Mouse cursor intelligent switching method and system based on automatic identification

Abstract

The invention relates to the technical field of graphic processing, and discloses a mouse cursor intelligent switching method and system based on automatic identification, and the method comprises the steps: dividing a current interface into a plurality of target plates, and carrying out the shape initialization of a mouse cursor; querying a plate function of the target plate, and analyzing a use demand of a cursor user of the mouse cursor on the mouse cursor by using the mouse use information; querying a current plate of the cursor user, dynamically updating the use demand to obtain an update demand, and mapping the update demand into a shape demand of the cursor user for the mouse cursor; generating partial data points of the mouse cursor, performing data point adjustment on the partial data points, and connecting the adjusted data points into the cursor shape of the mouse cursor; and when the current position of the mouse cursor is in the target plate, switching the initialized shape into the cursor shape so as to complete the intelligent switching process of the mouse cursor. According to the method and the device, the self-adaptability of mouse cursor switching can be improved.

Description

Technical Field

[0001] The present invention relates to a method and system for intelligent switching of a mouse cursor based on automatic recognition, belonging to the technical field of graphics processing. Background Art

[0002] Currently, when switching the shape of a mouse cursor in the prior art, the switching is only based on the fixed cursor shapes pre-stored in a database. For example, if only an arrow-shaped cursor and a cross-star-shaped cursor are stored in the database, then when actually switching the shape of the mouse cursor, the switching is only between the arrow-shaped cursor and the cross-star-shaped cursor. Secondly, when switching the shape of the mouse cursor in the prior art, a cursor trajectory prediction algorithm is used to predict the cursor trajectory, and then the shape of the cursor is dynamically adjusted according to different factors such as the speed, direction, proximity, or sliding motion in the mouse movement stage. The adjustment of the cursor shape adopted is purely realized from the perspective of improving mouse performance, such as improving the speed, direction, proximity, etc. of mouse movement, but the switching of the cursor shape is not realized from the perspective of user use. For example, from the user's perspective, the needs of users are different when using the mouse, and the tendency of the required cursor shape is also different. This makes the existing cursor shape switching unable to dynamically adjust the attention to the shape tendency of different needs from the user's perspective. Therefore, the adaptability of mouse cursor switching is insufficient. Summary of the Invention

[0003] The present invention provides a method and system for intelligent switching of a mouse cursor based on automatic recognition, and its main purpose is to improve the adaptability of mouse cursor switching.

[0004] To achieve the above object, a method for intelligent switching of a mouse cursor based on automatic recognition provided by the present invention includes:

[0005] Obtain the current interface where the mouse cursor is located, identify the interface functions in the current interface, divide the current interface into multiple target sections according to the interface functions, and initialize the shape of the mouse cursor to obtain an initial shape;

[0006] Query the section functions of the target sections, match the similar functions corresponding to the section functions from a preset historical section library, query the mouse usage information corresponding to the similar functions from the historical section library, and analyze the usage requirements of the cursor user of the mouse cursor by using the mouse usage information;

[0007] Query the current section of the cursor user, obtain the current usage information in the current section, dynamically update the usage requirements according to the current usage information to obtain updated requirements, and map the updated requirements to the shape requirements of the cursor user for the mouse cursor;

[0008] Generate partial data points of the mouse cursor based on the shape requirement, adjust the partial data points to obtain adjusted data points, connect the adjusted data points as the cursor shape of the mouse cursor, and detect whether the current position of the mouse cursor is in the target plate;

[0009] When the current position of the mouse cursor is in the target plate, switch the initialization shape to the cursor shape to complete the intelligent switching process of the mouse cursor.

[0010] Optionally, dividing the current interface into multiple target plates according to the interface function includes:

[0011] Divide the areas in the current interface with the same interface function and adjacent pixel positions into the same target plate;

[0012] Determine multiple target plates in the current interface from the same target plate.

[0013] Optionally, analyzing the usage requirements of the cursor user of the mouse cursor by using the mouse usage information includes:

[0014] Calculate the average speed, mis-touch rate, stay duration, and path tortuosity of the cursor user with respect to the mouse cursor by using the mouse usage information;

[0015] Determine the operation fluency requirement, positioning accuracy requirement, information focusing requirement, and navigation cognition requirement of the cursor user with respect to the mouse cursor respectively through the average speed, the mis-touch rate, the stay duration, and the path tortuosity;

[0016] Take the operation fluency requirement, the positioning accuracy requirement, the information focusing requirement, and the navigation cognition requirement as usage requirements.

[0017] Optionally, dynamically updating the usage requirements according to the current usage information to obtain updated requirements includes:

[0018] Obtain the current speed, current mis-touch rate, current stay duration, and current path tortuosity in the current usage information;

[0019] Perform standardization processing on the current speed, the current mis-touch rate, the current stay duration, and the current path tortuosity respectively to obtain a standard speed, a standard mis-touch rate, a standard stay duration, and a standard path tortuosity;

[0020] Query the plate interval between the current plate and the target plate;

[0021] Determine the standard weights of the standard speed, the standard false touch rate, the standard stay duration, and the standard path tortuosity by using the plate interval;

[0022] According to the standard weights, perform linear mapping on the standard speed, the standard false touch rate, the standard stay duration, and the standard path tortuosity respectively to obtain a speed error, a false touch rate error, a stay duration error, and a path tortuosity error;

[0023] Dynamically update the usage requirements by using the speed error, the false touch rate error, the stay duration error, and the path tortuosity error to obtain updated requirements.

[0024] Optionally, mapping the updated requirements to the shape requirements of the cursor user for the mouse cursor includes:

[0025] Obtain the updated text and updated numerical value in the updated requirements;

[0026] Obtain the updated operation fluency requirement, the updated positioning accuracy requirement, the updated information focusing requirement, and the updated navigation cognition requirement in the updated text;

[0027] Convert the updated operation fluency requirement, the updated positioning accuracy requirement, the updated information focusing requirement, and the updated navigation cognition requirement into a streamline design requirement, an enhanced directivity requirement, a multi-level visualization requirement, and a direction indication requirement respectively;

[0028] According to the updated numerical value, determine the shape requirements of the mouse cursor by using the streamline design requirement, the enhanced directivity requirement, the multi-level visualization requirement, and the direction indication requirement.

[0029] Optionally, generating some data points of the mouse cursor based on the shape requirements includes:

[0030] Obtain the text requirement and numerical requirement corresponding to the shape requirements;

[0031] Perform numerical normalization on the numerical requirement by using the following formula to obtain a normalized numerical value:

[0032]

[0033] Among them, represents the normalized numerical value, x i represents the i-th type of numerical requirement, μ i represents the mean value of the i-th type of numerical requirement within the time window, σ i represents the variance of the i-th type of numerical requirement within the time window;

[0034] Encode the normalized numerical value into a numerical vector;

[0035] Input the numerical vector into a preset two - layer fully - connected network to output the high - order feature vector corresponding to the numerical vector through the two - layer fully - connected network;

[0036] Input the text requirement into a preset BERT model to extract the semantic embedding vector of the text requirement through the BERT model;

[0037] Perform vector fusion on the high - order feature vector and the semantic embedding vector to obtain a fused feature vector;

[0038] Input the fused feature vector into a preset generator network to generate partial data points corresponding to the fused feature vector through the generator network.

[0039] Optionally, the performing vector fusion on the high - order feature vector and the semantic embedding vector to obtain a fused feature vector includes:

[0040] Map the high - order feature vector into the vector space corresponding to the semantic embedding vector to obtain a mapped feature vector;

[0041] Concatenate the mapped feature vector and the semantic embedding vector to obtain a concatenated vector;

[0042] Calculate the vector weight corresponding to the concatenated vector through a preset gating network;

[0043] Based on the vector weight, use the following formula to perform vector fusion on the mapped feature vector and the semantic embedding vector to obtain a fused feature vector:

[0044]

[0045] where S represents the fused feature vector, w represents the vector weight, represents the mapped feature vector, Y represents the semantic embedding vector, and ⊙ represents the element - wise multiplication symbol.

[0046] Optionally, the adjusting the partial data points to obtain adjusted data points includes:

[0047] Set the number threshold of the partial data points using the following formula:

[0048]

[0049] where N represents the number threshold, PPI represents the pixel density unit, width k represents the width of the k - th target plate, Height k represents the height of the k - th target plate, and 200 represents the reference PPI;

[0050] Determine whether the number of the partial data points is less than the number threshold;

[0051] When the number of the partial data points is less than the number threshold, perform data point completion on the partial data points to obtain completed data points;

[0052] Use the partial data points and the completed data points as adjusted data points.

[0053] Optionally, the performing data point completion on the partial data points to obtain completed data points includes:

[0054] Perform curve fitting on the partial data points by using the following formula to obtain a fitting curve:

[0055]

[0056] where Z(α j ) represents the fitting curve, m represents the curve order, N j,m (α j ) represents the B-spline basis function, j represents the index of the partial data points, N gen represents the number of the partial data points, P j represents the j-th partial data point, α j represents a constant between 0 and 1;

[0057] Query the completion number of the partial data points;

[0058] According to the completion number, calculate the completed data points of the partial data points by using the following formula:

[0059]

[0060] where, represents the completed data points, represents the coordinate points determined by the completion number N com and u represents the serial number of the completed coordinate points.

[0061] To solve the above problems, the present invention further provides a mouse cursor intelligent switching system based on automatic recognition, and the system includes:

[0062] A shape initialization module, configured to obtain the current interface where the mouse cursor is located, identify the interface functions in the current interface, divide the current interface into multiple target sections according to the interface functions, and perform shape initialization on the mouse cursor to obtain an initialized shape;

[0063] A requirements analysis module for querying the functions of the target section, matching similar functions corresponding to the section functions from a preset historical section library, querying the mouse usage information corresponding to the similar functions from the historical section library, and analyzing the usage requirements of the cursor user regarding the mouse cursor using the mouse usage information;

[0064] A requirements mapping module for querying the current section of the cursor user, obtaining the current usage information in the current section, dynamically updating the usage requirements according to the current usage information to obtain updated requirements, and mapping the updated requirements to the shape requirements of the cursor user for the mouse cursor;

[0065] A mouse detection module for generating partial data points of the mouse cursor based on the shape requirements, adjusting the partial data points to obtain adjusted data points, connecting the adjusted data points to form the cursor shape of the mouse cursor, and detecting whether the current position of the mouse cursor is in the target section;

[0066] A cursor switching module for switching the initialization shape to the cursor shape when the current position of the mouse cursor is in the target section to complete the intelligent switching process of the mouse cursor.

[0067] Compared with the problems described in the background art, in the embodiments of the present invention, according to the interface function, the current interface is divided into multiple target sections, so as to set the same-shaped cursor for the same target section in the follow-up. Further, in the embodiments of the present invention, by using the mouse usage information to analyze the cursor user's usage requirements for the mouse cursor, when encountering a new interface not stored in the historical database, the historical usage information of the user is used as the basic cursor requirement for the new interface. Further, in the embodiments of the present invention, according to the current usage information, the usage requirements are dynamically updated, so as to correct the basic cursor requirement for the new interface based on the mouse usage information within the initial period of time after the user enters the new interface. Further, in the embodiments of the present invention, the updated requirements are mapped to the cursor user's shape requirements for the mouse cursor, so as to generate a text description and a numerical description of the cursor shape for the user from the perspective of the user's requirements, laying a foundation for adaptively generating a cursor shape that meets the user's requirements based on the text description and the numerical description in the follow-up. Further, in the embodiments of the present invention, based on the shape requirements, some data points of the mouse cursor are generated, so as to generate a dynamic and non-fixed-shaped cursor through a generator. Further, in the embodiments of the present invention, the partial data points are complemented, so as to fill the data points based on a fitting curve, ensuring the invariance of the curve shape of the original partial data points, and at the same time, improving the visual appreciation of the cursor shape by increasing the data points. Therefore, the intelligent switching method and system of the mouse cursor based on automatic recognition provided by the embodiments of the present invention can improve the adaptability of the mouse cursor switching. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a schematic flowchart of an intelligent switching method of a mouse cursor based on automatic recognition provided by an embodiment of the present invention;

[0069] Figure 2 It is a schematic diagram of modules for implementing the intelligent switching system of the mouse cursor based on automatic recognition provided by an embodiment of the present invention.

[0070] The implementation, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0071] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0072] An embodiment of the present application provides a method for intelligent switching of a mouse cursor based on automatic recognition. The execution subject of the method for intelligent switching of a mouse cursor based on automatic recognition includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for intelligent switching of a mouse cursor based on automatic recognition can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0073] Embodiment 1:

[0074] Referring to Figure 1 As shown, it is a schematic flowchart of a method for intelligent switching of a mouse cursor based on automatic recognition provided by an embodiment of the present invention. In this embodiment, the method for intelligent switching of a mouse cursor based on automatic recognition includes:

[0075] S1. Obtain the current interface where the mouse cursor is located, identify the interface functions in the current interface, divide the current interface into multiple target sections according to the interface functions, and initialize the shape of the mouse cursor to obtain an initial shape.

[0076] In the embodiment of the present invention, the current interface refers to the interface that the cursor user is pausing and browsing when performing interface operations and interface browsing on the terminal. The interface function refers to the functions represented by different icons and texts in the current interface, such as navigation functions, tool functions, data input functions, information display functions, alarm functions, multimedia control functions. The navigation function is used to guide the user to perform operations such as interface backtracking and interface forward movement. The tool function is used to provide operation tools for the user, such as an eraser tool provided when drawing. The data input function is used to provide operations for the user to input data. The information display function is used to display information. The alarm function is used to perform system warnings. The multimedia control function is used to control audio and video, such as controlling the volume size.

[0077] Among them, the target section refers to other sections outside the section where the mouse is located at the first moment when the user enters the current interface. For example, when the user enters the current interface from other interfaces, if the mouse is in the first section of the current interface at the first moment, then the remaining 5 sections of the current interface are used as target sections. The target sections are, for example, a navigation area, a tool area, a data input area, an information display area, an alarm area, a multimedia control area, etc.

[0078] Furthermore, in the embodiment of the present invention, by dividing the current interface into multiple target sections according to the interface functions, the same-shaped cursor is set for the same target section in the subsequent process.

[0079] In an embodiment of the present invention, according to the interface function, the current interface is divided into multiple target sections, including: dividing the areas in the current interface with the same interface function and adjacent pixel positions into the same target section; determining multiple target sections in the current interface from the same target section.

[0080] Optionally, the process of initializing the shape of the mouse cursor to obtain the initialized shape refers to using the cursor shape within the section where the mouse is located at the first moment in the past period as the initialized shape. For example, when the user enters the current interface, if the section where the mouse is located at the first moment is the navigation area, then the default cursor shape (such as an arrow shape) in the navigation area in the past period is used as the initialized shape.

[0081] S2. Query the section function of the target section, match the similar function corresponding to the section function from the preset historical section library, query the mouse usage information corresponding to the similar function from the historical section library, and analyze the usage requirements of the cursor user of the mouse cursor regarding the mouse cursor using the mouse usage information.

[0082] In an embodiment of the present invention, the meaning of the section function is the same as that of the foregoing interface function. The historical section library refers to a database that collects the mouse usage information of cursor users regarding different interface functions and their corresponding information in the historical period. For example, the mouse usage information of users regarding the navigation function last month is collected. The mouse usage information includes information such as the coordinates where the mouse is located at different times, the number of times the interface is clicked, and the browsing duration within the interface. The similar function refers to a function that is the same as the section function. For example, if the section function is the navigation function, the similar function is also the navigation function.

[0083] Furthermore, in an embodiment of the present invention, by analyzing the usage requirements of the cursor user of the mouse cursor using the mouse usage information, when encountering a new interface not stored in the historical database, the historical usage information of the user is used as the basic cursor requirements for the new interface.

[0084] In an embodiment of the present invention, analyzing the usage requirements of the cursor user of the mouse cursor using the mouse usage information includes: calculating the average speed, mis-touch rate, stay duration, and path tortuosity of the cursor user regarding the mouse cursor using the mouse usage information; respectively determining the operation fluency requirement, positioning accuracy requirement, information focus requirement, and navigation cognition requirement of the cursor user regarding the mouse cursor through the average speed, the mis-touch rate, the stay duration, and the path tortuosity; and taking the operation fluency requirement, the positioning accuracy requirement, the information focus requirement, and the navigation cognition requirement as the usage requirements.

[0085] Among them, the average speed refers to the distance that the mouse moves per unit time, the mis-touch rate refers to the ratio of the number of mis-touch times to the total number of clicks, the path tortuosity refers to the ratio of the path length between the starting point and the ending point of the mouse movement within the plate to the straight-line length between these two points, the operation fluency requirement refers to the requirements for the average speed and the fluency of the cursor movement, the positioning accuracy requirement refers to the requirements for the mis-touch rate and the accuracy of clicking on a certain interface position, the information focusing requirement refers to the requirements for the residence time and the mouse focusing on a certain place, the navigation cognition requirement refers to the requirements for the path tortuosity and quickly finding the target point. It should be noted that the faster the average speed, the higher the user's operation fluency requirement; the larger the mis-touch rate, the higher the user's positioning accuracy requirement; the longer the residence time, the higher the user's information focusing requirement; the larger the path tortuosity, the higher the user's navigation cognition requirement. Since the values (average speed, mis-touch rate, residence time, path tortuosity) are in a direct proportional relationship with the operation fluency requirement, the positioning accuracy requirement, the information focusing requirement, and the navigation cognition requirement, the values are used as the representation values of the operation fluency requirement, the positioning accuracy requirement, the information focusing requirement, and the navigation cognition requirement.

[0086] S3. Query the current plate of the cursor user, obtain the current usage information in the current plate, and dynamically update the usage requirement according to the current usage information to obtain an updated requirement, and map the updated requirement to the shape requirement of the cursor user for the mouse cursor.

[0087] In the embodiment of the present invention, the current plate refers to the plate where the mouse is located at the first moment when the user enters the current interface, and the current usage information refers to the mouse usage information of the user within the current plate. The meaning of the mouse usage information corresponding to the similar functions described above is similar, except that the current usage information refers to the usage information of the user about the current new interface, and this new interface may be an interface that has never appeared in the historical period.

[0088] Furthermore, in the embodiment of the present invention, the usage requirement is dynamically updated according to the current usage information to correct the basic cursor requirement of the new interface based on the mouse usage information of the user within the initial period of time after entering the new interface.

[0089] In an embodiment of the present invention, dynamically updating the usage requirement according to the current usage information to obtain an updated requirement includes: obtaining the current speed, current mis-touch rate, current stay duration, and current path tortuosity in the current usage information; respectively performing standardization processing on the current speed, the current mis-touch rate, the current stay duration, and the current path tortuosity to obtain a standard speed, a standard mis-touch rate, a standard stay duration, and a standard path tortuosity; querying the section interval between the current section and the target section; determining the standard weights of the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity by using the section interval; performing linear mapping on the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity respectively according to the standard weights to obtain a speed error, a mis-touch rate error, a stay duration error, and a path tortuosity error; and dynamically updating the usage requirement by using the speed error, the mis-touch rate error, the stay duration error, and the path tortuosity error to obtain an updated requirement.

[0090] Wherein, the section interval refers to the number of other sections passed by the straight line between the current section and the target section.

[0091] Optionally, the process of respectively standardizing the current speed, the current mis-touch rate, the current stay duration, and the current path tortuosity to obtain the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity refers to the process of standardizing and normalizing the data. For example, for the current speed, the speed data of the user within a certain period of time is collected, and the speed is subtracted by the speed mean and then divided by the speed variance to obtain the standard speed. Further, the process of determining the standard weights of the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity using the plate interval refers to taking the ratio between 1 and the plate interval plus 1 as the standard weight. The formula for linearly mapping the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity according to the standard weights to obtain the speed error, the mis-touch rate error, the stay duration error, and the path tortuosity error is Δ = β(γh) + b, where β represents the characteristic weight of h, γ represents the standard weight, b represents the baseline adjustment term, Δ represents any error, and h represents any value among the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity. The process of dynamically updating the usage requirements using the speed error, the mis-touch rate error, the stay duration error, and the path tortuosity error to obtain the updated requirements refers to adding the speed error, the mis-touch rate error, the stay duration error, and the path tortuosity error to the existing average speed, mis-touch rate, stay duration, and path tortuosity within the target plate respectively to obtain the updated values. These updated values are standardized values, and it is also necessary to reverse-standardize these standardized values and use them together with the non-numerical data in the usage requirements as the updated requirements. The non-numerical data includes, for example, the requirement for the smoothness of the cursor movement in the operation smoothness requirement.

[0092] Further, the embodiment of the present invention maps the updated requirements to the shape requirements of the cursor user for the mouse cursor, so as to generate the text description and numerical description of the cursor shape from the perspective of the user's requirements, laying a foundation for adaptively generating the cursor shape that meets the user's requirements based on the text description and numerical description in the subsequent stage.

[0093] In an embodiment of the present invention, mapping the update requirement to the shape requirement of the cursor user for the mouse cursor includes: obtaining the update text and update value in the update requirement; obtaining the update operation fluency requirement, update positioning accuracy requirement, update information focusing requirement, and update navigation cognition requirement in the update text; respectively converting the update operation fluency requirement, the update positioning accuracy requirement, the update information focusing requirement, and the update navigation cognition requirement into a streamline design requirement, an enhanced directivity requirement, a multi-level visualization requirement, and a direction indication requirement; and determining the shape requirement of the mouse cursor according to the update value by using the streamline design requirement, the enhanced directivity requirement, the multi-level visualization requirement, and the direction indication requirement.

[0094] It should be noted that the essence of the update operation fluency requirement, the update positioning accuracy requirement, the update information focusing requirement, and the update navigation cognition requirement is the same as the essence of the non-numerical data in the foregoing usage requirements. The streamline design requirement refers to the requirement of whether the cursor is designed to be streamlined or non-streamlined with a thick and heavy feeling. The enhanced directivity requirement refers to the requirement of whether the cursor shape directly points to a certain interface position. For example, the center of the crosshair-shaped cursor reaches the corner of the square in the drawing interface. The multi-level visualization requirement refers to the requirement of expressing the cursor shape as a multi-level shape. For example, a small square is nested inside a large square, and the cursor composed of two squares of different sizes is a multi-level visualization cursor. The direction indication requirement refers to whether it is necessary to add or enhance the part with direction indication in the cursor shape. For example, the tip of the arrow can point in different directions and angles. Extending the tip or adding a tip to a shape without a tip originally is the direction indication requirement. Further, the shape requirement includes text requirements (the streamline design requirement, the enhanced directivity requirement, the multi-level visualization requirement, the direction indication requirement) and the update value. Finally, it should be noted that the streamline design requirement, the enhanced directivity requirement, the multi-level visualization requirement, and the direction indication requirement are text descriptions determined based on the update text. For example, for the streamline design requirement, if the update operation fluency requirement is divided into three levels: low, medium, and high, and if the user's update operation fluency requirement is at a high level, then the streamline design requirement is to be inclined to streamline and have a high inclination.

[0095] S4. Based on the shape requirement, generate some data points of the mouse cursor, adjust the data points to obtain adjusted data points, connect the adjusted data points to form the cursor shape of the mouse cursor, and detect whether the current position of the mouse cursor is in the target plate.

[0096] In an embodiment of the present invention, based on the shape requirement, partial data points of the mouse cursor are generated to generate a dynamic and non-fixed shape cursor through a generator.

[0097] Among them, the partial data points refer to a fixed number of coordinate points generated by a generator network, such as 128 coordinate points.

[0098] In an embodiment of the present invention, the generating partial data points of the mouse cursor based on the shape requirement includes: obtaining the text requirement and numerical requirement corresponding to the shape requirement; performing numerical normalization on the numerical requirement by using the following formula to obtain a normalized value:

[0099]

[0100] Among them, represents the normalized value, x i represents the i-th type of numerical requirement, μ i represents the mean of the i-th type of numerical requirement within a time window, and σ i represents the variance of the i-th type of numerical requirement within a time window;

[0101] Encoding the normalized value into a numerical vector; inputting the numerical vector into a preset two-layer fully connected network to output a high-order feature vector corresponding to the numerical vector through the two-layer fully connected network; inputting the text requirement into a preset BERT model to extract a semantic embedding vector of the text requirement through the BERT model; performing vector fusion on the high-order feature vector and the semantic embedding vector to obtain a fusion feature vector; inputting the fusion feature vector into a preset generator network to generate partial data points corresponding to the fusion feature vector through the generator network.

[0102] Among them, the numerical vector is represents the normalized values corresponding to n types of numerical requirements. The two-layer fully connected network refers to a two-layer fully connected network, and each layer of the fully connected network consists of multiple fully connected layers (FC layers) followed by a non-linear activation function. The BERT model refers to a trained natural language processing model that can take the original word vectors of each word in the text as input and output the vector representation of each word in the text after integrating the full-text semantic information, and can output a vector for downstream tasks separately at each layer. The BERT model mainly includes three basic components: an encoder (BertEmbeddings), a decoder (BertEncoder), and a pooling layer (BertPooler). The generator network refers to a network that generates a large number of pixel points or coordinate points, such as a generative adversarial network (GANs). It should be noted that when calculating the normalized value, the time window used needs to include x at the same time.i Data in the historical period and data in the current period. The data in the historical period is obtained from the historical block library, and the data in the current period is obtained from the error impact of the current block on the target block.

[0103] In another embodiment of the present invention, the vector fusion of the high-order feature vector and the semantic embedding vector to obtain a fused feature vector includes: mapping the high-order feature vector into the vector space corresponding to the semantic embedding vector to obtain a mapped feature vector; concatenating the mapped feature vector and the semantic embedding vector to obtain a concatenated vector; calculating the vector weight corresponding to the concatenated vector through a preset gating network; and based on the vector weight, performing vector fusion on the mapped feature vector and the semantic embedding vector using the following formula to obtain a fused feature vector:

[0104]

[0105] where S represents the fused feature vector, w represents the vector weight, represents the mapped feature vector, Y represents the semantic embedding vector, and ⊙ represents the element-wise multiplication symbol.

[0106] Among them, the partial data points can be understood as the skeletal key points of the human body, and fitting the partial data points into a curve can connect them into the shape of a cursor.

[0107] Optionally, the process of mapping the high-order feature vector into the vector space corresponding to the semantic embedding vector to obtain a mapped feature vector is determined by the formula wX + b, where X represents the high-order feature vector, w represents the projection matrix, and b represents the bias term. Further, in the process of calculating the vector weight corresponding to the concatenated vector through a preset gating network, the algorithm corresponding to the gating network is Sigmoid(w’M + b’), where w’ and b’ represent the weights and biases of the gating network, and M represents the concatenated vector.

[0108] In one embodiment of the present invention, the adjustment of the partial data points to obtain adjusted data points includes: setting the number threshold of the partial data points using the following formula:

[0109]

[0110] where N represents the number threshold, PPI represents the pixel density unit, width k represents the width of the kth target block, Height k represents the height of the kth target block, and 200 represents the reference PPI;

[0111] Determine whether the number of the partial data points is less than the number threshold; when the number of the partial data points is less than the number threshold, perform data point completion on the partial data points to obtain completed data points; use the partial data points and the completed data points as adjusted data points.

[0112] Wherein, the number threshold refers to a threshold of the number of personalized generated data points set for each target plate according to the size of the target plate. It should be noted that when the number of the partial data points is not less than the number threshold, there is no need to perform data point completion on the partial data points, and the partial data points can be directly used as adjusted data points.

[0113] Furthermore, in the embodiment of the present invention, by performing data point completion on the partial data points to fill the data points based on the fitting curve, the invariance of the curve shape of the original partial data points is ensured, and the visual appreciation of the cursor shape can be improved by increasing the number of data points.

[0114] In another embodiment of the present invention, the performing data point completion on the partial data points to obtain completed data points includes: performing curve fitting on the partial data points by using the following formula to obtain a fitting curve:

[0115]

[0116] Wherein, Z(α j ) represents the fitting curve, m represents the curve order, N j,m (α j ) represents the B-spline basis function, j represents the index of the partial data points, N gen represents the number of the partial data points, P j represents the j-th partial data point, α j represents a constant between 0 and 1;

[0117] Query the completion number of the partial data points; according to the completion number, calculate the completed data points of the partial data points by using the following formula:

[0118]

[0119] Wherein, represents the completed data points, represents the coordinate points determined by the completion number N com , and u represents the serial number of the completed coordinate points.

[0120] It should be noted that Z(α j ) represents α jThe corresponding coordinate points, and the set composed of the coordinate points is the curve function, and this curve function is the fitting curve. The supplemented data points are N com additional coordinate points, and each coordinate point is composed of a horizontal and a vertical coordinate.

[0121] Optionally, the process of connecting the supplemented data points into the cursor shape of the mouse cursor refers to the process of connecting the supplemented data points in sequence from the first data point to the last data point. Further, the process of detecting whether the current position of the mouse cursor is in the target plate refers to detecting whether the user moves the mouse from the current plate to the target plate. It should be noted that when the user moves the mouse to the A target plate, if the remaining other plates are the B, C, and D plates respectively, then the A target plate should be used as the current plate for the next round, and the B, C, and D plates should be used as the target plates for the next round. Then, analyze what shape of cursor should be used for the target plate in the next round. The analysis principle is similar to the previous one and will not be elaborated further here.

[0122] S5. When the current position of the mouse cursor is in the target plate, switch the initialization shape to the cursor shape to complete the intelligent switching process of the mouse cursor.

[0123] Compared with the problems described in the background art, in the embodiment of the present invention, according to the interface function, the current interface is divided into multiple target sections, so as to set the same-shaped cursor for the same target section subsequently. Further, in the embodiment of the present invention, by using the mouse usage information to analyze the usage requirements of the cursor user of the mouse cursor, when encountering a new interface not stored in the historical database, the historical usage information of the user is used as the basic cursor requirement for the new interface. Further, in the embodiment of the present invention, according to the current usage information, the usage requirements are dynamically updated, so as to correct the basic cursor requirement for the new interface based on the mouse usage information within an initial period of time after the user enters the new interface. Further, in the embodiment of the present invention, the updated requirements are mapped to the shape requirements of the cursor user for the mouse cursor, so as to generate a text description and a numerical description of the cursor shape for the user from the perspective of the user's requirements, laying a foundation for adaptively generating a cursor shape that meets the user's requirements based on the text description and the numerical description subsequently. Further, in the embodiment of the present invention, based on the shape requirements, some data points of the mouse cursor are generated, so as to generate a dynamic and non-fixed shape cursor through a generator. Further, in the embodiment of the present invention, the partial data points are complemented, so as to fill the data points based on a fitting curve, ensuring the invariance of the curve shape of the original partial data points and improving the visual appreciation of the cursor shape by increasing the data points. Therefore, the intelligent mouse cursor switching method and system based on automatic recognition provided by the embodiment of the present invention can improve the adaptability of mouse cursor switching.

[0124] Embodiment 2:

[0125] As Figure 2 shown, it is a functional module diagram of an intelligent mouse cursor switching system based on automatic recognition according to the present invention.

[0126] The intelligent mouse cursor switching system 200 based on automatic recognition according to the present invention can be installed in an electronic device. According to the functions implemented, the intelligent mouse cursor switching system based on automatic recognition can include a shape initial module 201, a requirement analysis module 202, a requirement mapping module 203, a mouse detection module 204, and a cursor switching module 205. The modules in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0127] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0128] The shape initialization module 201 is used to obtain the current interface where the mouse cursor is located, identify the interface functions in the current interface, divide the current interface into multiple target sections according to the interface functions, and perform shape initialization on the mouse cursor to obtain an initialized shape;

[0129] The requirement analysis module 202 is used to query the section functions of the target sections, match the similar functions corresponding to the section functions from a preset historical section library, query the mouse usage information corresponding to the similar functions from the historical section library, and analyze the usage requirements of the cursor user of the mouse cursor regarding the mouse cursor by using the mouse usage information;

[0130] The requirement mapping module 203 is used to query the current section of the cursor user, obtain the current usage information in the current section, dynamically update the usage requirements according to the current usage information to obtain updated requirements, and map the updated requirements to the shape requirements of the cursor user for the mouse cursor;

[0131] The mouse detection module 204 is used to generate some data points of the mouse cursor based on the shape requirements, perform data point adjustment on the some data points to obtain adjusted data points, connect the adjusted data points into the cursor shape of the mouse cursor, and detect whether the current position of the mouse cursor is in the target section;

[0132] The cursor switching module 205 is used to switch the initialized shape to the cursor shape when the current position of the mouse cursor is in the target section, so as to complete the intelligent switching process of the mouse cursor.

[0133] Specifically, each module in the intelligent mouse cursor switching system 200 based on automatic recognition in the embodiments of the present invention adopts the same technical means as those in the Figure 1 intelligent mouse cursor switching method based on automatic recognition described above, and can produce the same technical effects, which will not be elaborated here.

[0134] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent switching method for a mouse cursor based on automatic recognition, characterized in that, The method includes: Obtain the current interface where the mouse cursor is located, identify the interface functions in the current interface, divide the current interface into multiple target sections according to the interface functions, and initialize the shape of the mouse cursor to obtain an initial shape; Query the section functions of the target sections, match the similar functions corresponding to the section functions from a preset historical section library, query the mouse usage information corresponding to the similar functions from the historical section library, and analyze the usage requirements of the cursor user of the mouse cursor regarding the mouse cursor using the mouse usage information; Query the current section of the cursor user, obtain the current usage information in the current section, dynamically update the usage requirements according to the current usage information to obtain updated requirements, and map the updated requirements to the shape requirements of the cursor user for the mouse cursor; Generate partial data points of the mouse cursor based on the shape requirements, adjust the data points to obtain adjusted data points, connect the adjusted data points to form the cursor shape of the mouse cursor, and detect whether the current position of the mouse cursor is in the target section; When the current position of the mouse cursor is in the target section, switch the initial shape to the cursor shape to complete the intelligent switching process of the mouse cursor.

2. The intelligent mouse cursor switching method based on automatic recognition according to claim 1, wherein The step of dividing the current interface into multiple target sections according to the interface functions includes: Divide the areas in the current interface with consistent interface functions and adjacent pixel positions into the same target section; Determine multiple target sections in the current interface from the same target section.

3. The intelligent mouse cursor switching method based on automatic recognition according to claim 1, characterized in that The step of analyzing the usage requirements of the cursor user of the mouse cursor regarding the mouse cursor using the mouse usage information includes: Calculate the average speed, mis-touch rate, stay duration, and path tortuosity of the cursor user regarding the mouse cursor using the mouse usage information; Determine the operation fluency requirement, positioning accuracy requirement, information focusing requirement, and navigation cognition requirement of the cursor user regarding the mouse cursor through the average speed, the mis-touch rate, the stay duration, and the path tortuosity respectively; Regard the operation fluency requirement, the positioning accuracy requirement, the information focusing requirement, and the navigation cognition requirement as usage requirements.

4. The intelligent mouse cursor switching method based on automatic recognition according to claim 1, wherein, The step of dynamically updating the usage requirements according to the current usage information to obtain updated requirements includes: Obtain the current speed, current mis-touch rate, current stay duration, and current path tortuosity in the current usage information; Perform standardization processing on the current speed, the current mis-touch rate, the current stay duration, and the current path tortuosity respectively to obtain a standard speed, a standard mis-touch rate, a standard stay duration, and a standard path tortuosity; Query the section interval between the current section and the target section; Determine the standard weights of the standard speed, the standard mis-touch rate, the standard stay duration, and the standard path tortuosity using the section interval; According to the standard weights, linearly map the standard speed, the standard false touch rate, the standard stay duration, and the standard path tortuosity respectively to obtain a speed error, a false touch rate error, a stay duration error, and a path tortuosity error; Dynamically update the usage requirements by using the speed error, the false touch rate error, the stay duration error, and the path tortuosity error to obtain updated requirements.

5. The intelligent mouse cursor switching method based on automatic recognition according to claim 1, wherein The mapping of the updated requirements to the shape requirements of the cursor user for the mouse cursor includes: Obtain the updated text and updated values in the updated requirements; Obtain the updated operation fluency requirement, updated positioning accuracy requirement, updated information focus requirement, and updated navigation cognition requirement in the updated text; Convert the updated operation fluency requirement, the updated positioning accuracy requirement, the updated information focus requirement, and the updated navigation cognition requirement into a streamline design requirement, an enhanced directivity requirement, a multi-level visualization requirement, and a direction indication requirement respectively; Determine the shape requirements of the mouse cursor according to the updated values by using the streamline design requirement, the enhanced directivity requirement, the multi-level visualization requirement, and the direction indication requirement.

6. The intelligent mouse cursor switching method based on automatic recognition according to claim 1, wherein Generating some data points of the mouse cursor based on the shape requirements includes: Obtain the text requirements and numerical requirements corresponding to the shape requirements; Perform numerical normalization on the numerical requirements by using the following formula to obtain a normalized value: Among them, represents the normalized value, and x i represents the numerical requirement of the i-th category, and μ i represents the mean value of the numerical requirement of the i-th category within the time window, and σ i represents the variance of the numerical requirement of the i-th category within the time window; Encode the normalized value as a numerical vector; Input the numerical vector into a preset two-layer fully connected network to output a high-order feature vector corresponding to the numerical vector through the two-layer fully connected network; Input the text requirements into a preset BERT model to extract a semantic embedding vector of the text requirements through the BERT model; Perform vector fusion on the high-order feature vector and the semantic embedding vector to obtain a fusion feature vector; Input the fusion feature vector into a preset generator network to generate some data points corresponding to the fusion feature vector through the generator network.

7. The intelligent mouse cursor switching method based on automatic recognition according to claim 6, characterized in that The performing vector fusion on the high-order feature vector and the semantic embedding vector to obtain a fusion feature vector includes: Map the high-order feature vector to the vector space corresponding to the semantic embedding vector to obtain a mapped feature vector; Concatenate the mapped feature vector and the semantic embedding vector to obtain a concatenated vector; Calculate the vector weight corresponding to the concatenated vector through a preset gating network; Based on the vector weight, perform vector fusion on the mapped feature vector and the semantic embedding vector by using the following formula to obtain a fusion feature vector: Among them, S represents the fused feature vector, w represents the vector weight, represents the mapped feature vector, Y represents the semantic embedding vector, and ⊙ represents the symbol of element-wise multiplication.

8. The intelligent mouse cursor switching method based on automatic recognition according to claim 1, wherein The performing data point adjustment on the some data points to obtain adjusted data points includes: Set the number threshold of the some data points by using the following formula: where N represents a number threshold, PPI represents a pixel density unit, width k represents the width of the k-th target block, Height k represents the height of the k-th target block, and 200 represents the reference PPI; Judge whether the number of the some data points is less than the number threshold; When the number of the some data points is less than the number threshold, perform data point completion on the some data points to obtain completed data points; Use the some data points and the completed data points as adjusted data points.

9. The intelligent mouse cursor switching method based on automatic recognition according to claim 8, wherein Performing data point completion on the partial data points to obtain completed data points includes: Performing curve fitting on the partial data points using the following formula to obtain a fitting curve: Among them, Z(α j ) represents the fitting curve, m represents the curve order, N j,m (α j ) represents the B-spline basis function, j represents the index of partial data points, N gen represents the number of partial data points, P j represents the j-th partial data point, α j represents a constant between 0 and 1; Querying the completion number of the partial data points; Calculating the completed data points of the partial data points using the following formula according to the completion number; Among them, represents the completed data points, represents the coordinate points determined by the completion number N com where u represents the serial number of the completed coordinate points.

10. An intelligent mouse cursor switching system based on automatic recognition, characterized in that, The system includes: A shape initialization module, configured to obtain the current interface where the mouse cursor is located, identify the interface functions in the current interface, divide the current interface into multiple target sections according to the interface functions, and perform shape initialization on the mouse cursor to obtain an initialized shape; A requirement analysis module, configured to query the section functions of the target sections, match the similar functions corresponding to the section functions from a preset historical section library, query the mouse usage information corresponding to the similar functions from the historical section library, and analyze the usage requirements of the cursor user of the mouse cursor regarding the mouse cursor using the mouse usage information; A requirement mapping module, configured to query the current section of the cursor user, obtain the current usage information in the current section, dynamically update the usage requirements according to the current usage information to obtain updated requirements, and map the updated requirements to the shape requirements of the cursor user for the mouse cursor; A mouse detection module, configured to generate partial data points of the mouse cursor based on the shape requirements, perform data point adjustment on the partial data points to obtain adjusted data points, connect the adjusted data points into the cursor shape of the mouse cursor, and detect whether the current position of the mouse cursor is in the target section; A cursor switching module, configured to switch the initialized shape to the cursor shape when the current position of the mouse cursor is in the target section, so as to complete the intelligent switching process of the mouse cursor.

Images