An intelligent recognition method based on multilingual dictionary pen software

Abstract

The present invention relates to a field of intelligent recognition technology, and in particular to an intelligent recognition method based on a multilingual dictionary pen software, including collecting image information of text; preprocessing the image information; screening languages ​​whose text similarity meets the standard; determining the language to which the text feature belongs based on the number of matching languages; and after determining the language, interpreting, displaying and translating the text. The text recognition module in the present invention performs text recognition on the preprocessed image information, compares the image information with a plurality of languages ​​pre-stored in the text recognition module for similarity, screens languages ​​whose similarity meets the standard, and the central control module determines the number of languages ​​that meet the standard, thereby improving the accuracy of text recognition, effectively improving the text recognition rate of the dictionary pen, and reducing the influence of image information collection deviation on text recognition results.

Description

Technical Field

[0001] The present invention relates to the field of intelligent recognition technology, and in particular to an intelligent recognition method based on multilingual dictionary pen software. Background Art

[0002] A dictionary pen is a portable electronic device that combines the functions of a dictionary and a notebook, and can help users quickly look up information such as the meaning, spelling, pronunciation, and example sentences of words; intelligent recognition refers to the process of using artificial intelligence technology to identify, understand, and process input information;

[0003] Chinese patent application publication number: CN113642584A discloses a text recognition method, device, equipment, storage medium and intelligent dictionary pen, which relates to the field of artificial intelligence technology, specifically the field of computer vision and deep learning technology, and can be applied to scenarios such as optical character recognition (OCR). The specific implementation scheme is: obtaining multiple image sequences obtained by continuous scanning of documents; performing image splicing based on multiple image sequences to obtain corresponding multiple continuous splicing image frames, wherein there are overlapping areas between two continuous splicing image frames; performing text recognition based on multiple continuous splicing image frames to obtain corresponding multiple recognition results; and performing deduplication processing on multiple recognition results based on the overlapping areas between two continuous splicing image frames in multiple continuous splicing image frames to obtain text recognition results for the document.

[0004] It can be seen that the text recognition method, device, equipment, storage medium and smart dictionary pen have the following problems: if the image information scanned by the dictionary pen is not clear or has a certain tilt angle, the dictionary pen cannot accurately recognize the text, and the text recognition results will be biased or even output erroneous results. Summary of the invention

[0005] To this end, the present invention provides an intelligent recognition method based on a multi-language dictionary pen software to overcome the problem in the prior art that text recognition cannot be accurately performed due to deformation of image information.

[0006] To achieve the above object, the present invention provides an intelligent recognition method based on a multilingual dictionary pen software, comprising the following steps:

[0007] Step S1, the user terminal sends a collection instruction to the image collection module, and when the dictionary pen moves over the text, the image collection module collects image information of the text;

[0008] Step S2, the image processing module pre-processes the image information collected by the image collection module;

[0009] Step S3, a text recognition module performs text recognition on the pre-processed image information, and selects languages ​​whose text similarity meets the standard;

[0010] Step S4, after the text recognition module completes the determination of the similarity between the text feature and each language, the central control module counts the number of matching languages, and determines the language to which the text feature belongs based on the number of matching languages, or determines that it is necessary to add text features to be recognized to determine the language to which the text feature belongs for a second time, or determines the reason why the language to which the text feature belongs cannot be determined;

[0011] Step S5, the central control module re-determines the number of matching languages ​​whose similarity with the text feature meets the standard based on the newly acquired text feature, and completes the determination of the language to which the text feature belongs based on the number of matching languages, or determines the reason why the language to which the text feature belongs cannot be determined;

[0012] Step S6, after the central control module determines the language, the output module interprets, displays and translates the text scanned by the dictionary pen.

[0013] Furthermore, the text recognition module obtains text features in the preprocessed image information, and compares the obtained text features with various languages ​​pre-stored in the text recognition module in turn to obtain similarities between the text features and various languages ​​respectively. The text recognition module records the languages ​​with similarities to the text features higher than a preset standard as matching languages. After the text recognition module completes the determination of the similarities between the text features and each language, the central control module counts the number of matching languages:

[0014] If there is only a single matching language, the central control module completes the determination of the language to which the text feature belongs, and controls the output module to translate and interpret the scanned text in the language;

[0015] If there are multiple matching languages, the central control module determines that it is necessary to increase the text features to be recognized to determine the language to which the text features belong for a second time, and the central control module sends a command to the text recognition module to continue text recognition and determines the number of words that need to be recognized according to the number of languages ​​that meet the standard;

[0016] If there is no matching language, the central control module determines that the language to which the text feature belongs has not been determined, and the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the recognized text feature.

[0017] Furthermore, when determining that there are multiple matching languages, the central control module determines a word number determination method for words that need to be further recognized based on the number of languages ​​that meet the criteria, wherein:

[0018] The first word number determination method is that the central control module determines to use the preset word scan number set by it as the number of words to be continuously recognized; the first word number determination method satisfies that the number of matching languages ​​is less than or equal to the first preset matching language number set in the central control module;

[0019] The second word number determination method is that the central control module determines to use the first preset correction coefficient to correct the preset word scan number and uses the corrected preset word scan number as the number of words to be continuously recognized; the second word number determination method satisfies that the number of matching languages ​​is greater than the first preset matching language number and is less than or equal to the second preset matching language number set in the central control module;

[0020] The third word number determination method is that the central control module determines to use the second preset correction coefficient to correct the preset word scan number and uses the corrected preset word scan number as the number of words that need to continue to be recognized; the third word number determination method satisfies that the number of matching languages ​​is greater than the second preset matching language number.

[0021] Further, after the central control module determines the number of words to be further recognized, it controls the image processing module to obtain image information and controls the text recognition module to recognize and obtain a corresponding number of text features from the image information. The central control module re-determines the number of matching languages ​​whose similarity with the text features meets the standard based on the newly obtained text features, wherein:

[0022] If there is only a single matching language, the central control module completes the determination of the language to which the text feature belongs, and controls the output module to translate and interpret the scanned text in the language;

[0023] If there are multiple matching languages, the central control module records the matching language with the highest similarity between the newly acquired text features and the text features as the target language, and controls the output module to translate and interpret the scanned text in this language;

[0024] If there is no matching language, the central control module determines that the language to which the text feature belongs has not been determined, and the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the recognized text feature.

[0025] Furthermore, when there is no matching language that meets the standard of similarity with the text feature, the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the text feature obtained by the text recognition module:

[0026] If the length of the text feature is less than or equal to the first preset length set in the central control module, the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, and the central control module adjusts the preprocessing parameters of the image processing module for the image information to a corresponding value and controls the image processing module to re-preprocess the collected image information using the adjusted preprocessing parameters;

[0027] If the length of the text feature is greater than the first preset length set in the central control module and is less than or equal to the second preset length, the central control module determines that the reason why the text feature cannot be determined to belong to the language is that there is a problem with the angle and speed of the dictionary pen scanning the text, and the central control module controls the image processing module to perform scaling processing on the image information and controls the text recognition module to re-recognize the text on the image information after scaling processing;

[0028] If the length of the text feature is greater than the second preset length set in the central control module, the central control module determines that the reason for being unable to determine the language to which the text feature belongs is that the language does not exist in the language library or the corresponding font does not exist, and the central control module controls the output module to issue a notification that the language library needs to be updated.

[0029] Furthermore, when the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, the central control module is provided with several adjustment methods for the preprocessing parameters of the image processing module based on the average similarity between the image information and the language, wherein:

[0030] If the average similarity between the image information and the language is greater than the preset average similarity standard, the central control module selects the first preprocessing parameter adjustment method to adjust the initial preprocessing parameter; the first preprocessing parameter adjustment method refers to the central control module adjusting the grayscale of the preprocessing parameter, increasing the grayscale of the preprocessing parameter by 10%, and setting the preset average similarity standard to 90%;

[0031] If the average similarity between the image information and the language is less than or equal to the preset average similarity standard, the central control module selects the second preprocessing parameter adjustment method to adjust the initial preprocessing parameters; the second preprocessing parameter adjustment method refers to the central control module adjusting the resolution of the preprocessing parameters to increase the resolution of the preprocessing parameters by 200dpi.

[0032] Furthermore, when the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there are problems with the angle and speed at which the dictionary pen scans the text, the central control module performs scaling processing on the image information:

[0033] If the speed of the dictionary pen scanning the text is greater than the speed standard preset in the central control module, the central control module performs a horizontal stretching process on the image information;

[0034] If the speed at which the dictionary pen scans text is lower than the speed standard preset in the central control module, the central control module performs horizontal compression processing on the image information.

[0035] Further, after the central control module completes the scaling processing of the image information, if the central control module still determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the pre-processing of the image information by the image processing module, the central control module sets several adjustment methods for the scaling ratio of the scaling processing based on the character ratio of the image information, wherein:

[0036] If the character ratio of the image information is greater than the character ratio standard preset in the central control module, the central control module selects a first telescopic ratio adjustment method to adjust the preset telescopic ratio, wherein the first telescopic ratio adjustment method refers to increasing the telescopic ratio by 10%;

[0037] If the character ratio of the image information is less than the character ratio standard preset in the central control module, the central control module selects a second telescopic ratio adjustment method to adjust the preset telescopic ratio, and the second telescopic ratio adjustment method refers to reducing the telescopic ratio by 10%.

[0038] Further, the central control module performs scaling processing on the image information. If the adjustment range of the scaling ratio exceeds the scaling ratio adjustment range preset in the central control module, the central control module is provided with an adjustment method for the scaling angle based on the scaling ratio, wherein:

[0039] If the difference between the adjustment amplitude of the telescopic ratio and the preset adjustment amplitude is less than or equal to the adjustment amplitude difference standard preset in the central control module, the central control module selects a first telescopic angle adjustment method to adjust the telescopic angle, and the first telescopic angle adjustment method refers to increasing the angle between the initial telescopic direction and the horizontal direction of the image by 10°;

[0040] If the difference between the adjustment amplitude of the telescopic ratio and the preset adjustment amplitude is greater than the adjustment amplitude difference standard preset in the central control module, the central control module selects the second telescopic angle adjustment method to adjust the telescopic angle, and the second telescopic angle adjustment method refers to increasing the angle between the initial telescopic direction and the horizontal direction of the image by 20°.

[0041] Furthermore, when the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that the language does not exist in the language library or the corresponding font does not exist, the central control module uses all the languages ​​in the language library to perform a secondary semantic guessing judgment on the image information.

[0042] If the central control module determines for the second time that the guessed language information meets the guessing standard preset in the central control module, the central control module controls the output module to output the guessed language and specially marks it;

[0043] If the central control module determines for the second time that the guessed language information does not meet the guessing standard preset in the central control module, the central control module controls the output module to issue a notification that a rescan is required.

[0044] Compared with the prior art, the beneficial effect of the present invention lies in that the text recognition module in the present invention performs text recognition on the preprocessed image information, compares the image information with multiple languages ​​pre-stored in the text recognition module for similarity, and screens languages ​​whose similarity meets the standards, and the central control module determines the number of languages ​​that meet the standards, thereby improving the accuracy of text recognition, effectively improving the text recognition rate of the dictionary pen, and reducing the impact of image information collection deviation on text recognition results.

[0045] Furthermore, when the text recognition module identifies that there are two or more matching languages, the central control module determines the number of words that need to be further recognized based on the number of languages ​​that meet the standards, thereby improving the accuracy of text recognition, effectively improving the text recognition rate of the dictionary pen, and reducing the impact of image information acquisition deviation on text recognition results.

[0046] Furthermore, after the central control module completes the determination of the number of words that need to be further recognized, it sends a continuous recognition instruction to the text recognition module. After the text recognition module completes the recognition, a secondary similarity comparison is performed between the image information and multiple languages ​​pre-stored in the text recognition module, thereby further improving the accuracy of text recognition and reducing the impact of image information acquisition deviation on text recognition results.

[0047] Furthermore, after the text recognition module determines that there is no matching language for the image information, the central control module determines the reason why there is no matching language based on the length of the recognized characters, thereby realizing the function of analyzing the reason why the dictionary pen cannot recognize text, and improving the accuracy and efficiency of text recognition.

[0048] Furthermore, the central control module determines that the character length meets the standard and that there is a problem with the preprocessing of the image information by the image processing module. The central control module adjusts the preprocessing parameters of the image processing module based on the average similarity between the image information and the language, thereby further improving the accuracy of text recognition, effectively improving the text recognition rate of the dictionary pen, and reducing the impact of image information acquisition deviation on text recognition results.

[0049] Furthermore, the central control module preliminarily determines that the character length does not meet the standard, determines that the character is deformed or there are problems with the angle and speed of the dictionary pen scanning the text, and the central control module adjusts the telescopic ratio of the image information. If the adjustment range of the telescopic ratio exceeds the telescopic ratio adjustment range preset in the central control module, the central control module adjusts the telescopic angle based on the telescopic ratio, thereby further improving the accuracy of text recognition, effectively improving the text recognition rate of the dictionary pen, and reducing the impact of image information acquisition deviation on text recognition results. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of the steps of the intelligent recognition method based on the multi-language dictionary pen software of the present invention;

[0051] Figure 2 This is a flow chart of the determination of text features by the central control module according to the present invention based on the number of matching languages;

[0052] Figure 3 It is a flow chart of the central control module of the present invention determining the number of words that need to be further recognized based on the number of languages ​​that meet the standards;

[0053] Figure 4 This is a flow chart of secondary determination of text features by the central control module according to the present invention based on the number of newly acquired matching languages. DETAILED DESCRIPTION

[0054] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] The above and other technical features and advantages of the present invention are described in more detail below in conjunction with the accompanying drawings.

[0056] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0057] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0058] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0059] See also Figure 1 As shown, it is a flow chart of the steps of the intelligent recognition method based on the multi-language dictionary pen software of the present invention; including:

[0060] Step S1, the user terminal sends a collection instruction to the image collection module, and when the dictionary pen moves over the text, the image collection module collects image information of the text;

[0061] Step S2, the image processing module pre-processes the image information collected by the image collection module;

[0062] Step S3, a text recognition module performs text recognition on the pre-processed image information, and selects languages ​​whose text similarity meets the standard;

[0063] Step S4, after the text recognition module completes the determination of the similarity between the text feature and each language, the central control module counts the number of matching languages, and determines the language to which the text feature belongs based on the number of matching languages, or determines that it is necessary to add text features to be recognized to determine the language to which the text feature belongs for a second time, or determines the reason why the language to which the text feature belongs cannot be determined;

[0064] Step S5, the central control module re-determines the number of matching languages ​​whose similarity with the text feature meets the standard based on the newly acquired text feature, and completes the determination of the language to which the text feature belongs based on the number of matching languages, or determines the reason why the language to which the text feature belongs cannot be determined;

[0065] Step S6, after the central control module determines the language, the output module interprets, displays and translates the text scanned by the dictionary pen.

[0066] Please continue reading Figure 2As shown, it is a flow chart of the determination of text features by the central control module of the present invention based on the number of matching languages; the text recognition module obtains the text features in the preprocessed image information, and compares the obtained text features with various languages ​​pre-stored in the text recognition module in turn to obtain the similarity between the text features and various languages ​​respectively, and the text recognition module records the languages ​​with a similarity to the text features higher than a preset standard as matching languages, and the central control module counts the number of matching languages ​​after the text recognition module completes the determination of the similarity between the text features and each language:

[0067] If there is only a single matching language, the central control module completes the determination of the language to which the text feature belongs, and controls the output module to translate and interpret the scanned text in the language;

[0068] If there are multiple matching languages, the central control module determines that it is necessary to increase the text features to be recognized to determine the language to which the text features belong for a second time, and the central control module sends a command to the text recognition module to continue text recognition and determines the number of words that need to be recognized according to the number of languages ​​that meet the standard;

[0069] If there is no matching language, the central control module determines that the language to which the text feature belongs has not been determined, and the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the recognized text feature.

[0070] Please continue reading Figure 3 As shown, it is a flow chart of the central control module of the present invention determining the number of words that need to be further recognized based on the number of languages ​​that meet the standards;

[0071] The first word number determination method is that the central control module determines to use the preset word scan number set as the number of words to be continuously recognized; the first word number determination method satisfies that the number of matching languages ​​is less than or equal to the first preset matching language number set in the central control module; the first matching language number N1 is set to 4;

[0072] The second word number determination method is that the central control module determines to use the first preset correction coefficient to correct the preset word scan number and uses the corrected preset word scan number as the number of words to be continuously recognized; the second word number determination method satisfies that the number of matching languages ​​is greater than the first preset matching language number and is less than or equal to the second preset matching language number set in the central control module; the preset word scan number is set to 20, the first preset correction coefficient is 1.25, and the second preset matching language number N2=6;

[0073] The third word number determination method is that the central control module determines to use the second preset correction coefficient to correct the preset word scan number and uses the corrected preset word scan number as the number of words to be continuously recognized; the third word number determination method satisfies that the number of matching languages ​​is greater than the second preset matching language number; and the second preset correction coefficient is set to 1.5;

[0074] The central control module uses the correction coefficient to multiply the preset word scan number to obtain a new preset word scan number to complete the correction of the initial preset word scan number.

[0075] Please continue reading Figure 4 As shown, the central control module of the present invention performs a secondary determination flow chart on text features based on the number of newly acquired matching languages; after completing the determination of the number of words to be further recognized, the central control module controls the image processing module to obtain image information and controls the text recognition module to recognize and obtain a corresponding number of text features from the image information, and the central control module re-determines the number of matching languages ​​whose similarity with the text features meets the standard based on the newly acquired text features, wherein:

[0076] If there is only a single matching language, the central control module completes the determination of the language to which the text feature belongs, and controls the output module to translate and interpret the scanned text in the language;

[0077] If there are multiple matching languages, the central control module records the matching language with the highest similarity between the newly acquired text features and the text features as the target language, and controls the output module to translate and interpret the scanned text in this language;

[0078] If there is no matching language, the central control module determines that the language to which the text feature belongs has not been determined, and the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the recognized text feature.

[0079] Please continue reading Figures 1 to 4 As shown, when there is no matching language that meets the standard of similarity with the text feature, the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the text feature obtained by the text recognition module:

[0080] If the length of the text feature is less than or equal to the first preset length set in the central control module, the central control module determines that the reason why the language of the text feature cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, and the central control module adjusts the preprocessing parameters of the image processing module for the image information to the corresponding value and controls the image processing module to re-preprocess the collected image information using the adjusted preprocessing parameters; the first preset length is set to 10 characters;

[0081] If the length of the text feature is greater than the first preset length set in the central control module and is less than or equal to the second preset length, the central control module determines that the reason why the text feature cannot be determined to belong to the language is that there is a problem with the angle and speed of the dictionary pen scanning the text, and the central control module controls the image processing module to perform scaling processing on the image information and controls the text recognition module to re-recognize the text on the image information after scaling processing; the second preset length is set to 20 characters;

[0082] If the length of the text feature is greater than the second preset length set in the central control module, the central control module determines that the reason for being unable to determine the language to which the text feature belongs is that the language does not exist in the language library or the corresponding font does not exist, and the central control module controls the output module to issue a notification that the language library needs to be updated.

[0083] Please continue reading Figures 1 to 4 As shown, when the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, the central control module provides several adjustment methods for the preprocessing parameters of the image processing module based on the average similarity between the image information and the language.

[0084] If the average similarity between the image information and the language is greater than the preset average similarity standard, the central control module selects the first preprocessing parameter adjustment method to adjust the initial preprocessing parameter; the first preprocessing parameter adjustment method refers to the central control module adjusting the grayscale of the preprocessing parameter, increasing the grayscale of the preprocessing parameter by 10%, and setting the preset average similarity standard to 90%;

[0085] If the average similarity between the image information and the language is less than or equal to the preset average similarity standard, the central control module selects the second preprocessing parameter adjustment method to adjust the initial preprocessing parameters; the second preprocessing parameter adjustment method refers to the central control module adjusting the resolution of the preprocessing parameters to increase the resolution of the preprocessing parameters by 200dpi.

[0086] Please continue reading Figures 1 to 4 As shown, when the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there are problems with the angle and speed at which the dictionary pen scans the text, the central control module performs scaling processing on the image information:

[0087] If the speed of the dictionary pen scanning the text is greater than the speed standard preset in the central control module, the central control module performs a horizontal stretching process on the image information; the image information is magnified by 1.25 times;

[0088] If the speed of the dictionary pen scanning text is lower than the speed standard preset in the central control module, the central control module performs horizontal compression processing on the image information; the image information is reduced by 1.25 times.

[0089] Please continue reading Figures 1 to 4 As shown, after the central control module completes the scaling processing of the image information, if the central control module still determines that the reason why the language of the text feature cannot be determined is that there is a problem in the pre-processing of the image information by the image processing module, the central control module sets several adjustment methods for the scaling ratio of the scaling processing based on the character ratio of the image information, wherein:

[0090] If the character ratio of the image information is greater than the character ratio standard preset in the central control module, the central control module selects a first telescopic ratio adjustment method to adjust the preset telescopic ratio, wherein the first telescopic ratio adjustment method refers to increasing the telescopic ratio by 10%;

[0091] If the character ratio of the image information is less than the character ratio standard preset in the central control module, the central control module selects a second telescopic ratio adjustment method to adjust the preset telescopic ratio, and the second telescopic ratio adjustment method refers to reducing the telescopic ratio by 10%.

[0092] Please continue reading Figures 1 to 4 As shown, the central control module performs scaling processing on the image information. If the adjustment range of the scaling ratio exceeds the scaling ratio adjustment range preset in the central control module, the central control module is provided with an adjustment method for the scaling angle based on the scaling ratio, wherein:

[0093] If the difference between the adjustment amplitude of the telescopic ratio and the preset adjustment amplitude is less than or equal to the adjustment amplitude difference standard preset in the central control module, the central control module selects a first telescopic angle adjustment method to adjust the telescopic angle, and the first telescopic angle adjustment method refers to increasing the angle between the initial telescopic direction and the horizontal direction of the image by 10°;

[0094] If the difference between the adjustment amplitude of the telescopic ratio and the preset adjustment amplitude is greater than the adjustment amplitude difference standard preset in the central control module, the central control module selects the second telescopic angle adjustment method to adjust the telescopic angle, and the second telescopic angle adjustment method refers to increasing the angle between the initial telescopic direction and the horizontal direction of the image by 20°.

[0095] Please continue reading Figures 1 to 4 As shown, when the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that the language does not exist in the language library or the corresponding font does not exist, the central control module uses all the languages ​​in the language library to perform a secondary semantic guessing judgment on the image information.

[0096] If the central control module determines for the second time that the guessed language information meets the guessing standard preset in the central control module, the central control module controls the output module to output the guessed language and specially marks it;

[0097] If the central control module determines that the guessed language information does not meet the guessing standard preset in the central control module, the central control module controls the output module to issue a notification that a rescan is required;

[0098] The preset guessing standard is that the central control module performs grammatical analysis on the guessed language information, identifies the structure, part of speech and syntactic relationship of the sentence, and determines whether there are grammatical errors or non-compliance with grammatical rules, as well as whether there are ambiguities or non-common sense. The content with grammatical errors, non-compliance with grammatical rules, ambiguity and non-common sense does not exceed 90% of the guessed language information, which meets the preset guessing standard.

[0099] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0100] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent recognition method based on multilingual dictionary pen software, characterized in that: The following steps are involved: Step S1, the user terminal sends a collection instruction to the image collection module, and when the dictionary pen moves above the text, the image collection module collects image information of the text; Step S2, the image processing module pre-processes the image information collected by the image collection module; Step S3, a text recognition module performs text recognition on the pre-processed image information, and selects languages ​​whose text similarity meets the standard; Step S4, after the text recognition module completes the determination of the similarity between the text feature and each language, the central control module counts the number of matching languages, and determines the language to which the text feature belongs based on the number of matching languages, or determines that it is necessary to add text features to be recognized to determine the language to which the text feature belongs for a second time, or determines the reason why the language to which the text feature belongs cannot be determined; Step S5, the central control module re-determines the number of matching languages ​​whose similarity with the text feature meets the standard based on the newly acquired text feature, and completes the determination of the language to which the text feature belongs based on the number of matching languages, or determines the reason why the language to which the text feature belongs cannot be determined; Step S6, after the central control module determines the language, the output module interprets, displays and translates the text scanned by the dictionary pen; The text recognition module obtains text features in the preprocessed image information, and compares the obtained text features with various languages ​​pre-stored in the text recognition module in turn to obtain the similarity between the text features and various languages ​​respectively. The text recognition module records the languages ​​with a similarity to the text features higher than a preset standard as matching languages. After the text recognition module completes the determination of the similarity between the text features and each language, the central control module counts the number of matching languages: If there is only a single matching language, the central control module completes the determination of the language to which the text feature belongs, and controls the output module to translate and interpret the scanned text in the language; If there are multiple matching languages, the central control module determines that it is necessary to increase the text features to be recognized to determine the language to which the text features belong for a second time, and the central control module sends a command to the text recognition module to continue text recognition and determines the number of words that need to be recognized according to the number of languages ​​that meet the standard; If there is no matching language, the central control module determines that the language to which the text feature belongs has not been determined, and the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the recognized text feature.

2. The intelligent recognition method based on the multilingual dictionary pen software according to claim 1 is characterized in that: When determining that there are multiple matching languages, the central control module determines a word number determination method for words that need to be further recognized based on the number of languages ​​that meet the criteria, wherein: The first word number determination method is that the central control module determines to use the preset word scan number set by it as the number of words to be continuously recognized; the first word number determination method satisfies that the number of matching languages ​​is less than or equal to the first preset matching language number set in the central control module; The second word number determination method is that the central control module determines to use the first preset correction coefficient to correct the preset word scan number and uses the corrected preset word scan number as the number of words to be continuously recognized; the second word number determination method satisfies that the number of matching languages ​​is greater than the first preset matching language number and is less than or equal to the second preset matching language number set in the central control module; The third word number determination method is that the central control module determines to use the second preset correction coefficient to correct the preset word scan number and uses the corrected preset word scan number as the number of words that need to continue to be recognized; the third word number determination method satisfies that the number of matching languages ​​is greater than the second preset matching language number.

3. The intelligent recognition method based on the multilingual dictionary pen software according to claim 2 is characterized in that: After the central control module determines the number of words to be further recognized, it controls the image processing module to obtain image information and controls the text recognition module to recognize and obtain a corresponding number of text features from the image information. The central control module re-determines the number of matching languages ​​whose similarity with the text features meets the standard based on the newly obtained text features, wherein: If there is only a single matching language, the central control module completes the determination of the language to which the text feature belongs, and controls the output module to translate and interpret the scanned text in the language; If there are multiple matching languages, the central control module records the matching language with the highest similarity between the newly acquired text features and the text features as the target language, and controls the output module to translate and interpret the scanned text in this language; If there is no matching language, the central control module determines that the language to which the text feature belongs has not been determined, and the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the recognized text feature.

4. The intelligent recognition method based on the multilingual dictionary pen software according to claim 3 is characterized in that: When there is no matching language that meets the standard of similarity with the text feature, the central control module determines the reason why the language to which the text feature belongs cannot be determined based on the length of the text feature obtained by the text recognition module: If the length of the text feature is less than or equal to the first preset length set in the central control module, the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, and the central control module adjusts the preprocessing parameters of the image processing module for the image information to a corresponding value and controls the image processing module to re-preprocess the collected image information using the adjusted preprocessing parameters; If the length of the text feature is greater than the first preset length set in the central control module and is less than or equal to the second preset length, the central control module determines that the reason why the text feature cannot be determined to belong to the language is that there is a problem with the angle and speed of the dictionary pen scanning the text, and the central control module controls the image processing module to perform scaling processing on the image information and controls the text recognition module to re-recognize the text on the image information after scaling processing; If the length of the text feature is greater than the second preset length set in the central control module, the central control module determines that the reason for being unable to determine the language to which the text feature belongs is that the language does not exist in the language library or the corresponding font does not exist, and the central control module controls the output module to issue a notification that the language library needs to be updated.

5. The intelligent recognition method based on the multi-language dictionary pen software according to claim 4 is characterized in that: When the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, the central control module is provided with several adjustment methods for the preprocessing parameters of the image processing module based on the average similarity between the image information and the language, wherein: If the average similarity between the image information and the language is greater than the preset average similarity standard, the central control module selects the first preprocessing parameter adjustment method to adjust the initial preprocessing parameter; the first preprocessing parameter adjustment method refers to the central control module adjusting the grayscale of the preprocessing parameter, increasing the grayscale of the preprocessing parameter by 10%, and setting the preset average similarity standard to 90%; If the average similarity between the image information and the language is less than or equal to the preset average similarity standard, the central control module selects the second preprocessing parameter adjustment method to adjust the initial preprocessing parameters; the second preprocessing parameter adjustment method refers to the central control module adjusting the resolution of the preprocessing parameters to increase the resolution of the preprocessing parameters by 200dpi.

6. The intelligent recognition method based on the multi-language dictionary pen software according to claim 5 is characterized in that: When the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that there are problems with the angle and speed at which the dictionary pen scans the text, the central control module performs scaling processing on the image information: If the speed of the dictionary pen scanning the text is greater than the speed standard preset in the central control module, the central control module performs a horizontal stretching process on the image information; If the speed at which the dictionary pen scans text is lower than the speed standard preset in the central control module, the central control module performs horizontal compression processing on the image information.

7. The intelligent recognition method based on the multi-language dictionary pen software according to claim 6 is characterized in that: After the central control module completes the scaling process on the image information, if the central control module still determines that the reason why the language to which the text feature belongs cannot be determined is that there is a problem in the preprocessing of the image information by the image processing module, the central control module sets several adjustment methods for the scaling ratio of the scaling process based on the character ratio of the image information, wherein: If the character ratio of the image information is greater than the character ratio standard preset in the central control module, the central control module selects a first telescopic ratio adjustment method to adjust the preset telescopic ratio, wherein the first telescopic ratio adjustment method refers to increasing the telescopic ratio by 10%; If the character ratio of the image information is less than the character ratio standard preset in the central control module, the central control module selects a second telescopic ratio adjustment method to adjust the preset telescopic ratio, and the second telescopic ratio adjustment method refers to reducing the telescopic ratio by 10%.

8. The intelligent recognition method based on the multi-language dictionary pen software according to claim 7 is characterized in that: The central control module performs scaling processing on the image information. If the adjustment range of the scaling ratio exceeds the scaling ratio adjustment range preset in the central control module, the central control module is provided with an adjustment method for the scaling angle based on the scaling ratio, wherein: If the difference between the adjustment amplitude of the telescopic ratio and the preset adjustment amplitude is less than or equal to the adjustment amplitude difference standard preset in the central control module, the central control module selects a first telescopic angle adjustment method to adjust the telescopic angle, and the first telescopic angle adjustment method refers to increasing the angle between the initial telescopic direction and the horizontal direction of the image by 10°; If the difference between the adjustment amplitude of the telescopic ratio and the preset adjustment amplitude is greater than the adjustment amplitude difference standard preset in the central control module, the central control module selects the second telescopic angle adjustment method to adjust the telescopic angle, and the second telescopic angle adjustment method refers to increasing the angle between the initial telescopic direction and the horizontal direction of the image by 20°.

9. The intelligent recognition method based on the multi-language dictionary pen software according to claim 8, characterized in that: When the central control module determines that the reason why the language to which the text feature belongs cannot be determined is that the language does not exist in the language library or the corresponding font does not exist, the central control module uses all the languages ​​in the language library to perform a secondary semantic guessing judgment on the image information. If the central control module determines for the second time that the guessed language information meets the guessing standard preset in the central control module, the central control module controls the output module to output the guessed language and specially marks it; If the central control module determines that the guessed language information does not meet the guessing standard preset in the central control module, the central control module controls the output module to issue a notification that a rescan is required; The preset guessing standard is that the central control module performs grammatical analysis on the guessed language information, identifies the structure, part of speech and syntactic relationship of the sentence, and determines whether there are grammatical errors or non-compliance with grammatical rules, as well as whether there are ambiguities or non-common sense. The content with grammatical errors, non-compliance with grammatical rules, ambiguity and non-common sense does not exceed 90% of the guessed language information, which meets the preset guessing standard.

Images