Smart pen based on data collection

Abstract

The present invention provides a smart pen based on data collection, which relates to the field of smart pen technology. The smart pen includes: a controller, a communication module, a pressure sensor, a high-speed camera, an inertial navigation unit, an audio module, and an ink cartridge. The controller is used to: receive position information, pressure sensing signals, and reference content at multiple moments obtained from the high-speed camera, the inertial navigation unit, and the pressure sensor; determine a glyph similarity score, a writing style similarity score, and a writing proficiency score; obtain recognition results and send them to a server via the communication module, and play prompt messages fed back by the server via the audio module. According to the present invention, when a user writes with the smart pen, the glyph similarity, writing style similarity, and writing proficiency of the written content can be detected, and prompt messages can be generated and played, which helps the user understand the actual writing situation and improves the effect and efficiency of the user's writing practice using the smart pen.

Description

Technical Field

[0001] The present invention relates to the technical field of smart pens, and in particular to a smart pen based on data collection. Background Art

[0002] In related art, CN106739662B discloses a conference reminder recording pen, comprising a refill, a nib, a barrel, and a cap. The pen barrel is characterized by a dot matrix light source embedded in the barrel. The dot matrix light source, when at a certain brightness, can be refracted through the barrel. The front end of the barrel is connected to the nib, and the rear end has a sealing plate. A processor and a circular sliding rheostat are fixed to the sealing plate. The slider of the sliding rheostat is fixed to the cap. The fixed end of the sliding rheostat is electrically connected to the dot matrix light source, and the slider end is electrically connected to a power source. Data information requiring reminders is stored in the processor. The exterior of the cap has a read operation key for operating the data in the processor. The data information stored in the processor is displayed via the dot matrix light source. This solution has the following beneficial effects: While it looks like an ordinary signature pen, it can provide reminders to the speaker without others knowing, allowing for timely recording of important matters during client negotiations.

[0003] CN110539574A discloses a self-study pen device that helps students learn efficiently, comprising a connected writing pen segment and a reading pen segment. The writing pen segment comprises a first housing and a refill set disposed within the first housing, and the reading pen segment comprises a second housing and a reading head disposed at one end of the second housing. The reading head comprises an optical module circuit for identifying invisible codes on books, and the first and second housings are detachably connected. The present invention designs a detachably connected writing pen segment and reading pen segment. The writing pen segment enables writing and drawing, while the reading pen segment enables reading. This combines a traditional writing pen and a reading pen, overcoming the functional shortcomings of a separate writing pen or reading pen. Unlike traditional reading pens that store content on a local storage card, the smart writing pen manages media resources through an internet cloud platform, enabling real-time updates, modifications, and management of content.

[0004] When users of smart pens write, they may encounter problems such as irregular strokes, incorrect stroke order, or taking too long to write. According to relevant technologies, the user can only interact with the user through voice and other means, and it is difficult to detect the user's actual writing situation or generate targeted prompt messages, resulting in low quality of writing practice using the smart pen.

[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the Invention

[0006] The present invention provides a smart pen based on data collection, which can solve the technical problems that related technologies are difficult to determine the actual writing situation of the smart pen user and difficult to generate targeted prompt messages.

[0007] According to a first aspect of the present invention, there is provided a smart pen based on data collection, comprising: a controller, a communication module, a pressure sensor, a high-speed camera, an inertial navigation unit, an audio module, a battery, a charger, and an ink cartridge;

[0008] The high-speed camera is arranged at the front end of the smart pen to shoot the identification code on the code paper and transmit the captured video frame to the controller for identification;

[0009] The ink cartridge is used for writing in the area corresponding to the identification code, and the tip of the ink cartridge is located at the front end of the smart pen;

[0010] The pressure sensor is arranged at the rear end of the ink cartridge and is close to the rear end of the ink cartridge, and is used to generate a pressure sensing signal when the tip of the ink cartridge contacts the paving paper and generates pressure on the pressure sensor through the rear end of the ink cartridge, and send the pressure sensing signal to the controller;

[0011] The inertial navigation unit is used to determine the position information of the front end of the smart pen when writing;

[0012] The battery is used to provide power to the smart pen, and the charger is used to replenish power to the battery;

[0013] The controller is used to identify the written content through the pressure sensing signal and the position information, obtain the recognition result, and send it to the server through the communication module, and receive the prompt message corresponding to the recognition result fed back by the server, and play the prompt message through the audio module.

[0014] Technical Effect: According to the present invention, based on the position information of the front end of the smart pen at multiple moments received from the inertial navigation unit, the pressure sensing signal sent by the pressure sensor, and the video frames captured by the high-speed camera, the position information, pressure sensing signal, and reference content at multiple moments can be obtained. The controller can then determine the glyph similarity score, writing style similarity score, and writing proficiency score, and then determine the recognition result of the written content. The result is sent to the server and the prompt message received from the server is played at the same time. This can provide more accurate information such as glyph similarity, writing style similarity, and writing proficiency of the written content, helping users understand the writing situation and conduct more targeted writing exercises, thereby improving the effectiveness and efficiency of users' writing practice with the smart pen. The recognition result of the identification code can be sent to a server via a communication module, and reference content corresponding to the recognition result can be received from the server. A coordinate system is established when the smart pen user begins writing. Based on an inertial navigation unit and a pressure sensor, first position information of the front end of the smart pen at each moment while the user is writing, as well as second position information at each moment during the period when the front end of the smart pen leaves the paper between each stroke, is determined based on a preset duration threshold and a preset pressure threshold. This provides basic data for determining the recognition result of the written content. When determining a glyph similarity score, reference glyph information can be obtained based on the reference content. Trajectory information and the starting position of multiple writing strokes can be determined based on a preset pressure threshold. Reference glyph trajectory information and a reference glyph writing range can also be determined based on the reference glyph information. Target sampling points are determined based on sampling points set on the glyph reference trajectory and a preset neighborhood range. The degree of standardization of all trajectory information written by the smart pen user and the completeness of the glyph information written by the smart pen user are then determined. A glyph similarity score is then derived by combining these two aspects, improving the accuracy, comprehensiveness, and objectivity of the glyph similarity score. When determining a writing style similarity score, the glyph reference trajectory information can be determined based on the reference glyph information, and the standard writing trajectory of each stroke in the glyph reference trajectory information and the standard writing order of each stroke can be determined based on the reference writing style information. The first identification vector and the second identification vector are then sorted in a unified manner to avoid the problem of incorrect judgment of the strokes corresponding to the trajectory information due to the writing direction of the trajectory information being different from the writing direction of the standard writing trajectory, thereby improving the accuracy of determining the strokes corresponding to each trajectory information. In addition, the standard writing vector in the standard writing trajectory and the writing trajectory vector in the trajectory information can be determined, and then the writing style similarity score is obtained by using the strokes corresponding to the trajectory information, the standard writing vector, and the writing trajectory vector. This takes into account both the correctness of the writing style of the trajectory information and the correctness of the writing order of the trajectory information, so that the writing style similarity score can accurately represent the correctness of the writing style, thereby improving the accuracy and objectivity of the writing style similarity score.When determining a writing proficiency score, the first location information, the second location information, and the reference time information can be used to determine the degree of similarity between the actual writing duration information and the reference time information, thereby obtaining a writing proficiency score. This allows the writing proficiency score to accurately and objectively represent writing proficiency. Furthermore, based on the glyph similarity score, the writing style similarity score, and the writing proficiency score, the recognition results of the written content can be sent to a server, and prompt messages from the server can be received and played, thereby helping the smart pen user understand the actual writing status of the written content.

[0015] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and not limiting of the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can derive other embodiments based on these drawings without inventive efforts.

[0017] Figure 1 A schematic diagram of a smart pen based on data collection according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0019] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0020] Figure 1 A schematic diagram of a data collection-based smart pen according to an embodiment of the present invention is exemplarily shown, wherein the smart pen includes: a controller, a communication module, a pressure sensor, a high-speed camera, an inertial navigation unit, an audio module, a battery, a charger, and an ink cartridge;

[0021] The high-speed camera is arranged at the front end of the smart pen to shoot the identification code on the code paper and transmit the captured video frame to the controller for identification;

[0022] The ink cartridge is used for writing in the area corresponding to the identification code, and the tip of the ink cartridge is located at the front end of the smart pen;

[0023] The pressure sensor is arranged at the rear end of the ink cartridge and is close to the rear end of the ink cartridge, and is used to generate a pressure sensing signal when the tip of the ink cartridge contacts the paving paper and generates pressure on the pressure sensor through the rear end of the ink cartridge, and send the pressure sensing signal to the controller;

[0024] The inertial navigation unit is used to determine the position information of the front end of the smart pen when writing;

[0025] The battery is used to provide power to the smart pen, and the charger is used to replenish power to the battery;

[0026] The controller is used to identify the written content through the pressure sensing signal and the position information, obtain the recognition result, and send it to the server through the communication module, and receive the prompt message corresponding to the recognition result fed back by the server, and play the prompt message through the audio module.

[0027] According to an embodiment of the present invention, a data-collecting smart pen can obtain multiple moments of position information, pressure sensing signals, and reference content based on the received position information of the smart pen's front end transmitted by an inertial navigation unit, pressure sensing signals transmitted by a pressure sensor, and video frames captured by a high-speed camera. A controller is then used to determine a glyph similarity score, a writing style similarity score, and a writing proficiency score, thereby determining the recognition result of the written content. The result is then sent to a server and a prompt message received from the server is played simultaneously. This provides more accurate information on glyph similarity, writing style similarity, and writing proficiency of the written content, helping users understand their writing situation and conduct more targeted writing practice, thereby improving the effectiveness and efficiency of writing practice using the smart pen.

[0028] According to an embodiment of the present invention, the high-speed camera in the smart pen is arranged at the front end of the smart pen. Since the identification code on the coding paper is a very small QR code, an ordinary camera cannot capture the identification code on the coding paper. Therefore, when the user writes with the smart pen, the identification code on the coding paper can be captured by the high-speed camera arranged at the front end of the smart pen, and the captured video frame is transmitted to the controller for recognition.

[0029] According to an embodiment of the present invention, the pressure sensor is positioned at the rear end of the ink cartridge and in close proximity to the rear end. When a user writes with the smart pen, the tip of the ink cartridge contacts the paper, exerting a certain amount of pressure on the tip. The ink cartridge is cylindrical and positioned within the smart pen, with the tip located at the front end. When writing, pressure is transmitted from the front end of the ink cartridge to the rear end, whereupon the rear end exerts pressure on the pressure sensor, generating a pressure sensing signal that is transmitted to the controller.

[0030] According to an embodiment of the present invention, the inertial navigation unit is used to determine the position of the tip of the smart pen during writing. For example, when a user writes with the smart pen, they can use the ink cartridge within the smart pen to write in the area corresponding to the identification code. Simultaneously, the inertial navigation unit can record the position of the tip of the smart pen at each moment. By connecting this position information, the user can obtain the writing trajectory of the smart pen.

[0031] According to an embodiment of the present invention, the controller is used to identify the written content through the pressure sensing signal and the position information, obtain the recognition result, send it to the server through the communication module, and receive the prompt message corresponding to the recognition result fed back by the server, and play the prompt message through the audio module to help the user understand the writing situation.

[0032] According to an embodiment of the present invention, identifying written content using the pressure sensing signal and the position information to obtain a recognition result includes: identifying the identification code in the area corresponding to the identification code and sending the recognition result of the identification code to a server via the communication module; receiving reference content corresponding to the recognition result from the server; when the received pressure sensing signal is greater than or equal to a preset pressure threshold for the first time, establishing a coordinate system using the position information of the front end of the smart pen as the origin, wherein an xoy plane of the coordinate system is parallel to the horizontal direction and a z axis is parallel to the vertical direction; receiving the position information of the front end of the smart pen and the pressure sensing signal from the pressure sensor at multiple times from the inertial navigation unit; stopping receiving the pressure sensing signal and the position information when the duration of the pressure sensing signal being less than the preset pressure threshold is greater than or equal to a preset duration threshold; and determining the recognition result of the written content based on the position information, pressure sensing signal, and reference content received at the multiple times.

[0033] According to an embodiment of the present invention, the identification code in the area corresponding to the identification code is recognized, and the recognition result of the identification code is sent to the server via the communication module. The server can then generate reference content corresponding to the recognition result. For example, if the recognition result in the area corresponding to the identification code is the English letter "A", after the recognition result is sent to the server via the communication module, the server can generate information such as the standard writing method of the English letter "A", i.e., reference content corresponding to the recognition result, and send the reference content corresponding to the recognition result to the controller. Since the user does not apply pressure to the ink cartridge or applies very little pressure when not writing, when the received pressure sensor signal is greater than or equal to the preset pressure threshold for the first time, it can be considered that the user has applied pressure to the smart pen for the first time, and the ink cartridge in the smart pen will begin to release ink, that is, the user has started writing with the smart pen. At this time, the position information of the front end of the smart pen is used as the coordinate origin, that is, the position where the smart pen first contacts the paper when the user starts writing is used as the coordinate origin to establish a three-dimensional coordinate system. The xoy plane of the above three-dimensional coordinate system is parallel to the horizontal direction, that is, the plane where the paper is located, and the z axis is parallel to the vertical direction, that is, perpendicular to the plane where the paper is located.

[0034] According to an embodiment of the present invention, at multiple moments, for example, every 100 milliseconds, the coordinate information of the tip of the smart pen in the three-dimensional coordinate system, i.e., the position information of the tip of the smart pen, and the pressure sensing signal from the pressure sensor are received. The present invention does not limit the time intervals between these moments. Since when the user lifts the tip of the smart pen off the paper during writing (for example, when finishing one stroke of a character and lifting the smart pen to write the next), no pressure or very little pressure is generated, this can be considered to be the case if the pressure sensing signal is less than a preset pressure threshold for a short period of time (i.e., less than a preset duration threshold). If the pressure sensing signal is less than the preset pressure threshold for a period greater than or equal to a preset duration threshold, it can be considered that the user has not written for a long time, i.e., the user has stopped writing. At this point, receiving the pressure sensing signal and the position information is stopped. For example, if the pressure sensing signal is less than the preset pressure threshold for a period greater than or equal to a preset duration threshold of 1 second, receiving the pressure sensing signal and the position information can be stopped. The present invention does not limit the preset pressure threshold or the preset duration threshold.

[0035] According to an embodiment of the present invention, the position information includes: first position information for each moment within a time period when the pressure sensing signal is greater than or equal to a preset pressure threshold, and second position information for each moment within a time period when the pressure sensing signal is less than the preset pressure threshold. Specifically, from the time the received pressure sensing signal first becomes greater than or equal to the preset pressure threshold to the time the pressure sensing signal remains less than the preset pressure threshold for a duration greater than or equal to a preset duration threshold, the pressure sensing signal may be greater than or equal to the preset pressure threshold multiple times and less than the preset pressure threshold multiple times. The first position information for each moment within the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold is obtained, representing the position information of the front end of the smart pen at each moment while the user is writing a stroke. The second position information for each moment within the time period when the pressure sensing signal is less than the preset pressure threshold is representing the position information of the front end of the smart pen at each moment while the user is lifting the pen between strokes. During the writing process, between completing one stroke and writing the next, the tip of the smart pen may separate from the paper, generating a weaker pressure sensing signal. The inertial navigation unit then continues to determine the position of the tip of the smart pen. Using this position information during this time period, the inertial navigation unit can determine the next stroke's starting position, i.e., the next stroke's positional relationship with the previously written stroke. The time interval between these moments can be 100 milliseconds, and the present invention imposes no limitation on this time interval.

[0036] In this way, the recognition result of the identification code can be sent to the server through the communication module, and the reference content corresponding to the recognition result sent by the server is received. When the user of the smart pen starts writing, a coordinate system is established. Based on the inertial navigation unit and the pressure sensor, according to the preset time threshold and the preset pressure threshold, the first position information of the front end of the smart pen at each moment when the user is writing and the second position information of each moment in the time period when the front end of the smart pen leaves the coded paper between each stroke are determined, providing basic data for determining the recognition result of the written content.

[0037] According to an embodiment of the present invention, based on the position information, pressure sensing signals, and reference content received at multiple moments, a recognition result of the written content is determined, including: determining reference glyph information, reference writing style information, and reference time information for writing the reference content based on the reference content; determining a glyph similarity score based on the first position information, the second position information, and the reference glyph information; determining a writing style similarity score based on the first position information, the second position information, the reference glyph information, and the reference writing style information; determining a writing proficiency score based on the first position information, the second position information, and the reference time information; and determining a recognition result of the written content based on the glyph similarity score, the writing style similarity score, and the writing proficiency score. For example, based on the reference content of the English letter "A," the standard glyph, standard stroke order, and standard writing style of the corresponding character, as well as the theoretical time required to write the corresponding reference content, i.e., the reference glyph information, reference writing style information, and reference time information for writing the reference content of the corresponding character, can be determined.

[0038] According to an embodiment of the present invention, a glyph similarity score is determined based on the first position information, the second position information and the reference glyph information, including: determining the trajectory information of the first writing based on the first position information at each moment within the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the first time; determining the starting position of the second writing based on the second position information at the last moment within the time period when the pressure sensing signal is less than the preset pressure threshold for the first time; determining the trajectory information of the second writing based on the starting position of the second writing and the first position information at each moment within the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the second time; determining the trajectory information of the i+th writing based on the second position information at the last moment within the time period when the pressure sensing signal is less than the preset pressure threshold for the i-th time. The starting position of 1 writing, where i≥2 and i is a positive integer; determining the trajectory information of the i+1th writing based on the starting position of the i+1th writing and the first position information of each moment in the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the i+1th time; determining the glyph reference trajectory information and the reference glyph writing range based on the reference glyph information, wherein the reference glyph writing range is used to represent a reasonable range of the writing trajectory of each stroke of the reference glyph, the glyph reference trajectory information is used to represent the standard writing trajectory of each stroke of the reference glyph, and the glyph reference trajectory information is completely within the reference glyph writing range; determining the glyph similarity score based on the trajectory information of multiple writings, the glyph reference trajectory information and the reference glyph writing range.

[0039] According to an embodiment of the present invention, the moment when the pressure sensing signal first exceeds or equals a preset pressure threshold can be used as the time when the smart pen user begins writing. Therefore, the time period during which the pressure sensing signal first exceeds or equals the preset pressure threshold is the time period during which the smart pen user writes the first stroke. Therefore, the stroke trajectory of the first stroke, i.e., the first writing trajectory information, can be determined based on the first position information at each moment during the time period during which the pressure sensing signal first exceeds or equals the preset pressure threshold. The time period during which the pressure sensing signal first falls below the preset pressure threshold is the time period between when the smart pen user completes the first stroke and when they begin writing the second stroke. Therefore, the last moment during the time period during which the pressure sensing signal first falls below the preset pressure threshold can be used as the time when the smart pen user begins writing the second stroke. Therefore, the second position information at the last moment during the time period during which the pressure sensing signal first falls below the preset pressure threshold is the starting position of the second writing, in other words, the starting position of the second stroke. Similarly, the trajectory of the second stroke of the writing can be determined based on the starting position of the second writing and the first position information at each moment during the time period when the pressure sensor signal is greater than or equal to the preset pressure threshold for the second time, i.e., the trajectory information for the second writing. Similarly, the starting position of the (i+1)th writing can be determined based on the second position information at the last moment during the time period when the pressure sensor signal is less than the preset pressure threshold for the i-th time, i.e., the starting position of the (i+1)th writing stroke can be used as the starting position of the (i+1)th writing stroke. Furthermore, the trajectory of the (i+1)th stroke of the writing content can be determined based on the starting position of the (i+1)th writing and the first position information at each moment during the time period when the pressure sensor signal is greater than or equal to the preset pressure threshold for the i+1th time, i.e., the trajectory information for the (i+1)th writing stroke. Based on the above process, the trajectory of all strokes written by the smart pen user, i.e., the complete trajectory information for the writing, can be obtained.

[0040] According to an embodiment of the present invention, glyph reference trajectory information can be determined based on the reference glyph information, which can serve as a standard writing trajectory for written content. Since some deviations from the reference glyph information may occur when a smart pen user writes, but this does not affect the overall glyph shape, a reference glyph writing range can be determined based on the reference glyph information. The reference glyph writing range is a reasonable range set for the trajectory of each stroke, and the closest distance from each position in the glyph reference trajectory information to the edge of the reference glyph writing range is equal. The present invention does not impose any restrictions on the determination of the reference glyph writing range. The glyph reference trajectory information should be completely within the reference glyph writing range. When the trajectory information of each stroke written by the smart pen user is within the reference glyph writing range, the trajectory information of each stroke written by the smart pen user is within a reasonable writing range, and each stroke written by the smart pen user can be considered to be relatively consistent with the corresponding reference glyph information.

[0041] According to an embodiment of the present invention, based on the trajectory information of multiple writings, the glyph reference trajectory information and the reference glyph writing range, a glyph similarity score is determined, including: evenly setting sampling points on the glyph reference trajectory information; determining a target sampling point of the trajectory information within a preset neighborhood range centered on the sampling point; and determining the glyph similarity score S according to formula (1). F ,

[0042]

[0043] Among them, L tr，j is the jth written trajectory information, n is the number of trajectory information, Ω R N is the reference glyph writing range. M is the number of target sampling points, N is the number of sampling points, len(*) is the length function, j≤n, and both j and n are positive integers.

[0044] According to an embodiment of the present invention, sampling points are evenly set on the glyph reference trajectory information. For example, a sampling point is set every 2 mm on the glyph reference trajectory information. The present invention does not impose any restrictions on the spacing distance between sampling points. The circular range determined with each sampling point as the center can be used as a preset neighborhood range. When the trajectory information exists within the neighborhood range, it means that the trajectory written by the smart pen user exists within the neighborhood range. Therefore, the corresponding sampling points with trajectory information within the neighborhood range can be identified as target sampling points and used to determine the glyph similarity. If the trajectory information does not exist within the neighborhood range, it means that the trajectory written by the smart pen user does not exist within the neighborhood range. The user's writing trajectory may deviate greatly from the reference trajectory information, or the user's writing trajectory information is incomplete and not all strokes are written.

[0045] According to an embodiment of the present invention, in formula (1), since the writing trajectory information of the smart pen user is a point set, the writing range of the reference glyph is also a point set, therefore, (L tr，j ∩Ω R ) can represent the intersection of the j-th writing trajectory information and the reference glyph writing range, and then the sum of the lengths of the intersections of all the trajectory information written by the smart pen user and the reference glyph writing range can be obtained. therefore, It can represent the ratio between the sum of the lengths of the intersection of all the trajectory information written by the smart pen user and the reference glyph writing range and the sum of the lengths of all the trajectory information written by the smart pen user, that is, the proportion of all the trajectory information that is within the reference self-writing range. The closer the above ratio information is to 1, the higher the proportion of all the trajectory information that is within the reference glyph writing range, that is, the more standardized the trajectory information is. Therefore, the above ratio information can be used as the standardization degree of the glyph information written by the user. Since the user may only have a partial writing trajectory when writing, that is, the glyph is incomplete, but all or most of the written trajectory information is within the reference glyph writing range, resulting in the above ratio information being high, but the similarity between the trajectory information and the glyph reference trajectory information is not high. Therefore, it is necessary to further judge the completeness of the glyph information written by the smart pen user. In formula (1), represents the ratio of target sampling points to the number of sampling points. The larger this ratio, the greater the proportion of trajectory information written by the smart pen user within the neighborhood of each sampling point. In other words, the greater the proportion of target sampling points, the higher the completeness of the glyph information written by the smart pen user. Therefore, the ratio of target sampling points to sampling points can be used as the completeness of the glyph information written by the smart pen user. The product of the degree of standardization of all trajectory information written by the smart pen user and the completeness of the glyph information written by the smart pen user can be used as the glyph similarity score. The higher the glyph similarity score, the more similar the glyph information written by the smart pen user is to the reference glyph information.

[0046] In this way, reference glyph information can be obtained based on the reference content, and the trajectory information of multiple writings and the starting position of writing can be determined based on a preset pressure threshold. At the same time, the glyph reference trajectory information and the reference glyph writing range can be determined based on the reference glyph information. Based on the sampling points set on the glyph reference trajectory and the preset neighborhood range, the target sampling point can be determined, and then the degree of standardization of all trajectory information written by the smart pen user and the completeness of the glyph information written by the smart pen user can be obtained, thereby comprehensively combining the two aspects to obtain a glyph similarity score, thereby improving the accuracy, comprehensiveness and objectivity of the glyph similarity score.

[0047] According to an embodiment of the present invention, a writing style similarity score is determined based on the first position information, the second position information, the reference glyph information and the reference writing style information, including: determining the glyph reference trajectory information based on the reference glyph information; determining the standard writing trajectory of each stroke in the glyph reference trajectory information and the standard writing order of each stroke based on the reference writing style information; determining the first minimum bounding rectangle of the standard writing trajectory of multiple strokes and the first centroid of the first minimum bounding rectangle based on the glyph reference trajectory information; determining the second minimum bounding rectangle of multiple trajectory information and the second centroid of the second minimum bounding rectangle ; Determine the strokes corresponding to each trajectory information according to the glyph reference trajectory information, the first centroid, the trajectory information and the second centroid; Divide the standard writing trajectory equally according to the number of moments included in the time period corresponding to the trajectory information corresponding to the standard writing trajectory to obtain multiple standard writing trajectory segments; Determine the standard writing vector corresponding to each standard writing trajectory segment according to the reference writing method information; Determine the writing trajectory vector in the trajectory information according to each first position information in the trajectory information corresponding to the standard writing trajectory; Determine the writing method similarity score according to the standard writing vector, the writing trajectory vector, the standard writing order and the sequence number of the trajectory information.

[0048] According to an embodiment of the present invention, the glyph reference trajectory information can be obtained based on the reference glyph information, which can be used as the standard writing trajectory of the written content. Based on the reference writing method information, the standard writing method of each stroke and the order of writing each stroke under the standard writing method can be separated, that is, the standard writing trajectory of each stroke and the standard writing order of each stroke. Based on the glyph reference trajectory information, the first minimum bounding rectangle of the standard writing trajectory of multiple strokes is determined. For example, if the glyph reference trajectory information is the glyph reference trajectory information of the English letter "A", then a minimum bounding rectangle that selects all the strokes of the English letter "A" can be determined, that is, the first minimum bounding rectangle of the standard writing trajectory of multiple strokes, and the centroid of the above-mentioned first minimum bounding rectangle is determined at the same time, which is the first centroid. Similarly, the second minimum bounding rectangle of the multiple trajectory information of the content written by the smart pen user can be determined, and the centroid of the second minimum bounding rectangle can be determined at the same time, which is the second centroid.

[0049] According to an embodiment of the present invention, based on the glyph reference trajectory information, the first centroid, the trajectory information, and the second centroid, the stroke corresponding to each trajectory information is determined, including: evenly setting a preset number of first identification points on the standard writing trajectory of each stroke in the glyph reference trajectory information; determining a first identification vector between each first identification point and the first centroid; evenly setting the preset number of second identification points on the trajectory information; determining a second identification vector between each second identification point and the second centroid; and obtaining a stroke recognition score SE of the jth trajectory information according to formula (2): j ,

[0050]

[0051] Among them, I 1，y，x is the first identification vector corresponding to the xth first identification point on the standard writing trajectory of the yth stroke, For I 1，y，x The transposed vector of ||I 1，y，x || is I 1，y，x The model, I 2，j，x is the second identification vector corresponding to the xth second identification point on the jth trajectory information, ||I 2，j，x || is I 2，j，x The modulus, n p is the preset number, n is the number of trajectory information, j≤n, y≤n, x≤n p , and j, n, y, n, x and n p are all positive integers; the stroke corresponding to the standard writing trajectory that maximizes the stroke recognition score of the j-th trajectory information is determined as the stroke corresponding to the j-th trajectory information.

[0052] According to an embodiment of the present invention, a preset number of first identification points are evenly set on the standard writing trajectory of each stroke in the glyph reference trajectory information. For example, if the glyph reference trajectory information is the glyph reference trajectory information of the English letter "A", then three first identification points can be evenly set on the standard writing trajectory of each stroke in the corresponding glyph reference trajectory information. The present invention does not impose any restrictions on the preset number of identification points. Then, a first identification vector between each first identification point and the first centroid is determined. The first identification vector can be obtained by connecting each first identification point with the first centroid. In order to facilitate the determination of the strokes corresponding to each trajectory information, the obtained first identification points can be sorted in order from top to bottom. If the order cannot be determined from top to bottom, then the first identification points can be sorted in order from left to right, so that the serial number of each first identification point can be determined. When determining the serial number of the second identification point, the determination method is the same as the determination method of the first identification point, thereby avoiding the problem of different serial numbers due to the different writing direction of the trajectory information and the writing direction of the standard writing trajectory, which in turn causes errors in the similarity calculation. For example, when the glyph reference trajectory information is the glyph reference trajectory information of the English letter "A", the first stroke therein is a line segment from the upper right end to the lower left end, and three first identification points are evenly set on the standard writing trajectory information corresponding to the first stroke. When the first identification vectors are sorted in the order from top to bottom and then from left to right, it can be considered that the order of the three first identification points evenly set on the standard writing trajectory information corresponding to the first stroke is: the order of the first identification point at the upper right end is 1, the order of the first identification point in the middle is 2, and the order of the first identification point at the lower left end is 3. Similarly, the order of each first identification point corresponding to a preset number of identification points on the standard writing trajectory information corresponding to the remaining two strokes in the English letter "A" can be determined, and the order of the first identification point in the stroke can be used as the serial number of the first identification vector corresponding to the first identification point. Similarly, a preset number of second identification points can be evenly set on the trajectory information written by the smart pen user, and the second identification vector between each second identification point and the second centroid can be determined, and then the sequence number of the second identification vector can be determined, so as to be used to determine the strokes in the glyph reference trajectory information corresponding to each trajectory information written by the smart pen user.

[0053] According to an embodiment of the present invention, in formula (2), The cosine similarity between the first identification vector corresponding to the xth first identification point on the standard writing trajectory of the yth stroke and the second identification vector corresponding to the xth second identification point on the jth trajectory information is averaged to obtain the similarity between the standard writing trajectory of the yth stroke and the jth trajectory information. Based on this item, the similarity between the jth trajectory information and the standard writing trajectory of each stroke can be obtained. The maximum value of the similarity between the jth trajectory information and the standard writing trajectory of each stroke can be used as the stroke recognition score of the jth trajectory information. The stroke corresponding to the standard writing trajectory corresponding to the maximum value of the similarity can be considered as the stroke corresponding to the standard writing trajectory corresponding to the jth trajectory information. In the example, it is assumed that when the smart pen user writes, the order of the strokes is different from the standard writing order. For example, the horizontal line segment of the letter "A" is taken as the first trajectory information, and the standard writing trajectory of the first stroke in the glyph reference trajectory information is the line segment from the upper right end to the lower left end, and the horizontal line segment is the standard writing trajectory of the third stroke in the glyph reference trajectory information. Based on the above solution to the maximum value of similarity, the standard writing trajectory of the stroke with the highest similarity to the first trajectory information is not the standard writing trajectory of the first stroke, but the standard writing trajectory of the third stroke.

[0054] According to an embodiment of the present invention, the standard writing trajectory is divided equally according to the number of moments included in the time period corresponding to the trajectory information corresponding to the standard writing trajectory, and a plurality of standard writing trajectory segments are obtained. For example, in the glyph reference trajectory information of the English letter "A", the trajectory information corresponding to the standard writing trajectory of the first stroke is a line segment from the upper right end to the lower left end. The length of the writing time period of the trajectory information is 1 second, and the position information is obtained every 100 milliseconds. Then the number of moments is 10. Therefore, 10 equalization points can be set on the standard writing trajectory of the first stroke to obtain 9 standard writing trajectory segments. Since each standard writing trajectory segment has the position information of the starting endpoint and the ending endpoint (that is, the position information of the above-mentioned equalization points), and the standard writing direction of each stroke is included in the reference writing method information, a standard writing vector corresponding to each standard writing trajectory segment can be formed. Because the first position information in the trajectory information corresponding to the standard writing trajectory is determined at each moment, each piece of first position information in the trajectory information corresponding to the standard writing trajectory divides the corresponding trajectory information. Therefore, the vector between two adjacent pieces of first position information can be determined based on the two adjacent pieces of first position information. For example, the vector pointing from the first first position information to the second first position information in the trajectory information corresponding to the standard writing trajectory of the first stroke can be obtained. Similarly, the writing trajectory vector in the trajectory information of each stroke can be obtained. This can be used to determine the degree of similarity between the standard writing trajectory corresponding to the trajectory information of the smart pen user and the writing trajectory information.

[0055] According to an embodiment of the present invention, the writing style similarity score is determined based on the standard writing vector, the writing trajectory vector, the standard writing order and the sequence number of the trajectory information, including: determining the writing style similarity score S according to formula (3) W ,

[0056]

[0057] Among them, V j，k is the kth writing trajectory vector in the jth writing trajectory information, j S is the sequence number of the standard writing order of the strokes corresponding to the j-th writing trajectory information, V j，k The transposed vector of ||V j，k || is V j，k The model, V j，k，S is the standard writing vector corresponding to the kth standard writing trajectory segment in the standard writing trajectory corresponding to the jth writing trajectory information, ||V j，k，s || is V j，k，s The modulus, n j is the number of writing trajectory vectors in the trajectory information of the jth writing, n j Equal to the number of first position information in the j-th written track information minus 1, n is the number of track information, k≤n j , j≤n, and k, n j , j and n are all positive integers.

[0058] According to an embodiment of the present invention, in formula (3), The cosine similarity between the standard writing vector corresponding to the kth standard writing trajectory segment in the standard writing trajectory corresponding to the trajectory information of the jth writing and the kth writing trajectory vector in the trajectory information of the jth writing is represented. The above cosine similarities are averaged to obtain the similarity between each trajectory information and its corresponding standard writing trajectory.

[0059] According to an embodiment of the present invention, in formula (3), |jj S | can represent the difference between the sequence number of the standard writing order of the strokes corresponding to the trajectory information of the j-th writing and the sequence number of the actual writing order of the strokes corresponding to the trajectory information of the j-th writing. The degree of similarity between the sequence numbers of the standard writing order of the strokes corresponding to the j-th writing trajectory information and the sequence numbers of the actual writing order of the strokes corresponding to the j-th writing trajectory information can also be used as the standard degree of the order of the j-th writing trajectory information. The standard degree of the order of the j-th writing trajectory information can be used as a weight to weight the similarity between the j-th writing trajectory information and its corresponding standard writing trajectory. The similarity between multiple writing trajectory information and their corresponding standard writing trajectory can be weighted in the same manner, and a weighted average is obtained to determine a writing style similarity score. If the writing style similarity score is high, it can be indicated that the writing style of each piece of trajectory information is highly correct, and the writing order of each piece of trajectory information is highly correct.

[0060] In this way, the glyph reference trajectory information can be determined based on the reference glyph information, and the standard writing trajectory of each stroke in the glyph reference trajectory information and the standard writing order of each stroke can be determined based on the reference writing method information. The first identification vectors and the second identification vectors can be sorted in a unified manner to avoid the problem of incorrect judgment of the strokes corresponding to the trajectory information due to the writing direction of the trajectory information being different from the writing direction of the standard writing trajectory, thereby improving the accuracy of determining the strokes corresponding to each trajectory information. In addition, the standard writing vector in the standard writing trajectory and the writing trajectory vector in the trajectory information can be determined, and then the writing method similarity score is obtained by using the strokes corresponding to the trajectory information, the standard writing vector, and the writing trajectory vector. This takes into account both the correctness of the writing method of the trajectory information and the correctness of the writing order of the trajectory information, so that the writing method similarity score can accurately represent the correctness of the writing method, thereby improving the accuracy and objectivity of the writing method similarity score.

[0061] According to an embodiment of the present invention, determining a writing proficiency score based on the first position information, the second position information and the reference time information includes: determining a writing proficiency score S according to formula (4): S ,

[0062]

[0063] Where, Δt j is the duration of the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the jth time, Δt R is the reference time information, n is the number of trajectory information, j≤n, and both j and n are positive integers.

[0064] According to an embodiment of the present invention, in formula (4), It can represent the total duration of the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold, that is, the actual total writing duration information. Therefore, It can indicate the difference between the actual total writing time and the reference time information. The actual writing time information can be used to indicate the similarity between the reference time information and the actual writing time information, and can be used as a writing proficiency score. The higher the writing proficiency score, the closer the actual writing time of the smart pen user is to the reference time information, and the higher the writing proficiency of the smart pen user.

[0065] In this way, the degree of similarity between the actual writing duration information and the reference duration information can be determined based on the first position information, the second position information, and the reference duration information, thereby obtaining a writing proficiency score. This allows the writing proficiency score to accurately and objectively represent writing proficiency.

[0066] According to an embodiment of the present invention, a recognition result of written content can be determined based on a glyph similarity score, a writing style similarity score, and a writing proficiency score. The recognition result can describe the glyph similarity, writing style similarity, and writing proficiency of the content written by the smart pen user. These three scores can also be weighted and summed to obtain an overall score as the recognition result. The recognition result is sent to a server via a communication module, and a prompt message corresponding to the recognition result is received from the server. The prompt message is played via the audio module to help the smart pen user understand the actual writing status of the written content. For example, the prompt message may include a description of the overall correctness of the user's writing (i.e., a description generated based on the overall score); a description of glyph similarity (i.e., a description derived based on the glyph similarity score); a description of writing style correctness (i.e., a description derived based on the writing style similarity); and a description of writing proficiency (i.e., a description derived based on the writing proficiency score).

[0067] In this way, the recognition results of the written content can be sent to the server based on the glyph similarity score, writing style similarity score and writing proficiency score, and the prompt message feedback from the server can be received and played, thereby helping the smart pen user understand the actual writing situation of the written content.

[0068] According to an embodiment of the present invention, a data-collecting smart pen can obtain multiple moments of position information, pressure sensing signals, and reference content based on the received position information of the smart pen's front end transmitted by an inertial navigation unit, pressure sensing signals transmitted by a pressure sensor, and video frames captured by a high-speed camera. A controller is then used to determine a glyph similarity score, a writing style similarity score, and a writing proficiency score, thereby determining the recognition result of the written content. The result is then sent to a server and a prompt message received from the server is played simultaneously. This provides more accurate information on glyph similarity, writing style similarity, and writing proficiency of the written content, helping users understand their writing situation and conduct more targeted writing practice, thereby improving the effectiveness and efficiency of writing practice using the smart pen. The recognition result of the identification code can be sent to a server via a communication module, and reference content corresponding to the recognition result can be received from the server. A coordinate system is established when the smart pen user begins writing. Based on an inertial navigation unit and a pressure sensor, first position information of the front end of the smart pen at each moment while the user is writing, as well as second position information at each moment during the period when the front end of the smart pen leaves the paper between each stroke, is determined based on a preset duration threshold and a preset pressure threshold. This provides basic data for determining the recognition result of the written content. When determining a glyph similarity score, reference glyph information can be obtained based on the reference content. Trajectory information and the starting position of multiple writing strokes can be determined based on a preset pressure threshold. Reference glyph trajectory information and a reference glyph writing range can also be determined based on the reference glyph information. Target sampling points are determined based on sampling points set on the glyph reference trajectory and a preset neighborhood range. The degree of standardization of all trajectory information written by the smart pen user and the completeness of the glyph information written by the smart pen user are then determined. A glyph similarity score is then derived by combining these two aspects, improving the accuracy, comprehensiveness, and objectivity of the glyph similarity score. When determining a writing style similarity score, the glyph reference trajectory information can be determined based on the reference glyph information, and the standard writing trajectory of each stroke in the glyph reference trajectory information and the standard writing order of each stroke can be determined based on the reference writing style information. The first identification vector and the second identification vector are then sorted in a unified manner to avoid the problem of incorrect judgment of the strokes corresponding to the trajectory information due to the writing direction of the trajectory information being different from the writing direction of the standard writing trajectory, thereby improving the accuracy of determining the strokes corresponding to each trajectory information. In addition, the standard writing vector in the standard writing trajectory and the writing trajectory vector in the trajectory information can be determined, and then the writing style similarity score is obtained by using the strokes corresponding to the trajectory information, the standard writing vector, and the writing trajectory vector. This takes into account both the correctness of the writing style of the trajectory information and the correctness of the writing order of the trajectory information, so that the writing style similarity score can accurately represent the correctness of the writing style, thereby improving the accuracy and objectivity of the writing style similarity score.When determining a writing proficiency score, the first location information, the second location information, and the reference time information can be used to determine the degree of similarity between the actual writing duration information and the reference time information, thereby obtaining a writing proficiency score. This allows the writing proficiency score to accurately and objectively represent writing proficiency. Furthermore, based on the glyph similarity score, the writing style similarity score, and the writing proficiency score, the recognition results of the written content can be sent to a server, and prompt messages from the server can be received and played, thereby helping the smart pen user understand the actual writing status of the written content.

[0069] Those skilled in the art will appreciate that the embodiments of the present invention described above and shown in the accompanying drawings are intended to be illustrative only and are not intended to limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and illustrated in the embodiments. Any variations or modifications may be made to the embodiments of the present invention without departing from the principles described.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart pen based on data collection, characterized in that: include: Controller, communication module, pressure sensor, high-speed camera, inertial navigation unit, audio module, battery, charger and ink cartridge; The high-speed camera is arranged at the front end of the smart pen to shoot the identification code on the code paper and transmit the captured video frame to the controller for identification; The ink cartridge is used for writing in the area corresponding to the identification code, and the tip of the ink cartridge is located at the front end of the smart pen; The pressure sensor is arranged at the rear end of the ink cartridge and is close to the rear end of the ink cartridge, and is used to generate a pressure sensing signal when the tip of the ink cartridge contacts the paper and generates pressure on the pressure sensor through the rear end of the ink cartridge, and send it to the controller; The inertial navigation unit is used to determine the position information of the front end of the smart pen when writing; The battery is used to provide power to the smart pen, and the charger is used to replenish power to the battery; The controller is configured to recognize the written content using the pressure sensing signal and the position information, obtain a recognition result, and send the recognition result to the server via the communication module, receive a prompt message corresponding to the recognition result from the server, and play the prompt message via the audio module; Recognizing the written content using the pressure sensing signal and the position information to obtain a recognition result includes: Identify the identification code in the area corresponding to the identification code, and send the identification result of the identification code to the server through the communication module; receiving reference content corresponding to the recognition result sent by the server; When the received pressure sensing signal is greater than or equal to a preset pressure threshold for the first time, establishing a coordinate system using the position information of the front end of the smart pen as the origin, wherein the xoy plane of the coordinate system is parallel to the horizontal direction and the z axis is parallel to the vertical direction; receiving, at multiple moments, position information of the front end of the smart pen sent by the inertial navigation unit and a pressure sensing signal sent by a pressure sensor; When the duration for which the pressure sensing signal is less than the preset pressure threshold is greater than or equal to the preset duration threshold, stopping receiving the pressure sensing signal and the position information; Determining a recognition result of the written content based on the received position information at multiple moments, the pressure sensing signal, and the reference content; The position information includes: first position information at each moment in a time period when the pressure sensing signal is greater than or equal to a preset pressure threshold, and second position information at each moment in a time period when the pressure sensing signal is less than the preset pressure threshold; Determine the recognition result of the written content based on the received position information, pressure sensing signals and reference content at multiple moments, including: Determining reference glyph information, reference writing method information, and reference time information for writing the reference content based on the reference content; determining a glyph similarity score based on the first position information, the second position information, and the reference glyph information; determining a writing style similarity score based on the first position information, the second position information, the reference glyph information, and the reference writing style information; determining a writing proficiency score based on the first position information, the second position information, and the reference time information; Determining a recognition result of the written content based on the glyph similarity score, the writing style similarity score, and the writing proficiency score; Determining a glyph similarity score according to the first position information, the second position information, and the reference glyph information includes: determining the first writing trajectory information based on the first position information at each moment within the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the first time; determining a starting position for a second writing according to second position information at the last moment within a time period in which the pressure sensing signal is less than a preset pressure threshold for the first time; Determining the trajectory information of the second writing according to the starting position of the second writing and the first position information of each moment within the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the second time; Determining the starting position of the (i+1)th writing according to the second position information of the last moment in the time period in which the pressure sensing signal is less than the preset pressure threshold for the i-th time, where i ≥ 2 and i is a positive integer; Determine the trajectory information of the (i+1)th writing according to the starting position of the (i+1)th writing and the first position information of each moment in the time period when the pressure sensing signal is greater than or equal to the preset pressure threshold for the (i+1)th time; Determining, based on the reference glyph information, glyph reference trajectory information and a reference glyph writing range, wherein the reference glyph writing range is used to represent a reasonable range of a writing trajectory for each stroke of the reference glyph, the glyph reference trajectory information is used to represent a standard writing trajectory for each stroke of the reference glyph, and the glyph reference trajectory information is completely within the reference glyph writing range; The glyph similarity score is determined based on the track information of the multiple writings, the glyph reference track information and the reference glyph writing range.

2. The smart pen based on data collection according to claim 1, characterized in that: Determining a glyph similarity score based on the trajectory information of the multiple writings, the glyph reference trajectory information, and the reference glyph writing range includes: Evenly setting sampling points on the glyph reference trajectory information; Determining that a target sampling point of the trajectory information exists within a preset neighborhood range centered on the sampling point; According to the formula ； Determining glyph similarity scores ,in, is the jth written trajectory information, n is the number of trajectory information, For reference glyph writing range, is the number of target sampling points, is the number of sampling points, is a length function, , and j and n are both positive integers.

3. The smart pen based on data collection according to claim 1, characterized in that: Determining a writing style similarity score based on the first position information, the second position information, the reference glyph information, and the reference writing style information includes: Determining glyph reference trajectory information according to the reference glyph information; Determining, based on the reference writing method information, a standard writing trajectory for each stroke in the character reference trajectory information, and a standard writing order for each stroke; Determining, based on the glyph reference trajectory information, a first minimum circumscribed rectangle of a standard writing trajectory of a plurality of strokes and a first centroid of the first minimum circumscribed rectangle; Determine a second minimum bounding rectangle of the plurality of trajectory information and a second centroid of the second minimum bounding rectangle; Determining strokes corresponding to each piece of trajectory information based on the character reference trajectory information, the first centroid, the trajectory information, and the second centroid; According to the number of moments included in the time period corresponding to the trajectory information corresponding to the standard writing trajectory, the standard writing trajectory is evenly divided to obtain a plurality of standard writing trajectory segments; determining a standard writing vector corresponding to each standard writing trajectory segment according to the reference writing method information; determining a writing trajectory vector in the trajectory information according to each piece of first position information in the trajectory information corresponding to the standard writing trajectory; A writing style similarity score is determined according to the standard writing vector, the writing trajectory vector, the standard writing order and the sequence number of the trajectory information.

4. The smart pen based on data collection according to claim 3, characterized in that: Determining strokes corresponding to each piece of trajectory information according to the glyph reference trajectory information, the first centroid, the trajectory information, and the second centroid includes: Evenly setting a preset number of first marking points on the standard writing trajectory of each stroke in the glyph reference trajectory information; determining a first identification vector between each first identification point and the first centroid; Evenly setting the preset number of second marking points on the trajectory information; determining a second identification vector between each second identification point and the second centroid; According to the formula ； Get the stroke recognition score of the jth trajectory information ,in, is the first identification vector corresponding to the xth first identification point on the standard writing trajectory of the yth stroke, for The transposed vector of for The model, is the second identification vector corresponding to the xth second identification point on the jth trajectory information, for The model, is the preset number, n is the number of trajectory information, , , , and j, n, y, n, x and are all positive integers; The stroke corresponding to the standard writing trajectory that maximizes the stroke recognition score of the j-th trajectory information is determined as the stroke corresponding to the j-th trajectory information.

5. The smart pen based on data collection according to claim 3, characterized in that: Determining a writing style similarity score according to the standard writing vector, the writing trajectory vector, the standard writing order, and the sequence number of the trajectory information includes: According to the formula ； Determine writing style similarity score ,in, is the kth writing trajectory vector in the jth writing trajectory information, is the sequence number of the standard writing order of the strokes corresponding to the j-th writing trajectory information, for The transposed vector of for The model, is the standard writing vector corresponding to the kth standard writing trajectory segment in the standard writing trajectory corresponding to the jth writing trajectory information, for The model, is the number of writing trajectory vectors in the j-th writing trajectory information, = is equal to the number of first position information in the track information written for the jth time minus 1, where n is the number of track information. , , and k, , j and n are all positive integers.

6. The smart pen based on data collection according to claim 1, characterized in that: Determining a writing proficiency score according to the first position information, the second position information, and the reference time information includes: According to the formula ； Determining Writing Proficiency Scores ,in, is the duration of the time period during which the pressure sensing signal is greater than or equal to the preset pressure threshold for the jth time, is the reference time information, n is the number of trajectory information, , and j and n are both positive integers.

Images