Character recognition method and device, terminal equipment and computer readable storage medium

Abstract

The invention provides a character recognition method and device, terminal equipment and a computer readable storage medium. The method comprises the steps of obtaining a to-be-recognized character; determining whether the to-be-recognized character has a corresponding character correction record or not; and if the to-be-recognized character has the corresponding character correction record, recognizing the to-be-recognized character according to the character correction record to obtain a recognition result corresponding to the to-be-recognized character. According to the character recognition method provided by the invention, after the to-be-recognized character is obtained, whether the to-be-recognized character has the correction record or not is determined, if the to-be-recognized character has the correction record, the to-be-recognized character can be recognized by the correction record, and the accuracy of character recognition is improved under the condition that an algorithm of recognition equipment is not improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a character recognition method, device, terminal equipment and computer-readable storage medium. Background Art

[0002] With the development of text recognition technology, more and more electronic devices have the function of text recognition. For example: obtaining the handwritten text of the user and recognizing the handwritten text on the device. In the prior art, in order to improve the accuracy of text recognition, the recognition algorithm is usually improved so that the electronic device can better recognize the handwritten text of the user. Alternatively, the accuracy of the image sensor on the electronic device is improved so that the acquired characters to be recognized are clearer, which is more conducive to the electronic device to extract text features, thereby improving the accuracy of text recognition.

[0003] However, whether improving the recognition algorithm or using better hardware equipment, the accuracy of text recognition still depends on the basic requirement that the user's handwritten text should not be too different from the actual text. Otherwise, even if the algorithm is improved or the hardware equipment is improved, it is still impossible to effectively and correctly recognize handwritten text with poor fonts. Summary of the invention

[0004] The present application provides a character recognition method, which can correctly recognize the character by detecting whether there is a correction record for the character to be recognized, without improving the recognition algorithm or replacing the hardware device.

[0005] In a first aspect, the present application provides a character recognition method, the method comprising:

[0006] Get the character to be recognized;

[0007] Determine whether there is a corresponding character correction record for the character to be recognized;

[0008] If there is a corresponding character correction record for the character to be recognized, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0009] In some embodiments of the present application, determining whether there is a corresponding character correction record for the character to be recognized includes:

[0010] Determine whether there is a target character matching the character to be recognized in a preset character library;

[0011] If the target character exists in the character library, determining whether there is a corresponding target character correction record for the target character;

[0012] If the target character has a corresponding target character correction record, it is determined that the character to be recognized has a corresponding character correction record, wherein the target character correction record is the character correction record corresponding to the character to be recognized.

[0013] In some embodiments of the present application, the character correction record includes the number of character corrections, and the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized, including:

[0014] Determine a comparison result between the number of character corrections and a preset number threshold;

[0015] If the comparison result indicates that the number of character corrections is greater than the number threshold, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0016] In some embodiments of the present application, after obtaining the character to be recognized, the method further includes:

[0017] Recognize the character to be recognized to obtain an initial recognition result;

[0018] After determining whether the character to be recognized has a corresponding character correction record, the method further includes:

[0019] If there is no corresponding character correction record for the character to be recognized, a recognition result corresponding to the character to be recognized is obtained according to the initial recognition result.

[0020] In some embodiments of the present application, the character modification record includes a target character to be modified and a target replacement character corresponding to the target character to be modified. After obtaining the character to be recognized, the method further includes:

[0021] Recognize the character to be recognized to obtain an initial recognition result;

[0022] After determining whether the character to be recognized has a corresponding character correction record, the method further includes:

[0023] If there is a corresponding character correction record for the character to be recognized, the initial recognition character matching the target character to be modified in the initial recognition result is replaced with the target replacement character to obtain a recognition result corresponding to the character to be recognized.

[0024] In some embodiments of the present application, before obtaining the character to be recognized, the method further includes:

[0025] Get the reference character to be recognized;

[0026] Obtaining a character correction record corresponding to the reference character to be recognized;

[0027] The reference character to be recognized and the character correction record corresponding to the reference character to be recognized are stored.

[0028] In some embodiments of the present application, after determining whether the character to be recognized has a corresponding character correction record, the method further includes:

[0029] If there is no corresponding character correction record for the character to be recognized, the character to be recognized is recognized to obtain an initial recognition result of the character to be recognized;

[0030] Obtaining feedback information for identifying feedback on the initial recognition result;

[0031] If the feedback information indicates that the initial recognition result includes an erroneous recognition result, then obtaining a correction result of correcting the erroneous recognition result;

[0032] The initial recognition result is updated according to the modified result to obtain the recognition result corresponding to the character to be recognized.

[0033] In some embodiments of the present application, after obtaining the recognition result corresponding to the character to be recognized, the method further includes:

[0034] Storing the characters to be recognized in a preset character library;

[0035] The modified character in the initial recognition result is recorded as the target character to be modified, the modified replacement character is recorded as the target replacement character corresponding to the target character to be modified, and the number of character corrections, to obtain a character correction record of the character to be recognized.

[0036] In a second aspect, the present application further provides a character recognition device, the device comprising:

[0037] An acquisition module, used to acquire characters to be recognized;

[0038] A determination module, used to determine whether there is a corresponding character correction record for the character to be recognized;

[0039] The recognition module is used to recognize the character to be recognized according to the character correction record if there is a corresponding character correction record for the character to be recognized, so as to obtain a recognition result corresponding to the character to be recognized.

[0040] In a third aspect, the present application also provides a terminal device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in any one of the character recognition methods described.

[0041] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the steps in any one of the character recognition methods described.

[0042] The character recognition method provided by the present application determines whether there is a corresponding correction record for the character to be recognized after obtaining the character to be recognized. If there is a correction record, the character to be recognized can be recognized according to the correction record without using a recognition algorithm to recognize the character to be recognized. Therefore, the accuracy of character recognition is improved without improving the algorithm of the recognition device. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 is a scene schematic diagram of a character recognition system provided in an embodiment of the present application;

[0045] Figure 2 This is a flow chart of an embodiment of a character recognition method in an embodiment of the present application;

[0046] Figure 3 It is a functional module schematic diagram of a character recognition device in an embodiment of the present application;

[0047] Figure 4 It is a schematic diagram of the structure of the terminal device in the embodiment of the present application. DETAILED DESCRIPTION

[0048] 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 only 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 those skilled in the art without creative work are within the scope of protection of this application.

[0049] In the description of the present application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the feature. In the description of the present application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0050] In this application, the word "exemplary" is used to mean "used as an example, illustration or description". Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. At the same time, it is to be understood that in the specific implementation of this application, when user information, user data and other related data are involved, when the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data shall comply with relevant laws, regulations and standards of relevant countries and regions.

[0051] In order to enable any person skilled in the art to implement and use the present application, the following description is provided. In the following description, details are listed for the purpose of explanation. It should be understood that those of ordinary skill in the art will recognize that the present application can be implemented without using these specific details. In other examples, known structures and processes will not be elaborated in detail to avoid unnecessary details that make the description of the present application obscure. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest range of principles and features disclosed in the present application.

[0052] The present application provides a character recognition method, apparatus, device and storage medium, which are described in detail below.

[0053] See also Figure 1 , Figure 1 Schematic diagram of a scene of a character recognition system provided in an embodiment of the present application. The character recognition system may include a terminal device 100 and a storage device 200. The storage device 200 may transmit data to the terminal device 100. Figure 1 The terminal device 100 in the embodiment can obtain the characters to be recognized through the installed image sensor, and can also obtain the correction records stored in the storage device 200 to execute the character recognition method in the present application.

[0054] In the embodiment of the present application, the terminal device 100 includes but is not limited to a desktop computer, a portable computer, a network server, a PDA (Personal Digital Assistant), a tablet computer, a wireless terminal device, an embedded device, a mobile phone, etc.

[0055] In an embodiment of the present application, the storage device 200 may be a storage device of the terminal device 100 itself, or may be an external storage device. If it is an external storage device such as a mobile storage device or a cloud storage device, the terminal device 100 and the storage device 200 may communicate with each other through any communication method, including but not limited to mobile communications based on the 3rd Generation Partnership Project (3GPP), Long Term Evolution (LTE), and Worldwide Interoperability for Microwave Access (WiMAX), or computer network communications based on the TCP / IP Protocol Suite (TCP / IP) and User Datagram Protocol (UDP).

[0056] It should be noted that Figure 1 The scene diagram of the character recognition system shown is only an example. The character recognition system and scene described in the embodiment of the present application are intended to more clearly illustrate the technical solution of the embodiment of the present application, and do not constitute a limitation on the technical solution provided in the embodiment of the present application. Ordinary technicians in this field can know that with the evolution of the character recognition system and the emergence of new business scenarios, the technical solution provided in the embodiment of the present application is also applicable to similar technical problems.

[0057] like Figure 2 As shown, Figure 2 This is a flow chart of an embodiment of a character recognition method in an embodiment of the present application. The character recognition method may include the following steps 201 to 203:

[0058] 201. Obtain characters to be recognized.

[0059] The characters to be recognized in the embodiments of the present application may be words, symbols, etc. Among them, the characters to be recognized may include multiple words or symbols, etc., which are not limited in the embodiments of the present application. In addition, the method of obtaining the characters to be recognized may include: obtaining the characters written by the user on the touch screen by means of an electronic writing pen or fingers; or, after the user writes, the characters written by the user can be captured by an image sensor. Specifically, the embodiments of the present application do not limit the scenario for obtaining the characters to be recognized.

[0060] 202. Determine whether there is a corresponding character correction record for the character to be recognized.

[0061] In the embodiments of the present application, the character to be recognized can be any character written by the user on a writing device, including but not limited to Chinese characters, letters, punctuation marks, arithmetic symbols, special symbols, etc. The writing device can be a computer, a tablet, an e-ink screen, a reader display board, a touchpad, and so on. Since different users have different writing habits when using the writing device, there are some users who retain their own personalities in writing habits. When using the writing device, the handwriting recognition module on the device cannot effectively recognize the characters written by the user.

[0062] For example, when the user writes a string "你好", the handwriting recognition module recognizes "你好" as "竹好"; another example is that the user writes a character graph "菱形", but due to the non-standard writing handwriting, the handwriting recognition module recognizes this character graph that should be a rhombus as a parallelogram. The above examples all belong to the situation where the handwriting recognition module has recognition errors.

[0063] In the embodiments of the present application, in order to solve this recognition error problem and to perform more accurate content recognition on the characters written by the user, in a specific implementation, the character to be recognized written by the user can be uploaded to the cloud (cloud server). By querying in the cloud whether there is a corresponding character correction record for the character to be recognized by the user, the content recognition accuracy of the character to be recognized can be further improved based on this character correction record. In this implementation, the handwriting recognition module can be connected to the cloud. Whenever the user writes once, the writing device can upload the character to be recognized written by the user to the cloud in the form of an image file, and search in the cloud whether there is a corresponding character correction record for this character to be recognized. The advantage of this method is that compared with storing the character correction record on the local writing device, storing the character correction record in the cloud can relieve the local storage space.

[0064] In an example, the process of determining whether there is a corresponding character correction record for the character to be recognized by the cloud can specifically include: performing image recognition on the character image of the character to be recognized by the cloud to determine whether there is a matching character in the cloud's image library. Furthermore, it is only necessary to determine whether there is a corresponding character correction record for this matching character. It should be noted that when the user writes a character each time, the local writing device can perform initial recognition or pre-recognition on the user's character. If the user corrects the initial recognition result or pre-recognition result on the local writing device, the local writing device can generate a corresponding character correction record for the user's correction behavior. At the same time, the local writing device can upload the character image written by the user and the corresponding character correction record to the cloud, so that when the user writes the same character again later, the cloud can retrieve the same character image written by the user before and the character correction record corresponding to the previous character image based on the current character.

[0065] In another implementation, the recorded historical character correction library can be queried in the local cache of the user end, and the historical character correction library can be queried whether there is a corresponding character correction record for the character to be recognized written by the user, so as to further improve the content recognition accuracy of the character to be recognized based on the character correction record. For example: Similar to the above-mentioned cloud storage character correction record, in an embodiment of the present application, after the user writes a target character, the local writing device can also perform initial recognition or pre-recognition on the target character to generate an initial recognition result or pre-recognition result. If the user revise the initial recognition result or pre-recognition result, the local writing device can generate a character correction record corresponding to the target character, and store the character image of the target character and the corresponding character correction record locally. If the subsequent user writes the same target character again, based on the character image of the current target character, it can retrieve the matching character image in the local writing device, and then obtain the corresponding character correction record based on the matching character image.

[0066] 203. If there is a corresponding character correction record for the character to be recognized, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0067] According to the above embodiments, in one embodiment, when there is a character correction record that matches the character to be recognized, the local writing device does not need to call the handwriting recognition module built into the writing device or the recognition module in the cloud to perform character recognition on the character to be recognized in advance. Instead, the user's modification result in the character correction record can be directly used as the output result of the character to be recognized.

[0068] In another embodiment, if the local writing device or the cloud does not determine the character correction record corresponding to the character to be recognized, the local writing device or the cloud can call the recognition module to perform initial character recognition or pre-recognition based on the character to be recognized currently written by the user to obtain an initial recognition result. If the user's correction action on the initial recognition result is not detected, the initial recognition result can be used as the recognition result corresponding to the character to be recognized.

[0069] In another embodiment, if the local writing device or the cloud has not determined the character correction record corresponding to the character to be recognized, the local writing device or the cloud can call the recognition module to perform initial character recognition or pre-recognition based on the character to be recognized currently written by the user to obtain an initial recognition result. If the user's correction action on the initial recognition result is detected, the corresponding character correction record can be generated by the cloud or the local writing device, that is, the character image of the character to be recognized, the initial recognition result corresponding to the character to be recognized, and the corresponding character correction record can be stored locally or in the cloud.

[0070] In one example, after obtaining the characters to be recognized written by the user, the characters to be recognized can be recognized in advance. For example, the characters to be recognized can be recognized in an image file format to determine whether there is a character image matching the character image of the character to be recognized in the cloud or local writing device; if there is a matching character image, the character image of the character to be recognized may not be stored again; if there is no matching character image, the character image of the character to be recognized may be stored.

[0071] The character recognition method provided in the embodiment of the present application determines whether there is a corresponding correction record for the character to be recognized after obtaining the character to be recognized. If there is a correction record, the character to be recognized can be recognized based on the correction record, thereby improving the accuracy of character recognition without improving the algorithm of the recognition device.

[0072] In order to better implement the embodiment of the present application, in one embodiment of the present application, determining whether there is a corresponding character correction record for the character to be recognized includes:

[0073] Determine whether there is a target character matching the character to be recognized in the preset character library; if the target character exists in the character library, determine whether there is a corresponding target character correction record for the target character; if the target character exists a corresponding target character correction record, determine whether there is a corresponding character correction record for the character to be recognized, wherein the target character correction record is a character correction record corresponding to the character to be recognized. In this embodiment, the preset character library can be preset in a local writing device, or preset in the cloud, or can be synchronously set in the local writing device and the cloud, and this embodiment does not specifically limit this. It should be noted that the local writing device can store the character image of the character to be recognized written by the user each time, and store it in the local preset character library or the cloud preset character library according to the preset storage rules, so that the cloud or local writing device can preferentially search for the character image in the preset character library when the new character to be recognized written by the user is obtained next time, which not only narrows the search range, but also improves the search efficiency. In this way, it can solve the problem that in the process of retrieving character images, if the stored character images are stored irregularly, the search range of the cloud or local writing device is wider and the search efficiency is lower.

[0074] In this embodiment, similarly, when the cloud or local writing device generates a corresponding character correction record based on the user's correction of the initial recognition result or pre-recognition result of the character to be recognized, the cloud or writing device can also store the relevant character correction record, for example, store the character correction record in a preset correction record library, and associate the generated character correction record with the character image of the corresponding character to be recognized, for example, it can be marked with the same ID identifier, when the character image matching the character to be recognized is retrieved in the preset character library, the ID of the matched character image can be obtained, and then the character correction record with the same ID can be obtained through the ID, and then the character correction record corresponding to the character to be recognized can be obtained. Among them, the initial recognition result or the pre-recognition result can be the result obtained by the recognition module built into the local end of the writing device to recognize the character to be recognized, or it can be the result obtained by the recognition module in the cloud associated with the writing device to recognize the character to be recognized, and this embodiment does not make specific limitations on this.

[0075] Specifically, in an example of this embodiment, assume that the user has currently written a character to be recognized, "你好" (Hello in Chinese). The cloud or the local writing device can search for this character image in the character library. If the user has written "你好" before, the previously written "你好" character will be pre-stored in the character library in the form of an image file. At this time, based on the "你好" written by the user currently, the cloud or the writing device can find the "你好" image that matches the current "你好" in the character library, that is, it can be determined that there is a target character in the character library that matches the character to be recognized, "你好". If the user has not written "你好" before, the cloud or the local writing device will not find a matching "你好" image. It should be noted that when the cloud or the local writing device searches for character images, it can obtain the image of the character to be recognized written by the user currently and compare the similarity with various character images pre-stored in the character library. Specifically, it can include: extracting the image features between the two comparison images; calculating the similarity between the image features, including methods such as cosine similarity calculation and Euclidean distance calculation; and finally determining whether there is a matching target character in the character library through the similarity of the images.

[0076] As can be obtained from the above embodiment, the cloud or the local writing device can generate a character correction record based on the user's modification operation on the initial recognition result or pre-recognition result of the character to be recognized. Therefore, after generating the character correction record, the character correction record can be associated and saved with the character image of the character to be recognized written by the user. For example, if the initial recognition result or pre-recognition result corresponding to the previously written "你好" by the user is "竹好" (incorrectly recognized as "Bamboo Hello"), and this "竹好" is corrected by the user to "你好", that is, there is the above-mentioned correction operation for the character to be recognized, "你好", a character correction record corresponding to this character to be recognized can be generated based on this correction operation and the relevant character content; at this time, the cloud or the local writing device can add an ID identifier to the character image of the previously written "你好", and add the same ID identifier to the character correction record corresponding to this character. Then, when it is detected that the user writes "你好" again, through the index of the ID identifier marked on the character image, the corresponding character correction record can be finally obtained from the character image of the target character "你好" matched by image recognition in the character library.

[0077] If a character in the character correction record in the cloud or local writing device has only been modified once by the user, there is a possibility that the user may make a mistake in the modification. Therefore, the cloud or local writing device can record the number of times the user has corrected the same character and generate a character correction record including the number of character corrections. If the corresponding character correction record determined by the cloud or local writing device includes multiple modifications and the number of corrections exceeds the number threshold, the recognition result can be output based on the correction result in the character correction record.

[0078] In an embodiment of the present application, the character correction record may include the number of character corrections, and the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized, including:

[0079] Determine a comparison result between the number of character corrections and a preset number threshold; if the comparison result indicates that the number of character corrections is greater than the number threshold, recognize the character to be recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0080] Specifically, in one example, the user currently writes the character "Hello" to be recognized. If the cloud or local writing device has only one character correction record corresponding to "Hello". If the user's written "Hello" is obtained again, if there is a correction record for "Hello" at this time, it is necessary to compare the number 1 with the preset number threshold; if the comparison result indicates that the number of character corrections is less than or equal to the number threshold, the recognition module may not use the character correction record to recognize the character to be recognized, but can perform character recognition on the user's written "Hello" through its own configured recognition algorithm. It should also be noted that at this time, if the local writing device outputs the recognition result according to its own recognition algorithm, and the user still corrects the written "Hello" and feeds it back to the local writing device, at this time, the number of corrections in the character correction record corresponding to "Hello" is increased by 1, and the number of corrections is changed to 2. Afterwards, if the user writes "Hello" again, it is necessary to compare the two corrections of "Hello" with the number threshold; if the comparison result indicates that the number of character corrections is greater than the number threshold, the correction result of "Hello" in the character correction record can be directly used as the recognition result corresponding to the character to be recognized this time. In the embodiment of the present application, the specific value of the number threshold is not specifically limited.

[0081] In one embodiment of the present application, after obtaining the character to be recognized, the method may further include: recognizing the character to be recognized and obtaining an initial recognition result; in this embodiment, after obtaining the initial recognition result, determining whether there is a corresponding character correction record for the character to be recognized; if there is no corresponding character correction record for the character to be recognized, obtaining a recognition result corresponding to the character to be recognized based on the initial recognition result.

[0082] In another embodiment of the present application, after obtaining the character to be recognized, the method may further include: recognizing the character to be recognized to obtain an initial recognition result; in this embodiment, after obtaining the initial recognition result, determining whether there is a corresponding character correction record for the character to be recognized, wherein the character correction record includes a target character to be modified and a target replacement character corresponding to the target character to be modified; if there is a corresponding character correction record for the character to be recognized, then the initial recognition character that matches the target character to be modified in the character correction record in the initial recognition result is replaced with the target replacement character corresponding to the target character to be modified, and the recognition result corresponding to the character to be recognized is obtained. This embodiment provides a solution for secondary recognition of the acquired character to be recognized based on the corresponding correction record stored in the cloud or local writing device. In actual applications, there are many cases where the number of characters written by the user is large. For example: a sentence, a paragraph of text, an article, etc. Therefore, when the number of characters written by the user is large, it is necessary to correct the specific characters that need to be modified. Therefore, in this embodiment, the characters written by the user can be initially recognized or pre-recognized by the recognition module of the local writing device to obtain an initial recognition result. Assume that the characters written by the user are a paragraph of text, which may include "hello" or "diamond". However, the initial recognition result may only recognize "you" in a paragraph of text as "bamboo", or recognize "diamond" as "parallelogram". Since the character correction record includes the target character to be modified and the target replacement character corresponding to the target character to be modified, that is, the character correction record includes a correction record for correcting "bamboo" to "you", or the character correction record includes a correction record for correcting "parallelogram" to "diamond". Therefore, for the initial recognition result, "bamboo" or "parallelogram" in the character correction record, that is, the target character to be modified, is matched with each initial recognition character in the initial recognition result, and it is determined that the "bamboo" or "parallelogram" in the initial recognition result is the initial recognition character to be recognized for the second time. Afterwards, the "bamboo" in the initial recognition result is replaced with the initial recognition character "bamboo" in the initial recognition result according to the target replacement character "you" corresponding to the target character to be modified "bamboo" in the corresponding character correction record. Similarly, according to the target replacement character "diamond" corresponding to the target character to be modified "parallelogram" in the character correction record, the target replacement character "diamond" replaces the initial recognition character "parallelogram" in the initial recognition result. In this way, the entire recognition process is finally completed.

[0083] In an embodiment of the present application, before obtaining the character to be recognized, the character recognition method further includes:

[0084] Acquire a reference character to be recognized; acquire a character correction record corresponding to the reference character to be recognized; and store the reference character to be recognized and the character correction record corresponding to the reference character to be recognized.

[0085] In the embodiment of the present application, the purpose of obtaining reference characters to be recognized is to supplement samples for the preset character library. For example: each time the user writes a character, the character written by the user can be automatically regarded as a reference character to be recognized and stored. Among them, if the characters to be recognized written by the user include multiple characters, the storage strategy at this time can be to separate the multiple characters and store each character separately. Alternatively, when the user uses the recognition module for the first time, the writing device can prompt the user to establish a character library. For example: the terminal device can display the corresponding characters on the corresponding touch display device, allowing the user to write according to the displayed characters, and record them. In this way, the user's written characters corresponding to each character can be obtained. At this time, the user's written characters are stored to obtain reference characters to be recognized. Of course, there are many ways to obtain reference characters to be recognized, and the specific embodiments of the present application are not limited.

[0086] In addition, in the embodiment of the present application, obtaining the character correction record corresponding to the reference character to be recognized can be determined through the cloud in the manner of the above embodiment. For example, for each character in the reference character to be recognized, the recognition module can recognize it by itself, and then generate the corresponding character correction record, and then hand it over to the cloud to store the character correction record. After that, the reference character to be recognized and the corresponding character correction record can be stored in the cloud.

[0087] In one embodiment of the present application, after determining whether there is a corresponding character correction record for the character to be recognized, the character recognition method further includes:

[0088] If there is no corresponding character correction record for the character to be recognized, the character to be recognized is recognized to obtain an initial recognition result of the character to be recognized; feedback information for recognition feedback on the initial recognition result is obtained; if the feedback information indicates that the initial recognition result contains an erroneous recognition result, a correction result for correcting the erroneous recognition result is obtained; the initial recognition result is updated according to the corrected correction result to obtain a recognition result corresponding to the character to be recognized.

[0089] In the above embodiment, a solution is provided for obtaining a character correction record corresponding to a character to be recognized through a cloud or a local writing device. However, in practical applications, it is necessary to provide an interactive method with the user to collect the user's modification operation on the character. Therefore, in order to solve this problem, when the local writing device or the cloud detects that there is no corresponding character correction record for the character to be recognized written by the user, it can be handed over to the user for manual modification. For example: after the user writes "hello", the recognition module recognizes it as "bamboo good", that is, the initial recognition result is obtained. At this time, the user can find that there is an error in the initial recognition result, and the user can modify "bamboo" based on the error to obtain the recognition result of "hello". Therefore, every time the terminal device performs initial recognition, the user can be prompted on the corresponding display device so that the user can check the initial recognition result. For example: prompt the user through a pop-up window, the pop-up window may include words such as correct and wrong, and when the user clicks error or correct, feedback information for identifying feedback on the initial recognition result can be obtained. If the user clicks the error, the feedback information obtained indicates that the initial recognition result contains an error recognition result. At this time, the user can be prompted to make manual modifications and obtain a recognition result of "Hello", that is, a correction result of the erroneous recognition result is obtained. It should be noted that in the embodiment of the present application, the user's manual modification method can also record the number of modifications. When the number of manual modifications made by the user to a certain character exceeds a certain threshold, the next time the user writes the character, the recognition module will directly obtain the recognition result based on the correction record.

[0090] In one embodiment of the present application, the character recognition method further includes:

[0091] The characters to be recognized are stored in a preset character library; the modified characters in the initial recognition results are recorded as the target characters to be modified, the modified replacement characters are recorded as the target replacement characters corresponding to the target characters to be modified, and the number of character corrections of the characters to be recognized is recorded to obtain the character correction record corresponding to the characters to be recognized.

[0092] In an embodiment of the present application, when there is no corresponding character correction record for the character to be recognized written by the user, the cloud or local writing device can store the character to be recognized written by the user. The specific storage method is the same as that of any of the above embodiments, which will not be repeated here. At the same time, the user is prompted to make manual modifications, and the prompting method is the same as that of the above embodiment, which will not be repeated here. Similarly, a unified character correction record can be generated for storage of relevant information such as the number of character corrections, the target character to be modified, and the target replacement character.

[0093] In order to better implement the character recognition method in the embodiment of the present application, in addition to the character recognition method, the embodiment of the present application also provides a character recognition device, such as Figure 3 As shown, the device 300 includes:

[0094] An acquisition module 301 is used to acquire characters to be recognized;

[0095] A determination module 302 is used to determine whether there is a corresponding character correction record for the character to be recognized;

[0096] The recognition module 303 is used to recognize the character to be recognized according to the character correction record if there is a corresponding character correction record for the character to be recognized, so as to obtain a recognition result corresponding to the character to be recognized.

[0097] The character recognition device provided by the present application can, after acquiring the character to be recognized through the acquisition module 301, determine whether there is a correction record for the character to be recognized through the determination module 302. If there is a correction record, the character to be recognized can be recognized according to the correction record through the recognition module 303, thereby improving the accuracy of character recognition without improving the algorithm of the recognition device.

[0098] In some embodiments of the present application, the determination module 302 is specifically used to:

[0099] Determine whether there is a target character matching the character to be recognized in a preset character library;

[0100] If the target character exists in the character library, determining whether there is a corresponding target character correction record for the target character;

[0101] If the target character has a corresponding target character correction record, it is determined that the character to be recognized has a corresponding character correction record, wherein the target character correction record is the character correction record corresponding to the character to be recognized.

[0102] In some embodiments of the present application, the character correction record includes the number of character corrections, and the recognition module 303 is specifically used to:

[0103] Determine a comparison result of the number of character corrections and a preset number threshold;

[0104] If the comparison result indicates that the number of character corrections is greater than the number threshold, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0105] In some embodiments of the present application, the device further includes a first initial identification module, and the first initial identification module is specifically configured to:

[0106] Recognize the characters to be recognized and obtain an initial recognition result;

[0107] If there is no corresponding character correction record for the character to be recognized, a recognition result corresponding to the character to be recognized is obtained according to the initial recognition result.

[0108] In some embodiments of the present application, the character correction record includes a target character to be modified and a target replacement character corresponding to the target character to be modified, and the device further includes a first initial recognition module, wherein the first initial recognition module is further configured to:

[0109] Recognize the characters to be recognized and obtain an initial recognition result;

[0110] If there is a corresponding character correction record for the character to be recognized, the initial recognition character that matches the target character to be modified in the initial recognition result is replaced with the target replacement character to obtain a recognition result corresponding to the character to be recognized.

[0111] In some embodiments of the present application, the device further includes a storage module, and the storage module is specifically used to:

[0112] Get the reference character to be recognized;

[0113] Obtain a character correction record corresponding to the reference character to be recognized;

[0114] The reference character to be recognized and the character correction record corresponding to the reference character to be recognized are stored.

[0115] In some embodiments of the present application, the device further includes a second initial identification module, and the second initial identification module is specifically further used to:

[0116] If there is no corresponding character correction record for the character to be recognized, the character to be recognized is recognized to obtain an initial recognition result of the character to be recognized;

[0117] Obtaining feedback information for the initial recognition result;

[0118] If the feedback information indicates that the initial recognition result includes an erroneous recognition result, obtaining a correction result for correcting the erroneous recognition result;

[0119] The initial recognition result is updated according to the corrected correction result to obtain the recognition result corresponding to the character to be recognized.

[0120] In some embodiments of the present application, the device further includes a recording module, and the recording module is specifically used to:

[0121] Storing the characters to be recognized in a preset character library;

[0122] The modified character in the initial recognition result is recorded as the target character to be modified, the modified replacement character is recorded as the target replacement character corresponding to the target character to be modified, and the number of character corrections, to obtain a character correction record of the character to be recognized.

[0123] The present application also provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps in any one of the character recognition methods in the present application. The terminal device integrates any one of the character recognition methods provided in the present application. Figure 4 As shown, it shows a schematic diagram of the structure of the terminal device involved in the embodiment of the present application, specifically:

[0124] The terminal device may include one or more processing core processors 401, one or more computer-readable storage media memories 402, a power supply 403, an input unit 404 and other components. Those skilled in the art will appreciate that Figure 4 The terminal device structure shown in the figure does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0125] The processor 401 is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device. By running or executing software programs and / or modules stored in the memory 402, and calling data stored in the memory 402, the processor 401 executes various functions of the terminal device and processes data, thereby monitoring the terminal device as a whole. Optionally, the processor 401 may include one or more processing cores; the processor 401 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. Preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, the user interface and the application program, etc., and the modem processor mainly processes wireless communication. It is understandable that the above-mentioned modem processor may not be integrated into the processor 401.

[0126] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 402 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0127] The terminal device also includes a power supply 403 for supplying power to each component. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, so as to manage charging, discharging, power consumption management and other functions through the power management system. The power supply 403 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0128] The terminal device may further include an input unit 404, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0129] Although not shown, the terminal device may further include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 401 in the terminal device will load the executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 will run the application programs stored in the memory 402, thereby realizing various functions, such as:

[0130] Get the character to be recognized;

[0131] Determine whether there is a corresponding character correction record for the character to be recognized;

[0132] If there is a corresponding character correction record for the character to be recognized, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0133] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0134] To this end, the embodiment of the present application provides a computer-readable storage medium, which may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc. A computer program is stored thereon, and the computer program is loaded by a processor to execute the steps in any character recognition method provided in the embodiment of the present application. For example, the computer program loaded by the processor can execute the following steps:

[0135] Get the character to be recognized;

[0136] Determine whether there is a corresponding character correction record for the character to be recognized;

[0137] If there is a corresponding character correction record for the character to be recognized, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

[0138] In the above embodiments, the description of each embodiment has its own emphasis. For the parts that are not described in detail in a certain embodiment, please refer to the detailed description of other embodiments above, and will not be repeated here.

[0139] In specific implementation, the above units or structures can be implemented as independent entities, or can be arbitrarily combined to be implemented as the same or several entities. The specific implementation of the above units or structures can refer to the previous method embodiments, which will not be repeated here.

[0140] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0141] The above is a detailed introduction to a character recognition method and device provided in an embodiment of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A character recognition method, characterized in that: The method comprises: Get the character to be recognized; Determine whether there is a corresponding character correction record for the character to be recognized; If there is a corresponding character correction record for the character to be recognized, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

2. The character recognition method according to claim 1, characterized in that: The step of determining whether there is a corresponding character correction record for the character to be recognized comprises: Determine whether there is a target character matching the character to be recognized in a preset character library; If the target character exists in the character library, determining whether there is a corresponding target character correction record for the target character; If the target character has a corresponding target character correction record, it is determined that the character to be recognized has a corresponding character correction record, wherein the target character correction record is the character correction record corresponding to the character to be recognized.

3. The character recognition method according to claim 1, characterized in that: The character correction record includes the number of character corrections, and the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized, including: Determine a comparison result between the number of character corrections and a preset number threshold; If the comparison result indicates that the number of character corrections is greater than the number threshold, the character to be recognized is recognized according to the character correction record to obtain a recognition result corresponding to the character to be recognized.

4. The character recognition method according to claim 1, characterized in that: After obtaining the character to be recognized, the method further includes: Recognize the character to be recognized to obtain an initial recognition result; After determining whether the character to be recognized has a corresponding character correction record, the method further includes: If there is no corresponding character correction record for the character to be recognized, a recognition result corresponding to the character to be recognized is obtained according to the initial recognition result.

5. The character recognition method according to claim 1, characterized in that: The character modification record includes a target character to be modified and a target replacement character corresponding to the target character to be modified. After obtaining the character to be recognized, the method further includes: Recognize the character to be recognized to obtain an initial recognition result; After determining whether the character to be recognized has a corresponding character correction record, the method further includes: If there is a corresponding character correction record for the character to be recognized, the initial recognition character matching the target character to be modified in the initial recognition result is replaced with the target replacement character to obtain a recognition result corresponding to the character to be recognized.

6. The character recognition method according to any one of claims 1 to 5, characterized in that: Before obtaining the character to be recognized, the method further includes: Get the reference character to be recognized; Obtaining a character correction record corresponding to the reference character to be recognized; The reference character to be recognized and the character correction record corresponding to the reference character to be recognized are stored.

7. The character recognition method according to any one of claims 1 to 3, characterized in that: After determining whether the character to be recognized has a corresponding character correction record, the method further includes: If there is no corresponding character correction record for the character to be recognized, the character to be recognized is recognized to obtain an initial recognition result of the character to be recognized; Obtaining feedback information for identifying feedback on the initial recognition result; If the feedback information indicates that the initial recognition result includes an erroneous recognition result, obtaining a correction result for correcting the erroneous recognition result; The initial recognition result is updated according to the modified result to obtain the recognition result corresponding to the character to be recognized.

8. The character recognition method according to claim 7, characterized in that: The method further comprises: Storing the characters to be recognized in a preset character library; The modified character in the initial recognition result is recorded as the target character to be modified, the modified replacement character is recorded as the target replacement character corresponding to the target character to be modified, and the number of character corrections, to obtain a character correction record of the character to be recognized.

9. A character recognition device, characterized in that: The device comprises: An acquisition module, used to acquire characters to be recognized; A determination module, used to determine whether there is a corresponding character correction record for the character to be recognized; The recognition module is used to recognize the character to be recognized according to the character correction record if there is a corresponding character correction record for the character to be recognized, so as to obtain a recognition result corresponding to the character to be recognized.

10. A terminal device, characterized in that: The terminal device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps in the character recognition method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in the character recognition method according to any one of claims 1 to 8.

Images