LED display screen key switching method of gamepad, control system and storage medium

Abstract

The invention belongs to the technical field of data processing, and discloses an LED display screen key switching method of a gamepad, a control system and a storage medium. Comprising the steps that signal collection is conducted on key information in the using process of a target gamepad, and a corresponding gamepad operation signal is obtained; performing signal preprocessing on the collected handle operation signal to obtain a corresponding signal to be detected; inputting the obtained signal to be detected into a pre-constructed signal feature extraction model to obtain corresponding key feature information; performing instruction matching based on the obtained key feature information, and performing key switching and virtual key mapping based on an instruction matching result; the invention aims to realize accurate key switching and virtual key mapping and improve the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and more specifically to a method for switching buttons of an LED display screen of a game controller, a control system and a storage medium. Background Art

[0002] With the rapid development of the gaming industry, game controllers, as an important tool for gamers to interact with the gaming world, are increasingly required to have more functions and performance. Traditional game controllers mainly use physical buttons to input various operating instructions. However, as the complexity of games increases, players are increasingly demanding the convenience, flexibility, and personalization of controller operations.

[0003] Traditional game controllers often have some problems in key signal collection, such as signal errors caused by players operating too fast or accidentally pressing keys, and interference with signal collection caused by the vibration effect of the controller. These problems not only reduce the player's gaming experience, but may also affect the normal operation of the game.

[0004] In addition, the button functions of traditional game controllers are relatively fixed, which makes it difficult to meet the players' needs for personalized operations. For example, some players may want to set the button functions and layouts according to their own gaming habits and preferences, but traditional game controllers often cannot provide such customized functions.

[0005] In addition, with the continuous advancement of gaming technology, more and more games are beginning to support advanced features such as virtual buttons and LED displays; however, traditional game controllers have great limitations in virtual button mapping and LED display displays, and cannot meet players' demands for advanced gaming features.

[0006] In view of this, the present invention proposes a method for switching buttons of an LED display screen of a game controller, a control system and a storage medium to solve the above problems. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions:

[0008] The LED display button switching control system of the game controller includes:

[0009] The data acquisition module is used to collect the key information of the target game controller during use and obtain the corresponding controller operation signal;

[0010] The data processing module is used to perform signal preprocessing on the collected handle operation signal to obtain the corresponding signal to be detected;

[0011] The data analysis module is used to input the obtained signal to be detected into a pre-built signal feature extraction model to obtain corresponding key feature information;

[0012] The data control module is used to perform command matching based on the acquired key feature information, and perform key switching and virtual key mapping based on the command matching result.

[0013] Furthermore, the process of collecting signal of key information during the use of the target game controller and obtaining the corresponding controller operation signal includes:

[0014] The data acquisition module is composed of a number of sensor terminals, which are used to collect data on signals generated by operations such as button pressing, joystick movement, and display screen touch sliding during the use of the corresponding game controller, and verify whether the collected signal is a false touch signal. If it is not a false touch signal, the collected signal is converted to obtain a corresponding controller operation signal; if it is a false touch signal, the collected signal is discarded.

[0015] Furthermore, the process of performing signal preprocessing on the collected handle operation signal to obtain the corresponding signal to be detected includes:

[0016] Constructing standard white noise, adding the corresponding standard white noise to the obtained handle operation signal, and performing signal decomposition on the handle operation signal after adding the standard white noise to obtain signal components at several different scales; and constructing a corresponding scale signal sequence based on the components;

[0017] Perform secondary decomposition on the signal components in the corresponding scale signal sequence signal to obtain the corresponding first-order residual component and first-order IMF component;

[0018] Add the constructed standard white noise to the obtained first-order residual component, continue to decompose the signal, and so on, until the pre-set decomposition stop condition is met; obtain a number of signal modal components from high frequency to low frequency;

[0019] The obtained signal modal component is correlated with the original handle operation signal to obtain the corresponding signal correlation coefficient;

[0020] A correlation threshold is set, and the obtained signal correlation coefficient is compared with the correlation threshold. If the signal correlation coefficient is less than the correlation threshold, it indicates that the correlation between the corresponding signal modal component and the original handle signal is low, and the corresponding signal modal component is marked as an invalid component; if the signal correlation coefficient is not less than the correlation threshold, the corresponding signal modal component is marked as a component to be processed;

[0021] Read the obtained components to be processed, and perform wavelet packet decomposition on them to obtain corresponding detail sub-components and high-frequency sub-components;

[0022] Adaptively optimizing the signal of the obtained detail sub-components to obtain corresponding pure detail signals;

[0023] Obtain the wavelet packet coefficients corresponding to the high-frequency sub-components at different scales in the corresponding wavelet packet decomposition process;

[0024] Construct a time window with a time length of N, and add the corresponding time window to the wavelet packet coefficient corresponding to the corresponding high-frequency sub-component; and obtain the wavelet average value of the wavelet packet coefficient in the corresponding time window;

[0025] Get the wavelet average values ​​corresponding to all time windows in the corresponding high-frequency sub-component, sort them, select the wavelet packets corresponding to the n time windows at the end of the ranking and calculate the average value, and mark the obtained average value as the wavelet basis threshold; n is a fixed constant;

[0026] Based on the wavelet basis threshold, threshold processing is performed on the wavelet packet coefficients corresponding to the corresponding high-frequency sub-component; the high-frequency sub-component after the threshold processing and the corresponding pure detail sub-component are subjected to wavelet reconstruction to obtain the corresponding enhanced modal component;

[0027] The enhanced modal component and the invalid component are signal-fused to obtain a corresponding signal to be detected; the signal to be detected refers to an enhanced signal of the corresponding handle operation signal.

[0028] Furthermore, the formula for the secondary decomposition is: Where r1 represents the first-order residual component obtained, x(t) represents the original signal component; IMF1 represents the first-order IMF component obtained; E1(w 1 ) represents the signal component corresponding to the handle operation signal with standard white noise added, and x(t) represents the signal component corresponding to the original handle operation signal; In the formula, β0 represents the signal-to-noise ratio; σ() represents the mathematical expectation operation; δ0 is the preset signal amplitude;

[0029] The corresponding adaptive signal optimization processing formula is: Where T(k) represents the pure detail signal obtained after adaptive signal optimization processing; c1 and c2 represent the scaling factor and suppression factor respectively; Z k A represents the actual observed value corresponding to the corresponding detail sub-component at time k; k|k-1 represents the observed value predicted based on the observed value at time k-1; P k represents the Kalman gain coefficient corresponding to time k; tanh() represents a pre-selected nonlinear activation function; k represents the time index;

[0030] The formula for threshold processing is:

[0031] In the formula, d` (j，a) It represents the wavelet packet coefficient corresponding to the jth high-frequency subcomponent in the ath decomposition layer after threshold processing in the wavelet packet decomposition process; d (j，a) The wavelet packet coefficient corresponding to the jth high-frequency subcomponent in the ath decomposition layer during the wavelet packet decomposition process; (j，a) is the wavelet basis threshold; sgn() is the sign function; CD represents the scale factor of wavelet packet decomposition; a and j are natural numbers, and a and j represent the number of decomposition layers and the index of the high-frequency sub-component in the corresponding decomposition layer, respectively.

[0032] Furthermore, the obtained signal to be detected is read, and the obtained signal to be detected is input into a pre-constructed signal characteristic model to obtain a corresponding model output result, and the model output result is the key characteristic information corresponding to the corresponding signal to be detected.

[0033] Furthermore, the construction process of the signal feature model includes:

[0034] A key feature clustering tree is constructed, and a corresponding training data set is constructed based on the constructed key feature clustering tree; the training data set is composed of a plurality of training samples, and the training samples include key pressing signals and their corresponding clustering features.

[0035] The network framework of the signal feature model is defined as input layer, feature extraction layer, normalization layer and output layer;

[0036] The input layer is used to receive the training samples in the input training data set and perform data conversion on the corresponding training samples, wherein the data conversion refers to converting the input signal into a two-dimensional image with time correlation;

[0037] The feature extraction layer is composed of a convolution layer and a pooling layer; the convolution layer is used to extract local features of the input two-dimensional image to obtain a corresponding local feature map; the pooling layer is used to perform a pooling operation on the obtained local feature map;

[0038] The normalization layer is composed of a separable convolution layer and a batch normalization layer. The separable convolution layer and the separable convolution layer are used to extract the temporal features in the local feature map and perform feature fusion. The batch normalization layer is used to normalize the mean and variance of the input training samples.

[0039] The output layer is used to receive the feature information output by the normalization layer, flatten the information, and connect it to the softmax classifier through the built-in fully connected layer to perform final feature mapping and classification;

[0040] The training data set is divided into N3 training data subsets, and the training data subsets are planned as sequence data according to the chronology of historical data collection time points; AdaGrad is defined as the optimizer to continuously optimize the parameters of the signal feature extraction model during the training process, and the sequence data is input into the signal feature extraction model in batches according to the time step, and the value of the corresponding loss function is recorded. When the value of the loss function of consecutive L batches no longer decreases or changes, the parameters of the signal feature extraction model at this time are saved, and the training of the signal feature extraction model is completed; N3 is a constant.

[0041] Furthermore, the convolution kernel size in the convolution layer is defined as 3×3, and a convolution operation is performed on the corresponding two-dimensional image based on the convolution kernel to obtain a corresponding local feature map. The formula of the convolution operation is: In the formula, represents the local feature map output by the u-th convolution kernel in the v-th layer; f() represents the activation function; v represents the index of the convolution layer; u and v are natural numbers; Represents the input feature map corresponding to the previous layer of the v-th convolutional layer; Represents the weight parameter between the u-th convolution kernel and the h-th input feature map in the v-th layer; represents the bias term; C u Represents the feature area of ​​the local feature map corresponding to the u-th convolution kernel in the corresponding two-dimensional image;

[0042] The formula for defining the pooling operation is: CH = MAXP(cs0) + xbMAXP(cs1); where CH is the output result of the pooling layer, MAXP() represents the maximum pooling operation; cs0 and cs1 represent the real and imaginary parts of the local feature map output by the convolutional layer connected to the corresponding pooling layer, respectively; xb represents the imaginary unit; xb 2 = -1;

[0043] Define the loss function of the signal feature extraction model In the formula, M is the total number of feature categories, R ym` Represents the predicted value of category m in the prediction vector of the input training sample y; μ ym It represents the probability that the training sample y belongs to the feature category m after the training sample undergoes the softmax operation.

[0044] Furthermore, the process of performing command matching based on the obtained key feature information, and performing key switching and virtual key mapping based on the command matching result includes:

[0045] Based on the output result of the model, key feature information corresponding to the corresponding handle operation signal is obtained; the corresponding key feature information and the handle operation signal are respectively input into the corresponding cluster feature tree to obtain the signal similarity between the corresponding handle operation signal and the corresponding key pressing signal and the feature similarity between the key feature information and the corresponding cluster feature;

[0046] Based on the key feature clustering tree, signal similarity and key feature information, the handle key function to which the corresponding handle operation signal belongs is obtained, and a corresponding key instruction is generated based thereon;

[0047] Identify the key command obtained, compare it with the pre-stored key command, and determine whether it is a switching command based on the comparison result. If not, transmit the corresponding key command to the game device connected to the target game controller and execute it; if so, obtain the key function mode and LED light information that the corresponding user needs to switch based on the pre-stored key command; and transmit it to the corresponding game controller;

[0048] Furthermore, the game controller performs button mapping on the LED display screen of the corresponding game controller based on the received button function mode; at the same time, the light color is adjusted based on the LED light information.

[0049] Furthermore, the method for switching the LED display screen buttons of the game controller includes:

[0050] Step 1: Collect the key information during the use of the target game controller to obtain the corresponding controller operation signal;

[0051] Step 2: Preprocess the collected handle operation signal to obtain the corresponding signal to be detected;

[0052] Step 3: Input the obtained signal to be detected into a pre-built signal feature extraction model to obtain corresponding key feature information;

[0053] Step 4: Perform command matching based on the acquired key feature information, and perform key switching and virtual key mapping based on the command matching result.

[0054] Furthermore, a computer-readable storage medium is provided, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method for switching buttons of an LED display screen of the game controller is implemented.

[0055] The LED display screen button switching method, control system and storage medium of the game controller of the present invention have the following technical effects and advantages:

[0056] 1. Through the high-precision sensor terminal of the data acquisition module, the accurate collection of game controller button information can be achieved, effectively avoiding the influence of false touch signals, thereby improving the accuracy of signal collection; not only improving the player's operating experience, but also ensuring the accurate transmission of game commands;

[0057] 2. Through the signal preprocessing of the data processing module, the error caused by the vibration effect of the game controller on signal acquisition is reduced, and the accuracy of the controller operation signal is further enhanced; it provides a solid foundation for the subsequent button feature extraction and command matching, thereby ensuring the stability and reliability of the game controller operation; the use of the signal feature extraction model enables the system to accurately identify and extract button feature information, and then realize command matching based on feature information; it not only improves the response speed of the game controller, but also enables players to meet personalized needs by customizing button function modes and LED light information, greatly enriching the interaction mode and operation experience of the game controller. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a schematic diagram of the LED display screen button switching control system of the game controller of the present invention;

[0059] Figure 2 The present invention is a schematic diagram of a method for switching buttons on an LED display screen of a game controller. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] Example 1

[0062] See also Figure 1 As shown, the LED display screen button switching control system of the game controller described in this embodiment includes:

[0063] The data acquisition module is used to collect the key information of the target game controller during use and obtain the corresponding controller operation signal;

[0064] The data processing module is used to perform signal preprocessing on the collected handle operation signal to obtain the corresponding signal to be detected;

[0065] The data analysis module is used to input the obtained signal to be detected into a pre-built signal feature extraction model to obtain corresponding key feature information;

[0066] A data control module, used for performing command matching based on the acquired key feature information, and performing key switching and virtual key mapping based on the command matching result;

[0067] The modules are connected to each other via wired and / or wireless means to achieve data transmission between modules.

[0068] It should be further explained that, in the specific implementation process, the process of collecting signal of key information during the use of the target game controller and obtaining the corresponding controller operation signal includes:

[0069] The data acquisition module is composed of a number of sensor terminals, which are arranged in the physical buttons, handle remote sensing and handle LED display screen of the corresponding game controller; they are used to collect data on the signals generated by the button pressing, joystick movement and display screen touch sliding operations during the use of the corresponding game controller, and verify whether the collected signal is a false touch signal. If it is not a false touch signal, the collected signal is converted to obtain the corresponding handle operation signal; if it is a false touch signal, the collected signal is discarded; wherein the handle operation signal is an analog signal formed in one or more pressing processes corresponding to the corresponding collection time; wherein the sensor terminal can be a pressure terminal or a current terminal;

[0070] It should be further explained that, in the specific implementation process, the method of verifying whether the collected signal is a false touch signal is: obtain the duration of the corresponding button pressing and releasing process; and compare the duration of pressing and releasing with the pre-set teammate time duration threshold. If it is less than the time duration threshold, it is an error signal; if it is not less than the time duration threshold, it is not a false touch signal.

[0071] It should be further explained that, in the specific implementation process, the process of performing signal preprocessing on the collected handle operation signal to obtain the corresponding signal to be detected includes:

[0072] Constructing standard white noise, which refers to random noise with the same energy density at all frequencies; then, adding the corresponding standard white noise to the obtained handle operation signal, and performing signal decomposition on the handle operation signal after adding the standard white noise to obtain signal components at several different scales; and constructing a corresponding scale signal sequence based on the components;

[0073] Perform secondary decomposition on the signal components in the corresponding scale signal sequence signal to obtain the corresponding first-order residual component and first-order IMF component;

[0074] Among them, the formula for secondary decomposition is: Where r1 represents the first-order residual component obtained, x(t) represents the original signal component; IMF1 represents the first-order IMF component obtained; E1(w 1 ) represents the signal component corresponding to the handle operation signal with standard white noise added, and x(t) represents the signal component corresponding to the original handle operation signal (i.e., the signal component corresponding to the handle operation signal without adding standard white noise); In the formula, β0 represents the signal-to-noise ratio; σ() represents the mathematical expectation operation; δ0 is the preset signal amplitude;

[0075] Add the constructed standard white noise to the obtained first-order residual component, continue to decompose the signal, and so on, until the pre-set decomposition stop condition is met; obtain a number of signal modal components from high frequency to low frequency; wherein the decomposition stop condition is that the residual component obtained by decomposition is a monotonic signal;

[0076] The obtained signal modal component is correlated with the original handle operation signal to obtain the corresponding signal correlation coefficient;

[0077] A correlation threshold is set, and the obtained signal correlation coefficient is compared with the correlation threshold. If the signal correlation coefficient is less than the correlation threshold, it indicates that the correlation between the corresponding signal modal component and the original handle signal is low, and the corresponding signal modal component is marked as an invalid component; if the signal correlation coefficient is not less than the correlation threshold, the corresponding signal modal component is marked as a component to be processed;

[0078] Among them, the signal correlation coefficient In the formula, cov represents the covariance operation, f(t) and F(t) represent the signal modal component and the original handle operation signal respectively; the larger the signal correlation coefficient is, the greater the correlation between the corresponding signal modal component and the original handle operation signal is;

[0079] Read the obtained components to be processed, and perform wavelet packet decomposition on them to obtain corresponding detail sub-components and high-frequency sub-components;

[0080] Then, the obtained detail sub-component is subjected to adaptive signal optimization processing to obtain the corresponding pure detail signal; wherein the formula for the corresponding adaptive signal optimization processing is: Wherein, T(k) represents the pure detail signal obtained after adaptive signal optimization processing; c1 and c2 represent the scaling factor and suppression factor, respectively, which are set in advance by those skilled in the art; Z represents the actual observed value corresponding to the corresponding detail subcomponent at time k; A k|k-1 represents the observed value predicted based on the observed value at time k-1; P krepresents the Kalman gain coefficient corresponding to time k; tanh() represents a pre-selected nonlinear activation function; k represents the time index;

[0081] Obtain the wavelet packet coefficients corresponding to the high-frequency sub-components at different scales in the corresponding wavelet packet decomposition process; wherein the wavelet packet coefficients contain the information of the original signal at different scales and frequencies;

[0082] Then, a time window with a time length of N is constructed, and the corresponding time window is added to the wavelet packet coefficient corresponding to the corresponding high-frequency sub-component; and the wavelet average value of the wavelet packet coefficient in the corresponding time window is obtained. In the formula, represents the wavelet average value corresponding to the i-th time window in the j-th high-frequency sub-component; p = 1, 2, ..., N, N> 0 and N is an integer, N represents the time length of the time window; Represents the wavelet packet coefficient corresponding to the nth time in the ith time window of the jth high-frequency sub-component; where i and j are natural numbers; they represent the index of the time window and the high-frequency sub-component respectively; n is the time index in the corresponding time window;

[0083] Obtain the wavelet average values ​​corresponding to all time windows in the corresponding high-frequency sub-component, sort them, select the wavelet packet systems corresponding to the n time windows at the end of the ranking and calculate the average value, and mark the obtained average value as the wavelet basis threshold;

[0084] Then, threshold processing is performed on the wavelet packet coefficients corresponding to the corresponding high-frequency sub-component based on the wavelet basis threshold; the high-frequency sub-component after the threshold processing and the corresponding pure detail sub-component are subjected to wavelet reconstruction to obtain the corresponding enhanced modal component;

[0085] Furthermore, the enhanced modal component and the invalid component are subjected to signal fusion to obtain a corresponding signal to be detected; the signal to be detected refers to an enhanced signal of the corresponding handle operation signal, wherein the main purpose of the preprocessing is to reduce the signal error caused by the vibration effect of the game controller during the acquisition process of the corresponding handle operation signal, so as to maximize the accuracy of the handle operation signal during execution;

[0086] The formula for threshold processing is:

[0087] In the formula, d` (j，a) It represents the wavelet packet coefficient corresponding to the jth high-frequency subcomponent in the ath decomposition layer after threshold processing in the wavelet packet decomposition process; d (j,a) The wavelet packet coefficient corresponding to the jth high-frequency subcomponent in the ath decomposition layer during the wavelet packet decomposition process; (j,a)is the wavelet basis threshold; sgn() is the sign function; CD represents the scale factor of wavelet packet decomposition; a and j are natural numbers, and a and j represent the number of decomposition layers and the index of the high-frequency sub-component in the corresponding decomposition layer, respectively.

[0088] It should be further explained that, in the specific implementation process, the obtained signal to be detected is input into the pre-built signal feature extraction model, and the process of obtaining the corresponding key feature information includes:

[0089] Read the obtained signal to be detected, and input the obtained signal to be detected into a pre-built signal feature model to obtain a corresponding model output result, wherein the model output result is key feature information corresponding to the corresponding signal to be detected, wherein the key feature information includes feature data composed of signal features corresponding to one or more handle operation signals;

[0090] It should be further explained that, in the specific implementation process, the construction process of the signal feature model includes:

[0091] A key feature clustering tree is constructed, and a corresponding training data set is constructed based on the constructed key feature clustering tree; the training data set is composed of a plurality of training samples, and the training samples include key pressing signals and their corresponding clustering features.

[0092] The network framework of the signal feature model is defined as input layer, feature extraction layer, normalization layer and output layer;

[0093] The input layer is used to receive the training samples in the input training data set and perform data conversion on the corresponding training samples, wherein the data conversion refers to converting the input signal into a two-dimensional image with time correlation;

[0094] The feature extraction layer is composed of a convolution layer and a pooling layer; the convolution layer is used to extract local features of the input two-dimensional image to obtain a corresponding local feature map; the pooling layer is used to perform a pooling operation on the obtained local feature map to reduce the spatial dimension of the local feature map while retaining the main features of the feature map;

[0095] The normalization layer consists of a separable convolution layer and a batch normalization layer. The separable convolution layer and the separable convolution layer are used to extract the temporal features in the local feature map and perform feature linear fusion. The batch normalization layer is used to normalize the mean and variance of the input training samples to make the input distribution of the network layer more stable. This helps alleviate the problem of internal covariate shift and allows the use of a higher learning rate, thereby accelerating the training process.

[0096] The output layer is used to receive the feature information output by the normalization layer, flatten the information, and connect it to the softmax classifier through the built-in fully connected layer to perform final feature mapping and classification; wherein the information flattening refers to converting multidimensional data into one-dimensional data;

[0097] The convolution kernel size in the convolution layer is defined as 3×3, and a convolution operation is performed on the corresponding two-dimensional image based on the convolution kernel to obtain a corresponding local feature map. The formula of the convolution operation is: In the formula, represents the local feature map output by the u-th convolution kernel in the v-th layer; f() represents the activation function, which is used to perform nonlinear mapping on the output of the convolution layer so that the network can learn more complex features; v represents the index of the convolution layer; u and v are natural numbers; Represents the input feature map corresponding to the previous layer of the v-th convolutional layer; Represents the weight parameter between the u-th convolution kernel and the h-th input feature map in the v-th layer; represents the bias term; C u Represents the feature area of ​​the local feature map corresponding to the u-th convolution kernel in the corresponding two-dimensional image;

[0098] The formula for defining the pooling operation is: CH = MAXP(cs0) + xbMAXP(cs1); where CH is the output result of the pooling layer, MAXP() represents the maximum pooling operation; cs0 and cs1 represent the real and imaginary parts of the local feature map output by the convolutional layer connected to the corresponding pooling layer, respectively; xb represents the imaginary unit; xb 2 =-1; one convolutional layer is connected to one pooling layer, and the layer index of the corresponding pooling layer is the same as the convolutional layer it is connected to;

[0099] Define the loss function of the signal feature extraction model In the formula, M is the total number of feature categories, R ym` Represents the predicted value of category m in the prediction vector of the input training sample y; μ ym It indicates the probability that the training sample y belongs to the feature category m after the training sample is subjected to the softmax operation;

[0100] The training data set is divided into N3 training data subsets, and the training data subsets are planned as sequence data according to the chronology of historical data collection time points; AdaGrad is defined as the optimizer to continuously optimize the parameters of the signal feature extraction model during the training process, and the sequence data is input into the signal feature extraction model in batches according to the time step, and the value of the corresponding loss function is recorded. When the value of the loss function of consecutive L batches no longer decreases or changes, the parameters of the signal feature extraction model at this time are saved, and the training of the signal feature extraction model is completed. N3 is a constant.

[0101] It should be further explained that, in the specific implementation process, the process of constructing the key feature clustering tree includes:

[0102] Acquire known handle button functions of the target game controller, and acquire several groups of button pressing signals corresponding to the corresponding handle button functions; wherein the handle button functions include single button functions and combined button functions; based on the historical usage records of the same type of game controllers, acquire several button pressing information corresponding to different handle button functions, wherein the button pressing signals are handle operation signals corresponding to the target game controller when the corresponding handle button functions are realized during the historical usage process;

[0103] Based on the signal preprocessing process, the collected key pressing signals are preprocessed to obtain the signal modal components corresponding to the corresponding key pressing signals, the energy entropy corresponding to the corresponding signal modal components is obtained, and a multidimensional feature vector corresponding to the corresponding key pressing signals is constructed based on the energy entropy;

[0104] Taking a certain handle button function as an example, obtain the multi-dimensional feature vector of the button pressing signal corresponding to the corresponding handle button function, and construct the corresponding feature vector set {xl α}; Among them, xl α represents the αth multidimensional feature vector in the feature vector set; α=1, 2, ..., XL; XL>0 and XL is an integer, XL represents the total number of multidimensional feature vectors, which is determined by the number of corresponding key pressing signals;

[0105] Define the multidimensional feature vector xl α The density function of is: In the formula, xl z1 represents the z1th multidimensional feature vector in the feature vector set, z1∈XL and z1≠α; Where z2 represents the corresponding index of the multidimensional feature, and z2∈XL but z2≠α; || || 2 Indicates the Euclidean distance operation; dens(xl α ) represents the density value corresponding to the αth density function;

[0106] Obtain the density values ​​corresponding to all multidimensional feature vectors in the corresponding feature vector set; at the same time, obtain the Euclidean distances between all multidimensional feature vectors, and then select the centroid based on the density value and the Euclidean distance between the multidimensional feature vectors to obtain ZX initial centroids; wherein, the basis for selecting the centroid is that the density value of the corresponding multidimensional feature vector must be the maximum value, and the Euclidean distance between it and other initial centroids is not less than a preset distance threshold; ZX is a constant;

[0107] Then, obtain the Euclidean distances between other multidimensional feature vectors and the corresponding initial centroids respectively, and divide the corresponding multidimensional feature vectors and the latest initial centroids into the same cluster set based on the Euclidean distances; obtain ZX cluster sets, repeat the corresponding centroid selection process, cluster the obtained ZX cluster sets again, and so on, until the obtained centroids no longer change; obtain several feature cluster sets;

[0108] Then, the density value corresponding to the corresponding multidimensional feature vector set is obtained, and based on the density value, the multidimensional feature vectors in the corresponding feature cluster set are weighted, and then the multidimensional feature vectors in the corresponding feature cluster set are feature-fused based on the assigned feature weights to obtain corresponding cluster features, and then, based on the cluster features, a corresponding feature bifurcation tree is constructed, and the basic framework of the feature bifurcation tree is handle button function-button pressing signal-cluster feature;

[0109] The feature bifurcation graphs corresponding to all the handle button functions are counted and merged into trees to obtain the corresponding button feature clustering tree.

[0110] It should be further explained that, in a specific implementation process, the process of performing command matching based on the obtained key feature information, and performing key switching and virtual key mapping based on the command matching result includes:

[0111] Based on the output result of the model, the key feature information corresponding to the corresponding handle operation signal is obtained; the corresponding key feature information and the handle operation signal are respectively input into the corresponding cluster feature tree to obtain the signal similarity between the corresponding handle operation signal and the corresponding key pressing signal and the similarity between the key feature information and the corresponding cluster feature; wherein, the process of obtaining the similarity is prior art and will not be elaborated in detail in the present invention;

[0112] Then, based on the key feature clustering tree, signal similarity and key feature information, the handle key function to which the corresponding handle operation signal belongs is obtained, and a corresponding key instruction is generated based thereon;

[0113] Identify the key command obtained, compare it with the pre-stored key command, and determine whether it is a switching command based on the comparison result. If not, transmit the corresponding key command to the game device connected to the target game controller and execute it; if so, obtain the key function mode and LED light information that the corresponding user needs to switch based on the pre-stored key command; and transmit it to the corresponding game controller;

[0114] Furthermore, the game controller performs button mapping on the LED display screen of the corresponding game controller based on the received button function mode; at the same time, the light color is adjusted based on the LED light information, wherein different LED light colors correspond to different button modes, and there are partial deviations in the button functions of the controllers under different button modes and the display positions in the LED display screen;

[0115] It should be further explained that the key function mode includes information such as the key function of the virtual key and the mapping position in the corresponding LED display screen; wherein, the key function mode includes a variety of default function modes set when the game controller leaves the factory and personalized function modes involved by the user based on personal habits.

[0116] The present invention collects data on the button pressing process of the game controller. The data collection module can accurately collect signals generated by the button operation of the game controller during use, pre-process the collected controller operation signals, and finally fuse them to obtain the signal to be detected. By constructing a signal feature extraction model, the signal to be detected is input into the model for feature extraction to obtain key feature information; command matching is performed based on the obtained key feature information, and key switching and virtual key mapping are performed according to the command matching results; the purpose is to achieve accurate and personalized key switching and virtual key mapping, and improve user experience; and successfully solves the shortcomings of traditional game controllers in signal collection, processing and operation methods, and provides game players with a more accurate, stable, personalized and interactive game controller operation experience.

[0117] Example 2

[0118] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, and a method for switching buttons of an LED display screen of a game controller is provided, including:

[0119] Step 1: Collect the key information during the use of the target game controller to obtain the corresponding controller operation signal;

[0120] Step 2: Preprocess the collected handle operation signal to obtain the corresponding signal to be detected;

[0121] Step 3: Input the obtained signal to be detected into a pre-built signal feature extraction model to obtain corresponding key feature information;

[0122] Step 4: Perform command matching based on the acquired key feature information, and perform key switching and virtual key mapping based on the command matching result.

[0123] Example 3

[0124] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, the LED display screen button switching control system of the game controller according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0125] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided by the present application, such as: switching of the LED display button of the game controller. When the computer program is executed by the central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0126] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0127] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. The LED display button switching control system of the game controller is characterized by: include: The data acquisition module is used to collect the key information of the target game controller during use and obtain the corresponding controller operation signal; The data processing module is used to perform signal preprocessing on the collected handle operation signal to obtain the corresponding signal to be detected; The data analysis module is used to input the obtained signal to be detected into a pre-built signal feature extraction model to obtain corresponding key feature information; The data control module is used to perform command matching based on the acquired key feature information, and perform key switching and virtual key mapping based on the command matching result.

2. The LED display screen button switching control system of the game controller according to claim 1, characterized in that: The process of collecting signal of key information during the use of the target game controller and obtaining the corresponding controller operation signal includes: The data acquisition module is composed of a number of sensor terminals, which are used to collect data on signals generated by operations such as button pressing, joystick movement, and display screen touch sliding during the use of the corresponding game controller, and verify whether the collected signal is a false touch signal. If it is not a false touch signal, the collected signal is converted to obtain a corresponding controller operation signal; if it is a false touch signal, the collected signal is discarded.

3. The LED display screen button switching control system of the game controller according to claim 2, characterized in that: The process of preprocessing the collected handle operation signal to obtain the corresponding signal to be detected includes: Constructing standard white noise, adding the corresponding standard white noise to the obtained handle operation signal, and performing signal decomposition on the handle operation signal after adding the standard white noise to obtain signal components at several different scales; and constructing a corresponding scale signal sequence based on the components; Perform secondary decomposition on the signal components in the corresponding scale signal sequence signal to obtain the corresponding first-order residual component and first-order IMF component; Add the constructed standard white noise to the obtained first-order residual component, continue to decompose the signal, and so on, until the pre-set decomposition stop condition is met; obtain a number of signal modal components from high frequency to low frequency; The obtained signal modal component is correlated with the original handle operation signal to obtain the corresponding signal correlation coefficient; A correlation threshold is set, and the obtained signal correlation coefficient is compared with the correlation threshold. If the signal correlation coefficient is less than the correlation threshold, it indicates that the correlation between the corresponding signal modal component and the original handle signal is low, and the corresponding signal modal component is marked as an invalid component; if the signal correlation coefficient is not less than the correlation threshold, the corresponding signal modal component is marked as a component to be processed; Read the obtained components to be processed, and perform wavelet packet decomposition on them to obtain corresponding detail sub-components and high-frequency sub-components; Adaptively optimizing the signal of the obtained detail sub-components to obtain corresponding pure detail signals; Obtain the wavelet packet coefficients corresponding to the high-frequency sub-components at different scales in the corresponding wavelet packet decomposition process; Construct a time window with a time length of N, and add the corresponding time window to the wavelet packet coefficient corresponding to the corresponding high-frequency sub-component; and obtain the wavelet average value of the wavelet packet coefficient in the corresponding time window; Get the wavelet average values ​​corresponding to all time windows in the corresponding high-frequency sub-component, sort them, select the wavelet packets corresponding to the n time windows at the end of the ranking and calculate the average value, and mark the obtained average value as the wavelet basis threshold; n is a fixed constant; Based on the wavelet base threshold, threshold processing is performed on the wavelet packet coefficients corresponding to the corresponding high-frequency sub-component; the high-frequency sub-component after the threshold processing and the corresponding pure detail sub-component are subjected to wavelet reconstruction to obtain the corresponding enhanced modal component; The enhanced modal component and the invalid component are signal-fused to obtain a corresponding signal to be detected; the signal to be detected refers to an enhanced signal of the corresponding handle operation signal.

4. The LED display screen button switching control system of the game controller according to claim 3, characterized in that: The formula for the quadratic decomposition is: Where r1 represents the first-order residual component obtained, x(t) represents the original signal component; IMF1 represents the first-order IMF component obtained; E1(w 1 ) represents the signal component corresponding to the handle operation signal with standard white noise added, and x(t) represents the signal component corresponding to the original handle operation signal; In the formula, β0 represents the signal-to-noise ratio; σ() represents the mathematical expectation operation; δ0 is the preset signal amplitude; The corresponding adaptive signal optimization processing formula is: Where T(k) represents the pure detail signal obtained after adaptive signal optimization processing; c1 and c2 represent the scaling factor and suppression factor respectively; Z k A represents the actual observed value corresponding to the corresponding detail sub-component at time k; k|k-1 represents the observed value predicted based on the observed value at time k-1; P k represents the Kalman gain coefficient corresponding to time k; tanh() represents a pre-selected nonlinear activation function; k represents the time index; The formula for threshold processing is: In the formula, d` (j，a) It represents the wavelet packet coefficient corresponding to the jth high-frequency subcomponent in the ath decomposition layer after threshold processing in the wavelet packet decomposition process; d (j，a) The wavelet packet coefficient corresponding to the jth high-frequency subcomponent in the ath decomposition layer during the wavelet packet decomposition process; (j，a) is the wavelet basis threshold; sgn() is the sign function; CD represents the scale factor of wavelet packet decomposition; a and j are natural numbers, and a and j represent the number of decomposition layers and the index of the high-frequency sub-component in the corresponding decomposition layer, respectively.

5. The LED display screen button switching control system of the game controller according to claim 3, characterized in that: The obtained signal to be detected is read, and the obtained signal to be detected is input into a pre-constructed signal characteristic model to obtain a corresponding model output result, wherein the model output result is key characteristic information corresponding to the corresponding signal to be detected.

6. The LED display screen button switching control system of the game controller according to claim 5, characterized in that: The construction process of the signal feature model includes: A key feature clustering tree is constructed, and a corresponding training data set is constructed based on the constructed key feature clustering tree; the training data set is composed of a plurality of training samples, and the training samples include key pressing signals and their corresponding clustering features. The network framework of the signal feature model is defined as input layer, feature extraction layer, normalization layer and output layer; The input layer is used to receive the training samples in the input training data set and perform data conversion on the corresponding training samples, wherein the data conversion refers to converting the input signal into a two-dimensional image with time correlation; The feature extraction layer is composed of a convolution layer and a pooling layer; the convolution layer is used to extract local features of the input two-dimensional image to obtain a corresponding local feature map; the pooling layer is used to perform a pooling operation on the obtained local feature map; The normalization layer is composed of a separable convolution layer and a batch normalization layer. The separable convolution layer and the separable convolution layer are used to extract the temporal features in the local feature map and perform feature fusion. The batch normalization layer is used to normalize the mean and variance of the input training samples. The output layer is used to receive the feature information output by the normalization layer, flatten the information, and connect it to the softmax classifier through the built-in fully connected layer to perform final feature mapping and classification; The training data set is divided into N3 training data subsets, and the training data subsets are planned as sequence data according to the chronology of historical data collection time points; AdaGrad is defined as the optimizer to continuously optimize the parameters of the signal feature extraction model during the training process, and the sequence data is input into the signal feature extraction model in batches according to the time step, and the value of the corresponding loss function is recorded. When the value of the loss function of consecutive L batches no longer decreases or changes, the parameters of the signal feature extraction model at this time are saved, and the training of the signal feature extraction model is completed; N3 is a constant.

7. The LED display screen button switching control system of the game controller according to claim 6, characterized in that: The convolution kernel size in the convolution layer is defined as 3×3, and a convolution operation is performed on the corresponding two-dimensional image based on the convolution kernel to obtain a corresponding local feature map. The formula of the convolution operation is: In the formula, represents the local feature map output by the u-th convolution kernel in the v-th layer; f() represents the activation function; v represents the index of the convolution layer; u and v are natural numbers; Represents the input feature map corresponding to the previous layer of the v-th convolutional layer; Represents the weight parameter between the u-th convolution kernel and the h-th input feature map in the v-th layer; represents the bias term; C u Represents the feature area of ​​the local feature map corresponding to the u-th convolution kernel in the corresponding two-dimensional image; The formula for defining the pooling operation is: CH = MAXP(cs0) + xbMAXP(cs1); where CH is the output result of the pooling layer, MAXP() represents the maximum pooling operation; cs0 and cs1 represent the real and imaginary parts of the local feature map output by the convolutional layer connected to the corresponding pooling layer, respectively; xb represents the imaginary unit; xb 2 = -1; Define the loss function of the signal feature extraction model In the formula, M is the total number of feature categories, R ym` Represents the predicted value of category m in the prediction vector of the input training sample y; μ ym It represents the probability that the training sample y belongs to the feature category m after the training sample undergoes the softmax operation.

8. The LED display screen button switching control system of the game controller according to claim 6, characterized in that: The process of performing command matching based on the obtained key feature information, and performing key switching and virtual key mapping based on the command matching result includes: Based on the output result of the model, key feature information corresponding to the corresponding handle operation signal is obtained; the corresponding key feature information and the handle operation signal are respectively input into the corresponding cluster feature tree to obtain the signal similarity between the corresponding handle operation signal and the corresponding key pressing signal and the feature similarity between the key feature information and the corresponding cluster feature; Based on the key feature clustering tree, signal similarity and key feature information, the handle key function to which the corresponding handle operation signal belongs is obtained, and a corresponding key instruction is generated based thereon; Identify the key command obtained, compare it with the pre-stored key command, and determine whether it is a switching command based on the comparison result. If not, transmit the corresponding key command to the game device connected to the target game controller and execute it; if so, obtain the key function mode and LED light information that the corresponding user needs to switch based on the pre-stored key command; and transmit it to the corresponding game controller; The game controller performs button mapping on the LED display screen of the corresponding game controller based on the received button function mode; at the same time, the light color is adjusted based on the LED light information.

9. A method for switching buttons of an LED display screen of a game controller, which is implemented based on the control system for switching buttons of an LED display screen of a game controller according to any one of claims 1 to 8, characterized in that: include: Step 1: Collect the key information during the use of the target game controller to obtain the corresponding controller operation signal; Step 2: Preprocess the collected handle operation signal to obtain the corresponding signal to be detected; Step 3: Input the obtained signal to be detected into a pre-built signal feature extraction model to obtain corresponding key feature information; Step 4: Perform command matching based on the acquired key feature information, and perform key switching and virtual key mapping based on the command matching result.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the method for switching the LED display screen buttons of the game controller according to claim 9 is implemented.

Images