verified, and obtain a text sequence of the target template as a first text sequence;

[0012] A second character sequence obtaining module, configured to recognize the to-be-verified transcribed text based on the target template, and obtain a character sequence of the to-be-verified transcribed text as a second character sequence;

[0013] The verification result obtaining module is used to obtain the verification result of the transcribed text to be verified based on the first character sequence and the second character sequence.

[0014] In a third aspect, the present application also provides a computer device, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement the handwritten transcript text verification method as described above when executing the computer program.

[0015] In a fourth aspect, the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the handwritten transcript text verification method as described above.

[0016] The present application discloses a verification method, device, computer equipment and storage medium for handwritten transcript text, which obtains a customer's handwritten transcript video, identifies the handwritten transcript video, obtains the transcript text to be verified and the table area corresponding to the transcript text to be verified; matches the table area in a preset transcript template library to obtain a target template corresponding to the transcript text to be verified, and obtains the text sequence of the target template as a first text sequence; identifies the transcript text to be verified based on the target template, obtains the text sequence of the transcript text to be verified as a second text sequence; obtains the verification result of the transcript text to be verified based on the first text sequence and the second text sequence. The present application determines the target template corresponding to the transcript text to be verified through the table area of ​​the transcript text to be verified, and then obtains the verification result based on the text sequence of the target template and the text sequence of the transcript text to be verified, thereby realizing intelligent verification, eliminating the need for manual verification, avoiding manual misdetection, and improving the efficiency and accuracy of verification. Secondly, through target template matching, different verification text sequences can be matched to the transcribed texts to be verified in different scenarios, achieving accurate verification, avoiding the situation where general text recognition solutions incorrectly recognize the transcribed texts to be verified in different scenarios, and improving the accuracy of verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 is a schematic flow chart of a first embodiment of a method for verifying a handwritten transcript provided in an embodiment of the present application;

[0019] Figure 2 is a schematic flow chart of a second embodiment of a handwritten transcript text verification method provided by an embodiment of the present application;

[0020] Figure 3 is a schematic flow chart of a third embodiment of a method for verifying a handwritten transcript provided in an embodiment of the present application;

[0021] Figure 4 A schematic block diagram of a handwritten transcript text verification device provided in an embodiment of the present application;

[0022] Figure 5 A schematic block diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0025] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0026] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0027] The embodiments of the present application provide a verification method, device, computer equipment and storage medium for handwritten transcripts. The verification method for handwritten transcripts can be applied to a server, and the target template corresponding to the transcript to be verified is determined by the table area of ​​the transcript to be verified, and then the verification result is obtained based on the text sequence of the target template and the text sequence of the transcript to be verified, thereby realizing intelligent verification without manual verification, avoiding manual misdetection, and improving the efficiency and accuracy of verification. The server can be an independent server or a server cluster.

[0028] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0029] See also Figure 1 , Figure 1 This is a schematic flow chart of a handwritten transcript verification method provided by an embodiment of the present application. The handwritten transcript verification method can be applied to a server, and is used to determine a target template corresponding to the transcript to be verified through a table area of ​​the transcript to be verified, and then obtain a verification result based on the text sequence of the target template and the text sequence of the transcript to be verified, thereby realizing intelligent verification without manual verification, avoiding manual misdetection, and improving the efficiency and accuracy of verification.

[0030] like Figure 1 As shown, the handwritten transcript text verification method specifically includes steps S101 to S104.

[0031] S101, obtaining a handwritten transcription video of a customer, identifying the handwritten transcription video, and obtaining a transcription text to be verified and a table area corresponding to the transcription text to be verified;

[0032] In one embodiment, the transcribed text is a text that the customer needs to transcribe when conducting a financial transaction in order to confirm whether the customer has read the relevant terms. The text obtained by the customer after handwriting transcription is the transcribed text, such as "I have read the insurance terms."

[0033] In one embodiment, the handwriting transcription video is a video recorded by the customer when performing handwriting transcription, and may be a surveillance video or a video recorded on-site by a business person.

[0034] In one embodiment, the handwriting transcription video can be obtained from a recorded video library uploaded by a business person, or from a stored surveillance video.

[0035] In one embodiment, an image frame containing handwritten text in a handwriting transcription video is identified, and image processing is performed on the image frame to identify a table area and a transcription text to be verified.

[0036] Furthermore, the identifying the handwritten transcription video to obtain the transcription text to be verified and the table area corresponding to the transcription text to be verified includes: performing text recognition on the handwritten transcription video to obtain an image containing text; performing table line extraction on the image to determine the table area; performing text recognition on the table area to obtain the transcription text to be verified.

[0037] In one embodiment, for the handwritten transcription video, OCR (Optical Character Recognition) technology may be used to perform text recognition, determine the video frame containing the text, and obtain the image of the video frame.

[0038] In one embodiment, after obtaining an image containing text, a table line extraction algorithm is used to determine the table area. The table line extraction algorithm can detect lines in the table based on image processing techniques, such as edge detection, line detection, etc., to determine the position and size of the table area.

[0039] In one embodiment, OCR technology may be used to perform text recognition on the table area to obtain the transcribed text to be verified.

[0040] S102, matching the table area in a preset transcription template library to obtain a target template corresponding to the to-be-verified transcription text, and obtaining a character sequence of the target template as a first character sequence;

[0041] In one embodiment, image processing is performed on the table area, including denoising, binarization, edge detection, etc., to clearly identify the boundaries and internal structure of the table. Key features such as the number of rows and columns, cell size, text position, etc. of the table are extracted from the processed image.

[0042] In one embodiment, the extracted table area features are matched with the template features in the template library, and the template that best matches the table area is determined as the target template. Specifically, pattern recognition technology can be used to identify the similarity between the table area and the template. Machine learning models, such as support vector machines or neural networks, can also be applied to predict which template the table area best matches. The distance between the table area features and the template features (such as the Euclidean distance) can also be calculated to select the template with the smallest distance.

[0043] In one embodiment, a text sequence is extracted from the target template as the first text sequence, such as V1V2V3...Vn.

[0044] Furthermore, before matching the table area in a preset transcription template library to obtain a target template corresponding to the to-be-verified transcription text, and obtaining a text sequence of the target template as the first text sequence, the method further includes: obtaining a historical transcription table of at least one scene; performing feature recognition on the historical transcription table of each scene respectively to determine a transcribed text area and a printed text area; generating a transcription template corresponding to each scene based on the transcribed text area and the printed text area to obtain the transcription template library.

[0045] In one embodiment, a historical transcription form of at least one scene is collected, which may be a scanned paper document or an existing electronic format document. The collected form may be pre-processed, including denoising, adjusting resolution, converting to a unified format, etc.

[0046] In one embodiment, the transcribed text area and the printed text area in the transcribed form are separated by using image segmentation technology. Features are extracted from the two areas respectively, and the features may include the font, size, color, position, etc. of the text. Based on the extracted features, the transcribed text area, i.e., the customer's handwritten text area, is identified and marked. The printed text area, i.e., the text portion printed on the form, is identified and marked.

[0047] In one embodiment, based on the identified text area, a transcription template is designed. The template reflects the layout and format of the transcribed text and printed text. Specifically, an automated tool can be used according to the layout and format of the text area and printed text, and the automated tool can be a script or software.

[0048] In one embodiment, the generated transcription templates are stored in a centralized database or file system for easy management and retrieval. The templates are classified according to the scene to ensure that the templates for each scene can be correctly identified and used.

[0049] S103, recognizing the to-be-verified transcribed text based on the target template, and obtaining a character sequence of the to-be-verified transcribed text as a second character sequence;

[0050] In one embodiment, the transcript to be verified is segmented according to the structure of the target template to obtain handwritten characters. Features are extracted from each segmented handwritten character, and a search is performed using a preset feature search tree and the extracted character features to obtain a second character sequence. Specifically, the extracted handwritten character features are matched with the nodes in the feature search tree to find the most similar preset characters. Based on the search results, a second character sequence is generated, which corresponds to the handwritten part of the transcript to be verified, such as W1W2W3...Wn.

[0051] In one embodiment, according to the layout of the target template, the most similar preset characters found are organized into a second character sequence in a correct order.

[0052] S104: Obtain a verification result of the transcribed text to be verified based on the first character sequence and the second character sequence.

[0053] In one embodiment, the first text sequence is the expected correct text, which is derived from the target template.

[0054] In one embodiment, the second text sequence is text recognized in the transcribed text to be verified, and may contain recognition errors or be not completely consistent with the first text sequence.

[0055] In one embodiment, whether the transcript to be verified has passed verification is determined by comparing whether the first character sequence and the second character sequence are consistent. If the first character sequence and the second character sequence are consistent, it indicates that the transcript to be verified has passed verification. If the first character sequence and the second character sequence are inconsistent, the similarity between the transcript to be verified and the printed text is calculated, and whether the transcript to be verified has passed verification is determined based on the similarity.

[0056] The above-mentioned embodiment provides a verification method, device, computer equipment and storage medium for handwritten transcript text, obtains a customer's handwritten transcript video, recognizes the handwritten transcript video, obtains the transcript text to be verified and the table area corresponding to the transcript text to be verified; matches in a preset transcript template library based on the table area, obtains a target template corresponding to the transcript text to be verified, and obtains the text sequence of the target template as a first text sequence; recognizes the transcript text to be verified based on the target template, obtains the text sequence of the transcript text to be verified as a second text sequence; obtains the verification result of the transcript text to be verified based on the first text sequence and the second text sequence. The present application determines the target template corresponding to the transcript text to be verified through the table area of ​​the transcript text to be verified, and then obtains the verification result based on the text sequence of the target template and the text sequence of the transcript text to be verified, thereby realizing intelligent verification, eliminating the need for manual verification, avoiding manual misdetection, and improving the efficiency and accuracy of verification. Secondly, through target template matching, different verification text sequences can be matched to the transcribed texts to be verified in different scenarios, achieving accurate verification, avoiding the situation where general text recognition solutions incorrectly recognize the transcribed texts to be verified in different scenarios, and improving the accuracy of verification.

[0057] See also Figure 2 , Figure 2 This is a schematic flow chart of a handwritten transcript verification method provided by an embodiment of the present application. The handwritten transcript verification method can be applied to a server to extract the features of handwritten text from the transcript to be verified, and use a preset feature search tree for matching to obtain a second text sequence, thereby improving the efficiency and accuracy of transcript verification.

[0058] like Figure 2 As shown, the handwritten transcript text verification method specifically includes steps S201 to S203.

[0059] S201, segmenting the to-be-verified transcribed text based on the target template to obtain at least one handwritten character;

[0060] S202, extracting features of the handwritten characters respectively to obtain text features of the handwritten characters;

[0061] S203: Perform feature search based on a preset feature search tree and the text features to obtain the second text sequence.

[0062] In one embodiment, the transcribed text to be verified is segmented according to the structure of the target template to identify the handwritten text portion. Specifically, the transcribed text is aligned with the target template to determine the layout and structure of the text. According to the layout information of the template, the transcribed text to be verified is segmented into separate handwritten texts.

[0063] In one embodiment, features are extracted from each segmented handwritten character, specifically, shape features of the characters, such as strokes, contours, and geometric structures are extracted. Texture features of the characters, such as line thickness, direction, and spacing, are analyzed. Statistical features of the characters, such as the number of strokes, character size, and tilt, are calculated.

[0064] In one embodiment, a preset feature search tree and extracted text features are used to search for a second text sequence. Specifically, the extracted handwritten text features are matched with nodes in the feature search tree to find the most similar preset text. Based on the search results, a second text sequence is generated, which corresponds to the handwritten portion of the transcript to be verified, such as W1W2W3...Wn.

[0065] Furthermore, before the feature search based on the preset feature search tree and the text features is performed to obtain the second text sequence, it also includes: obtaining at least one printed text to be transcribed; performing scene classification on the printed text to be transcribed to obtain a text set of at least one scene; and performing feature extraction on the text set of each scene respectively to obtain a feature search tree for each of the scenes.

[0066] In one embodiment, the printed text to be transcribed is text that requires the customer to transcribe and confirm in handwriting, such as “I have read the insurance terms”.

[0067] In one embodiment, at least one printed text to be transcribed is obtained to ensure the diversity of text sources to cover different transcription requirements and scenarios. Texts in different formats are converted into a unified format for easy processing.

[0068] In one embodiment, a machine learning or deep learning method may be used to classify scenes of digital images. For example, a convolutional neural network may be used to classify images into different financial transaction scenes, such as "buying insurance" or "buying financial management".

[0069] In one embodiment, computer vision technology may be used to extract features from the text set of each scene, for example, a SIFT (Scale Invariant Feature Transform) algorithm or a SURF (Speeded-Up Robust Features) algorithm may be used to extract key points and descriptors of text in an image.

[0070] In one embodiment, a feature search tree is obtained for each scene. Specifically, a feature search tree can be constructed for the feature vector of each scene. The feature search tree is a data structure based on a tree structure, which can be used to quickly search for similar feature vectors. The feature search tree can be constructed using algorithms such as kd-tree or FLANN (Fast Library for Approximate Nearest Ne i ghbors).

[0071] The above embodiment can effectively extract the features of handwritten text from the transcribed text to be verified, and use a preset feature search tree for matching to obtain a second text sequence, thereby improving the efficiency and accuracy of transcribed text verification.

[0072] See also Figure 3 , Figure 3 This is a schematic flow chart of a handwritten transcript verification method provided by an embodiment of the present application. The handwritten transcript verification method can be applied to a server to effectively verify the transcript to be verified by comparing whether the text sequence is consistent and calculating the similarity between the printed text and the text to be verified, thereby ensuring the accuracy and consistency of the data and improving the efficiency and accuracy of the transcript verification.

[0073] like Figure 3 As shown, the handwritten transcript text verification method specifically includes steps S301 to S304.

[0074] S301, comparing the first character sequence with the second character sequence;

[0075] S302: When the first text sequence is inconsistent with the second text sequence, identifying the table area to obtain a printed text corresponding to the transcribed text to be verified;

[0076] S303, calculating the similarity between the transcribed text to be verified and the printed text to obtain the similarity between the transcribed text to be verified and the printed text;

[0077] S304: When the similarity is greater than or equal to a preset similarity threshold, taking verification passed as the verification result.

[0078] Furthermore, after comparing the first character sequence with the second character sequence, the method further includes: when the first character sequence is consistent with the second character sequence, taking verification passed as a verification result of the transcript to be verified.

[0079] In one embodiment, the first character sequence is compared with the second character sequence. Specifically, a string comparison or difference marking method can be used to compare the first character sequence corresponding to the target template with the second character sequence extracted from the handwritten transcript text.

[0080] In one embodiment, when the first text sequence is inconsistent with the second text sequence, a table area is identified to obtain printed text, and the printed text is identified from the table area using OCR technology.

[0081] In one embodiment, the similarity between the transcribed text to be verified and the printed text is evaluated. Specifically, a text similarity algorithm (such as cosine similarity, Jaccard similarity) can be used to calculate the similarity of the two texts. The similarity is quantified to quantify the similarity into a numerical value for easy judgment and comparison.

[0082] In one embodiment, the obtained similarity is compared with a preset similarity threshold, and if the similarity is greater than or equal to the preset threshold, the transcribed text is considered to have passed the verification. If the similarity is lower than the preset threshold, the transcribed text is considered to have failed the verification.

[0083] The above embodiment can effectively verify the transcribed text to be verified by comparing whether the text sequence is consistent and calculating the similarity between the printed text and the text to be verified, thereby ensuring the accuracy and consistency of the data and improving the efficiency and accuracy of the transcribed text verification.

[0084] See also Figure 4 , Figure 4 The embodiment of the present application provides a schematic block diagram of a handwritten transcript text verification device, which is used to execute the handwritten transcript text verification method described above. The handwritten transcript text verification device can be configured on a server.

[0085] like Figure 4 As shown, the handwritten transcript verification device 400 includes:

[0086] The handwriting transcription video recognition module 401 is used to obtain the customer's handwriting transcription video, recognize the handwriting transcription video, and obtain the transcription text to be verified and the table area corresponding to the transcription text to be verified;

[0087] A first character sequence obtaining module 402 is used to perform matching in a preset transcription template library based on the table area to obtain a target template corresponding to the transcription text to be verified, and obtain a character sequence of the target template as a first character sequence;

[0088] A second character sequence obtaining module 403 is used to recognize the to-be-verified transcribed text based on the target template, and obtain a character sequence of the to-be-verified transcribed text as a second character sequence;

[0089] The verification result obtaining module 404 is used to obtain the verification result of the transcribed text to be verified based on the first character sequence and the second character sequence.

[0090] Furthermore, the handwriting transcription video recognition module 401 includes:

[0091] A text image obtaining unit, used to perform text recognition on the handwritten transcription video to obtain an image containing text;

[0092] A table area determination unit, used for extracting table lines from the image to determine the table area;

[0093] The unit for obtaining the transcript to be verified is used for performing text recognition on the table area to obtain the transcript to be verified.

[0094] Furthermore, the second character sequence obtaining module 403 includes:

[0095] A handwritten character obtaining unit, configured to segment the to-be-verified transcribed text based on the target template to obtain at least one handwritten character;

[0096] A character feature obtaining unit, used for extracting features of the handwritten characters respectively to obtain character features of the handwritten characters;

[0097] The text sequence obtaining unit is used to perform feature search based on a preset feature search tree and the text features to obtain the second text sequence.

[0098] Furthermore, the second character sequence obtaining module 403 further includes:

[0099] A printed text acquisition unit, used to acquire at least one printed text to be transcribed;

[0100] A character set obtaining unit, used for classifying the printed text to be transcribed into scenes, and obtaining a character set of at least one scene;

[0101] The feature search tree obtaining unit is used to extract features from the text set of each scene respectively to obtain the feature search tree of each scene.

[0102] Furthermore, the verification result obtaining module 404 includes:

[0103] a character sequence comparison unit, configured to compare the first character sequence with the second character sequence;

[0104] a printed text obtaining unit, configured to identify the table area and obtain a printed text corresponding to the transcribed text to be verified when the first text sequence is inconsistent with the second text sequence;

[0105] a similarity calculation unit, used for calculating the similarity between the transcribed text to be verified and the printed text, so as to obtain the similarity between the transcribed text to be verified and the printed text;

[0106] The verification result obtaining unit is used to take verification pass as the verification result when the similarity is greater than or equal to a preset similarity threshold.

[0107] Furthermore, the verification result obtaining module 404 further includes:

[0108] The verification result obtaining unit is used to take verification passed as the verification result of the transcript to be verified when the first character sequence is consistent with the second character sequence.

[0109] Furthermore, the handwritten transcript verification device 400 further includes a transcript template library acquisition module, and the transcript template library acquisition module includes:

[0110] A history transcription table acquisition unit, used to acquire a history transcription table of at least one scene;

[0111] A text area determination unit, used to perform feature recognition on the historical transcription table of each scene, and determine a transcription text area and a printed text area;

[0112] The transcription template library obtaining unit is used to generate a transcription template corresponding to each scene based on the transcription text area and the printed text area to obtain the transcription template library.

[0113] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0114] The above-mentioned device can be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.

[0115] See also Figure 5 , Figure 5 1 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0116] See also Figure 5The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0117] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any one of the verification methods for handwritten transcripts.

[0118] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0119] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any verification method of the handwritten transcript text.

[0120] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0121] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0122] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0123] Obtaining a handwritten transcription video of a customer, recognizing the handwritten transcription video, and obtaining a transcription text to be verified and a table area corresponding to the transcription text to be verified;

[0124] Matching is performed in a preset transcription template library based on the table area to obtain a target template corresponding to the transcription text to be verified, and a character sequence of the target template is obtained as a first character sequence;

[0125] Recognize the to-be-verified transcribed text based on the target template to obtain a character sequence of the to-be-verified transcribed text as a second character sequence;

[0126] Based on the first character sequence and the second character sequence, a verification result of the transcribed text to be verified is obtained.

[0127] In one embodiment, when the processor recognizes the handwritten transcription video and obtains the transcription text to be verified and the table area corresponding to the transcription text to be verified, it is used to implement:

[0128] Performing text recognition on the handwritten transcription video to obtain an image containing text;

[0129] Extracting table lines from the image to determine the table area;

[0130] Perform text recognition on the table area to obtain the transcribed text to be verified.

[0131] In one embodiment, when the processor recognizes the to-be-verified transcribed text based on the target template and obtains a character sequence of the to-be-verified transcribed text as the second character sequence, it is configured to implement:

[0132] Based on the target template, segment the transcribed text to be verified to obtain at least one handwritten character;

[0133] Extracting features of the handwritten characters respectively to obtain text features of the handwritten characters;

[0134] Based on a preset feature search tree and the text features, a feature search is performed to obtain the second text sequence.

[0135] In one embodiment, before the processor performs feature search based on a preset feature search tree and the text features to obtain the second text sequence, the processor is further configured to implement:

[0136] obtaining at least one printed text to be transcribed;

[0137] Classifying the printed text to be transcribed by scene to obtain a text set of at least one scene;

[0138] Feature extraction is performed on the text set of each scene respectively to obtain a feature search tree for each scene.

[0139] In one embodiment, when obtaining the verification result of the transcribed text to be verified based on the first character sequence and the second character sequence, the processor is configured to implement:

[0140] comparing the first character sequence with the second character sequence;

[0141] When the first text sequence is inconsistent with the second text sequence, identifying the table area to obtain a printed text corresponding to the transcribed text to be verified;

[0142] Calculating the similarity between the transcribed text to be verified and the printed text to obtain the similarity between the transcribed text to be verified and the printed text;

[0143] When the similarity is greater than or equal to a preset similarity threshold, verification pass is taken as the verification result.

[0144] In one embodiment, after comparing the first character sequence with the second character sequence, the processor is further configured to:

[0145] When the first character sequence is consistent with the second character sequence, verification passed is taken as the verification result of the transcript to be verified.

[0146] In one embodiment, before the processor performs matching in a preset transcription template library based on the table area to obtain a target template corresponding to the to-be-verified transcription text, and obtains a text sequence of the target template as the first text sequence, it is further configured to implement:

[0147] Obtain a historical transcription form of at least one scenario;

[0148] Performing feature recognition on the historical transcription table of each scene respectively to determine the transcription text area and the printed text area;

[0149] Based on the transcribed text area and the printed text area, a transcribed template corresponding to each scene is generated to obtain the transcribed template library.

[0150] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores a computer program. The computer program includes program instructions. The processor executes the program instructions to implement any handwritten transcript text verification method provided in the embodiment of the present application.

[0151] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (Smart Medi a Card, SMC), a secure digital (Secure Digital, SD) card, a flash memory card (Flash Card), etc., equipped on the computer device.

[0152] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for verifying handwritten transcripts, characterized in that: include: Obtaining a handwritten transcription video of a customer, recognizing the handwritten transcription video, and obtaining a transcription text to be verified and a table area corresponding to the transcription text to be verified; Matching is performed in a preset transcription template library based on the table area to obtain a target template corresponding to the transcription text to be verified, and a character sequence of the target template is obtained as a first character sequence; Recognize the to-be-verified transcribed text based on the target template to obtain a character sequence of the to-be-verified transcribed text as a second character sequence; Based on the first character sequence and the second character sequence, a verification result of the transcribed text to be verified is obtained.

2. The method for verifying handwritten transcripts according to claim 1, characterized in that: The step of recognizing the handwritten transcription video to obtain the transcription text to be verified and the table area corresponding to the transcription text to be verified includes: Performing text recognition on the handwritten transcription video to obtain an image containing text; Extracting table lines from the image to determine the table area; Perform text recognition on the table area to obtain the transcribed text to be verified.

3. The method for verifying a handwritten transcript according to claim 1, characterized in that: The step of recognizing the to-be-verified transcribed text based on the target template to obtain a character sequence of the to-be-verified transcribed text as a second character sequence includes: Based on the target template, segment the transcribed text to be verified to obtain at least one handwritten character; Extracting features of the handwritten characters respectively to obtain text features of the handwritten characters; Based on a preset feature search tree and the text features, a feature search is performed to obtain the second text sequence.

4. The method for verifying a handwritten transcript according to claim 3, characterized in that: Before performing feature search based on the preset feature search tree and the text features to obtain the second text sequence, the method further includes: obtaining at least one printed text to be transcribed; Classifying the printed text to be transcribed by scene to obtain a text set of at least one scene; Feature extraction is performed on the text set of each scene respectively to obtain a feature search tree for each scene.

5. The method for verifying handwritten transcripts according to claim 1, characterized in that: The obtaining, based on the first character sequence and the second character sequence, a verification result of the transcribed text to be verified includes: comparing the first character sequence with the second character sequence; When the first text sequence is inconsistent with the second text sequence, identifying the table area to obtain a printed text corresponding to the transcribed text to be verified; Calculating the similarity between the transcribed text to be verified and the printed text to obtain the similarity between the transcribed text to be verified and the printed text; When the similarity is greater than or equal to a preset similarity threshold, verification pass is taken as the verification result.

6. The method for verifying a handwritten transcript according to claim 5, characterized in that: After comparing the first character sequence with the second character sequence, the method further includes: When the first character sequence is consistent with the second character sequence, verification passed is taken as the verification result of the transcript to be verified.

7. The method for verifying a handwritten transcript according to any one of claims 1 to 6, characterized in that: Before matching the table area in a preset transcription template library to obtain a target template corresponding to the to-be-verified transcription text, and obtaining a character sequence of the target template as the first character sequence, the method further includes: Obtain a historical transcription form of at least one scenario; Performing feature recognition on the historical transcription table of each scene respectively to determine the transcription text area and the printed text area; Based on the transcribed text area and the printed text area, a transcribed template corresponding to each scene is generated to obtain the transcribed template library.

8. A device for checking handwritten transcripts, characterized in that: include: A handwriting transcription video recognition module is used to obtain a customer's handwriting transcription video, recognize the handwriting transcription video, and obtain a transcription text to be verified and a table area corresponding to the transcription text to be verified; A first text sequence obtaining module, configured to perform matching in a preset transcription template library based on the table area, obtain a target template corresponding to the transcription text to be verified, and obtain a text sequence of the target template as a first text sequence; A second character sequence obtaining module, configured to recognize the to-be-verified transcribed text based on the target template, and obtain a character sequence of the to-be-verified transcribed text as a second character sequence; The verification result obtaining module is used to obtain the verification result of the transcribed text to be verified based on the first character sequence and the second character sequence.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the handwritten transcript text verification method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the handwritten transcript verification method according to any one of claims 1 to 7.

Images