Concentration improving method and system based on keyboard lighting effect guidance

Abstract

The invention relates to the technical field of user experience optimization, and discloses a concentration improvement method and system based on keyboard lighting effect guidance, and the method comprises the steps: carrying out the information preprocessing of manual use information, obtaining the preprocessing information, and recognizing a manual behavior mode of a user through the preprocessing information; detecting whether a computer scene connected with the keyboard is a new scene or not; when the computer scene is a new scene, inquiring the current use habit of the user about the lighting effect of the keyboard, establishing a basic keyboard lighting effect of the computer scene by utilizing a manual behavior mode, and guiding the lighting effect of the keyboard on the basis of the basic keyboard lighting effect by utilizing the current use habit; when the computer scene is not the new scene, keyboard lighting effect guiding is carried out on the user in a manual behavior mode, a second lighting effect guiding result is obtained, concentration improvement is carried out on the user through the second lighting effect guiding result, and a second concentration improvement result is obtained. The method and the device can meet individual requirements of users.

Description

Technical Field

[0001] The present invention relates to a method and system for improving concentration based on keyboard light effect guidance, belonging to the technical field of user experience optimization. Background Art

[0002] The process of the method for improving concentration based on keyboard light effect guidance refers to a technical method of optimizing the user's concentration by dynamically adjusting the brightness and hue of the keyboard, combined with the user's behavior data and historical habits.

[0003] Currently, many existing patents, such as the backlit keyboard patent of Lite-On Technology, although improve the uniformity and brightness of the backlight, usually adopt a fixed light effect mode and cannot be dynamically adjusted according to the user's personal preferences. Secondly, the backlight device of Depu Electronics mainly focuses on the adjustment of light uniformity and brightness, but fails to dynamically optimize the light effect according to the user's usage habits or scenarios. Thirdly, although the backlit keyboard of Lite-On Technology introduces an intelligent control system, it mainly relies on the ambient light sensor and fails to make full use of the user's historical data for personalized adjustment. Finally, although Apple's True Tone technology improves visual comfort, it lacks in-depth analysis of the user's historical behavior and is difficult to meet the diverse needs of users. Therefore, the current technology is difficult to meet the personalized needs of users. Summary of the Invention

[0004] The present invention provides a method and system for improving concentration based on keyboard light effect guidance, and its main purpose is to meet the personalized needs of users.

[0005] To achieve the above object, a method for improving concentration based on keyboard light effect guidance provided by the present invention includes:

[0006] During a historical period, store the user's manual usage information about the light effect of the keyboard in the keyboard, perform information preprocessing on the manual usage information to obtain preprocessed information, and use the preprocessed information to identify the user's manual behavior pattern, where the manual usage information includes spatial information, time information, and behavior information;

[0007] During the current period, detect whether the computer scenario connected to the keyboard is a new scenario;

[0008] When the computer scenario is a new scenario, query the user's current usage habit about the light effect of the keyboard, establish the basic keyboard light effect of the computer scenario by using the manual behavior pattern, perform keyboard light effect guidance on the basic keyboard light effect by using the current usage habit to obtain a first light effect guidance result, so as to improve the concentration of the user through the first light effect guidance result to obtain a first concentration improvement result;

[0009] When the computer scenario is not a new scenario, use the manual behavior mode to guide the user for the keyboard light effect, obtain a second light effect guidance result, and improve the user's concentration through the second light effect guidance result to obtain a second concentration improvement result;

[0010] Use the first concentration improvement result and the second concentration improvement result as the concentration improvement result of the user regarding the keyboard.

[0011] Optionally, the preprocessing the manual usage information to obtain preprocessing information includes:

[0012] Obtain the key position, brightness data, hue data, operation timestamp, and operation period in the manual usage information, and obtain the keyboard corresponding to the manual usage information;

[0013] Map the overall keys of the keyboard to a preset grid coordinate system to obtain keyboard grid coordinates;

[0014] Query the key coordinates of the key position from the keyboard grid coordinates;

[0015] Determine the grayscale image of the key coordinates based on the ratio of the number of adjustments to the total number of adjustments on the key coordinates;

[0016] Perform color initialization on the keyboard grid coordinates to obtain initialization colors;

[0017] Based on the brightness data and the hue data, perform color adjustment on the initialization colors to obtain adjusted colors;

[0018] Use the adjusted colors to analyze the gradient diffusion colors of the neighborhood coordinates of the key position;

[0019] Perform a color overlay operation on the gradient diffusion colors to obtain overlay colors;

[0020] Convert the overlay colors into atomic operations;

[0021] Calculate the migration probability of adjacent operations in the atomic operations;

[0022] Convert the operation timestamp into a frequency domain feature;

[0023] Perform window sliding on the operation period to obtain time features;

[0024] Use the grayscale image, the atomic operations, the migration probability, the frequency domain feature, and the time features as preprocessing information.

[0025] Optionally, the identifying the user's manual behavior mode using the preprocessing information includes:

[0026] Prepare a feature extraction model for the preprocessed information and a Transformer encoder;

[0027] Among them, the feature extraction model includes a deep residual network, a global pooling layer, a word vector model, a concatenation layer, a long short-term memory network, a fully connected layer, an activation function, and a non-linear mapping layer;

[0028] Use the feature extraction model to extract information features from the preprocessed information to obtain extracted features;

[0029] Perform feature concatenation on the extracted features to obtain concatenated features;

[0030] Use the Transformer encoder to perform feature fusion on the concatenated features to obtain a classification probability vector;

[0031] Use the classification probability vector to determine the manual behavior pattern of the user.

[0032] Optionally, detecting whether the computer scenario connected to the keyboard is a new scenario includes:

[0033] Query the software information and peripheral information in the computer scenario;

[0034] Determine whether similar software and similar peripherals matching the software information and the peripheral information can be found in the historical period;

[0035] When similar software and similar peripherals matching the software information and the peripheral information can be found in the historical period, determine that the computer scenario is not a new scenario;

[0036] When similar software and similar peripherals matching the software information and the peripheral information cannot be found in the historical period, determine that the computer scenario is a new scenario.

[0037] Optionally, querying the user's current usage habit regarding the lighting effect of the keyboard includes:

[0038] Query the number of times the user adjusts the key positions regarding the lighting effect of the keyboard;

[0039] Generate scene coordinates of the computer scenario using the software information and peripheral information of the computer scenario;

[0040] Calculate the Euclidean distance between the scene coordinates and the preset coordinates;

[0041] According to the Euclidean distance, calculate the current probability vector of the computer scenario using the following formula;

[0042]

[0043] Among them, P j represents the current probability vector, and d j represents the Euclidean distance between the scene coordinates and the preset coordinates of the j-th category, and M represents the number of preset coordinates;

[0044] Take the number of key position adjustments and the current probability vector as the current usage habit.

[0045] Optionally, establishing the basic keyboard lighting effect of the computer scene by using the manual behavior mode includes:

[0046] Calculate the software matching degree between the software information of the computer scene and the software types in the preset software library by using the following formula:

[0047]

[0048] Among them, M app represents the software matching degree, S i represents the keyword coincidence degree between the current software and the software type, and W i represents the category weight of the software type;

[0049] Select the initial lighting effect in the software library corresponding to the highest matching degree of the software matching degree;

[0050] Calculate the peripheral influence coefficient of the computer scene by using the following formula:

[0051]

[0052] Among them, C device,k represents the peripheral influence coefficient corresponding to the k-th type of manual behavior mode, D num represents the number of connected peripherals normalized to the 0-1 interval, D intensity represents the peripheral usage intensity;

[0053] According to the initial lighting effect, the peripheral influence coefficient and the manual behavior mode, calculate the basic keyboard brightness of the computer scene by using the following formula:

[0054]

[0055] Among them, L base represents the basic keyboard brightness, H hour represents the current time period, C device,k represents the peripheral influence coefficient corresponding to the k-th type of manual behavior mode, L init represents the initial brightness in the initial lighting effect, p k represents the classification probability vector of the k-th manual behavior mode, and m represents the number of manual behavior modes;

[0056] Calculate the base keyboard hue of the computer scenario using the following formula based on the initial light effect and the peripheral influence coefficient:

[0057]

[0058] Where, H base represents the base keyboard hue, H hour represents the current time period, H k represents the main hue corresponding to the k-th type of manual behavior pattern, H init represents the initial hue in the initial light effect, p k represents the classification probability vector of the k-th manual behavior pattern, and m represents the number of manual behavior patterns;

[0059] Use the base keyboard brightness and the base keyboard hue as the base keyboard light effect.

[0060] Optionally, guiding the keyboard light effect on the base keyboard light effect using the current usage habit to obtain a first light effect guiding result, including:

[0061] Obtain the number of key position adjustments and the current probability vector in the current usage habit;

[0062] Calculate the current behavior entropy corresponding to the current usage habit using the following formula based on the number of key position adjustments:

[0063]

[0064] Where, E (t) represents the current behavior entropy, C var represents the hue fluctuation index, H i represents the hue of the i-th adjustment in the number of key position adjustments, H i-1 represents the hue of the (i - 1)-th adjustment in the number of key position adjustments, n represents the number of times of adjusting the hue in the number of key position adjustments, i represents the index of adjusting the hue in the number of key position adjustments, D key represents the key position dispersion, N1 represents the number of key position adjustments, and N2 represents the total number of key position adjustments;

[0065] Calculate the behavior entropy threshold of the current behavior entropy using the following formula based on the current probability vector:

[0066] E j = α(μ j + 1.5σ j ) + (1 - α)(μ E + 1.5σ E )

[0067]

[0068] Among them, E threshold represents the behavioral entropy threshold, and μ E represents the moving window mean of E (t) and σ E represents the standard deviation of E (t) and μ j represents the mean of the behavioral entropy in the historical period, and σ j represents the standard deviation of the behavioral entropy in the historical period. P j represents the j-th current probability vector, M represents the number of preset coordinates, α represents a weight with a value of 0.7, and E j represents the updated behavioral entropy;

[0069] Using the entropy comparison result between the current behavioral entropy and the behavioral entropy threshold, keyboard lighting effect guidance is performed on the basic keyboard lighting effect to obtain a first lighting effect guidance result.

[0070] Optionally, the user is guided to improve concentration through the first lighting effect guidance result to obtain a first concentration improvement result, including:

[0071] Changing the color in the keyboard according to the first lighting effect guidance result to guide the user to improve concentration and obtain a first concentration improvement result.

[0072] Optionally, using the manual behavior mode to perform keyboard lighting effect guidance on the user to obtain a second lighting effect guidance result, including:

[0073] Based on the manual behavior mode, the following formula is used to perform keyboard brightness guidance on the user to obtain a brightness guidance result:

[0074]

[0075] Among them, L new represents the brightness guidance result, p k represents the classification probability vector of the k-th manual behavior mode, m represents the number of manual behavior modes, λ represents the amplitude coefficient of brightness rhythm adjustment, T represents the period length of brightness rhythm, and L k represents the main brightness corresponding to the k-th type of manual behavior mode, and t represents the time variable;

[0076] Based on the manual behavior mode, the following formula is used to perform keyboard hue guidance on the user to obtain a hue guidance result:

[0077]

[0078] Among them, ΔH represents the hue guidance result, p k represents the classification probability vector of the k-th manual behavior mode, and ΔH kThe hue change step representing the classification probability vector of the k-th manual behavior pattern, β represents the hue correction weight manually adjusted by the user, ΔH user represents the direction and amplitude of the user's most recent manual hue adjustment, and Sign() represents the sign function;

[0079] Based on the luminance guidance result and the hue guidance result, the user is guided for the keyboard light effect to obtain a second light effect guidance result.

[0080] To solve the above problems, the present invention further provides a concentration improvement system based on keyboard light effect guidance, and the system includes:

[0081] A pattern recognition module, which is used to store the user's manual usage information about the keyboard light effect in the keyboard during a historical period, perform information preprocessing on the manual usage information to obtain preprocessed information, and use the preprocessed information to identify the user's manual behavior pattern. Among them, the manual usage information includes spatial information, time information, and behavior information;

[0082] A scene detection module, which is used to detect whether the computer scene connected to the keyboard is a new scene during the current period;

[0083] A first guidance module, which is used to query the user's current usage habit about the keyboard light effect when the computer scene is a new scene, establish a basic keyboard light effect for the computer scene by using the manual behavior pattern, and perform keyboard light effect guidance on the basic keyboard light effect by using the current usage habit to obtain a first light effect guidance result, so as to improve the user's concentration through the first light effect guidance result to obtain a first concentration improvement result;

[0084] A second guidance module, which is used to perform keyboard light effect guidance on the user by using the manual behavior pattern when the computer scene is not a new scene to obtain a second light effect guidance result, so as to improve the user's concentration through the second light effect guidance result to obtain a second concentration improvement result;

[0085] A result determination module, which is used to use the first concentration improvement result and the second concentration improvement result as the concentration improvement result of the user about the keyboard.

[0086] Compared with the problems described in the background art, in the embodiments of the present invention, the manual usage information is preprocessed to process the data form into data that can be autonomously analyzed by the subsequent neural network model. Further, in the embodiments of the present invention, the preprocessed information is used to identify the manual behavior pattern of the user, so as to adjust the light effect in real time according to the user's behavior pattern in the follow-up, helping the user to stay focused or stimulate creativity. In the embodiments of the present invention, it is detected whether the computer scenario connected to the keyboard is a new scenario to distinguish between new and old scenarios, and different historical data influences are set in the new and old scenarios. Further, in the embodiments of the present invention, the current usage habit of the user regarding the light effect of the keyboard is queried to update the old usage habit through the current usage habit, and improve the adaptability of the old usage habit to the user in the new scenario. Further, in the embodiments of the present invention, the basic keyboard light effect of the computer scenario is established by using the manual behavior pattern, so as to dynamically optimize the brightness and hue of the keyboard by analyzing the user's historical adjustment habits, and provide a light effect that better meets the user's preferences, thereby improving the usage experience. Further, in the embodiments of the present invention, the keyboard light effect is guided on the basis of the basic keyboard light effect by using the current usage habit to meet the personalized needs of the user by combining the user's historical data and real-time behavior. Therefore, the focus improvement method and system based on keyboard light effect guidance provided by the embodiments of the present invention can meet the personalized needs of the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 FIG. is a schematic flowchart of a focus improvement method based on keyboard light effect guidance provided by an embodiment of the present invention;

[0088] Figure 2 FIG. is a schematic diagram of modules for implementing the focus improvement system based on keyboard light effect guidance provided by an embodiment of the present invention.

[0089] 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

[0090] 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.

[0091] An embodiment of the present application provides a focus improvement method based on keyboard light effect guidance. The execution subject of the focus improvement method based on keyboard light effect guidance 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 by the embodiment of the present application. In other words, the focus improvement method based on keyboard light effect guidance 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.

[0092] Example 1:

[0093] Refer to Figure 1 As shown, it is a flowchart of a focus improvement method based on keyboard light effect guidance provided by an embodiment of the present invention. In this embodiment, the focus improvement method based on keyboard light effect guidance includes:

[0094] S1. During the historical period, store the manual usage information of the user about the keyboard light effect in the keyboard, perform information preprocessing on the manual usage information to obtain preprocessing information, and use the preprocessing information to identify the manual behavior pattern of the user. Among them, the manual usage information includes spatial information, time information, and behavior information.

[0095] In the embodiment of the present invention, the keyboard refers to an input device of a computer or other electronic device, which is used to convert the information input by the user into a signal that can be recognized by the computer or other electronic device. The user refers to the user of the keyboard. The manual usage information refers to the information that the user manually adjusts the brightness, hue, and key position of the keyboard, including key position, brightness data, hue data, operation timestamp, and operation period. Among them, the key position refers to a certain key on the keyboard that the user manually adjusts. The brightness data refers to the brightness result of a certain key on the keyboard that the user manually adjusts. For example, the brightness of a certain key is adjusted to 50, and 50 is the final result after adjustment, that is, the brightness data. The hue data is the same as the brightness data. The operation timestamp refers to the end time when the color data of the keyboard is adjusted to the brightness data or the hue data. The operation period refers to the duration of each operation. For example, the 5ms duration consumed when the brightness of a certain key is adjusted to 50.

[0096] Furthermore, in the embodiment of the present invention, the manual usage information is subjected to information preprocessing to process the data form into data that can be autonomously analyzed by the subsequent neural network model.

[0097] In one embodiment of the present invention, the preprocessing of the manual usage information to obtain preprocessed information includes: obtaining the key position, brightness data, hue data, operation timestamp, and operation period in the manual usage information, and obtaining the keyboard corresponding to the manual usage information; mapping the overall keys of the keyboard to a preset grid coordinate system to obtain keyboard grid coordinates; querying the key coordinates of the key position from the keyboard grid coordinates; determining the grayscale image of the key coordinates based on the ratio of the number of adjustment times to the total number of adjustment times on the key coordinates; performing color initialization on the keyboard grid coordinates to obtain an initial color; performing color adjustment on the initial color based on the brightness data and the hue data to obtain an adjusted color; analyzing the gradient diffusion color of the neighborhood coordinates of the key position using the adjusted color; performing a color overlay operation on the gradient diffusion color to obtain an overlaid color; converting the overlaid color into an atomic operation; calculating the transition probability between adjacent operations in the atomic operation; converting the operation timestamp into a frequency domain feature; performing window sliding on the operation period to obtain a time feature; and using the grayscale image, the atomic operation, the transition probability, the frequency domain feature, and the time feature as preprocessed information.

[0098] Wherein, the keyboard grid coordinates refer to the two-dimensional coordinates of each key on the keyboard in the grid coordinate system, the key coordinates refer to the two-dimensional coordinates of the key position in the grid coordinate system, the number of adjustment times refers to the number of times the color parameters of a single key coordinate are manually adjusted by the user, the total number of adjustment times refers to the number of times the color parameters of all key coordinates are manually adjusted by the user, and the grayscale image refers to a multi-row and multi-column grayscale matrix composed of grayscale values on the key coordinates.

[0099] Optionally, the process of mapping the overall keys of the keyboard to a preset grid coordinate system is as follows: By observing the physical structure of the keyboard or referring to the layout diagram of the keyboard to clarify the physical layout of the keyboard, including the arrangement of keys and the position of each key. According to the physical layout of the keyboard, a corresponding grid coordinate system is established. This coordinate system is a two-dimensional grid, where each grid cell corresponds to the position of a key. Mapping each key on the keyboard to the corresponding position in the grid coordinate system can be achieved by writing a program or using specific software tools. Further, the process of determining the grayscale image of the key coordinates based on the ratio of the number of adjustments at the key coordinates to the total number of adjustments refers to calculating the ratio of the number of times each key position (i.e., each key is adjusted) to the number of times all keys are adjusted, obtaining a value in the range of 0 to 1. Construct a matrix with several rows and columns according to the key coordinates. The values in this matrix are the calculated values in the range of 0 to 1. The number of rows of this matrix is the maximum number of rows of the key coordinates in each row on the keyboard. When the number of keys in some rows does not reach the maximum number of rows, the grayscale value of this part is replaced with the value 0. Further, the process of initializing the color of the keyboard grid coordinates to obtain the initial color refers to the color set during the factory setting of the keyboard. The color settings for each type of keyboard are different. The color set during the factory setting refers to the default color in the keyboard when the user first purchases the keyboard. Further, the process of adjusting the initial color based on the brightness data and the hue data to obtain the adjusted color refers to the process in which the user manually adjusts the color of a certain key. Further, the process of analyzing the gradient diffusion color of the neighborhood coordinates of the key position using the adjusted color refers to the situation that after the user manually adjusts the color of a certain key, since the color of the key will be inconsistent with the colors of the surrounding keys after being adjusted, at this time, the colors of the surrounding keys also need to be changed. First, it is necessary to calculate the Gaussian attenuation of the surrounding keys of this key. The specific formula is Among them, I(r) represents the color change amount of the peripheral keys of the key, I0 represents the color change amount manually set by the user. For example, the aforementioned brightness data and color data are the final results manually set by the user, and I0 is the difference between any one of the brightness data or color data and the previous data (the data before the user's adjustment). For example, when the keyboard is initially at a brightness of 50 and the brightness data manually set by the user is 70, then I0 is 70 minus 50 to get 20. r represents the Euclidean distance from the peripheral key positions to the key, and σ represents the diffusion coefficient. The diffusion coefficient is used to control the attenuation rate of the influence range. The larger the value, the wider the influence range. For example, when σ = 3, it means that the distance from the peripheral key positions to the key is 3 grids. Further, the process of performing color superposition operation on the gradient-diffused color to obtain the superimposed color is as follows: When both key A and key B are manually adjusted, both key A and key B cause Gaussian attenuation to key C. Then, the Gaussian attenuation caused by key A to key C needs to be superimposed and summed with the Gaussian attenuation caused by key B to key C to obtain the total color change amount of key C. Then, based on the initial color of key C, the total color change amount of key C is adjusted. For example, if the total brightness change amount of key C is +20, then on the basis of the initial brightness of key C, the brightness value is increased by 20. It should be noted that when other keys cause color Gaussian attenuation to key A and key B, key A and key B are also the peripheral keys of other keys. Therefore, the peripheral keys corresponding to this superimposed color are relative to other keys, not specifically referring to certain keys, but including all keys. Further, the process of converting the superimposed color into atomic operations refers to converting the text description into a numbered representation. These atomic operations include OP001 (indicating a brightness increase of 0 to 5%), OP002 (indicating a brightness decrease of 0 to 5%), OP003 (indicating a hue clockwise rotation of 0 to 5%. When the hue rotates clockwise, the hue value increases, and when it rotates counterclockwise, the hue value decreases), OP004 (indicating a hue counterclockwise rotation of 0 to 5%), etc. That is, each atomic operation includes the direction (increase or decrease) and amplitude of the operation. Further, the process of calculating the transition probability between adjacent operations in the atomic operations is as follows: After the A atomic operation is completed, the user performs the B atomic operation. At this time, the A atomic operation and the B atomic operation are adjacent. That is to say, the transition probability here is only the probability between two adjacent operations. All atomic operations are mutually constructed into operation pairs (these operation pairs, for example, the A atomic operation - the C atomic operation, the A atomic operation - the C atomic operation are not necessarily adjacent operations). Then, calculate the ratio of the number of occurrences of the A atomic operation - the B atomic operation pair to the number of occurrences of the A atomic operation in all operation pairs to obtain the transition probability. Further, the process of converting the operation timestamp into a frequency-domain feature is realized through Fourier transform. The Fourier transform refers to an algorithm for converting a signal from the time domain (or spatial domain) to the frequency domain. Further, perform window sliding on the operation period,The process of obtaining time features refers to starting from the starting position of the time series data composed of operation periods, moving the window sequentially according to the sliding step size, and extracting the features within each window. These features include, for example, the mean, variance, etc. within each window.

[0100] Furthermore, in the embodiment of the present invention, the manual behavior pattern of the user is identified by using the preprocessing information, so as to adjust the light effect in real time according to the user's behavior pattern in the subsequent process, helping the user to maintain concentration or stimulate creativity.

[0101] Among them, the manual behavior pattern refers to the keyboard color adjustment method of the user's needs analyzed according to the user's adjustment of the color of the keyboard. The manual behavior pattern includes multiple patterns such as deep concentration type, multi-task distraction type, environment adaptation type, creative divergence type, etc. The deep concentration type refers to a mode with low brightness to avoid interference from overly bright light to concentration, a single hue (such as blue or green) to create a steady and concentrated atmosphere. The multi-task distraction type refers to a mode with medium brightness to ensure that the light is sufficient but not dazzling, and a dynamically changing hue to remind the user to concentrate through visual stimulation. The environment adaptation type refers to a mode that automatically adjusts the brightness according to the day and night light, avoiding being too bright or too dark, and a hue matching the day and night light (such as warm hue for adapting to warm light environment, cold hue for adapting to cold light environment). The creative divergence type refers to a mode with high brightness (stimulating vitality) and multi-color gradient excitation.

[0102] In an embodiment of the present invention, the identifying the manual behavior pattern of the user by using the preprocessing information includes: preparing a feature extraction model and a Transformer encoder for the preprocessing information; wherein, the feature extraction model includes a deep residual network, a global pooling layer, a word vector model, a splicing layer, a long short-term memory network, a fully connected layer, an activation function, and a non-linear mapping layer; using the feature extraction model to perform information feature extraction on the preprocessing information to obtain extracted features; performing feature splicing on the extracted features to obtain spliced features; using the Transformer encoder to perform feature fusion on the spliced features to obtain a classification probability vector; and determining the manual behavior pattern of the user by using the classification probability vector.

[0103] It should be noted that there is a front-back layer relationship between the deep residual network (the deep residual network is ResNet-18 with the last fully connected layer removed) and the global pooling layer, that is, the grayscale image in the preprocessed information is input into the deep residual network (to extract high-dimensional features), and then the result output by the deep residual network is processed by the global pooling layer. The word vector model (Word2Vec) is in a parallel relationship with the deep residual network, and there is a front-back layer relationship between the word vector model and the splicing layer, that is, the atomic operation is input into the word vector model, and then the output result of the word vector model and the transfer probability are spliced by the splicing layer. The long short-term memory network receives the splicing result of the output result of the word vector model and the transfer probability and outputs a vector. The fully connected layer and the activation function are in a front-back layer relationship (a fully connected layer with 32 neurons and a ReLU activation function). The fully connected layer and the activation function are used to process the spliced frequency domain features and time features. Since the feature extraction model receives data within a certain duration, the frequency domain features and time features here are actually frequency domain feature sequences and time feature sequences. The spliced features refer to the result of splicing all the features extracted by the feature extraction model. The classification probability vector is the probability value p, and this probability value p is the maximum p value. For example, multiple classification probability vectors are output, the probability value p regarding deep focus type, the probability value p regarding multi-task distraction type, the probability value p regarding environment adaptation type, the probability value p regarding creative divergence type. The classification label at the output end of the Transformer encoder is the manual behavior pattern, and the label corresponding to the maximum probability value p is used as the manual behavior pattern. In addition, it should be noted that the manual behavior patterns in different time periods are different. For example, the historical period is the past month in total, and the manual behavior patterns are divided into the manual behavior pattern in the first week of the past month, the manual behavior pattern in the second week, the manual behavior pattern in the third week, etc.

[0104] Optionally, the process of using the Transformer encoder to perform feature fusion on the spliced features to obtain the classification probability vector refers to the process of inputting the spliced features into the Transformer encoder and outputting the classification probability vector by the Transformer encoder.

[0105] S2. During the current period, detect whether the computer scenario connected to the keyboard is a new scenario.

[0106] In the embodiment of the present invention, by detecting whether the computer scenario connected to the keyboard is a new scenario, new and old scenarios are distinguished, and different historical data influence powers are set in new and old scenarios.

[0107] Among them, the new scenario includes the software type and the connected peripheral device. The software type refers to the software within the computer interface where the current keyboard inputs data, such as painting software. The connected peripheral device refers to the peripheral device connected when the current keyboard inputs data, such as an external painting device connected to the keyboard, and the user is inputting data to this painting device through the keyboard.

[0108] In an embodiment of the present invention, detecting whether the computer scenario connected to the keyboard is a new scenario includes: querying the software information and peripheral device information in the computer scenario; determining whether similar software and similar peripheral devices matching the software information and the peripheral device information can be found in the historical period; when similar software and similar peripheral devices matching the software information and the peripheral device information can be found in the historical period, determining that the computer scenario is not a new scenario; when similar software and similar peripheral devices matching the software information and the peripheral device information cannot be found in the historical period, determining that the computer scenario is a new scenario.

[0109] Exemplarily, the process of determining whether similar software and similar peripheral devices matching the software information and the peripheral device information can be found in the historical period is as follows: when the software information is painting software and the peripheral device information is a painting device, and in the historical period, the similar software in the user's computer and external device is painting software and the similar peripheral device is a painting device, it is determined that similar software and similar peripheral devices matching the software information and the peripheral device information can be found in the historical period. When at least one of the software information and the peripheral device information can find similar information in the historical period, it can be determined that the matching is successful, otherwise the matching is unsuccessful.

[0110] S3. When the computer scenario is a new scenario, query the current usage habit of the user regarding the light effect of the keyboard, establish the basic keyboard light effect of the computer scenario using the manual behavior pattern, perform keyboard light effect guidance on the basis of the basic keyboard light effect using the current usage habit, obtain the first light effect guidance result, and improve the user's concentration through the first light effect guidance result to obtain the first concentration improvement result.

[0111] When the computer scenario is a new scenario, it is necessary to dynamically adjust the keyboard color parameters according to the change data of the current usage habit at each moment, rather than only relying on the data in the past period to dynamically adjust the keyboard color parameters.

[0112] Furthermore, the embodiment of the present invention queries the current usage habit of the user regarding the light effect of the keyboard to update the old usage habit through the current usage habit and improve the adaptability of the old usage habit to the user in the new scenario.

[0113] Among them, the current usage habit refers to the data newly generated by the user within the new scenario.

[0114] In one embodiment of the present invention, querying the current usage habits of the user regarding the lighting effects of the keyboard includes: querying the number of key position adjustments of the user regarding the lighting effects of the keyboard; generating the scene coordinates of the computer scene using the software information and peripheral information of the computer scene; calculating the Euclidean distance between the scene coordinates and the preset coordinates; and calculating the current probability vector of the computer scene according to the Euclidean distance using the following formula:

[0115]

[0116] where P j represents the current probability vector, d j represents the Euclidean distance between the scene coordinates and the j-th type of preset coordinates, and M represents the number of preset coordinates;

[0117] Taking the number of key position adjustments and the current probability vector as the current usage habits.

[0118] Wherein, the number of key position adjustments refers to the number of adjustments made by the user for each key on each keyboard. For example, the number of color adjustments for key A. It should be noted that the color parameters of the present invention only involve brightness and hue. The scene coordinates refer to the coordinates composed of the numbers of software information and peripheral information, and the numbers of software information and peripheral information can be obtained through one-hot encoding. The one-hot encoding is a method of converting discrete features into binary features. The preset coordinates refer to the numbers of the data in the software type and peripheral type databases that have been recorded and stored during the historical period.

[0119] Further, in the embodiment of the present invention, the basic keyboard lighting effect of the computer scene is established by using the manual behavior pattern, so as to dynamically optimize the brightness and hue of the keyboard by analyzing the historical adjustment habits of the user, provide a lighting effect that better meets the user's preferences, and thus improve the usage experience.

[0120] Wherein, the basic keyboard lighting effect refers to the keyboard lighting effect not affected by the current usage habits.

[0121] In one embodiment of the present invention, establishing the basic keyboard lighting effect of the computer scene by using the manual behavior pattern includes: calculating the software matching degree between the software information of the computer scene and the software types in the preset software library using the following formula:

[0122]

[0123] where M app represents the software matching degree, S i represents the keyword coincidence degree between the current software and the software type, Wi Represents the category weight of the software type;

[0124] Select the initial light effect in the software library corresponding to the highest matching degree of the software; calculate the peripheral influence coefficient of the computer scenario using the following formula:

[0125]

[0126] Where, C device,k Represents the peripheral influence coefficient corresponding to the kth type of manual behavior pattern, D num Represents the number of connected peripherals normalized to the 0-1 interval, D intensity Represents the peripheral usage intensity;

[0127] According to the initial light effect, the peripheral influence coefficient, and the manual behavior pattern, calculate the basic keyboard brightness of the computer scenario using the following formula:

[0128]

[0129] Where, L base Represents the basic keyboard brightness, H hour Represents the current time period, C device,k Represents the peripheral influence coefficient corresponding to the kth type of manual behavior pattern, L init Represents the initial brightness in the initial light effect, p k Represents the classification probability vector of the kth manual behavior pattern, and m represents the number of manual behavior patterns;

[0130] According to the initial light effect and the peripheral influence coefficient, calculate the basic keyboard hue of the computer scenario using the following formula:

[0131]

[0132] Where, H base Represents the basic keyboard hue, H hour Represents the current time period, H k Represents the main hue corresponding to the kth type of manual behavior pattern, H init Represents the initial hue in the initial light effect, p k Represents the classification probability vector of the kth manual behavior pattern, and m represents the number of manual behavior patterns;

[0133] Take the basic keyboard brightness and the basic keyboard hue as the basic keyboard light effect.

[0134] Among them, the preset software library is consistent with the software type and peripheral type database that has been recorded and stored in the historical period corresponding to the aforementioned preset coordinates. It is mainly used to store the software types and peripheral types that have appeared in the historical period. In addition, the color parameters of the keyboard are stored each time the software or device is used. The software types in the preset software library are the data stored in the database. The initial light effect refers to the default keyboard light effect (i.e., the color parameters of each key) when the user uses the software type and peripheral type corresponding to the highest matching degree. For example, when the software type and peripheral type corresponding to the highest matching degree are office types, the user adjusted the brightness and hue of the keyboard to 50, 50 the most times in the historical period, then the 50, 50 can be directly referenced as the initial light effect.

[0135] It should be noted that regarding S i Indicates the keyword overlap between the current software and the software type. The keyword overlap can be calculated by cosine similarity, TF-IDF and vector space model, and Levenshtein Distance. i The weight of the software type is obtained in advance through sample data and parameter optimization algorithm. For example, the least square method of calculating the regression coefficient of the regression model can be used to fit the regression coefficient, that is, the weight parameter of the regression model. The fitting result is a value between 0 and 1, such as office software = 0.4, design software = 0.3, programming software = 0.3. num 、D intensity , these two parameters are parameters that appear when the user is predicted to be in the kth manual behavior pattern in the historical period. It should be noted that the manual behavior pattern refers to the prediction result within a period of time, not the prediction result within all historical periods. For example, the historical period is the past month, and the manual behavior pattern is divided into the prediction results of the manual behavior pattern of the first week of the past month, the manual behavior pattern of the second week, the manual behavior pattern of the third week, and so on. Secondly, the manual behavior pattern corresponding to the maximum probability among the above-mentioned probability values p of the numerical prediction predicted to be the kth manual behavior pattern here, that is, there is only one manual behavior pattern in the first week of the past month, only one manual behavior pattern in the second week, and only one manual behavior pattern in the third week. D intensity The default value is the normalized power value of the peripheral device, that is, the ratio of the actual power of the peripheral device to the reference power is D intensity , about H k The main hue corresponding to the k-th manual behavior mode refers to the hue that appears most frequently during a period of time when the k-th manual behavior mode is predicted. There is not one main hue here, but each key on the keyboard has a hue that appears most frequently. The formula is The 24 in it represents 24 hours.

[0136] Furthermore, in the embodiment of the present invention, keyboard lighting effect guidance is carried out on the basis keyboard lighting effect by using the current usage habit, so as to combine the historical data and real-time behavior of the user and meet the personalized needs of the user.

[0137] In an embodiment of the present invention, the keyboard lighting effect guidance is carried out on the basis keyboard lighting effect by using the current usage habit to obtain a first lighting effect guidance result, including: obtaining the key position adjustment times and the current probability vector in the current usage habit; according to the key position adjustment times, calculating the current behavior entropy corresponding to the current usage habit by using the following formula:

[0138]

[0139] where, E (t) represents the current behavior entropy, C var represents the hue fluctuation index, H i represents the hue of the i-th adjustment in the key position adjustment times, H i-1 represents the hue of the (i - 1)-th adjustment in the key position adjustment times, n represents the number of times of adjusting the hue in the key position adjustment times, i represents the index of adjusting the hue in the key position adjustment times, D key represents the key position dispersion, N1 represents the key position adjustment times, and N2 represents the total key position adjustment times;

[0140] According to the current probability vector, calculate the behavior entropy threshold of the current behavior entropy by using the following formula:

[0141] E j = α(μ j + 1.5σ j ) + (1 - α)(μ E + 1.5σ E )

[0142]

[0143] where, E threshold represents the behavior entropy threshold, μ E represents the moving window mean of E (t) , σ E represents the standard deviation of E (t) , μ j represents the mean value of the behavior entropy in the historical period, σ j represents the standard deviation of the behavior entropy in the historical period, P j represents the j-th current probability vector, M represents the number of preset coordinates, α represents a weight with a value of 0.7, and E j represents the updated behavior entropy;

[0144] Using the entropy comparison result between the current behavior entropy and the behavior entropy threshold, keyboard lighting effect guidance is performed on the basic keyboard lighting effect to obtain a first lighting effect guidance result.

[0145] It should be noted that parameters such as the calculated current behavior entropy and behavior entropy threshold here are all relative to a certain key on the keyboard. For example, calculate the current behavior entropy and behavior entropy threshold corresponding to a certain key, and then based on the current behavior entropy and behavior entropy threshold, adjust the color of this key, and then adjust other color parameters in turn. This is similar to the principle of establishing the basic keyboard lighting effect of the computer scene using the manual behavior mode mentioned above, and it is a formula related to a certain key. And the total number of key position adjustments refers to the number of adjustments for all key positions.

[0146] Optionally, the process of using the entropy comparison result between the current behavior entropy and the behavior entropy threshold to perform keyboard lighting effect guidance on the basic keyboard lighting effect to obtain a first lighting effect guidance result means that when E (t) ≤E threshold it indicates that the influence of the behavior entropy in the historical period is relatively large, so the keyboard lighting effect in the historical period is maintained, such as maintaining the keyboard lighting effect in the third week of the past month. When E (t) >E threshold it indicates that the influence of the current behavior entropy is relatively large, so the parameters manually adjusted by the current user are maintained, that is, the hue and brightness of the key position manually adjusted by the user corresponding to the number of key position adjustments in the front.

[0147] In an embodiment of the present invention, the process of improving the user's concentration through the first lighting effect guidance result to obtain a first concentration improvement result includes: performing color changes in the keyboard according to the first lighting effect guidance result to improve the user's concentration and obtain a first concentration improvement result.

[0148] S4. When the computer scene is not a new scene, use the manual behavior mode to perform keyboard lighting effect guidance on the user to obtain a second lighting effect guidance result, so as to improve the user's concentration through the second lighting effect guidance result to obtain a second concentration improvement result.

[0149] In an embodiment of the present invention, the process of using the manual behavior mode to perform keyboard lighting effect guidance on the user to obtain a second lighting effect guidance result includes: based on the manual behavior mode, using the following formula to perform keyboard brightness guidance on the user to obtain a brightness guidance result:

[0150]

[0151] where L new represents the brightness guidance result, and p kDenote the classification probability vector of the k-th manual behavior pattern, m represents the number of manual behavior patterns, λ represents the amplitude coefficient of the brightness rhythm adjustment, T represents the period length of the brightness rhythm, and L k Denote the main brightness corresponding to the k-th type of manual behavior pattern, and t represents the time variable;

[0152] Based on the manual behavior pattern, use the following formula to perform keyboard hue guidance on the user to obtain a hue guidance result:

[0153]

[0154] where, ΔH represents the hue guidance result, and p k Denote the classification probability vector of the k-th manual behavior pattern, and ΔH k Denote the hue change step of the classification probability vector of the k-th manual behavior pattern, β represents the hue correction weight manually adjusted by the user, and ΔH user Denote the direction and amplitude of the user's most recent manual hue adjustment, and Sign() represents the sign function;

[0155] Perform keyboard light effect guidance on the user through the brightness guidance result and the hue guidance result to obtain a second light effect guidance result.

[0156] where, the brightness guidance result refers to the final brightness of the keyboard, and the hue guidance result is the hue change amplitude. λ = 5, T = 14400, β = 0.3. The main brightness corresponding to the k-th type of manual behavior pattern here refers to taking the brightness parameter that appears the most times during a certain period when it is predicted to be the k-th type of manual behavior pattern during this period as the main brightness. The hue change step of the classification probability vector of the k-th manual behavior pattern refers to taking the hue change step that appears the most times during a certain period when it is predicted to be the k-th type of manual behavior pattern during this period.

[0157] Optionally, the process of performing keyboard light effect guidance on the user through the brightness guidance result and the hue guidance result to obtain a second light effect guidance result refers to adjusting the current brightness of the keyboard to the brightness guidance result, and adding the hue guidance result to the current hue parameter of the keyboard to obtain an adjusted hue parameter.

[0158] S5. Take the first focus improvement result and the second focus improvement result as the focus improvement result of the user regarding the keyboard.

[0159] Compared with the problems described in the background art, in the embodiments of the present invention, the manual usage information is preprocessed to process the data form into data that can be autonomously analyzed by the subsequent neural network model. Further, in the embodiments of the present invention, the preprocessing information is used to identify the manual behavior pattern of the user, so as to adjust the light effect in real time according to the user's behavior pattern in the subsequent stage, helping the user to stay focused or stimulate creativity. In the embodiments of the present invention, it is detected whether the computer scenario connected to the keyboard is a new scenario to distinguish between new and old scenarios, and different historical data influence levels are set in the new and old scenarios. Further, in the embodiments of the present invention, the current usage habit of the user regarding the light effect of the keyboard is queried to update the old usage habit through the current usage habit, improving the adaptability of the old usage habit to the user in the new scenario. Further, in the embodiments of the present invention, the basic keyboard light effect of the computer scenario is established by using the manual behavior pattern, so as to dynamically optimize the brightness and hue of the keyboard by analyzing the user's historical adjustment habits, providing a light effect that better meets the user's preferences, thereby improving the usage experience. Further, in the embodiments of the present invention, keyboard light effect guidance is performed on the basis of the basic keyboard light effect by using the current usage habit to meet the personalized needs of the user by combining the user's historical data and real-time behavior. Therefore, the focus improvement method and system based on keyboard light effect guidance provided by the embodiments of the present invention can meet the personalized needs of the user.

[0160] Embodiment 2:

[0161] As Figure 2 shown, it is a functional module diagram of a focus improvement system based on keyboard light effect guidance according to the present invention.

[0162] The focus improvement system 200 based on keyboard light effect guidance according to the present invention can be installed in an electronic device. According to the functions achieved, the focus improvement system based on keyboard light effect guidance can include a pattern recognition module 201, a scenario detection module 202, a first guidance module 203, a second guidance module 204, and a result determination 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 the processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0163] In the embodiments of the present invention, the functions of each module / unit are as follows:

[0164] The pattern recognition module 201 is used to store the manual usage information of the user regarding the light effect of the keyboard in the keyboard during a historical period, preprocess the manual usage information to obtain preprocessing information, and use the preprocessing information to identify the manual behavior pattern of the user, wherein the manual usage information includes spatial information, time information, and behavior information;

[0165] The scene detection module 202 is configured to detect whether the computer scene connected to the keyboard is a new scene during the current period;

[0166] The first guidance module 203 is configured to, when the computer scene is a new scene, query the user's current usage habit regarding the lighting effect of the keyboard, establish the basic keyboard lighting effect of the computer scene using the manual behavior pattern, perform keyboard lighting effect guidance on the basis of the basic keyboard lighting effect using the current usage habit, obtain a first lighting effect guidance result, so as to improve the user's concentration through the first lighting effect guidance result and obtain a first concentration improvement result;

[0167] The second guidance module 204 is configured to, when the computer scene is not a new scene, perform keyboard lighting effect guidance on the user using the manual behavior pattern, obtain a second lighting effect guidance result, so as to improve the user's concentration through the second lighting effect guidance result and obtain a second concentration improvement result;

[0168] The result determination module 205 is configured to use the first concentration improvement result and the second concentration improvement result as the concentration improvement result of the user regarding the keyboard.

[0169] Specifically, each module in the concentration improvement system 200 based on keyboard lighting effect guidance in the embodiments of the present invention adopts the same technical means as those in the Figure 1 concentration improvement method based on keyboard lighting effect guidance described above, and can produce the same technical effects, which will not be elaborated here.

[0170] 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 without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms.

[0171] 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. A focus improvement method based on keyboard light effect guidance, characterized in that, The method includes: During a historical period, storing in the keyboard the manual usage information of the user regarding the lighting effect of the keyboard, preprocessing the manual usage information to obtain preprocessed information, and using the preprocessed information to identify the manual behavior pattern of the user, where the manual usage information includes spatial information, time information, and behavior information; During the current period, detecting whether the computer scenario connected to the keyboard is a new scenario; When the computer scenario is a new scenario, querying the current usage habit of the user regarding the lighting effect of the keyboard, establishing the basic keyboard lighting effect of the computer scenario using the manual behavior pattern, and performing keyboard lighting effect guidance on the basis of the basic keyboard lighting effect using the current usage habit to obtain a first lighting effect guidance result, so as to improve the concentration of the user through the first lighting effect guidance result and obtain a first concentration improvement result; When the computer scenario is not a new scenario, performing keyboard lighting effect guidance on the user using the manual behavior pattern to obtain a second lighting effect guidance result, so as to improve the concentration of the user through the second lighting effect guidance result and obtain a second concentration improvement result; Taking the first concentration improvement result and the second concentration improvement result as the concentration improvement result of the user regarding the keyboard.

2. The focus improvement method based on keyboard lighting effect guidance according to claim 1, wherein The preprocessing of the manual usage information to obtain preprocessed information includes: Obtaining the key position, brightness data, hue data, operation timestamp, and operation period in the manual usage information, and obtaining the keyboard corresponding to the manual usage information; Mapping the overall keys of the keyboard to a preset grid coordinate system to obtain keyboard grid coordinates; Querying the key coordinates of the key position from the keyboard grid coordinates; Determining the grayscale image of the key coordinates based on the ratio of the number of adjustments to the total number of adjustments on the key coordinates; Performing color initialization on the keyboard grid coordinates to obtain an initialized color; Performing color adjustment on the initialized color based on the brightness data and the hue data to obtain an adjusted color; Analyzing the gradient diffusion color of the neighborhood coordinates of the key position using the adjusted color; Performing a color overlay operation on the gradient diffusion color to obtain an overlaid color; Converting the overlaid color into atomic operations; Calculating the transition probability between adjacent operations in the atomic operations; Converting the operation timestamp into a frequency domain feature; Performing window sliding on the operation period to obtain a time feature; Taking the grayscale image, the atomic operations, the transition probability, the frequency domain feature, and the time feature as preprocessed information.

3. The focus improvement method based on keyboard light effect guidance according to claim 1, wherein The identifying the manual behavior pattern of the user using the preprocessed information includes: Preparing a feature extraction model and a Transformer encoder for the preprocessed information; Among them, the feature extraction model includes a deep residual network, a global pooling layer, a word vector model, a splicing layer, a long short-term memory network, a fully connected layer, an activation function, and a non-linear mapping layer; Using the feature extraction model to extract information features from the preprocessed information to obtain extracted features; Performing feature splicing on the extracted features to obtain spliced features; Feature fusion is performed on the spliced features using the Transformer encoder to obtain a classification probability vector; The manual behavior pattern of the user is determined using the classification probability vector.

4. The focus improvement method based on keyboard light effect guidance according to claim 1, wherein, Detecting whether the computer scenario connected to the keyboard is a new scenario includes: Querying the software information and peripheral information in the computer scenario; Determining whether similar software and similar peripherals that match the software information and the peripheral information can be found in the historical period; When similar software and similar peripherals that match the software information and the peripheral information can be found in the historical period, it is determined that the computer scenario is not a new scenario; When similar software and similar peripherals that match the software information and the peripheral information cannot be found in the historical period, it is determined that the computer scenario is a new scenario.

5. The focus improvement method based on keyboard light effect guidance according to claim 1, wherein Querying the user's current usage habit regarding the lighting effect of the keyboard includes: Querying the number of key position adjustments of the user regarding the lighting effect of the keyboard; Generating the scene coordinates of the computer scenario using the software information and peripheral information of the computer scenario; Calculating the Euclidean distance between the scene coordinates and the preset coordinates; According to the Euclidean distance, calculating the current probability vector of the computer scenario using the following formula; Among them, P j represents the current probability vector, and d j represents the Euclidean distance between the scene coordinates and the preset coordinates of the j-th class, and M represents the number of preset coordinates; Taking the number of key position adjustments and the current probability vector as the current usage habit.

6. The focus improvement method based on keyboard light effect guidance according to claim 1, wherein, Establishing the basic keyboard lighting effect of the computer scenario using the manual behavior pattern includes: Calculating the software matching degree between the software information of the computer scenario and the software types in the preset software library using the following formula: Among them, M app represents the software matching degree, S i represents the keyword overlap degree between the current software and the software type, W i represents the category weight of the software type; Selecting the initial lighting effect in the software library corresponding to the highest matching degree of the software matching degree; Calculating the peripheral influence coefficient of the computer scenario using the following formula; Among them, C device,k represents the peripheral influence coefficient corresponding to the k-th type of manual behavior pattern, D num represents the number of connected peripherals normalized to the range of 0 - 1, D intensity represents the peripheral usage intensity; According to the initial lighting effect, the peripheral influence coefficient, and the manual behavior pattern, calculating the basic keyboard brightness of the computer scenario using the following formula; Among them, L base represents the basic keyboard brightness, H hour represents the current time period, C device,k represents the peripheral influence coefficient corresponding to the k-th type of manual behavior pattern, L init represents the initial brightness in the initial light effect, p k represents the classification probability vector of the k-th manual behavior pattern, and m represents the number of manual behavior patterns; According to the initial lighting effect and the peripheral influence coefficient, calculating the basic keyboard hue of the computer scenario using the following formula; Among them, H base represents the base keyboard hue, H hour represents the current time period, H k represents the main hue corresponding to the k-th type of manual behavior pattern, H init represents the initial hue in the initial light effect, p k represents the classification probability vector of the k-th manual behavior pattern, and m represents the number of manual behavior patterns; Taking the basic keyboard brightness and the basic keyboard hue as the basic keyboard lighting effect.

7. The focus improvement method based on keyboard lighting effect guidance according to claim 1, characterized in that, Performing keyboard lighting effect guidance on the basis of the basic keyboard lighting effect using the current usage habit to obtain a first lighting effect guidance result includes: Obtaining the number of key position adjustments and the current probability vector in the current usage habit; Calculating the current behavior entropy corresponding to the current usage habit using the following formula according to the number of key position adjustments; Among them, E (t) represents the current behavior entropy, C var represents the hue fluctuation index, H i represents the hue of the i-th adjustment in the number of key position adjustments, H i-1 represents the hue of the (i - 1)-th adjustment in the number of key position adjustments, n represents the number of times of adjusting the hue in the number of key position adjustments, i represents the index of adjusting the hue in the number of key position adjustments, D key represents the key position dispersion, N1 represents the number of key position adjustments, N2 represents the total number of key position adjustments; Calculating the behavior entropy threshold of the current behavior entropy using the following formula according to the current probability vector; E i = α(μ j + 1.5σ j ) + (1 - α)(μ E + 1.5σ E ) Among them, E threshold represents the behavior entropy threshold, μ E represents the moving window mean of E (t) and σ E represents the standard deviation of E (t) ; μ j represents the mean behavior entropy of the historical period, and σ j represents the standard deviation of the behavior entropy of the historical period; P j represents the j-th current probability vector, M represents the number of preset coordinates, α represents the weight with a value of 0.7, and E j represents the updated behavior entropy; Using the entropy comparison result between the current behavior entropy and the behavior entropy threshold to perform keyboard lighting effect guidance on the basis of the basic keyboard lighting effect to obtain a first lighting effect guidance result.

8. The focus improvement method based on keyboard lighting effect guidance according to claim 1, wherein Improving the user's concentration through the first lighting effect guidance result to obtain a first concentration improvement result includes: Changing the color in the keyboard according to the first lighting effect guidance result to improve the user's concentration and obtain a first concentration improvement result.

9. The focus improvement method based on keyboard light effect guidance according to claim 1, wherein Performing keyboard lighting effect guidance on the user using the manual behavior pattern to obtain a second lighting effect guidance result includes: Based on the manual behavior pattern, the following formula is used to guide the keyboard brightness of the user to obtain a brightness guidance result: Among them, L new represents the luminance guiding result, p k represents the classification probability vector of the k-th manual behavior pattern, m represents the number of manual behavior patterns, λ represents the amplitude coefficient of luminance rhythm adjustment, T represents the period length of the luminance rhythm, and L k represents the main luminance corresponding to the k-th type of manual behavior pattern, and t represents the time variable; Based on the manual behavior pattern, the following formula is used to guide the keyboard hue of the user to obtain a hue guidance result: Among them, ΔH represents the hue guidance result, and p k represents the classification probability vector of the k-th manual behavior pattern, and ΔH k represents the hue change step of the classification probability vector of the k-th manual behavior pattern, β represents the hue correction weight manually adjusted by the user, and ΔH user represents the direction and amplitude of the user's most recent manual hue adjustment, and Sign() represents the sign function; Based on the brightness guidance result and the hue guidance result, the keyboard light effect is guided for the user to obtain a second light effect guidance result.

10. A concentration improvement system based on keyboard lighting effect guidance, characterized in that, The system includes: A pattern recognition module, which is used to store the manual usage information of the user about the light effect of the keyboard in the keyboard during a historical period, preprocess the manual usage information to obtain preprocessed information, and use the preprocessed information to identify the manual behavior pattern of the user, where the manual usage information includes spatial information, time information, and behavior information; A scene detection module, which is used to detect whether the computer scene connected to the keyboard is a new scene during the current period; A first guidance module, which is used to query the current usage habit of the user about the light effect of the keyboard when the computer scene is a new scene, establish a basic keyboard light effect for the computer scene using the manual behavior pattern, and perform keyboard light effect guidance on the basis of the basic keyboard light effect using the current usage habit to obtain a first light effect guidance result, so as to improve the concentration of the user through the first light effect guidance result to obtain a first concentration improvement result; A second guidance module, which is used to perform keyboard light effect guidance on the user using the manual behavior pattern when the computer scene is not a new scene to obtain a second light effect guidance result, so as to improve the concentration of the user through the second light effect guidance result to obtain a second concentration improvement result; A result determination module, which is used to use the first concentration improvement result and the second concentration improvement result as the concentration improvement result of the user about the keyboard.

Images