Intelligent reading recommendation method and system for assisting learning

Abstract

The invention relates to an intelligent reading recommendation method and system for assisting learning, and relates to the field of artificial intelligence, and the method comprises the steps: collecting account login information; obtaining education background information according to the account login information; matching an English recommended article based on the educational background information, performing article recommendation based on the English recommended article, and collecting a word scanning image at the same time; in response to the word scanning image, identifying a required query word, and generating a synonym word based on the required query word; and combining the required query word, the synonym word, the English recommendation article and the education information to generate query word information, and highlighting the query word information for viewing and learning. The method has the effect of improving the learning efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, in particular to an intelligent reading recommendation method and system for assisting learning. BACKGROUND

[0002] Intelligent reading recommendation refers to a technology that analyzes a user's reading behavior, learning goals, knowledge level and interest preferences to accurately recommend reading materials that meet individual needs, thereby improving learning efficiency and effectiveness.

[0003] Currently, intelligent reading recommendation technology is mostly based on collaborative filtering, content feature matching algorithms, etc., which analyzes user reading history, click preferences, dwell time and other behavior data, and combines text themes, keywords, difficulty and other features to recommend similar or related content to users.

[0004] However, the existing intelligent reading recommendation technology fails to fully consider the user's educational level in the context of assisting learning, and only relies on general behavior data and content features for recommendation, making it difficult to accurately match the difficulty of learning materials with the user's knowledge level, thereby reducing learning efficiency and requiring improvement. SUMMARY

[0005] In order to improve learning efficiency, the present application provides an intelligent reading recommendation method and system for assisting learning.

[0006] In a first aspect, the present application provides an intelligent reading recommendation method for assisting learning, which adopts the following technical solution: An intelligent reading recommendation method for assisting learning, comprising: Collecting account login information; Obtaining educational information based on the account login information; Matching English recommended articles based on the educational information, and recommending articles based on the English recommended articles, while collecting word scan images; Identifying demand query words in response to the word scan images, and generating synonymous words based on the demand query words; Generating query word information by combining the demand query words, the synonymous words, the English recommended articles and the educational information, and highlighting the query word information for viewing and learning.

[0007] By adopting the technical scheme, the system first acquires the educational background information of the user through the account login information, and matches the recommended English article according to the educational background information, so as to ensure that the difficulty of the recommended content matches the knowledge level of the user, and avoid learning frustration or inefficient repetition caused by too difficult or too easy materials. In the learning process, the system supports the user to query the word by scanning the image, can not only identify the required query word, but also generate the synonym of the word, integrate the query word information into the context of the recommended English article and the educational background information of the user, and highlight the query word information, so as to consider the meaning of the word in the specific context, expand the vocabulary of the user through the synonym, help the user to efficiently accumulate vocabulary and understand the article in the content suitable for the user's level, and thus improve the learning efficiency.

[0008] Optionally, the method further comprises: generating an examination scene category based on the educational background information, and collecting category trigger information; generating a specific examination scene in response to the category trigger information and the examination scene category; generating a word example sentence based on the required query word; filtering out a display example sentence based on the word example sentence and the specific examination scene; highlighting the display example sentence for viewing and learning.

[0009] Optionally, the method further comprises: collecting an example sentence click signal; matching a specific clicked example sentence in response to the example sentence click signal; performing syntax analysis on the specific clicked example sentence to obtain example sentence related syntax; generating a syntax explanation based on the example sentence related syntax; analyzing the example sentence related syntax based on the syntax explanation, and broadcasting the analysis.

[0010] Optionally, the method further comprises: collecting a collection content and a collection content quantity; knowing a collection word and a collection example sentence based on the collection content; determining whether the collection content quantity exceeds a preset reference collection quantity; if not, matching a recommended article in combination with the collection word, the collection example sentence, and the specific examination scene; performing article recommendation based on the recommended article for viewing and learning; if so, generating a learning article by a preset article generation method for viewing and learning.

[0011] Optionally, the article generation method comprises: extract core semantics according to the collected words and the collected example sentences; generate semantic directions based on the core semantics and the educational background information; generate semantic learning scenarios by combining the semantic directions and the specific test scenarios; create a basic article segment according to the semantic directions and the semantic learning scenarios; integrate the collected words and the collected example sentences into the basic article segment to form an initial learning article; perform language fluency verification on the initial learning article, and after the verification is passed, mark the specific meanings of the collected words in the article and the contextual interpretation of the collected example sentences for viewing and learning.

[0012] Optionally, the method further comprises a method for identifying the demand query words: extract word contours and character information from the word scanning image; perform integrity detection on the extracted character information to determine whether the word is complete; if it is detected that the word is not complete, perform scanning adjustment by using a preset scanning adjustment method; update the word scanning image after the scanning adjustment is completed; if it is detected that the word is complete, generate demand query words based on the word contours.

[0013] Optionally, the scanning adjustment method comprises: acquire an adjustment switching signal; when the adjustment switching signal is a preset automatic adjustment signal, identify the stroke thickness of the word from the word scanning image to obtain stroke thickness parameters; combine the stroke thickness parameters and preset reference stroke thickness parameters to calculate a current text proportion; generate a focal length adjustment parameter according to the current text proportion; adjust the focal length of a preset scanning device based on the focal length adjustment parameter, thereby completing the scanning adjustment.

[0014] Optionally, the scanning adjustment method further comprises: when the adjustment switching signal is a preset manual adjustment signal, start a manual adjustment mode of the scanning device, and acquire an image scanning proportion; generate an illumination parameter according to the image scanning proportion; control the scanning device to emit auxiliary illumination light with the illumination parameter, and acquire an illumination coverage image; determine whether the auxiliary illumination light completely covers the word to be identified based on the illumination coverage image; If not fully covered, a preset adjustment prompting method is used for adjustment prompting; If fully covered, an adjustment completion signal is generated; An updated word scanning image is generated in response to the adjustment completion signal.

[0015] Optionally, the adjustment prompting method comprises: A current illumination range of the auxiliary illumination light is identified from the illumination coverage image to obtain the current illumination range, and a range of the word to be identified is identified from the illumination coverage image to obtain a required coverage range; When the current illumination range does not exceed the required coverage range, the current illumination range and the required coverage range are combined to generate an illumination adjustment parameter; The adjustment prompting information is generated according to the illumination adjustment parameter, and adjustment prompting is performed based on the adjustment prompting information; When the current illumination range exceeds the required coverage range, the illumination coverage image and a preset character feature are combined to analyze a missing coverage position; The direction prompting information is generated according to the missing coverage position, and adjustment prompting is performed based on the direction prompting information.

[0016] In a second aspect, the present application provides an intelligent reading recommendation system for assisting learning, which adopts the following technical solution: An intelligent reading recommendation system for assisting learning, comprising: A collection module for collecting account login information and a word scanning image; A storage for storing a program for implementing any of the intelligent reading recommendation methods for assisting learning; A processor for loading and executing the program stored in the storage.

[0017] In summary, the present application has at least one of the following beneficial technical effects: 1. The system first obtains the user's educational information through the account login information, and matches the appropriate English recommendation article based on this, to ensure that the difficulty of the recommended content matches the user's knowledge level, and to avoid learning frustration or inefficiency caused by materials that are too difficult or too easy. During the learning process, the system supports users to query words through scanning images, which not only identifies the required query words, but also generates their synonyms, integrates comprehensive query word information and highlights them in combination with the context of the English recommendation article and the user's educational information, to consider the meaning of the word in the specific context and expand the user's vocabulary through synonyms, helping users efficiently accumulate vocabulary and understand articles in content suitable for their level, thereby effectively improving learning efficiency; 2. The system uses the user's favorite words and favorite sentences as core materials, extracts the core semantics, and ensures that the generated content is closely related to the user's learning focus. Combined with the user's academic information, a semantic direction is generated, and then combined with specific examination scenarios, a targeted semantic learning scenario is generated. Finally, after creating basic article fragments based on the semantic direction and semantic learning scenario, the favorite words and favorite sentences are naturally integrated into them to form an initial learning article. After the article is generated, a language fluency check is performed to ensure the naturalness of the writing. After the check is passed, the specific meaning of the favorite words and the contextual interpretation of the favorite sentences are marked to facilitate users to understand and learn at any time while reading. In this way, the generated learning articles can not only closely follow the user's learning focus, but also adapt to their academic level and examination needs, helping users to expand their learning based on familiar vocabulary and examples, and improve their efficiency in mastering knowledge and their ability to apply it; 3. When switching to manual adjustment mode, the system collects the image scan ratio and generates corresponding lighting parameters, controlling the scanning device to emit auxiliary lighting. Simultaneously, the lighting coverage image determines whether the illumination completely covers the word to be recognized. If it does not, adjustment prompts guide the user to make adjustments to ensure that there are no blind spots in the lighting. If it does, an adjustment completion signal is generated and the word scan image is updated. This precise lighting control and feedback mechanism provides stable image input for word scanning and recognition, ensuring the smooth progress of subsequent word search, article generation, and other processes, further enhancing the consistency and effectiveness of assisted learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a method flow chart of an intelligent reading recommendation method for assisting learning; Figure 2 It is a method flow chart of the article generation method; Figure 3 The scanning adjustment method is the process Figure 1 ; Figure 4 The scanning adjustment method is the process Figure 2 . DETAILED DESCRIPTION

[0019] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0020] Reference Figure 1 , the embodiment of the present application discloses an intelligent reading recommendation method for assisting learning, comprising the following steps: S1: Collect account login information.

[0021] Account login information refers to the identification information entered by a user when logging into the system. Account login information is collected in real time by the system's preset collection module when the user logs into the system. The collection module is pre-configured by those skilled in the art and will not be described in detail here.

[0022] S2: obtaining education information according to the account login information.

[0023] The education information refers to the user's educational background data associated with the account login information. The education information is obtained by querying after understanding the account login information. The user will pre-enter the education information corresponding to the account login information.

[0024] S3: matching English recommended articles based on the education information, and recommending articles based on the English recommended articles, and collecting word scan images.

[0025] The English recommended articles refer to English reading materials recommended for the user to learn. The English recommended articles are obtained by matching a preset article library. The article library stores English reading materials classified by education levels (such as simple picture book original texts for primary school stage, selected sections of textbooks for junior high school stage, extracurricular reading articles for high school stage, and professional related English documents for university stage), and the system selects English articles of corresponding difficulty level from the article library based on the user's education information (such as "high school"), that is, the English recommended articles.

[0026] The word scan image refers to an image containing a word that needs to be queried, which is taken by a preset scanning device. The scanning device refers to an electronic device for taking and scanning a word to be queried to obtain a word image.

[0027] After the English recommended articles are matched, the preset learning device is controlled to display the English recommended articles for recommendation, and the word scan image is collected for subsequent steps.

[0028] The learning device refers to an electronic device for the user to learn English.

[0029] S4: identifying the demand query word in response to the word scan image, and generating a synonym word based on the demand query word.

[0030] The demand query word refers to a word that the user needs to query the meaning or usage of. The specific identification method of the demand query word is described in detail in subsequent S40 to S44, and is not repeated here.

[0031] Synonyms are words that have similar meanings to the query word. Synonyms are generated by matching a pre-set synonym library. The synonym library is built by those skilled in the art according to different levels of English learning syllabus and vocabulary requirements, and stores a large number of English words and their corresponding synonym entries (e.g. "happy" corresponds to "glad", "pleased", "difficult" corresponds to "hard", "tough", etc.). After the query word is understood, the synonym library is automatically searched to match words with similar meanings to form a synonym list, which provides accurate vocabulary expansion reference for users to help them expand their vocabulary and understand the subtle differences between words.

[0032] When the query word is identified, synonyms need to be generated first for subsequent steps.

[0033] S5: Combine the query word, synonyms, recommended English articles, and educational information to generate query word information, and highlight the query word information for viewing and learning.

[0034] Query word information refers to a comprehensive information package that integrates the meaning of the query word, synonym comparison, contextual usage in recommended English articles, and difficulty level according to educational information. Query word information is generated by integrating the basic information of the query word (such as part of speech, phonetic symbols, core meaning), the analysis of synonyms (such as word meaning focus, usage scenario differences), the original sentence and context analysis in recommended English articles, and the difficulty level according to educational information (such as "core vocabulary for junior high school" and "high-frequency words for college level 6").

[0035] When the query word information is obtained, the query word information needs to be highlighted for users to view and learn.

[0036] The following steps are also included: S60: Based on the educational information, generate the exam scenario category, and collect category trigger information.

[0037] The exam scenario category refers to the classification of exams that the user may participate in. The exam scenario category is generated by matching a pre-set exam scenario database. The database stores exam types corresponding to different educational information (e.g. "primary school English final exam" and "primary school entrance exam English mock exam" for primary school education; "College English Test 4" and "IELTS exam" for college education, etc.), and the system retrieves the appropriate exam type set from the database based on the user's educational information, which is the exam scenario category. The exam scenario database is formed by sequentially recording the exam types corresponding to different educational information by those skilled in the art, which is not described here.

[0038] The category trigger information refers to an operation signal of a user triggering a specific test scene. The category trigger information is collected through interactive operation of the user. For example, the user clicks the "postgraduate entrance examination English" option on the system interface, searches for the "TOEFL reading" keyword, or inputs the "trigger the CET-4 / CET-6 scene" voice, and the system collection module captures these operation signals, that is, the category trigger information.

[0039] S61: generating a specific test scene in response to the category trigger information and the test scene category.

[0040] The specific test scene refers to a specific test situation selected from the test scene category and matched with the category trigger information. By accurately comparing the category trigger information (such as "click postgraduate entrance examination English") triggered by the user with the test scene category (such as the university stage test scene), the specific situation with the highest matching degree (such as "postgraduate entrance examination English reading comprehension true test scene") is selected to obtain the specific test scene.

[0041] S62: generating a word example sentence according to the demand query word.

[0042] The word example sentence refers to a typical example sentence containing a demand query word. The word example sentence is generated by retrieving a preset example sentence library. The example sentence library stores a large number of typical example sentences containing various words (such as the example sentence "One can never acquire enough knowledge" of "acquire"). The system retrieves the corresponding example sentence from the example sentence library according to the demand query word, which is the word example sentence. The example sentence library is constructed by a person skilled in the art according to different educational stage English learning goals and vocabulary mastery requirements, and will not be described here.

[0043] S63: filtering out a display example sentence according to the word example sentence and the specific test scene.

[0044] The display example sentence refers to an example sentence that meets the focus of the specific test scene selected from the word example sentence. The display example sentence is selected from the word example sentence by understanding the focus of the specific test scene (such as "high school entrance examination English writing scene" focusing on sentence diversity, and "IELTS listening scene" focusing on daily conversation usage), and the example sentence that meets the focus (such as selecting an example sentence containing a complex sentence pattern for "high school entrance examination writing scene") is the display example sentence.

[0045] S64: highlighting the display example sentence for viewing and learning.

[0046] The selected display example sentence is highlighted for the user to view and learn.

[0047] Further comprising the following steps: S65: collecting an example sentence click signal.

[0048] The example sentence click signal refers to an interactive instruction of a user pointing to a specific displayed example sentence. The example sentence click signal is collected by a preset touch sensor. When the user performs a click operation on the interface displaying the example sentence, the touch sensor captures the position coordinates and triggering time of the operation in real time, and then generates the corresponding example sentence click signal.

[0049] S66: In response to the example sentence click signal, a specific clicked example sentence is matched out.

[0050] The specific clicked example sentence refers to the displayed example sentence selected by the user through the example sentence click signal. The specific clicked example sentence matches and associates the interactive position corresponding to the example sentence click signal with the display area of the displayed example sentence. The specific matching and associating method is well known in the art and will not be described here.

[0051] S67: The specific clicked example sentence is analyzed to obtain example sentence related grammar.

[0052] The example sentence related grammar refers to the grammar knowledge points parsed from the specific clicked example sentence, including tense, sentence structure, fixed collocation, and word class usage. The example sentence related grammar is obtained by analyzing the specific clicked example sentence. The specific grammar analysis method is well known in the art and will not be described here.

[0053] S68: Based on the example sentence related grammar, a grammar explanation is generated.

[0054] The grammar explanation refers to a standardized explanatory text of the example sentence related grammar, including grammar rule definitions, constitutive forms, and usage scenarios and examples. The grammar explanation is generated by a preset grammar knowledge base. The knowledge base stores standardized explanations of various grammar knowledge points, and the system retrieves the corresponding explanatory text from the knowledge base according to the parsed example sentence related grammar, and integrates it to form the grammar explanation. The grammar knowledge base is constructed by a person skilled in the art according to the English grammar system, and will not be described here.

[0055] S69: According to the grammar explanation, the example sentence related grammar is analyzed and broadcasted.

[0056] According to the grammar explanation, the example sentence related grammar is analyzed, so that the abstract grammar knowledge is combined with the actual example sentence, and the analysis content is broadcasted by voice to facilitate the user to understand.

[0057] Further comprising the following steps: S70: Collecting the collection content and the collection content quantity.

[0058] The collection content refers to the key learning content marked by the user through the collection function of the system. The collection content quantity refers to the total number of contents collected by the user. The collection content and the collection content quantity are collected in real time by the collection module of the system.

[0059] S71: Knowing the collection words and collection example sentences based on the collection content.

[0060] The collection words refer to the words extracted from the collection content and marked as important by the user. The collection example sentences refer to the example sentences extracted from the collection content and marked as important by the user. The collection content contains the collection words and the collection example sentences.

[0061] S72: Determining whether the number of collection content exceeds a preset reference collection number.

[0062] The reference collection number refers to a reference threshold for understanding whether the number of collection content of the user is too large. The reference collection number is set by a person skilled in the art in advance, and is not described here.

[0063] The specific way of subsequent recommended articles is distinguished by determining whether the number of collection content exceeds the reference collection number.

[0064] S73: If not, the recommended articles are matched by combining the collection words, the collection example sentences, and the specific test scene.

[0065] The recommended articles refer to reading materials containing multiple collection words and collection example sentences and adapting to the specific test scene. The recommended articles are matched by a preset associated article library. The associated article library stores a large number of reading materials associated with different words, example sentences, and test scenes. The system filters the reading materials containing multiple collection words, collection example sentences, and meeting the requirements of the specific test scene from the associated article library based on the core content of the collection words, the collection example sentences (such as word meaning, example sentence theme), and the examination range of the specific test scene. The reading materials are the recommended articles. The associated article library is constructed by a person skilled in the art according to the association between words, example sentences, and articles and the characteristics of the test scene, and is not described here.

[0066] When the number of collection content does not exceed the reference collection number, the recommended articles need to be matched first for subsequent steps.

[0067] S730: Based on the recommended articles, the article recommendation is performed for viewing and learning.

[0068] The matched recommended articles are displayed for article recommendation and learning.

[0069] S74: If it exceeds, the learning articles are generated by a preset article generation method for viewing and learning.

[0070] The article generation method refers to a standardized process of automatically creating a customized learning article containing all the collected words and example sentences when the number of collected contents exceeds the reference collection number. The specific article generation method is described in detail in S740 to S745, which is not repeated here.

[0071] When the number of collected contents exceeds the reference collection number, the article generation method is needed to generate learning articles, so as to provide users with learning.

[0072] Reference Figure 2 The article generation method includes the following steps: S740: Extracting core semantics according to the collected words and the collected example sentences.

[0073] The core semantics refers to the core meaning extracted from the collected words and the collected example sentences (for example, the collected words are mostly "environment" and "protect", and the collected example sentences are around environmental protection, so the core semantics is "environmental protection"). By extracting the core meaning of each collected word and summarizing the common theme of the collected example sentences, the common core meaning of these information is extracted through a semantic clustering algorithm, which is the core semantics. The semantic clustering algorithm is a common knowledge in the art, which is not repeated here.

[0074] S741: Generating semantic direction based on the core semantics and the educational information.

[0075] The semantic direction refers to the detailed direction of the core semantics (for example, when the core semantics is "environmental protection", the semantic direction can be defined as "teenagers' environmental protection action"). The core semantics is first dimensionally split and scenically refined, and then combined with the learning focus corresponding to the user's educational information (for example, middle school students focus on "school environmental protection" and college students focus on "global environmental protection"), the specific refinement direction is determined from the theme range, application scene and other dimensions, which is the semantic direction.

[0076] S742: Generating semantic learning scene by combining the semantic direction and the specific examination scene.

[0077] The semantic learning scenario refers to generating a specific context that adapts to the user's learning needs around the semantic direction and the specific test scenario. The semantic learning scenario is generated by matching a preset scenario matching database. The scenario matching database stores specific context templates corresponding to different semantic directions and specific test scenarios (for example, when the semantic direction is "environmental protection" and the specific test scenario is "college entrance examination English writing", the corresponding "environmental protection themed argumentative writing context" template). The system retrieves the adaptive context template from the scenario matching database based on the current semantic direction and specific test scenario, and then generates the semantic learning scenario. The scenario matching database is constructed by those skilled in the art according to the classification of semantic directions, the examination requirements of specific test scenarios, and the practicability of learning scenarios, and will not be described here.

[0078] S743: creating a basic article segment according to the semantic direction and the semantic learning scenario.

[0079] The basic article segment refers to the first draft of the article created according to the semantic direction and the semantic learning scenario, which includes the framework of introduction, main body paragraph, conclusion, etc., and the language difficulty adapts to the user's educational information. The basic article segment is generated by understanding the semantic direction to know the core theme of the article (such as "the role of teenagers in environmental protection"), and then according to the structural requirements of the semantic learning scenario (such as introducing the background in the introduction, describing the action in the main body, and summarizing the significance in the conclusion), the language difficulty of the article first draft framework that adapts to the user's educational stage is generated, which is the basic article segment. The specific generation means is well known in the art and will not be described here.

[0080] S744: integrating the collected words and the collected example sentences into the basic article segment to form an initial learning article.

[0081] The initial learning article refers to a complete article formed by naturally integrating all collected words and collected example sentences into the basic article segment. The collected words and collected example sentences are embedded in the basic article segment according to semantic logic. The specific embedding method is well known in the art and will not be described here.

[0082] S745: checking the language fluency of the initial learning article, and after passing the check, marking the specific meaning of the collected words in the article and the contextual interpretation of the collected example sentences for viewing and learning.

[0083] The system checks the initial learning article through a preset language fluency checking algorithm, and after passing the check, automatically marks the specific meaning of the collected words in the text (such as "protect" in the sentence indicating "protection") and the contextual interpretation of the collected example sentences (such as explaining the argumentation role of the example sentence in the text), and finally presents them to the user for learning. The language fluency checking algorithm is well known in the art and will not be described here.

[0084] The method also comprises identifying the demand query word: S40: extracting word contour and character information from the word scanning image.

[0085] The word contour refers to the external edge shape of the characters constituting the word. The character information refers to the specific letters identified from the word scanning image. The word contour and the character information are extracted from the word scanning image through image recognition technology. Image recognition technology is well known in the art and will not be described here.

[0086] S41: integrity detection of the extracted character information to determine whether the word is complete.

[0087] The integrity of the character information is detected to determine whether the word is complete, and then to know whether scanning adjustment is needed. The method of integrity detection is well known in the art and will not be described here.

[0088] S42: if the word is detected to be incomplete, scanning adjustment is performed by a preset scanning adjustment method.

[0089] The scanning adjustment method refers to the operation method of adjusting the scanning device to ensure complete recognition of the word when the character information is incomplete. The specific scanning adjustment method is described in detail in S420 to S4212 and S422 to S42240, which will not be described here.

[0090] If the word is detected to be incomplete, the scanning device needs to be adjusted by the scanning adjustment method.

[0091] S43: after completing the scanning adjustment, updating the word scanning image.

[0092] After completing the scanning adjustment, the word scanning image needs to be reacquired to update the word scanning image.

[0093] S44: if the word is detected to be complete, generating a demand query word based on the word contour.

[0094] If the word is detected to be complete, the demand query word can be directly generated. The method of generating a specific word by understanding the word contour is well known in the art and will not be described here.

[0095] Referring to Figure 3 , the scanning adjustment method comprises the following steps: S420: acquiring an adjustment switching signal.

[0096] The adjustment switching signal refers to an operation signal for selecting a scanning adjustment mode. The adjustment switching signal is acquired through the interactive operation of the user.

[0097] S421: When the adjustment switching signal is a preset automatic adjustment signal, the stroke thickness of a word in the word scanning image is recognized to obtain a stroke thickness parameter.

[0098] The automatic adjustment signal refers to an operation signal generated when the user selects an automatic adjustment mode, which is used to trigger the system to automatically complete the parameter adjustment of the scanning device. The automatic adjustment signal is preset by those skilled in the art, and will not be described here.

[0099] The stroke thickness parameter refers to the width value of the character stroke in the word scanning image. The stroke thickness parameter is obtained by image measurement of the stroke in the word scanning image. The image measurement technology is well known in the art, and will not be described here.

[0100] When the adjustment switching signal is an automatic adjustment signal, it means that the user wants the scanning device to automatically perform scanning adjustment, and the stroke thickness parameter needs to be obtained first for subsequent steps.

[0101] S4210: The stroke thickness parameter and a preset reference stroke thickness parameter are combined to calculate a current text ratio.

[0102] The reference stroke thickness parameter refers to the reference value of the stroke thickness of the characters in the text without ratio adjustment. The reference stroke thickness parameter is preset by those skilled in the art, and will not be described here.

[0103] The current text ratio refers to the ratio of the current text expansion. The current text ratio is calculated by the ratio of the stroke thickness parameter to the reference stroke thickness parameter.

[0104] S4211: The current text ratio is used to generate a focal length adjustment parameter.

[0105] The focal length adjustment parameter refers to the specific value of the focal length adjustment required by the scanning device. The focal length adjustment parameter is generated by the difference between the current text ratio and a preset standard ratio (in this embodiment, the standard ratio is 1.0). If the current text ratio is 1.5 (greater than the standard ratio), the parameter "reduce focal length by 0.5 times" is generated; if it is 0.8 (less than the standard ratio), the parameter "enlarge focal length by 0.2 times" is generated.

[0106] S4212: Based on the focal length adjustment parameter, the preset scanning device is adjusted in focal length, and the scanning adjustment is completed.

[0107] The scanning device is controlled to adjust the focal length with the focal length adjustment parameter, so as to complete the scanning adjustment for subsequent image acquisition.

[0108] Reference Figure 4 , the scanning adjustment method further comprises the following steps: S422: When the adjustment switching signal is a preset manual adjustment signal, starting a manual adjustment mode of the scanning device, and collecting an image scanning ratio.

[0109] The manual adjustment signal refers to an operation signal generated when a user selects a manual adjustment mode. The manual adjustment signal is preset by a person skilled in the art, and is not described herein.

[0110] The image scanning ratio refers to a proportional relationship in size between a scanned image and an original image in an image scanning process. The image scanning ratio is obtained by understanding image scanning parameters of the scanning device. The image scanning parameters include the image scanning ratio.

[0111] When the adjustment switching signal is the manual adjustment signal, it indicates that the user wants to manually adjust the scanning, and the manual adjustment mode needs to be started first, and the image scanning ratio needs to be collected for subsequent steps.

[0112] S4220: Generating an illumination parameter according to the image scanning ratio.

[0113] The illumination parameter refers to a parameter for controlling the size of the illumination range of the auxiliary illumination light. The auxiliary illumination light is used to prompt the user whether to continue to adjust the light of the scanning device by whether the illumination coverage range completely covers the word to be recognized in the manual adjustment mode.

[0114] The illumination parameter is generated by matching a preset illumination parameter database. The database stores illumination parameters corresponding to different image scanning ratios. By understanding the image scanning ratio, the adaptive illumination parameter is retrieved from the illumination parameter database. The illumination parameter database is constructed by a person skilled in the art according to the lighting requirements and device performance under different scanning ratios, and is not described herein.

[0115] S4221: Controlling the scanning device to emit auxiliary illumination light with the illumination parameter, and collecting an illumination coverage image.

[0116] The illumination coverage image refers to an image taken by the scanning device after the auxiliary illumination light is turned on, and the image contains the illumination area and the word to be recognized, which is used to judge the illumination coverage. The word to be recognized refers to a word that the user needs to query and that is not completely recognized in the current scanning image.

[0117] When a word is not completely recognized in the current scanning image, the word is marked as the word to be recognized.

[0118] The scanning device is controlled to emit auxiliary illumination light with the illumination parameter, and the illumination coverage image is collected for subsequent steps.

[0119] S4222: judging whether the auxiliary illuminating light completely covers the word to be recognized based on the light coverage image.

[0120] The light coverage range of the auxiliary illuminating light and the range of the word to be recognized are identified from the light coverage image, and whether the auxiliary illuminating light completely covers the word to be recognized is judged by comparing the ranges, so as to know whether the user needs to be prompted for adjustment.

[0121] S4223: if not completely covered, prompting for adjustment by a preset adjustment prompting method.

[0122] The adjustment prompting method refers to the adjustment guide process provided by the system to the user when the auxiliary illuminating light does not completely cover the word to be recognized. The specific adjustment prompting method is described in detail in subsequent S42230 to S42234, which is not repeated here.

[0123] If the auxiliary illuminating light does not completely cover the word to be recognized, the user needs to be prompted for adjustment by the adjustment prompting method.

[0124] S4224: if completely covered, generating an adjustment completion signal.

[0125] The adjustment completion signal refers to a signal indicating that the scanning adjustment is qualified when the auxiliary illuminating light completely covers the word to be recognized.

[0126] If the auxiliary illuminating light completely covers the word to be recognized, it means that the scanning adjustment is qualified, and the system can directly generate an adjustment completion signal.

[0127] S42240: updating the word scanning image in response to the adjustment completion signal.

[0128] When the adjustment completion signal is received, the word scanning image can be re-acquired to update the word scanning image.

[0129] The adjustment prompting method comprises the following steps: S42230: identifying the light coverage range of the auxiliary illuminating light from the light coverage image to obtain a current light coverage range, and identifying the range of the word to be recognized from the light coverage image to obtain a required coverage range.

[0130] The current light coverage range refers to the range of the area actually irradiated by the auxiliary illuminating light identified from the light coverage image. The required coverage range refers to the complete area range identified from the light coverage image and containing the word to be recognized. Both the current light coverage range and the required coverage range are obtained by image recognition technology. The image recognition technology is well known in the art, which is not repeated here.

[0131] S42231: based on the current illumination range not exceeding the demand coverage range, combining the current illumination range and the demand coverage range to generate an illumination adjustment parameter.

[0132] The illumination adjustment parameter refers to a specific parameter for expanding the illumination range when the current illumination range does not exceed the demand coverage range. The illumination adjustment parameter is obtained by using image recognition technology to obtain the boundary data of the current illumination range and the demand coverage range, and is converted into a specific parameter that can expand the illumination range through calculation. The algorithm and parameter conversion logic followed in this process are constructed by those skilled in the art, and the relevant details are not repeated here.

[0133] Based on the current illumination range not exceeding the demand coverage range, the illumination adjustment parameter needs to be calculated first, in order to facilitate the subsequent steps.

[0134] S42232: generating an adjustment prompt information according to the illumination adjustment parameter, and making an adjustment prompt based on the adjustment prompt information.

[0135] The adjustment prompt information refers to a text prompt generated based on the illumination adjustment parameter, which is used to guide the user to adjust the illumination. The adjustment prompt information is generated by converting the illumination adjustment parameter into text instructions that the user can understand. The specific conversion method is well known in the art and is not repeated here.

[0136] When the adjustment prompt information is generated, the scanning device needs to be controlled to make an adjustment prompt according to the adjustment prompt information.

[0137] S42233: based on the current illumination range exceeding the demand coverage range, combining the illumination coverage image and the preset character feature to analyze the coverage missing direction.

[0138] The character feature refers to the contour feature of a word character. The character feature is set by those skilled in the art in advance and is not repeated here.

[0139] The coverage missing direction refers to the specific direction of the un-illuminated area in the word to be recognized when the current illumination range exceeds the demand coverage range. By identifying the character contour of the word to be recognized from the illumination coverage image, the range of the complete word is located according to the character feature; then by comparing the current illumination range, the specific area of the word that is not illuminated is determined, and the direction of the area is specified, which is the coverage missing direction.

[0140] Based on the current illumination range exceeding the demand coverage range, the coverage missing direction needs to be obtained first, in order to facilitate the subsequent steps.

[0141] S42234: generating a direction prompt information according to the coverage missing direction, and making an adjustment prompt based on the direction prompt information.

[0142] The direction prompt information refers to a direction guide for instructing the user to move the scanning device. The distance prompt information refers to a distance guide for instructing the user to move the scanning device. The direction prompt information is obtained by knowing the position of the coverage gap, and thus knowing the specific direction of the word to be recognized that is not covered by the auxiliary lighting light.

[0143] When the direction prompt information is obtained, the scanning device needs to be controlled to adjust the prompt according to the direction prompt information.

[0144] Optionally, the method further comprises the following steps: S80: judging whether the word has a preset marking feature based on the word scanning image.

[0145] The marking feature refers to a specific marking trace of the word made by the user for the convenience of learning and distinguishing the key points. In this embodiment, the marking feature includes the underlined (single underline, double underline), wavy line, square circle note, triangular symbol note, different color highlighter painting, star mark and other visual marking forms that can be captured by image recognition.

[0146] By judging whether the word has a marking feature from the word scanning image, the attention degree and learning needs of the user to the word are known.

[0147] S81: When and only when the marking feature exists, the specific marking type and marking color are obtained by identifying the marking feature from the word scanning image.

[0148] The specific marking type refers to the subdivision of the specific form of the marking feature. The marking color refers to the color information presented by the marking feature.

[0149] The specific marking type is obtained by recognizing the contour form of the marking feature through edge detection technology, and then combining the feature template matching of the marking feature. The feature template of the marking feature is set by the person skilled in the art in advance, and is not described here.

[0150] The marking color is obtained by color recognition on the word scanning image. The specific color recognition method is well known in the art, and is not described here.

[0151] S82: generating a marked word classification according to the specific marking type and marking color.

[0152] The marked word classification refers to the classification of the word with the marking feature. The marked word classification is obtained by generating a preset marking classification database. The database stores the mapping relationship between the combination of the specific marking type and the marking color and the learning attribute, such as single underline plus red corresponding to “high-frequency exam point vocabulary”, wavy line plus blue corresponding to “easily confused vocabulary”, etc. These mapping rules can be pre-set by the person skilled in the art, and are not described here.

[0153] S83: classifying the tagged word based on the tagged word classification to classify the word with the tagged feature.

[0154] The word with the tagged feature is classified according to the generated tagged word classification for subsequent learning of the user.

[0155] Based on the same inventive concept, the embodiments of the present application provide an intelligent reading recommendation system for assisting learning, comprising: The acquisition module is configured to acquire account login information, word scanning images, category trigger information, example sentence click signals, collected content, the number of collected content, adjustment switching signals, image scanning ratios, and light coverage images. The memory is configured to store a program for implementing an intelligent reading recommendation method for assisting learning. The processor is configured to load and execute the program stored in the memory.

[0156] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is exemplified, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0157] The above is only the preferred embodiment of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solution falling within the concept of the present application shall be considered as falling within the protection scope of the present application. It should be noted that, for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application shall also be considered as falling within the protection scope of the present application.

Claims

1. An intelligent reading recommendation method for assisting learning, characterized by, The method comprises the following steps: collecting account login information; obtaining education information according to the account login information; matching English recommended articles based on the education information, and recommending articles based on the English recommended articles, while collecting word scanning images; responding to the word scanning images to identify demand query words, and generating synonyms based on the demand query words; combining the demand query words, the synonyms, the English recommended articles, and the education information to generate query word information, and highlighting the query word information for viewing and learning. 2.The intelligent reading recommendation method for assisting learning according to claim 1, characterized in that, Further comprising: generating test scene categories based on the education information, and collecting category trigger information; generating specific test scenes in response to the category trigger information and the test scene categories; generating word example sentences according to the demand query words; screening display example sentences according to the word example sentences and the specific test scenes; highlighting the display example sentences for viewing and learning. 3.The intelligent reading recommendation method for assisting learning according to claim 2, characterized in that, Further comprising: collecting example sentence click signals; matching specific clicked example sentences in response to the example sentence click signals; performing syntax analysis on the specific clicked example sentences to obtain example sentence related syntax; generating syntax interpretation based on the example sentence related syntax; analyzing the example sentence related syntax according to the syntax interpretation, and broadcasting. 4.The intelligent reading recommendation method for assisting learning according to claim 2, characterized in that, Further comprising: collecting collection content and collection content quantity; knowing collection words and collection example sentences based on the collection content; determining whether the collection content quantity exceeds a preset reference collection quantity; if not, combining the collection words, the collection example sentences, and the specific test scenes to match recommended articles; performing article recommendation based on the recommended articles for viewing and learning; if so, generating learning articles using a preset article generation method for viewing and learning. 5.The intelligent reading recommendation method for assisting learning according to claim 4, characterized in that, The article generation method comprises: extracting core semantics from the collection words and the collection example sentences; generating semantic directions based on the core semantics and the education information; combining the semantic directions and the specific test scenes to generate semantic learning scenes; creating basic article segments according to the semantic directions and the semantic learning scenes; integrating the collection words and the collection example sentences into the basic article segments to form initial learning articles; performing language fluency verification on the initial learning articles, and after passing the verification, marking the specific meanings of the collection words in the articles and the contextual interpretation of the collection example sentences for viewing and learning. 6.The intelligent reading recommendation method for assisting learning of claim 1, wherein, Further comprising the identification method of the demand query words: extracting word contours and character information from the word scanning images; performing integrity detection on the extracted character information to determine whether the word is complete; if the word is detected to be incomplete, performing scanning adjustment using a preset scanning adjustment method; updating the word scanning images after completing the scanning adjustment; if the word is detected to be complete, generating demand query words based on the word contours. 7.The intelligent reading recommendation method for assisting learning according to claim 6, wherein, The scanning adjustment method comprises: collecting adjustment switching signals; When the adjustment switching signal is a preset automatic adjustment signal, stroke thickness of a word in the word scanning image is recognized to obtain a stroke thickness parameter; The stroke thickness parameter and a preset reference stroke thickness parameter are combined to calculate a current text scale; The current text scale is used to generate a focal length adjustment parameter; The focal length adjustment parameter is used to adjust the focal length of a preset scanning device, thereby completing scanning adjustment. 8.The intelligent reading recommendation method for assisting learning of claim 7, wherein, The scanning adjustment method further includes: When the adjustment switching signal is a preset manual adjustment signal, a manual adjustment mode of the scanning device is started, and an image scanning scale is collected; The image scanning scale is used to generate an illumination parameter; The scanning device is controlled to emit auxiliary illumination light according to the illumination parameter, and an illumination coverage image is collected; The illumination coverage image is used to determine whether the auxiliary illumination light completely covers the word to be recognized; If not, a preset adjustment prompting method is used for adjustment prompting; If yes, an adjustment completion signal is generated; The word scanning image is updated in response to the adjustment completion signal. 9.The intelligent reading recommendation method for assisting learning of claim 8, wherein, The adjustment prompting method includes: The illumination range of the auxiliary illumination light in the illumination coverage image is recognized to obtain a current illumination range, and the word to be recognized in the illumination coverage image is range-recognized to obtain a required coverage range; When the current illumination range does not exceed the required coverage range, the current illumination range and the required coverage range are combined to generate an illumination adjustment parameter; The illumination adjustment parameter is used to generate adjustment prompting information, and adjustment prompting is performed based on the adjustment prompting information; When the current illumination range exceeds the required coverage range, the illumination coverage image and a preset character feature are combined to analyze a coverage missing position; The coverage missing position is used to generate direction prompting information, and adjustment prompting is performed based on the direction prompting information.

10. An intelligent reading recommendation system for assisting learning, characterized by, It includes: The acquisition module is used to collect account login information and a word scanning image; The memory is used to store a program for implementing the intelligent reading recommendation method for auxiliary learning according to any one of claims 1 to 9; The processor is used to load and execute the program stored in the memory.

Images