Character recognition method, display information control method and device and electronic equipment

Abstract

The embodiment of the invention provides a character recognition method, a display information control method and device and electronic equipment. The character recognition method comprises the following steps: acquiring a track point sequence corresponding to a character to be recognized; based on the track point sequence, multi-dimensional feature information of the character to be recognized is determined, and the multi-dimensional feature information comprises at least one of geometric features, motion features, shape features and direction features; based on the multi-dimensional feature information, determining at least one target character template matched with the to-be-recognized character in a plurality of preset types of character templates; and determining a target character of the to-be-recognized character based on the at least one target character template. In the embodiment of the invention, the high-performance character recognition operation can be realized through the multi-dimensional feature information, the maximum matching rate of the target character can be ensured, and when the method is applied to an application scene for displaying information, the jump regulation and control operation of the landing page can be carried out based on a relatively accurate character matching result, so that the operation efficiency is improved. And the information conversion rate can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a character recognition method, a control method and device for display information and electronic equipment. BACKGROUND

[0002] With the rapid development of human-computer interaction technology, the application of character recognition system is more and more widely. The current character recognition system often realizes character recognition operation through strict path matching operation. However, when the above character recognition system performs character recognition operation, the accuracy of character recognition is closely related to the writing style of characters. When facing complex characters or non-standard handwritten characters, the accuracy of character recognition is low. SUMMARY

[0003] The embodiments of the present application provide a character recognition method, a control method and device for display information and electronic equipment, which can ensure the accuracy of character recognition operation.

[0004] The embodiments of the present application provide a character recognition method, comprising:

[0005] obtaining a trajectory point sequence corresponding to a to-be-recognized character;

[0006] determining multi-dimensional feature information of the to-be-recognized character based on the trajectory point sequence, wherein the multi-dimensional feature information comprises at least one of the following: geometric feature, motion feature, shape feature and direction feature;

[0007] determining at least one target character template matched with the to-be-recognized character from a plurality of preset types of character templates based on the multi-dimensional feature information;

[0008] determining a target character of the to-be-recognized character based on the at least one target character template.

[0009] The embodiments of the present application provide a control method for display information, comprising:

[0010] obtaining a trajectory point sequence corresponding to a to-be-recognized character in display information;

[0011] determining multi-dimensional feature information of the to-be-recognized character based on the trajectory point sequence, wherein the multi-dimensional feature information comprises at least one of the following: geometric feature, motion feature, shape feature and direction feature;

[0012] determining a target character of the to-be-recognized character based on the multi-dimensional feature information and a plurality of preset types of character templates;

[0013] controlling the display information based on the target character.

[0014] The embodiment of the present application provides a character recognition device, comprising:

[0015] The first obtaining module is used for obtaining a track point sequence corresponding to a to-be-recognized character;

[0016] The first determining module is used for determining multi-dimensional feature information of the to-be-recognized character based on the track point sequence, wherein the multi-dimensional feature information comprises at least one of the following: geometric feature, motion feature, shape feature and direction feature;

[0017] The first processing module is used for determining at least one target character template matched with the to-be-recognized character in a plurality of preset type character templates based on the multi-dimensional feature information.

[0018] The first processing module is further used for determining a target character of the to-be-recognized character based on the at least one target character template.

[0019] The embodiment of the present application provides a control device of display information, comprising:

[0020] The second obtaining module is used for obtaining a track point sequence corresponding to a to-be-recognized character in display information;

[0021] The second determining module is used for determining multi-dimensional feature information of the to-be-recognized character based on the track point sequence, wherein the multi-dimensional feature information comprises at least one of the following: geometric feature, motion feature, shape feature and direction feature;

[0022] The second determining module is further used for determining a target character of the to-be-recognized character based on the multi-dimensional feature information and a plurality of preset type character templates.

[0023] The second processing module is used for controlling the display information based on the target character.

[0024] The embodiment of the present application provides an electronic device, comprising a memory and a processor, wherein the memory is used for storing one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method in the first aspect or the second aspect.

[0025] The embodiment of the present application provides a computer storage medium for storing a computer program, wherein the computer program makes a computer execute the method in the first aspect or the second aspect.

[0026] The embodiment of the present application provides a computer program product, comprising a computer readable storage medium storing computer instructions, wherein the computer instructions are executed by one or more processors to make the one or more processors execute the steps in the method in the first aspect or the second aspect.

[0027] The character recognition method, the control method and device of display information and the electronic equipment provided by the embodiment can obtain a sequence of track points corresponding to a to-be-recognized character, determine multi-dimensional feature information of the to-be-recognized character based on the sequence of track points, determine at least one target character template matching the to-be-recognized character in a plurality of preset types of character templates based on the multi-dimensional feature information, and determine a target character of the to-be-recognized character based on the at least one target character template. In this way, the character recognition operation is performed by comprehensively analyzing the multi-dimensional feature information (including geometric features, motion features, shape features, direction features, etc.) of the to-be-recognized character, which not only effectively realizes high performance, maximum matching rate, and high fault-tolerant matching of gestures input by a user on a mobile device, but also realizes more intelligent and robust character recognition operation. When applied to an application scenario of display information, the landing page jump control operation can be performed based on the target character of the to-be-recognized character, thereby ensuring the practicability of the method and being beneficial to improving the information conversion rate and user experience. BRIEF DESCRIPTION OF DRAWINGS

[0028] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and serve to explain the principles of the application, and do not limit the application. In the drawings:

[0029] Figure 1 A scene schematic diagram of a character recognition method provided by an exemplary embodiment of the application;

[0030] Figure 2 A flowchart of a character recognition method provided by an exemplary embodiment of the application;

[0031] Figure 3 A schematic diagram of determining a plurality of adjacent first track vectors and second track vectors provided by an exemplary embodiment of the application;

[0032] Figure 4 A flowchart of determining at least one target character template matching the to-be-recognized character in the character template based on the multi-dimensional feature information provided by an exemplary embodiment of the application;

[0033] Figure 5 A flowchart of another character recognition method provided by an exemplary embodiment of the application;

[0034] Figure 6 A flowchart of a control method of display information provided by an exemplary application embodiment of the application;

[0035] Figure 7A schematic diagram of multi-dimensional feature information provided for an exemplary application embodiment of the present application;

[0036] Figure 8 A flowchart of a control method for displaying information provided for an exemplary embodiment of the present application;

[0037] Figure 9 A structural schematic diagram of a character recognition device provided for an exemplary embodiment of the present application;

[0038] Figure 10 A structural schematic diagram of an electronic device provided for an exemplary embodiment of the present application;

[0039] Figure 11 A structural schematic diagram of a control device for displaying information provided for an exemplary embodiment of the present application;

[0040] Figure 12 A structural schematic diagram of an electronic device provided for an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be described below in conjunction with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0042] It should be noted that, in the case where the embodiments of the present application involve user information, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the embodiments of the present application are all information and data authorized by the user or authorized by all parties, and the collection, use, and processing of related data need to comply with relevant laws, regulations, and standards of relevant countries and regions, and provide corresponding operation portals for the user to choose authorization or rejection. In addition, the various models (including but not limited to language models or large models) involved in the present application comply with relevant legal and standard regulations.

[0043] In addition, it should be noted that, in the case where the embodiments of the present application involve user interaction operations or trigger operations, the user interaction operations or trigger operations involved in the embodiments of the present application include but are not limited to: touch operations, gesture operations, voice operations, head movement operations, eye movement operations, and various modes of interaction operations; wherein the touch operations include but are not limited to: click operations, double-click operations, long-press operations, sliding operations, pinch operations, or mouse hovering operations, etc. The sliding operation includes but is not limited to: straight-line sliding, curve sliding, etc.

[0044] For the convenience of understanding the character recognition method, the control method and the device of display information and the electronic equipment provided by the embodiments of the present application, the related art will be briefly described as follows:

[0045] The opening screen advertisement refers to the full-screen advertisement displayed when the application App in the terminal device (for example, a smart phone, a tablet computer, a smart watch, etc.) is started or switched between the foreground and the background. In the application scenario of the opening screen advertisement of a certain brand, the merchant can customize the landing page to which the user jumps after triggering a specific track. The landing page refers to the page to which the opening screen advertisement jumps after meeting the interaction condition.

[0046] For example, for the brand V, after the user draws a hook, if the drawn hook strictly matches the character shape of the jump character corresponding to the brand V, the landing page corresponding to the brand V will be jumped to. For the brand C, after the user draws a C character, if the drawn C character strictly matches the character shape of the jump character corresponding to the brand C, the landing page corresponding to the brand C will be jumped to. Such user-deterministic input not only improves the interest of the advertisement, but also realizes a screening operation for the user, so that it can be determined that the user who needs to enter the activity page is definitely a person interested in the content, and the mis-touch behavior of the traditional advertisement is also avoided, which is beneficial to improving the post-link conversion effect of the merchant and the user experience.

[0047] However, the character recognition operation realized by the strict matching operation of the path is closely related to the writing style of the character, and when facing a complex character or a non-standard hand-written character, the accuracy of the character recognition is low.

[0048] To solve the above technical problems, the embodiments of the present application provide a character recognition method, a control method and a device of display information and an electronic equipment, which are described in detail with reference to the accompanying drawings. Figure 1As shown, the execution subject of the character recognition method can be a character recognition device 200, which can be implemented as a local server or a cloud server. In the case of the character recognition device 200 being implemented as a cloud server, the character recognition method can be executed in the cloud, and a plurality of computing nodes (cloud servers) can be deployed in the cloud, each of which has computing, storage, and other processing resources. In the cloud, a plurality of computing nodes can be organized to provide a certain service, and of course, one computing node can also provide one or more services. The cloud provides the service in the form of a service interface that can be called by a user to use the corresponding service. The service interface includes a software development kit (SDK), an application programming interface (API), and the like.

[0049] The character recognition device 200 is in communication connection with the client 100, wherein the client 100 is used by a user to perform an application to trigger a character recognition operation. The client 100 can be any computing device with certain information interaction capabilities. In actual implementation, the client 100 can be a mobile phone, a personal computer (PC), a tablet computer, a set application, and the like. In addition, the basic structure of the client 100 can include at least one processor. The number of processors depends on the configuration and type of the client. The client 100 can also include a memory, which can be volatile, such as a random access memory (RAM), or non-volatile, such as a read-only memory (ROM), a flash memory, or both. The memory usually stores an operating system (OS), one or more application programs, and program data. In addition to the processing unit and the memory, the client 100 also includes some basic configurations, such as a network card chip, an IO bus, a display component, and some peripheral devices. Optionally, some peripheral devices can include, for example, a keyboard, a mouse, a stylus, a printer, and the like. Other peripheral devices are well known in the art and will not be described here.

[0050] The character recognition device 200 refers to a device that can implement a character recognition operation in a network virtual environment, and generally refers to a device that plans information and performs a character recognition operation using a network. The character recognition device 200 can be implemented as a character recognition model for implementing a character recognition operation, and in physical implementation, the character recognition device 200 can be any device that can provide computing services and perform a corresponding character recognition operation, such as a processor, a server, and the like. The character recognition device 200 mainly includes a processor, a hard disk, a memory, a system bus, and the like, which is similar to the general computer architecture.

[0051] In the above embodiment, the character recognition device 200 and the client 100 are connected through a network, which can be a wireless or wired network connection. If the character recognition device 200 and the client 100 are connected through a communication connection, the network mode of the mobile network can be any one of 2G (Global System for Mobile Communications GSM), 2.5G (General Packet Radio Service GPRS), 3G (Wideband Code Division Multiple Access WCDMA, Time Division Synchronous Code Division Multiple Access TD-SCDMA), 4G (Long Term Evolution LTE), 4G+ (Enhanced Long Term Evolution LTE+), Worldwide Interoperability for Microwave Access (WiMax), 5G, 6G, and the like.

[0052] In the embodiment of the present application, the client 100 is used by a user to trigger a character recognition operation on a to-be-recognized character, wherein the to-be-recognized character can be composed of a plurality of track points, and the plurality of track points form a track point sequence. In some examples, the to-be-recognized character can be obtained through a human-computer interaction operation. Specifically, an interaction interface can be displayed on the client 100, and the user can perform a character drawing operation in the interaction interface, and then the to-be-recognized character can be obtained. In order to accurately recognize the to-be-recognized character, the to-be-recognized character can be sent to the character recognition device 200 for analysis and processing.

[0053] The character recognition device 200 is configured to acquire the trajectory point sequence corresponding to the to-be-recognized character sent by the client 100, wherein the trajectory point sequence comprises a plurality of trajectory point information used for identifying the trajectory attribute of the to-be-recognized character. In order to accurately recognize the to-be-recognized character, the trajectory point sequence is analyzed and processed to determine the multi-dimensional feature information of the to-be-recognized character. The multi-dimensional feature information can include at least one of the following: geometric features, motion features, shape features, and direction features. In some examples, the multi-dimensional feature information can include geometric features, motion features, shape features, and direction features. The geometric features can include at least one of the following: curvature information, aspect ratio information, and closeness information. The motion features can include at least one of the following: direction change information, smoothness information, and speed information. The shape features can include at least one of the following: arc information, straight line information, and wave count information. The direction information can include at least one of the following: dominant direction information, angle histogram information, and direction consistency information.

[0054] After the multi-dimensional feature information of the to-be-recognized character is acquired, at least one target character template that matches the to-be-recognized character can be determined from a plurality of preset types of character templates based on the multi-dimensional feature information. The preset types of character templates can include at least one of the following: a number template, a letter template, a direction template, and the like. After the at least one target character template that matches the to-be-recognized character is determined, the target character of the to-be-recognized character can be determined based on the at least one target character template. In this way, accurate recognition of the to-be-recognized character is effectively achieved.

[0055] In this embodiment, the multi-dimensional feature information (including geometric features, motion features, shape features, and direction features) of the to-be-recognized character is comprehensively analyzed to perform the character recognition operation. This not only effectively realizes high performance, maximum matching rate, and high fault tolerance for matching gestures input by a user on a mobile device, but also realizes more intelligent and robust character recognition operations. When applied to an application scenario for displaying information, the landing page can be controlled to jump based on the target character of the to-be-recognized character that is recognized. In this way, the practicability of the method is ensured, and information conversion rate and user experience are improved.

[0056] The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0057] Figure 2 A flowchart of a character recognition method provided by an example embodiment of the present application is shown in FIG. 1. The details are described below with reference to the accompanying drawings. Figure 2As shown, the embodiment provides a character recognition method, and an execution subject of the method is a character recognition device. The character recognition device can be implemented as software or a combination of software and hardware. When the character recognition device is implemented as hardware, it can be various electronic devices capable of implementing character recognition operations, including but not limited to personal computers, servers, and the like. When the character recognition device is implemented as software, it can be installed in the above-mentioned electronic devices. Specifically, the character recognition method provided in the embodiment can include the following steps.

[0058] Step S201: Obtain a sequence of trajectory points corresponding to a character to be recognized.

[0059] Step S202: Determine multi-dimensional feature information of the character to be recognized based on the sequence of trajectory points. The multi-dimensional feature information includes at least one of the following: geometric features, motion features, shape features, and direction features.

[0060] Step S203: Determine at least one target character template matching the character to be recognized from a plurality of pre-set types of character templates based on the multi-dimensional feature information.

[0061] Step S204: Determine a target character of the character to be recognized based on the at least one target character template.

[0062] The specific implementation and principles of each of the above steps are described in detail as follows:

[0063] Step S201: Obtain a sequence of trajectory points corresponding to a character to be recognized.

[0064] When a user has a character recognition requirement, the character recognition device can obtain a sequence of trajectory points corresponding to a character to be recognized. The sequence of trajectory points can include a plurality of trajectory point information used to form the character to be recognized. The number of characters to be recognized corresponding to the sequence of trajectory points can be one or more. When the number of characters to be recognized corresponding to the sequence of trajectory points is one, the character recognition device can implement a single character recognition operation. When the number of characters to be recognized corresponding to the sequence of trajectory points is more than one, the character recognition device can implement a multi-character recognition operation.

[0065] Specifically, the embodiment is not limited to the specific acquisition manner of the trajectory point sequence corresponding to the character to be recognized. In some examples, the trajectory point sequence corresponding to the character to be recognized can be obtained through a data transmission operation. At this time, the trajectory point sequence corresponding to the character to be recognized can be obtained by determining a client in communication connection with the character recognition device, wherein the client stores the trajectory point sequence corresponding to the character to be recognized. The character recognition device can actively or passively obtain the trajectory point sequence corresponding to the character to be recognized through the client, thereby effectively ensuring the accuracy and reliability of obtaining the trajectory point sequence.

[0066] In other examples, the trajectory point sequence corresponding to the character to be recognized can be obtained not only through interoperation with the client, but also through human-computer interaction. At this time, the trajectory point sequence corresponding to the character to be recognized can be obtained by displaying a human-computer interaction interface, obtaining a character drawing operation input by a user in the human-computer interaction interface, and obtaining the trajectory point sequence corresponding to the character to be recognized based on the character drawing operation, thereby also ensuring the accuracy and reliability of obtaining the trajectory point sequence.

[0067] Step S202: determining multi-dimensional feature information of the character to be recognized based on the trajectory point sequence, wherein the multi-dimensional feature information includes at least one of the following: geometric feature, motion feature, shape feature, and direction feature.

[0068] After obtaining the trajectory point sequence, the trajectory point sequence can be analyzed and processed to determine the multi-dimensional feature information of the character to be recognized. The multi-dimensional feature information can include at least one of the following: geometric feature, motion feature, shape feature, and direction feature, etc. In some examples, the multi-dimensional feature information can include geometric feature, motion feature, shape feature, or direction feature, or the multi-dimensional feature information can include geometric feature and motion feature, or the multi-dimensional feature information can include geometric feature, motion feature, and shape feature; or the multi-dimensional feature information can include geometric feature, motion feature, shape feature, and direction feature, etc. The above-mentioned geometric feature can include at least one of the following: curvature information, aspect ratio information, and closeness information. The motion feature can include at least one of the following: direction change information, smoothness information, and speed information. The shape feature can include at least one of the following: arc line information, straight line information, and wave count information. The direction information includes at least one of the following: dominant direction information, angle histogram information, and direction consistency information, etc.

[0069] As for the multi-dimensional feature information of the character to be recognized, the embodiment is not limited to the specific determination manner of the multi-dimensional feature information. In some examples, the multi-dimensional feature information can be determined based on a pre-trained multi-dimensional feature extractor (or multi-dimensional feature extraction model). At this time, based on the trajectory point sequence, determining the multi-dimensional feature information of the character to be recognized can include: determining a multi-dimensional feature extractor (or multi-dimensional feature extraction model) for implementing a feature extraction operation; and performing a feature extraction operation on the trajectory point sequence by using the multi-dimensional feature extractor (or multi-dimensional feature extraction model) to obtain the multi-dimensional feature information of the trajectory point sequence. In this way, the accuracy and reliability of determining the multi-dimensional feature information are effectively ensured.

[0070] In other examples, the multi-dimensional feature information can be determined not only based on a pre-trained multi-dimensional feature extractor, but also based on a feature extraction operation directly performed on the trajectory point sequence. As for the character to be recognized, since the aspect ratio can reflect the basic geometric properties of the character shape, different characters have a characteristic aspect ratio range. For example, the letter "I" and the number "1" are usually very narrow, with an aspect ratio <0.5; the letter "W" is usually very wide, with an aspect ratio >2.0; and the letter "O" is close to a square, with an aspect ratio ≈1.0. Specifically, in the case where the geometric feature includes the aspect ratio, based on the trajectory point sequence, determining the multi-dimensional feature information of the character to be recognized can include: based on the trajectory point sequence, determining width information and height information of the character to be recognized; and based on the width information and the height information, determining an aspect ratio of the character to be recognized.

[0071] In this case, after the trajectory point sequence is obtained, the bounding box of the trajectory point can be calculated based on the trajectory point sequence, the maximum horizontal coordinate maxX, the maximum vertical coordinate maxY, the minimum horizontal coordinate minX, and the minimum vertical coordinate minY can be determined based on the bounding box, and then the width information of the character to be recognized can be determined based on the maximum horizontal coordinate and the minimum vertical coordinate, i.e., width information = maxX-minX. Similarly, the height information of the character to be recognized can be determined based on the maximum vertical coordinate and the minimum vertical coordinate, i.e., height information = maxY-minY, and then the aspect ratio of the character to be recognized can be determined based on the width information and the height information, i.e., aspect ratio = width information / height information, or aspect ratio = width information / (height information+1e-6). In this way, the accuracy and reliability of determining the aspect ratio are effectively ensured.

[0072] Similarly, for the to-be-recognized character, since the closeness degree can reflect the closeness degree of the trajectory, the start-end point distance of the completely closed graph (such as O, 0) is close to 0, and the start-end point distance of the open graph (such as C, U) is larger, and this feature is crucial for distinguishing closed and open characters. Therefore, in order to accurately recognize the character, the geometric features can not only include the aspect ratio, but also include the closeness degree, and in some examples, the closeness degree can be determined by detecting the bounding box corresponding to the to-be-recognized character, and at this time, the determining the multi-dimensional feature information of the to-be-recognized character based on the trajectory point sequence can include: determining the bounding box corresponding to the to-be-recognized character based on the trajectory point sequence, which can be an axis-aligned or arbitrary-direction rectangle (2D), and then the closeness degree of the to-be-recognized character can be determined based on the determined bounding box, and specifically, the closeness degree of the to-be-recognized character can be determined by detecting the bounding box through the closeness detection model.

[0073] Alternatively, the closeness degree can also be determined based on the start point position, the end point position and the total length of the trajectory of the to-be-recognized character, and at this time, the determining the multi-dimensional feature information of the to-be-recognized character based on the trajectory point sequence can include: determining the total length of the trajectory, the start point position and the end point position of the to-be-recognized character based on the trajectory point sequence, wherein the total length of the trajectory can be the sum of the distances between all adjacent trajectory points in the to-be-recognized character; and determining the closeness degree of the to-be-recognized character based on the start point position, the end point position and the total length of the trajectory.

[0074] For the to-be-recognized character, the closeness degree of the to-be-recognized character is related to the total length of the trajectory, the start point position and the end point position of the to-be-recognized character, so after obtaining the trajectory point sequence, the total length of the trajectory, the start point position and the end point position of the to-be-recognized character can be determined based on the trajectory point sequence, and then the start point position, the end point position and the total length of the trajectory can be analyzed and processed to determine the closeness degree of the to-be-recognized character.

[0075] In some examples, the closeness degree of the to-be-recognized character can be determined by analyzing and processing the start point position, the end point position and the total length of the trajectory through the closeness degree detection model, and at this time, the determining the closeness degree of the to-be-recognized character based on the start point position, the end point position and the total length of the trajectory can include: determining the closeness degree detection model for implementing the closeness degree detection operation; and inputting the start point position, the end point position and the total length of the trajectory into the closeness degree detection model for analysis and processing to obtain the closeness degree of the to-be-recognized character output by the closeness degree detection model, which effectively ensures the accuracy and reliability of determining the closeness degree of the to-be-recognized character.

[0076] In yet some examples, the closeness of the character to be recognized can be determined not only by the closeness detection model analyzing the start point position, the end point position and the total length of the trajectory, but also by analyzing the start point position, the end point position and the total length of the trajectory. At this time, determining the closeness of the character to be recognized based on the start point position, the end point position and the total length of the trajectory can include: determining a start-end point distance based on the start point position and the end point position; determining the closeness of the character to be recognized based on the total length of the trajectory and the start-end point distance.

[0077] Wherein, after obtaining the start point position and the end point position, the start point position and the end point position can be analyzed to determine the start-end point distance, which is the straight-line distance between the start point position and the end point position, and then the total length of the trajectory and the start-end point distance are analyzed to determine the closeness of the character to be recognized. In some examples, determining the closeness of the character to be recognized based on the total length of the trajectory and the start-end point distance can include: determining a distance ratio between the start-end point distance and the total length of the trajectory; determining the closeness of the character to be recognized based on the distance ratio, the closeness being negatively correlated with the distance ratio, specifically, closeness = 1 - distance ratio; or, closeness = 1 - (distance ratio x 3), which effectively ensures the accuracy and reliability of determining the closeness.

[0078] Similarly, the geometric feature can not only include the closeness, but also include the closeness trend. At this time, determining the multi-dimensional feature information of the character to be recognized based on the trajectory point sequence can include: determining a start point orientation and an end point orientation of the character to be recognized based on the trajectory point sequence; determining a closeness trend of the trajectory point sequence based on the start point orientation and the end point orientation; specifically, determining the closeness trend of the trajectory point sequence based on the start point orientation and the end point orientation can include: determining that the character to be recognized has a closing trend in the case that the start point orientation is opposite to the end point orientation; determining that the character to be recognized does not have a closing trend in the case that the start point orientation is not opposite to the end point orientation, which effectively ensures the accuracy and reliability of determining the closeness trend.

[0079] In some examples, the curvature information can be used to distinguish different characters. For example, a straight line has a curvature close to 0, while a circle has a constant curvature. Therefore, the geometric features can include the curvature information in addition to the width-height ratio. In this case, based on the sequence of trajectory points, determining the multi-dimensional feature information of the character to be recognized can include: determining a plurality of adjacent first trajectory vectors and second trajectory vectors based on adjacent trajectory points in the sequence of trajectory points; determining a plurality of vector angles formed by the first trajectory vectors and the second trajectory vectors based on the plurality of adjacent first trajectory vectors and second trajectory vectors; and determining the curvature information of the character to be recognized based on the plurality of vector angles.

[0080] Specifically, after obtaining the sequence of trajectory points, a plurality of adjacent first trajectory vectors and second trajectory vectors can be determined based on adjacent trajectory points in the sequence of trajectory points. Different trajectory points can correspond to different first trajectory vectors and second trajectory vectors. For example, as shown in FIG. 4, for a trajectory point A in the sequence of trajectory points, the trajectory point A can correspond to a first trajectory vector and a second trajectory vector. The first trajectory vector can be determined by a previous trajectory point and the trajectory point A, and the second trajectory vector can be determined by the trajectory point A and a next trajectory point. For a trajectory point B in the sequence of trajectory points, the trajectory point B can correspond to a first trajectory vector and a second trajectory vector. The first trajectory vector can be determined by a previous trajectory point and the trajectory point B, and the second trajectory vector can be determined by the trajectory point B and a next trajectory point. Figure 3

[0081] After determining the plurality of adjacent first trajectory vectors and second trajectory vectors, the plurality of adjacent first trajectory vectors and second trajectory vectors can be analyzed and processed to determine a plurality of vector angles formed by the first trajectory vectors and the second trajectory vectors. Specifically, a cosine value of the vector angle can be determined based on the first trajectory vector and the second trajectory vector, and then an inverse cosine operation can be performed on the cosine value, so that the plurality of vector angles formed by the first trajectory vectors and the second trajectory vectors can be obtained. As shown in FIG. 4, a vector angle a can be determined based on the first trajectory vector and the second trajectory vector corresponding to the trajectory point A, and a vector angle b can be determined based on the first trajectory vector and the second trajectory vector corresponding to the trajectory point B. Figure 3

[0082] ​​After the plurality of vector angles between the first trajectory vector and the second trajectory vector are determined, the plurality of vector angles can be analyzed to determine the curvature information of the character to be recognized. In some examples, the curvature information of the character to be recognized can be determined by analyzing the plurality of vector angles using a pre-trained curvature calculation model. In this case, determining the curvature information of the character to be recognized based on the plurality of vector angles can include determining the pre-trained curvature calculation model, inputting the plurality of vector angles into the curvature calculation model for analysis, and obtaining the curvature information of the character to be recognized output by the curvature calculation model. This effectively ensures the accuracy and reliability of determining the curvature information.

[0083] In other examples, the curvature information can be determined based on not only the pre-trained curvature calculation model, but also an average angle of the plurality of vector angles. In this case, determining the curvature information of the character to be recognized based on the plurality of vector angles can include determining the average angle of the plurality of vector angles, and determining the average angle as the curvature information of the character to be recognized. This effectively ensures the accuracy and reliability of determining the curvature information.

[0084] In addition, in the case where the geometric feature includes the curvature information, the curvature information can be determined based on not only the average angle of the plurality of vector angles, but also a center point (or a center point) determined by the sequence of trajectory points. In this case, determining the multi-dimensional feature information of the character to be recognized based on the sequence of trajectory points can include determining the center point corresponding to the sequence of trajectory points, and determining the curvature information of the sequence of trajectory points based on the center point and the sequence of trajectory points.

[0085] After the sequence of trajectory points is obtained, the positions of the trajectory points in the sequence of trajectory points can be obtained first, and then the positions of the trajectory points can be averaged to obtain the center point corresponding to the sequence of trajectory points. After the center point corresponding to the sequence of trajectory points is determined, the center point and the sequence of trajectory points can be analyzed to determine the curvature information of the sequence of trajectory points. In some examples, determining the curvature information of the sequence of trajectory points based on the center point and the sequence of trajectory points can include obtaining any two adjacent trajectory points in the sequence of trajectory points, determining a plurality of virtual triangles formed by the any two adjacent trajectory points and the center point, and determining the curvature information of the sequence of trajectory points based on the plurality of virtual triangles.

[0086] Specifically, determining the curvature information of the sequence of trajectory points based on the plurality of virtual triangles can include: determining angle change information corresponding to each trajectory point based on adjacent virtual triangles in the plurality of virtual triangles; determining quantity information of the virtual triangles; obtaining a ratio between the angle change information corresponding to each trajectory point and the quantity information; and accumulating the ratios of the plurality of trajectory points in the sequence of trajectory points, thereby obtaining the curvature information of the sequence of trajectory points, which also ensures the accuracy and reliability of determining the curvature information.

[0087] Similar to the implementation of determining the geometric features of the character to be recognized, the shape features of the character to be recognized can also be directly extracted based on the sequence of trajectory points. Specifically, since the radian information can measure the similarity between the trajectory of the character to be recognized and a circular arc, for example, when the radian information corresponding to a circular arc is implemented as a score, the radian score corresponding to the circular arc is close to 1, while the radian score corresponding to a straight line or an irregular curve is relatively low, which is important for recognizing characters such as C, O, and U. Therefore, in order to accurately recognize the character to be recognized, the shape features can include radian information, and at this time, determining the multi-dimensional feature information of the character to be recognized based on the sequence of trajectory points can include: determining a centroid corresponding to the character to be recognized based on the sequence of trajectory points; determining a plurality of distance information between each point in the character to be recognized and the centroid; and determining radian information of the character to be recognized based on the plurality of distance information.

[0088] After obtaining the sequence of trajectory points, the sequence of trajectory points can be analyzed and processed to determine the centroid corresponding to the character to be recognized. In some examples, determining the centroid corresponding to the character to be recognized can include: using a least squares method to perform a fitting processing operation on the sequence of trajectory points to obtain a fitted circle corresponding to the character to be recognized, and then determining the center of the fitted circle as the centroid corresponding to the character to be recognized; determining a plurality of distance information between each point in the character to be recognized and the centroid; and then analyzing and processing the plurality of distance information to determine the radian information of the character to be recognized.

[0089] Specifically, determining the radian information of the character to be recognized based on the plurality of distance information can include: determining an average distance corresponding to the plurality of distance information; calculating a relative distance error between each distance information and the average distance; determining an average error of the plurality of relative distance errors; and determining the radian information of the character to be recognized based on the average error, wherein the radian information is negatively correlated with the average error. In some examples, the radian information can be implemented as score information, for example, radian information = 1 / (1+average error), which effectively ensures the accuracy and reliability of determining the radian information.

[0090] In addition, for the shape feature of the character to be recognized, since the smoothness information can reflect the degree of angle change of the character to be recognized, for example, a smooth curve has a consistent angle change, and a jagged or irregular trajectory has a large angle change, the smoothness information can distinguish the handwriting quality and identify the specific character feature (e.g., the smooth curve of S vs. the sharp angle of Z). Therefore, in order to accurately recognize the character, the shape feature can include not only the curvature information but also the smoothness information. At this time, determining the multi-dimensional feature information of the character to be recognized based on the trajectory point sequence can include: determining a plurality of adjacent first line segments and second line segments based on adjacent trajectory points in the trajectory point sequence; determining a plurality of angle change information formed between the plurality of adjacent first line segments and second line segments; and determining the smoothness information of the character to be recognized based on the plurality of angle change information.

[0091] After obtaining the trajectory point sequence, the adjacent trajectory points in the trajectory point sequence can be analyzed and processed to determine a plurality of adjacent first line segments and second line segments, and then a plurality of angle change information formed between the plurality of adjacent first line segments and second line segments can be determined. For example, the plurality of adjacent first line segments and second line segments can include: [line segment a1, line segment a2], [line segment a2, line segment a3], [line segment a3, line segment a4], and [line segment a4, line segment a5]. The above line segment a1 and line segment a2 can form an angle A1, the line segment a2 and the line segment a3 can form an angle A2, the line segment a3 and the line segment a4 can form an angle A3, and the line segment a4 and the line segment a5 can form an angle A4. Then, the angle change information 1 between the angle A1 and the angle A2, the angle change information 2 between the angle A2 and the angle A3, the angle change information 3 between the angle A3 and the angle A4, and the angle change information 4 between the angle A4 and the angle A5 can be determined. In this way, the plurality of angle change information can be obtained.

[0092] After obtaining the plurality of angle change information, the plurality of angle change information can be analyzed and processed to determine the smoothness information of the character to be recognized. In some examples, the smoothness information can be determined based on a pre-trained smoothness calculation model to analyze and process the plurality of angle change information, or the smoothness information can be determined based on the average value of the angle change corresponding to the angle change information and the standard deviation of the angle change. At this time, determining the smoothness information of the character to be recognized based on the plurality of angle change information can include: determining the average value of the angle change based on the plurality of angle change information; determining the standard deviation of the angle change based on the average value of the angle change; and determining the smoothness information of the character to be recognized based on the standard deviation of the angle change, wherein the smoothness information is negatively correlated with the standard deviation of the angle change. Specifically, the smoothness information is determined by the following formula: smoothness information = 1 - standard deviation of the angle change. In this way, the accuracy and reliability of determining the smoothness information are effectively ensured.

[0093] In addition, for the trajectory point sequence, different trajectory point sequences can correspond to different to-be-recognized characters, and different trajectory point sequences can include different numbers of trajectory points, for example, 2 trajectory points, 3 trajectory points, 5 trajectory points, 10 trajectory points, etc. Among them, the trajectory point sequence including a smaller number of trajectory points can be obtained based on the mis-touch operation character. Therefore, in order to accurately implement the character recognition operation while avoiding the identification operation of the mis-touch operation character, after obtaining the trajectory point sequence corresponding to the to-be-recognized character, it can be determined whether the trajectory point sequence can trigger the character recognition operation. At this time, the method in the embodiment can further include: determining the number of trajectory points in the trajectory point sequence; in the case that the number of trajectory points is greater than or equal to a preset number threshold, allowing the determination of the multi-dimensional feature information of the trajectory point sequence; in the case that the number of trajectory points is less than the preset number threshold, prohibiting the determination of the multi-dimensional feature information of the trajectory point sequence.

[0094] After obtaining the trajectory point sequence, the trajectory point sequence can be scanned or statistically processed, so that the number of trajectory points in the trajectory point sequence can be determined. Then, the number of trajectory points can be compared with the preset number threshold. In the case that the number of trajectory points is greater than or equal to the preset number threshold, it indicates that the to-be-recognized character corresponding to the trajectory point sequence at this time is a character that needs to be normally recognized. Therefore, the multi-dimensional feature information of the trajectory point sequence is allowed to be determined. Correspondingly, in the case that the number of trajectory points is less than the preset number threshold, it indicates that the to-be-recognized character corresponding to the trajectory point sequence at this time is a mis-touch operation character that does not need to be normally recognized. Therefore, the multi-dimensional feature information of the trajectory point sequence can be prohibited to be determined. In this way, it is effectively avoided to perform the character recognition operation on the mis-touch operation character, thereby improving the effectiveness and accuracy of the character recognition operation.

[0095] Step S203: determining at least one target character template matching the to-be-recognized character from a plurality of preset types of character templates based on the multi-dimensional feature information.

[0096] Since the multi-dimensional feature information can describe the attribute features of the to-be-recognized character from multiple dimensions, after obtaining the multi-dimensional feature information, at least one target character template matching the to-be-recognized character can be determined from a plurality of preset types of character templates. The at least one target character template can be a character template with a higher matching degree from the plurality of preset types of character templates.

[0097] In some examples, when the multi-dimensional feature information includes any one of the geometric feature, the motion feature, the shape feature or the direction feature, at least one target character template matching the character to be recognized can be determined in the plurality of preset types of character templates based on the obtained multi-dimensional feature information including one type of feature information; or, when the multi-dimensional feature information includes any two of the geometric feature, the motion feature, the shape feature or the direction feature, at least one target character template matching the character to be recognized can be determined in the plurality of preset types of character templates based on the obtained multi-dimensional feature information including two types of feature information; or, when the multi-dimensional feature information includes any three of the geometric feature, the motion feature, the shape feature or the direction feature, at least one target character template matching the character to be recognized can be determined in the plurality of preset types of character templates based on the obtained multi-dimensional feature information including three types of feature information; or, when the multi-dimensional feature information includes the geometric feature, the motion feature, the shape feature and the direction feature, at least one target character template matching the character to be recognized can be determined in the plurality of preset types of character templates based on the obtained multi-dimensional feature information including four types of feature information, which effectively ensures the flexibility and reliability of determining the at least one target character template.

[0098] In other examples, regardless of the number of types of feature information included in the multi-dimensional feature information, the at least one target character template can be determined based on a pre-trained character detection model. At this time, determining the at least one target character template matching the character to be recognized in the plurality of preset types of character templates based on the multi-dimensional feature information can include: determining the pre-trained character detection model; inputting the multi-dimensional feature information and the plurality of preset types of character templates into the character detection model for a character recognition operation to obtain at least one target character template matching the character to be recognized output by the character detection model, which effectively ensures the accuracy and reliability of determining the at least one target character template.

[0099] Step S204: determining a target character of the character to be recognized based on the at least one target character template.

[0100] Since different target character templates correspond to different characters, after determining the at least one target character template matching the character to be recognized, the target character of the character to be recognized can be directly determined based on the at least one target character template, which effectively realizes accurate character recognition operation on the character to be recognized.

[0101] The character recognition method provided in the embodiment comprises the following steps: acquiring a track point sequence corresponding to a to-be-recognized character; determining multi-dimensional feature information of the to-be-recognized character based on the track point sequence; determining at least one target character template matched with the to-be-recognized character in a plurality of preset types of character templates based on the multi-dimensional feature information; and determining a target character of the to-be-recognized character based on the at least one target character template. In this way, the multi-dimensional feature information (including geometric features, motion features, shape features, direction features, etc.) of the to-be-recognized character is comprehensively analyzed to perform the character recognition operation, which not only effectively realizes high performance, maximum matching rate, high fault-tolerant matching of the gesture input by the user on the mobile device, but also realizes more intelligent and robust character recognition operation. When applied to an application scenario of displaying information (for example, a webpage, an advertisement or an opening screen advertisement), the landing page jump control operation can be performed based on the target character of the to-be-recognized character, thereby ensuring the practicability of the method and being conducive to improving the advertisement conversion rate and user experience.

[0102] Figure 4 The flowchart for determining at least one target character template matched with the to-be-recognized character in the character templates based on the multi-dimensional feature information is provided for an exemplary embodiment of the present application. On the basis of the above-mentioned embodiment, reference is made to the accompanying drawings Figure 4 For the target character template, it can be determined based on the pre-trained character detection model, or can be directly determined based on the character matching degree between the letter template and the to-be-recognized character. At this time, the step of determining at least one target character template matched with the to-be-recognized character in the character templates based on the multi-dimensional feature information can comprise the following steps:

[0103] Step S401: determining a plurality of preset types of character templates for analyzing and processing the to-be-recognized character, wherein the plurality of preset types of character templates at least comprise a number type character template, a letter type character template and a direction type character template.

[0104] The plurality of preset types of character templates for analyzing and processing the to-be-recognized character are preconfigured for the to-be-recognized character, and the plurality of preset types of character templates at least include: a number type of character template, a letter type of character template, and a direction type of character template. The number type of character template can include: "0", "1", "2", "3", "4", and the like. The letter type of character template can include: "A", "B", "C", "a", "b", "c", and the like. The direction type of character template can include: "→", and the like. Specifically, the plurality of preset types of character templates can be stored in a preset area or a preset device. When a character recognition operation is needed, the plurality of preset types of character templates can be determined by accessing the preset area or the preset device, so that the accuracy and reliability of determining the plurality of preset types of character templates are stably achieved.

[0105] Step S402: determining a character matching degree between the to-be-recognized character and each preset type of character template based on the multi-dimensional feature information.

[0106] Since the multi-dimensional feature information can describe the character features of the to-be-recognized character from multiple dimensions, after obtaining the multi-dimensional feature information and the plurality of preset types of character templates, the multi-dimensional feature information and the plurality of preset types of character templates can be analyzed and processed to determine the character matching degree between the to-be-recognized character and each preset type of character template. In some examples, the character matching degree can be determined based on a pre-trained matching degree detection model. At this time, determining the character matching degree between the to-be-recognized character and each preset type of character template based on the multi-dimensional feature information can include: determining the pre-trained matching degree detection model; inputting the multi-dimensional feature information and the plurality of preset types of character templates into the matching degree detection model for analysis and processing, and determining the character matching degree between the to-be-recognized character and each preset type of character template output by the matching degree detection model, so that the character matching degree between the to-be-recognized character and each character template can be stably determined.

[0107] In other examples, the character matching degree can be determined not only based on the pre-trained matching degree detection model, but also based on fusion feature information corresponding to the multi-dimensional feature information. At this time, determining the character matching degree between the to-be-recognized character and each preset type of character template based on the multi-dimensional feature information can include: fusing the multi-dimensional feature information to obtain the fusion feature information; and determining the character matching degree between the to-be-recognized character and each preset type of character template based on the fusion feature information.

[0108] The multi-dimensional feature information can include at least one of the following: geometric features, motion features, shape features, and direction features. In order to accurately implement character recognition, the multi-dimensional feature information can be fused to determine fused feature information corresponding to the multi-dimensional feature information. The fused feature information can be represented as a vector. In some examples, the fused feature information can be determined by analyzing the multi-dimensional feature information using a pre-trained feature extraction model, or the fused feature information can be determined based on the character type of the multi-dimensional feature information. In this case, the multi-dimensional feature information is fused to obtain the fused feature information, which includes: determining at least one character type to which the to-be-recognized character belongs based on the multi-dimensional feature information; determining a feature fusion weight corresponding to each of the multi-dimensional feature information based on the at least one character type; and fusing the multi-dimensional feature information based on the feature fusion weight to obtain the fused feature information.

[0109] Different types of to-be-recognized characters correspond to different dimensions of feature information, and different dimensions of feature information correspond to different feature fusion weights. Therefore, in order to accurately obtain the fused feature information, at least one character type to which the to-be-recognized character belongs can be determined based on the multi-dimensional feature information. Different character types correspond to different feature fusion weights. For example, in the case of an arc character (C, O, U), the feature fusion weight corresponding to the radian information can be configured to be high (for example, *4); in the case of a straight line character (I, L, 7), the feature fusion weight corresponding to the straight line feature can be configured to be high (for example, *3-5); in the case of a wave character (S, W, M), the feature fusion weight corresponding to the number of waves can be configured to be high (for example, *2-3); and in the case of a complex character, the weights of the multiple dimensions of features can be equal. Therefore, after determining the at least one character type, the at least one character type can be analyzed to determine a feature fusion weight corresponding to each of the multi-dimensional feature information. The feature fusion weight can be determined based on the to-be-recognized character and the application scenario corresponding to the multi-dimensional feature information, or the feature fusion weight can be determined based on a preset mapping relationship, the to-be-recognized character, and the multi-dimensional feature information.

[0110] After determining the feature fusion weight corresponding to each of the multi-dimensional feature information, the multi-dimensional feature information can be fused based on the feature fusion weight, i.e., the multi-dimensional feature information can be weighted and fused based on the feature fusion weight to obtain the fused feature information. Then, the character matching degree between the to-be-recognized character and each preset type of character template can be determined based on the fused feature information, which effectively ensures the accuracy and reliability of determining the character matching degree.

[0111] Step S403: determining at least one target character template matched with the to-be-recognized character from the plurality of preset types of character templates based on the character matching degree.

[0112] Since the character matching degree can identify the similarity between the to-be-recognized character and each character template, the higher the character matching degree is, the higher the similarity between the to-be-recognized character and the character template is; the lower the character matching degree is, the lower the similarity between the to-be-recognized character and the character template is. Therefore, after obtaining the character matching degree, at least one target character template matched with the to-be-recognized character can be determined from the plurality of preset types of character templates based on the character matching degree.

[0113] In some examples, the target character template can be determined directly based on the size of the character matching degree, and at this time, determining at least one target character template matched with the to-be-recognized character from the plurality of preset types of character templates based on the character matching degree can include: comparing the character matching degree with a preset threshold; in a case where the character matching degree is greater than or equal to the preset threshold, determining the character template corresponding to the character matching degree as at least one target character template matched with the to-be-recognized character, which effectively ensures the accuracy and reliability of determining the target character template.

[0114] In another example, in the process of identifying whether the preset type of character template is the target character template, different types of character templates can correspond to different preset thresholds, and at this time, determining at least one target character template matched with the to-be-recognized character from the plurality of preset types of character templates based on the character matching degree can include: determining a character matching threshold corresponding to each preset type of character template; determining at least one target character template matched with the to-be-recognized character from the plurality of preset types of character templates based on the character matching degree and the character matching threshold.

[0115] In which, different types of character templates can correspond to different character matching thresholds, in order to accurately determine the target character template matched with the to-be-recognized character, the character matching threshold corresponding to each preset type of character template can be determined first, specifically, different preset types of character templates can correspond to the same or different character matching thresholds, the character matching threshold can be determined based on a preset mapping relationship and the type of character template, after determining the character matching threshold corresponding to each preset type of character template, the character matching degree and the character matching threshold can be analyzed and matched to determine at least one target character template matched with the to-be-recognized character from the plurality of preset types of character templates according to the analysis and matching result.

[0116] In some examples, based on the character matching degree and the character matching threshold, determining at least one target character template matching the to-be-recognized character from the plurality of preset types of character templates can include: in a case where the character matching degree is greater than or equal to the character matching threshold, determining the character template corresponding to the character matching threshold as the target character template; in a case where the character matching degree is less than the character matching threshold, determining the character template corresponding to the character matching threshold as not the target character template, so as to achieve the accuracy and reliability of determining the target character template.

[0117] It can be understood that the determination manner of the target character template can not only be determined by comparing the character matching degree with the character matching threshold, but also can be determined by analyzing and processing the character matching degree and the character matching threshold by using a pre-trained network model, as long as the accuracy and reliability of determining the target character template can be ensured, and details are not repeated here.

[0118] In this embodiment, by determining a plurality of preset types of character templates for analyzing and processing the to-be-recognized character, then determining the character matching degree between the to-be-recognized character and each preset type of character template based on the multi-dimensional feature information, and determining at least one target character template matching the to-be-recognized character from the plurality of preset types of character templates based on the character matching degree, the accuracy and reliability of determining the target character template are effectively ensured.

[0119] Figure 5 Another flowchart of a character recognition method provided by an example embodiment of the present application is provided; based on any one of the above embodiments, referring to FIG. 8, after the target character of the to-be-recognized character is determined, the method in this embodiment can further include: Figure 5 After the target character of the to-be-recognized character is determined, the method in this embodiment can further include:

[0120] Step S501: In a case where the to-be-recognized character is a to-be-recognized character in the display information, determining a jump character corresponding to the display information.

[0121] After the target character of the to-be-recognized character is determined, the application scenario corresponding to the to-be-recognized character can be determined, and in a case where the application scenario is that the to-be-recognized character is a to-be-recognized character in the display information, in order to accurately and accurately regulate the display information, a jump character corresponding to the display information can be determined first, the jump character can be determined based on a preset mapping relationship and an information identifier of the display information, specifically, different display information can correspond to different jump characters, and the display information can be implemented as information displayed in a webpage, an advertisement, or an opening screen advertisement.

[0122] Step S502: Controlling the display information based on the target character and the jump character.

[0123] Since the jump character is a condition for determining whether the input character in the display information meets the page jump operation, after obtaining the jump character and the target character, the target character and the jump character can be analyzed and processed to enable the display information to be controlled according to the analysis and processing result.

[0124] In some examples, the control operation on the display information can be implemented based on a pre-trained regulation network model, and at this time, controlling the display information based on the target character and the jump character can include: determining a regulation network model for controlling the display information; inputting the target character and the jump character into the regulation network model for processing, which can not only determine the matching result between the target character and the jump character, but also generate a regulation instruction based on the matching result and control the display information based on the regulation instruction. Specifically, in the case of a first type of instruction, the jump page corresponding to the display information is determined and displayed; in the case of a second type of instruction, the page jump operation on the display information is not required, i.e., the state of continuously displaying the display information is maintained, which realizes accurate control operation on the display information.

[0125] In other examples, the control operation on the display information can be implemented not only based on a pre-trained regulation network model, but also directly based on the matching result of the target character and the jump character, and at this time, controlling the display information based on the target character and the jump character can include: in the case that the target character and the jump character match, determining and displaying the jump page corresponding to the display information; in the case that the target character and the jump character do not match, maintaining the continuous display of the display information.

[0126] In this embodiment, in the case that the to-be-recognized character is a to-be-recognized character in the display information, the jump character corresponding to the display information is determined, and then the display information is controlled based on the target character and the jump character, which also realizes flexible and accurate control operation on the display information, and further improves the practicability of the method.

[0127] In a specific application, the application scenario of the opening screen advertisement is taken as an example, and as shown in FIG. 1, the application embodiment provides a regulation method of the opening screen advertisement, which can include the following steps: Figure 6

[0128] Step 1: In the application scenario of the opening screen advertisement, obtain a sequence of track points of inputting a to-be-recognized character in the opening screen advertisement.

[0129] ​The trajectory point sequence input in the opening screen advertisement can be implemented as a two-dimensional array, and the trajectory point sequence can be implemented as {(x1, y1), (x2, y2), (x3, y3), (x4, y4), (x5, y5), (x6, y6)} and the like. In order to accurately regulate and operate the opening screen advertisement, after obtaining the trajectory point sequence input in the opening screen advertisement, the trajectory point sequence can be preprocessed, and the preprocessing operation can include at least one of the following: noise removal operation, missing value filling operation and the like, to ensure the accuracy and reliability of the determination of the trajectory point sequence.

[0130] In addition, in order to avoid the recognition operation of the mis-touch character, after obtaining the trajectory point sequence, the number of trajectory points included in the trajectory point sequence can be determined, and then the number of trajectory points can be compared with the preset threshold (for example: 3, 4 and the like). If the number of trajectory points is greater than or equal to the preset threshold, it means that the trajectory point sequence includes a large number of trajectory points, and the character corresponding to the trajectory point sequence at this time is probably a character that needs to be normally recognized, and the next step is allowed to be executed; if the number of trajectory points is less than the preset threshold, it means that the trajectory point sequence includes a small number of trajectory points, and the character corresponding to the trajectory point sequence at this time is probably a character of mis-touch operation, and the next step is prohibited to be executed.

[0131] Step 2: The multi-dimensional feature extractor (which can be set on the client or terminal device) is used to perform feature extraction operation on the trajectory point sequence to obtain multi-dimensional feature information corresponding to the to-be-recognized character, wherein the multi-dimensional feature information at least includes: geometric feature, motion feature, shape feature, direction feature.

[0132] As shown in the reference Figure 7 The geometric feature can include at least one of the following: curvature information, aspect ratio information, and closeness information. The motion feature can include at least one of the following: direction change information, smoothness information, and speed change information. The shape feature can include at least one of the following: arc fitting information, straight line fitting information, and wave feature information. The direction feature can include at least one of the following: dominant direction information, angle histogram information, and direction consistency information.

[0133] Specifically, the implementation principle of the multi-dimensional feature extractor for extracting multi-dimensional feature information is as follows:

[0134] In the case that the multi-dimensional feature information includes the aspect ratio, the bounding box of the to-be-recognized character can be calculated based on the trajectory point sequence, and the maximum X coordinate value, the minimum X coordinate value, the maximum Y coordinate value, and the minimum Y coordinate value can be obtained based on the bounding box. Then, the width information of the to-be-recognized character can be determined as max X-min X, the height information can be determined as max Y-min Y, and the aspect ratio can be determined as the width information / height information. In this way, the accurate determination of the aspect ratio is stably achieved.

[0135] In the case that the multi-dimensional feature information includes the closedness information, the straight-line distance from the starting point to the ending point of the to-be-recognized character can be calculated based on the trajectory point sequence, and the total length of the trajectory of the to-be-recognized character (the sum of the distances between all adjacent trajectory points) can be calculated. Then, the distance ratio can be determined based on the straight-line distance and the total length of the trajectory, which is the starting and ending point distance / total length of the trajectory. After that, the closedness information can be determined based on the distance ratio, which can be implemented as score information. At this time, the closedness information is 1-(distance ratio*3). In this way, the accurate determination of the closedness information is stably achieved.

[0136] In the case that the multi-dimensional feature information includes the curvature information, each point (except the first and last points) in the trajectory corresponding to the to-be-recognized character can be determined based on the trajectory point sequence, a plurality of adjacent points before and after the to-be-recognized character can be determined, two adjacent vectors can be calculated based on the plurality of adjacent points before and after, which can include v1 (from the previous point to the current point) and v2 (from the current point to the next point). Then, the cosine value of the angle between the two adjacent vectors v1 and v2 can be calculated based on the two adjacent vectors v1 and v2. The angle (in radian) between the two vectors v1 and v2 can be obtained through the inverse cosine function. By sequentially processing the sequence of trajectory points, a plurality of angles can be determined. Then, the average of all angles can be determined as the average curvature corresponding to the to-be-recognized character. In this way, the accurate determination of the curvature information is stably achieved.

[0137] In the case that the multi-dimensional feature information includes the curvature information, each point (except the first and last points) in the trajectory corresponding to the to-be-recognized character can be determined based on the trajectory point sequence, a plurality of adjacent points before and after the to-be-recognized character can be determined, two adjacent vectors can be calculated based on the plurality of adjacent points before and after, which can include v1 (from the previous point to the current point) and v2 (from the current point to the next point). Then, the cosine value of the angle between the two adjacent vectors v1 and v2 can be calculated based on the two adjacent vectors v1 and v2. The angle (in radian) between the two vectors v1 and v2 can be obtained through the inverse cosine function. By sequentially processing the sequence of trajectory points, a plurality of angles can be determined. Then, the average of all angles can be determined as the average curvature corresponding to the to-be-recognized character. In this way, the accurate determination of the curvature information is stably achieved.

[0138] In the case that the multi-dimensional feature information includes smoothness information, a plurality of adjacent trajectory points included in the to-be-recognized character can be determined based on the trajectory point sequence, and a plurality of adjacent line segments can be determined based on the plurality of adjacent trajectory points. Then, the angle change information between all adjacent line segments in the to-be-recognized character can be calculated, and the average value and variance information of the angle change can be determined based on the angle change information. The angle change standard deviation corresponding to the to-be-recognized character can be determined based on the variance information. Finally, the smoothness information can be determined based on the angle change standard deviation. At this time, the smoothness information = 1 - angle change standard deviation. In this way, the accurate determination of the smoothness information is stably realized.

[0139] Since the dominant direction information can reflect the overall trend of the to-be-recognized character, for example, the vertical stroke (I, 1) mainly goes downward, the horizontal stroke mainly goes right, and the diagonal stroke ( / , \) has a specific inclination direction, therefore, the multi-dimensional feature information can include the dominant direction information. At this time, the starting point and the ending point of the to-be-recognized character can be determined based on the trajectory point sequence, and the vector information from the starting point to the ending point can be calculated. The vector angle (-180° to 180°) can be obtained by using the atan2 function to calculate the vector information. Then, the angle range is divided into 8 direction intervals. The dominant direction information of the to-be-recognized character can be determined based on the vector angle and the 8 direction intervals. Specifically, the direction interval in which the vector angle is located can be identified, and the dominant direction information corresponding to the to-be-recognized character can be determined based on the direction interval. In this way, the accurate determination of the dominant direction information is stably realized.

[0140] Since the number of direction changes can reflect the complexity of the trajectory corresponding to the to-be-recognized character, for example, the number of direction changes of a simple character (such as I, O) is small, and the number of direction changes of a complex character (such as M, W, Z) is multiple and obvious. Therefore, the multi-dimensional feature information can include the number of direction changes. At this time, a plurality of adjacent trajectory points can be determined based on the trajectory point sequence, and a plurality of line segments can be determined based on the plurality of adjacent trajectory points. Then, the unit direction vector of each line segment can be calculated, and the dot product information can be obtained by comparing the direction vectors of adjacent line segments. If the dot product information is less than a preset 0.7 (corresponding to about 45°), it is determined that a direction change has occurred. If the dot product information is greater than or equal to the preset 0.7, it is determined that no direction change has occurred, which can be ignored. In the process of direction vector change for the plurality of line segments, the line segments that are too short (length < 5 pixels) can be ignored. According to the above manner, the direction statistics operation is sequentially performed on all line segments, and the number of direction changes corresponding to the to-be-recognized character is obtained. In this way, the accurate determination of the number of direction changes is stably realized.

[0141] Since the angle histogram can provide statistical information of the direction distribution of the character to be recognized, different characters have characteristic direction distribution, for example, the vertical character (I, 1) is mainly distributed near 90 degrees, the horizontal character is mainly distributed at 0 degrees and 180 degrees, and the circular character (O, 0) is uniformly distributed in each direction. Therefore, the multi-dimensional feature information can include the angle histogram, at this time, 360 degrees can be divided into 8 intervals of 45 degrees, and then the number of line segments in each angle interval can be counted based on the direction angle of the line segment of the trajectory point sequence, and the angle histogram information of the probability distribution can be obtained by normalizing the number of line segments in each angle interval, so that the angle histogram information is accurately determined.

[0142] Since the number of waves is a key feature for recognizing wave-shaped characters such as S, W, and M, the characteristic wave pattern of such characters can be accurately recognized by detecting the fluctuation of the trajectory, therefore, the multi-dimensional feature information can include the number of waves, at this time, the sliding window is used to slide the trajectory point sequence to determine the X coordinate sequence and the Y coordinate sequence; the local extreme points (peaks and valleys) in the Y direction are detected based on the Y coordinate sequence, the number of direction changes is calculated based on the local extreme points, and the number of waves is obtained by summing the Y direction extreme value and the number of direction changes, so that the number of waves is accurately determined.

[0143] Since the straight line fitting information (or called straight line score) can measure the similarity between the trajectory of the character to be recognized and the straight line, for example, the straight line type character (such as I, L, and part of the strokes of 7) has a higher straight line score, and the curved line type character has a lower straight line score, therefore, the multi-dimensional feature information can include the straight line fitting information, at this time, the starting point and the ending point are determined based on the trajectory point sequence; the straight line is defined as the connecting line from the starting point to the ending point; the perpendicular distance of each intermediate point in the character to be recognized to the straight line is calculated; the relative average distance of the plurality of perpendicular distances is determined, and then the straight line fitting information can be determined based on the relative average distance, in some examples, the straight line fitting information can be implemented as score information, i.e. straight line fitting information = 1 / (1+relative average distance), so that the straight line fitting information (or called straight line score) is accurately determined.

[0144] It is worth noting that the multi-dimensional feature extractor can not only obtain multi-dimensional feature information corresponding to the character to be recognized, but also can perform a specific shape detection operation on the character to be recognized. The specific shape detection operation is designed for characters that are difficult to distinguish using general features. These rules are based on the unique geometric characteristics of the characters, which improve the recognition accuracy. Specifically, the multi-dimensional feature information can include specific shape detection score information, for example, the specific shape detection operation can include a "3" character shape detection operation score, a "7" character shape detection operation score, a "C" character shape detection operation score, etc. The "7" character shape detection operation can include at least one of the following: detecting the horizontal start part, detecting the downward diagonal end part, and detecting the curvature feature at the corner. The "C" character shape detection operation can include checking the opening direction (should open to the right), checking the angle coverage range (120-240 degrees), checking the up-down symmetry, checking the trajectory continuity, etc. Through the above operations, the detection score information of the character to be recognized and the specific shape can be obtained.

[0145] Step 3: Based on the multi-dimensional feature information, a multi-dimensional feature vector is constructed.

[0146] After obtaining the multi-dimensional feature information, the vector representation and weight information of each dimension of the feature information can be determined. Then, the vector representation of the multi-dimensional feature information can be fused based on the weight information to obtain a multi-dimensional feature vector.

[0147] Step 4: Based on the multi-dimensional feature vector, traverse the character template library to obtain at least one target character template that matches the character to be recognized.

[0148] The character template library can include uppercase letter templates, lowercase letter templates, number templates, direction gesture templates, etc. The character templates in the character template library can support self-customization operations of the character templates. Users can flexibly adjust or configure the templates in the character template library according to their needs. Generally, the character template library can abstract four dimensions of features, i.e., geometric dimension features, motion dimension features, shape dimension features, and direction dimension features, etc.

[0149] Specifically, in the process of traversing the character template library, in view of the irregularity of the handwritten character, the following measures can be taken to realize the maximum matching operation: different types of character templates can correspond to different preset thresholds, for example: the number character template corresponds to a preset threshold 1, the letter character template corresponds to a preset threshold 2, and the direction character template corresponds to a preset threshold 3. When using the number character template to determine the type of the to-be-recognized character, a number-specific matching algorithm can be used to calculate the multi-dimensional feature vector and the number character template to obtain a number matching score of the to-be-recognized character; when using the letter character template to determine the type of the to-be-recognized character, a letter-specific matching algorithm can be used to calculate the multi-dimensional feature vector and the letter character template to obtain a letter matching score of the to-be-recognized character; and when using the direction character template to determine the type of the to-be-recognized character, a direction-specific matching algorithm can be used to calculate the multi-dimensional feature vector and the direction character template to obtain a direction matching score.

[0150] After obtaining the matching scores of the to-be-recognized character and each different type of character template (including: number matching score, letter matching score, and direction matching score), the obtained matching scores can be compared with the preset thresholds corresponding to each type of character template, that is, the number matching score is compared with the preset threshold 1, the letter matching score is compared with the preset threshold 2, and the direction matching score is compared with the preset threshold 3. In the case that the analysis and comparison results are that the matching scores are greater than the corresponding preset thresholds, that is, number matching score > preset threshold 1, letter matching score > preset threshold 2, and direction matching score > preset threshold 3, it indicates that the to-be-recognized character is compatible with the current number character template, letter character template, and direction character template, and thus the above adapted number character template, letter character template, and direction character template can be determined as the target character template corresponding to the to-be-recognized character, which stably realizes the accuracy and reliability of determining the target character template.

[0151] It should be noted that after obtaining at least one target character template matched with the to-be-recognized character, different dimensional feature vectors can be used for character optimization recognition operation for different character templates. For the number character template, different dimensional feature vectors can be used for character optimization recognition operation for different types of number character templates, for example: in the case that the target character template includes the number 1, vertical bias detection operation and aspect ratio constraint can be used for character optimization recognition operation; in the case that the target character template includes the number 3, arc feature weight enhancement operation and S-shaped detection can be used for character optimization recognition operation; and in the case that the target character template includes the number 7, horizontal-diagonal line combination mode recognition operation can be used for character optimization recognition operation.

[0152] Similarly, for the letter character template, different dimensional feature vectors can also be used for different types of letter character templates for character optimization recognition operation, for example: in the case of including the letter w in the target character template, the arc wave pattern and width features can be used for character optimization recognition operation; in the case of including the letter d / p in the target character template, the circular + straight line combination detection operation can be used for character optimization recognition operation; in the case of including the number e in the target character template, the open arc feature recognition operation can be used for character optimization recognition operation, which is beneficial to improve the accuracy and reliability of determining the target character template.

[0153] Step 4: Based on at least one target character template matched with the to-be-recognized character, determining the target character of the to-be-recognized character.

[0154] After obtaining at least one target character template, the target character corresponding to the at least one target character template can be determined as the target character of the to-be-recognized character. In some examples, the number of target character templates can be multiple, for example: the number of target character templates is 20, and then the target characters corresponding to the 20 target character templates can be determined as the target character of the to-be-recognized character, which is beneficial to improve the recognition success rate of the to-be-recognized character.

[0155] Step 5: In the case that the to-be-recognized character is in the opening screen advertisement, determining a jump character corresponding to the opening screen advertisement; controlling the opening screen advertisement based on the target character and the jump character.

[0156] In the case that the target character and the jump character match, a jump page corresponding to the opening screen advertisement is determined and displayed; in the case that the target character and the jump character do not match, the opening screen advertisement is continuously displayed.

[0157] The technical scheme provided by the application embodiment can accurately implement character recognition operation based on multi-dimensional feature information, which not only can achieve extremely low delay (within 10ms), maximize the matching rate of characters (top N results), but also can ensure that the obtained recognition result has high fault tolerance (angle and arc interchange), in addition, when the device capable of implementing character recognition operation is deployed on a mobile device, most gesture characters can be recognized within 10ms, thereby ensuring the practicability of the scheme, which is beneficial to market promotion and application.

[0158] Figure 8 A flowchart of a control method for displaying information is provided for an exemplary embodiment of the present application. Reference is made to the accompanying drawings Figure 8As shown, the embodiment provides a control method of display information, an execution subject of the method is a control device of display information, the control device of display information can be implemented as software or a combination of software and hardware, when the control device of display information is implemented as hardware, it can be various electronic devices capable of realizing the control operation of display information, including but not limited to personal computers, servers and the like, when the control device of display information is implemented as software, it can be installed in the above-mentioned electronic devices. Specifically, the control method of display information provided in the embodiment can include:

[0159] Step S801: obtaining a track point sequence corresponding to a to-be-recognized character in display information.

[0160] Step S802: determining multi-dimensional feature information of the to-be-recognized character based on the track point sequence, the multi-dimensional feature information including at least one of the following: geometric features, motion features, shape features, and direction features.

[0161] Step S803: determining a target character of the to-be-recognized character based on the multi-dimensional feature information and a plurality of preset types of character templates.

[0162] Determining the target character of the to-be-recognized character based on the multi-dimensional feature information and the plurality of preset types of character templates can include: determining at least one target character template matching the to-be-recognized character in the plurality of preset types of character templates based on the multi-dimensional feature information; and determining the target character of the to-be-recognized character based on the at least one target character template. Specifically, the specific implementation manner and implementation effect of steps S801-S803 in the embodiment are similar to those of steps S201-S204 in the above-mentioned embodiment, and details are not repeated here. Figure 2 The specific implementation manner and implementation effect of steps S201-S204 in the embodiment are similar to those of steps S201-S204 in the above-mentioned embodiment, and details are not repeated here.

[0163] Step S804: controlling the display information based on the target character.

[0164] After obtaining the target character, the display information can be controlled based on the target character, in some instances, controlling the display information based on the target character can include: determining a jump character corresponding to the display information; and controlling the display information based on the target character and the jump character.

[0165] Controlling the display information based on the target character and the jump character can include: in a case where the target character matches the jump character, determining and displaying a jump page corresponding to the display information; and in a case where the target character does not match the jump character, maintaining the display of the display information.

[0166] Specifically, the specific implementation manner and implementation effect of the above step in this embodiment are similar to those of the step "controlling the display information based on the target character and the jump character" in the above embodiment, and reference can be made to the above statements. Details are not described herein again.

[0167] The method for controlling display information provided in this embodiment can obtain a track point sequence corresponding to a to-be-recognized character in display information, determine multi-dimensional feature information of the to-be-recognized character based on the track point sequence, then determine a target character of the to-be-recognized character based on the multi-dimensional feature information and a plurality of preset types of character templates, and can control the display information based on the target character. In this way, in an application scenario capable of enhancing the interest of an advertisement, not only can the conversion efficiency of the display information be improved and the occurrence of a mis-touch behavior be avoided, but also a precise user screening operation can be implemented, that is, a user who enters a display information jump page is a user interested in the content of the display information, thereby being beneficial to improving the post-link conversion rate of the display information and further improving the practicability of the method, and being beneficial to the promotion and application in the market.

[0168] Figure 9 FIG. 1 is a structural schematic diagram of a character recognition device provided for an exemplary embodiment of the present application; refer to FIG. 1 Figure 9 The character recognition device provided in this embodiment is used for executing the character recognition method shown in FIG. 1, and the device can include: Figure 2

[0169] The first obtaining module 11 is configured to obtain a track point sequence corresponding to a to-be-recognized character.

[0170] The first determining module 12 is configured to determine multi-dimensional feature information of the to-be-recognized character based on the track point sequence, the multi-dimensional feature information including at least one of the following: geometric features, motion features, shape features, and direction features.

[0171] The first processing module 13 is configured to determine at least one target character template matching the to-be-recognized character from a plurality of preset types of character templates based on the multi-dimensional feature information.

[0172] The first processing module 13 is further configured to determine a target character of the to-be-recognized character based on the at least one target character template.

[0173] The character recognition device in this embodiment can also perform the description of the above Figures 1-8 embodiment shown in FIG. 1, and reference can be made to the detailed description in the above embodiments. Details are not described herein again.

[0174] ​Furthermore, in some of the processes described in the above embodiments and accompanying drawings, multiple operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or they may be executed in parallel. The operation numbers, such as 11, 12, etc., are merely used to distinguish different operations and do not represent any execution order. Additionally, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first" and "second" in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0175] Figure 10 A schematic diagram of the structure of an electronic device provided in an exemplary embodiment of this application; as shown Figure 10 As shown, this embodiment provides an electronic device for performing the above-described... Figure 2 The character recognition method shown includes an electronic device that may include a memory 24 and a processor 25.

[0176] Memory 24 is used to store computer programs and can be configured to store various other data to support operation on the electronic device. Examples of this data include instructions for any application or method used to operate on the electronic device, data structures, contact data, phone book data, messages, pictures, videos, etc.

[0177] The processor 25, coupled to the memory 24, is used to execute a computer program in the memory 24 for: acquiring a sequence of trajectory points corresponding to a character to be recognized; determining multidimensional feature information of the character to be recognized based on the sequence of trajectory points, the multidimensional feature information including at least one of the following: geometric features, motion features, shape features, and orientation features; determining at least one target character template that matches the character to be recognized from multiple preset character templates based on the multidimensional feature information; and determining the target character of the character to be recognized based on the at least one target character template.

[0178] Furthermore, such as Figure 10 As shown, the electronic device also includes other components such as a communication component 26, a display 27, a power supply component 28, and an audio component 29. Figure 10 The diagram only shows some components and does not mean that the electronic device includes only these components. Figure 10 The components shown. Additionally... Figure 10The components in the midline frame are optional components, not mandatory components, and can be determined according to the product form of the working node. The working node of the embodiment can be implemented as a terminal device such as a desktop computer, a notebook computer, a smart phone, or an IOT device, or can be a server device such as a general server, a cloud server, or a server array. If the working node of the embodiment is implemented as a terminal device such as a desktop computer, a notebook computer, or a smart phone, it can include Figure 10 components in the midline frame; if the working node of the embodiment is implemented as a server device such as a general server, a cloud server, or a server array, it can not include Figure 10 components in the midline frame.

[0179] The above-mentioned memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, a magnetic disk, or an optical disk.

[0180] The above-mentioned communication component is configured to facilitate wired or wireless communication between the device where the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as a 2G, 3G, 4G / LTE, 5G, or the like mobile communication network, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel.

[0181] The above-mentioned display includes a screen, which can include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a swipe, and a gesture on the touch panel. The touch sensor can not only sense the boundary of a touch or swipe action, but also detect the duration and pressure associated with the touch or swipe operation.

[0182] The aforementioned power supply components provide power to various components within the device in which they reside. These power supply components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device in which they reside.

[0183] The aforementioned audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive external audio signals when the device containing the audio component is in an operating mode, such as call mode, recording mode, or voice recognition mode. The received audio signals can be further stored in memory or transmitted via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0184] Figure 11 A schematic diagram of the structure of a control device for displaying information is provided for an exemplary embodiment of this application; see attached drawing. Figure 11 As shown, this embodiment provides a control device for displaying information, which is used to perform the above-described... Figure 2 The control method for displaying the information shown, the device may include:

[0185] The second acquisition module 31 is used to acquire the trajectory point sequence corresponding to the character to be recognized in the displayed information;

[0186] The second determining module 32 is used to determine the multidimensional feature information of the character to be identified based on the trajectory point sequence. The multidimensional feature information includes at least one of the following: geometric features, motion features, shape features, and orientation features.

[0187] The second determining module 32 is also used to determine the target character of the character to be identified based on multi-dimensional feature information and multiple preset character templates;

[0188] The second processing module 33 is used to control the displayed information based on the target character.

[0189] The control device for displaying information in this embodiment can also perform the above-described actions. Figures 1-8 The description of the embodiments shown is for reference only, and will not be elaborated upon here.

[0190] Figure 12 A schematic diagram of the structure of an electronic device provided in an exemplary embodiment of this application; as shown Figure 12 As shown, this embodiment provides an electronic device for performing the above-described... Figure 8 The method for controlling the display of information shown includes an electronic device that may include a memory 44 and a processor 45.

[0191] The memory 44 is configured to store computer programs and can be configured to store various data to support operations on the electronic device. Examples of the data include instructions, data structures, contact data, phonebook data, messages, pictures, videos, etc. of any application or method for operating on the electronic device.

[0192] The processor 45 is coupled with the memory 44 and is configured to execute the computer programs in the memory 44 to: acquire a sequence of track points corresponding to a to-be-recognized character in the presentation information; determine multi-dimensional feature information of the to-be-recognized character based on the sequence of track points, the multi-dimensional feature information including at least one of: geometric features, motion features, shape features, and direction features; determine a target character of the to-be-recognized character based on the multi-dimensional feature information and a plurality of preset type character templates; and control the presentation information based on the target character.

[0193] Further, as shown in Figure 12 , the electronic device further includes a communication component 46, a display 47, a power supply component 48, an audio component 49, and other components. Figure 12 The partial components shown in Figure 12 do not mean that the electronic device only includes the components shown in Figure 12 . In addition, Figure 12 the components within the dashed line in Figure 12 are optional components rather than mandatory components, and the specific product form of the working node can be determined according to the actual situation. The working node of the present embodiment can be implemented as a terminal device such as a desktop computer, a notebook computer, a smart phone, or an IOT device, or a server device such as a conventional server, a cloud server, or a server array. If the working node of the present embodiment is implemented as a terminal device such as a desktop computer, a notebook computer, a smart phone, etc., it can include the components within the dashed line in ; if the working node of the present embodiment is implemented as a server device such as a conventional server, a cloud server, or a server array, it can not include the components within the dashed line in

[0194] .The above-described memory can be implemented by any type of volatile or nonvolatile memory devices or a combination thereof, such as a Static Random-Access Memory (SRAM), an Electrically Erasable Programmable Read Only Memory (EEPROM), an Erasable Programmable Read Only Memory (EPROM), a Programmable Read-Only Memory (PROM), a Read-Only Memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or a compact disk.

[0195] The above-described communication component is configured to facilitate wired or wireless communication between the device in which the communication component is located and other devices. The device in which the communication component is located can access a wireless network based on a communication standard, such as a 2G, 3G, 4G / LTE, 5G, or the like mobile communication network, or a combination thereof. In an example embodiment, the communication component receives a broadcast signal or broadcast-related information from an external broadcast managing system via a broadcast channel.

[0196] The above-described display includes a screen, which can include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide, and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touching or a sliding action, but also detect a duration and a pressure associated with the touching or the sliding action.

[0197] The above-described power component provides power to various components of the device in which the power component is located. The power component can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device in which the power component is located.

[0198] The above-described audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) that is configured to receive an external audio signal when the device in which the audio component is located is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal can be further stored in the memory or transmitted via the communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0199] Accordingly, the embodiments of the present application also provide a computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to implement the steps of the above-mentioned method embodiments. The computer readable storage medium includes volatile or non-volatile or a combination thereof, and can be removable or non-removable. Examples of the computer readable storage medium include, but are not limited to, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital video disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store program code in the form of instructions or data structures and that can be accessed by a general purpose or special purpose computer, or both.

[0200] Accordingly, the embodiments of the present application also provide a computer program product, which includes a computer program or instructions, and when the computer program or instructions are executed by a processor, causes the processor to implement the steps of the above-mentioned method embodiments. It should be understood that each of the above-mentioned method processes or a combination thereof can be implemented by the computer program or instructions. In addition, these computer programs or instructions can be applied to the processor of a general purpose computer, a special purpose computer, an embedded processor or other programmable data processing devices, so that the processor of the general purpose computer, the special purpose computer, the embedded processor or other programmable data processing devices can be implemented as a device for implementing the corresponding functions in the above-mentioned method embodiments.

[0201] It should also be noted that the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed, or further include elements inherent in such processes, methods, articles or devices. Without more limitations, the element defined by the statement "including a" does not exclude the presence of additional identical elements in the process, method, article or device including the element.

[0202] The above merely provides an example of the present application, and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall fall into the scope of claims of the present application.

Claims

1. A character recognition method, characterized in that, include: Obtain the sequence of trajectory points corresponding to the character to be recognized; Based on the trajectory point sequence, multidimensional feature information of the character to be identified is determined, wherein the multidimensional feature information includes at least one of the following: geometric features, motion features, shape features, and orientation features; Based on the multidimensional feature information, at least one target character template that matches the character to be identified is determined from multiple preset character template types; Based on the at least one target character template, the target character of the character to be identified is determined.

2. The method according to claim 1, characterized in that, Based on the multidimensional feature information, at least one target character template matching the character to be identified is determined from the character template, including: A plurality of preset character templates of various types are determined for analyzing and processing the character to be identified. The plurality of preset character templates of various types include at least: character templates of number type, character templates of letter type, and character templates of direction type. Based on the multidimensional feature information, the character matching degree between the character to be identified and each preset type of character template is determined; Based on the character matching degree, at least one target character template that matches the character to be identified is determined from multiple preset type character templates.

3. The method according to claim 2, characterized in that, Based on the multidimensional feature information, the character matching degree between the character to be identified and each preset type of character template is determined, including: The multidimensional feature information is fused to obtain fused feature information; Based on the fused feature information, the character matching degree between the character to be identified and each preset type of character template is determined.

4. The method according to claim 3, characterized in that, The multidimensional feature information is fused to obtain fused feature information, including: Based on the multidimensional feature information, at least one character type to which the character to be identified belongs is determined; Based on the at least one character type, determine the feature fusion weights corresponding to each of the multidimensional feature information; The multidimensional feature information is fused based on the feature fusion weights to obtain the fused feature information.

5. The method according to claim 2, characterized in that, Based on the character matching degree, at least one target character template that matches the character to be identified is determined from multiple preset type character templates, including: Determine the character matching threshold corresponding to each preset type of character template; Based on the character matching degree and the character matching threshold, at least one target character template that matches the character to be identified is determined from multiple preset type character templates.

6. The method according to claim 5, characterized in that, Based on the character matching degree and the character matching threshold, at least one target character template that matches the character to be identified is determined from multiple preset type character templates, including: If the character matching degree is greater than or equal to the character matching threshold, the character template corresponding to the character matching threshold is determined as the target character template; If the character matching degree is less than the character matching threshold, the character template corresponding to the character matching threshold is determined to be not the target character template.

7. The method according to any one of claims 1-6, characterized in that, After determining the target character of the character to be recognized, the method further includes: If the character to be identified is a character to be identified in the displayed information, determine the jump character corresponding to the displayed information; The displayed information is controlled based on the target character and the jump character.

8. The method according to claim 7, characterized in that, Controlling the displayed information based on the target character and the jump character includes: If the target character matches the jump character, determine and display the jump page corresponding to the displayed information; If the target character does not match the jump character, the displayed information will continue to be displayed.

9. A method for controlling the display of information, characterized in that, include: Obtain the sequence of trajectory points corresponding to the characters to be identified in the displayed information; Based on the trajectory point sequence, multidimensional feature information of the character to be identified is determined, wherein the multidimensional feature information includes at least one of the following: geometric features, motion features, shape features, and orientation features; Based on the multidimensional feature information and multiple preset character templates, the target character of the character to be identified is determined; The displayed information is controlled based on the target character.

10. The method according to claim 9, characterized in that, Controlling the displayed information based on the target character includes: Determine the jump character corresponding to the displayed information; The displayed information is controlled based on the target character and the jump character.

11. The method according to claim 10, characterized in that, Controlling the displayed information based on the target character and the jump character includes: If the target character matches the jump character, determine and display the jump page corresponding to the displayed information; If the target character does not match the jump character, the displayed information will continue to be displayed.

12. A character recognition device, characterized in that, include: The first acquisition module is used to acquire the sequence of trajectory points corresponding to the character to be recognized; The first determining module is used to determine the multidimensional feature information of the character to be identified based on the trajectory point sequence, wherein the multidimensional feature information includes at least one of the following: geometric features, motion features, shape features, and orientation features; The first processing module is used to determine at least one target character template that matches the character to be identified from multiple preset type character templates based on the multidimensional feature information. The first processing module is further configured to determine the target character of the character to be identified based on the at least one target character template.

13. A control device for displaying information, characterized in that, include: The second acquisition module is used to acquire the trajectory point sequence corresponding to the character to be recognized in the displayed information; The second determining module is used to determine the multidimensional feature information of the character to be identified based on the trajectory point sequence, wherein the multidimensional feature information includes at least one of the following: geometric features, motion features, shape features, and orientation features; The second determining module is further configured to determine the target character of the character to be identified based on the multidimensional feature information and multiple preset character templates; The second processing module is used to control the displayed information based on the target character.

14. An electronic device, characterized in that, include: A memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions, when executed by the processor, implement the method of any one of claims 1-11.

15. A computer storage medium, characterized in that, Used to store a computer program that, when executed by a computer, implements the method of any one of claims 1-11.

16. A computer program product, characterized in that, include: A computer program, when executed by a processor of an electronic device, causes the processor to perform the steps of the method of any one of claims 1-11.

Images