Sliding touch behavior recognition method, device and equipment based on television remote controller and storage medium

Abstract

The invention discloses a sliding touch behavior recognition method, device and equipment based on a television remote controller and a storage medium, and the method comprises the steps: judging whether a user mistaken touch behavior exists or not according to obtained touch coordinate data and a touch event state; if the behavior is a normal sliding touch behavior of the user, performing Kalman filtering processing on the touch coordinate data to remove noise influence possibly existing during data acquisition so as to achieve a track smoothing effect; and judging the sliding touch intention of the user, and identifying the sliding touch moving direction of the user. According to the method, the problems of mistaken touch behaviors of a television remote controller, noise errors of a sliding touch track and difficulty in recognizing a complex sliding touch moving direction are avoided; the pre-trained self-organizing mapping network is adopted to recognize the complex sliding touch moving direction of the user, and the recognition precision of the sliding touch behavior of the television remote controller is improved.

Description

Technical Field

[0001] The present invention relates to the field of home appliance control technology, and in particular to a method, device, equipment and storage medium for identifying sliding touch behavior based on a television remote controller. Background Art

[0002] The scale of the smart remote control industry continues to expand. In recent years, the annual compound growth rate of my country's smart remote control output has reached 10%, showing the continuous growth of market demand. As the hub of smart home, the smart remote control can control other smart home appliances and realize the all-round intelligence of home entertainment and life. As the control device of the home entertainment center, the design and function of the TV touch remote control are constantly evolving with the development of smart home technology.

[0003] The touch screen of a TV remote control is usually small and highly sensitive, and users can easily make accidental touch operations during use. In addition, during the sliding touch process, the collected touch data may have certain deviations due to electronic noise and sensor errors, which is reflected in the uneven sliding touch trajectory. These situations may cause incorrect control feedback from the TV, resulting in a poor user experience of controlling the TV device; secondly, some sliding touch directions and control functions with complex trajectories, such as two-finger sliding and gesture control, usually require a large amount of information data to be processed. Using logical judgment or classification to solve this problem is inefficient and has weak scalability, making it difficult to apply to various actual situations to provide users with an excellent device usage experience. Summary of the invention

[0004] In order to solve the above problems, this patent proposes a sliding touch behavior recognition method, device, equipment and storage medium based on a TV remote control. First, it determines whether the user's sliding line is accidentally touched, screens out the accidental touch, and performs Kalman filtering on the sliding touch trajectory of the user's normal touch behavior to eliminate noise and smooth the trajectory curve; finally, it combines the pre-trained self-organizing map network (Self-organizing Maps, SOM) to identify the user's complex sliding touch movement direction, provide more accurate operation feedback, and enhance the user's TV control experience.

[0005] The present invention solves the above problems through the following technical solutions:

[0006] The invention relates to a sliding touch behavior recognition method based on a TV remote control, comprising:

[0007] Step 1: Get the usage status of the touchpad on the TV remote control in real time;

[0008] Step 2: When the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained;

[0009] Step 3: Determine whether the touch is an accidental touch based on the acquired touch coordinate data and touch event status; if it is determined to be an accidental touch, return to step 1;

[0010] Step 4: If the user's sliding touch behavior is normal, Kalman filtering is used to process the touch coordinate data to remove the noise that may exist during data collection and achieve a smooth trajectory effect;

[0011] Step 5: Determine the user's sliding touch intention and identify the user's sliding touch movement direction.

[0012] A further technical solution is that the method for determining whether the user's touch behavior is an accidental touch comprises the following steps:

[0013] Step 3.1: When the touch event is detected as a pressed state, the monitoring end starts to record the initial touch coordinates (x_start, y_start) at this time;

[0014] Step 3.2: When the touch event is detected as a sliding state, the sliding touch coordinates (x t ,y t ) until the touch event is lifted, and record the final touch coordinates (x_final, y_final);

[0015] Step 3.3: If the initial and final touch coordinates are at the same coordinates as the sliding touch coordinates, and the sliding state and the lifting state do not change the touch coordinates, then the touch event is determined to be an error touch behavior that does not generate a sliding touch track.

[0016] A further technical solution is that the method for determining whether the user's touch behavior is an accidental touch further includes:

[0017] Step 3.4: If the sliding state changes the touch coordinates in a single touch event, that is, a sliding touch track is generated, it is regarded as a user sliding operation, and the sliding touch coordinates (x t ,y t ) and the distance d between the starting point of the touch track, that is, the initial touch coordinates (x_start, y_start), and the distance d between the starting point of the sliding touch, that is, the initial touch coordinates (x_start, y_start) and the end point, that is, the final touch coordinates (x_final, y_final) c ;

[0018] Step 3.5: If the distance d is less than the threshold τ1 until the lifting state, the distance d c If both are less than the threshold τ2, the current sliding touch event is considered to be an accidental sliding touch by the user.

[0019] A further technical solution is that the distance d and the distance d c The calculation formula is:

[0020]

[0021] A further technical solution is that in step 4, the specific steps include:

[0022] Step 4.1: Collect complete sliding touch data, including the initial touch coordinates (x_start, y_start) and final touch coordinates (x_final, y_final) when the touch event is pressed and lifted, as well as the sliding touch coordinates (x_final, y_final) collected at time intervals Δt during the sliding process. t ,y t );

[0023] Step 4.2: Initialize the state vector and covariance matrix of the Kalman filter. The state vector contains the coordinates of each touch point in the touch trajectory and the vertical and horizontal velocities. The covariance matrix is ​​used to represent the uncertainty of the state estimation.

[0024] Step 4.3: Use the Kalman filter prediction equation to predict the state estimate of the touch point at the next moment t+Δt, combined with the actual collected sliding touch coordinates (x t+Δt ,y t+Δt ), the update equation compares the predicted value with the actual collected value, and uses the update equation to update the state estimation and covariance matrix, completely predicting and updating the touch trajectory to obtain a smoothed sliding touch trajectory.

[0025] A further technical solution is that in step 5, the user's sliding touch intention is determined and the user's sliding touch moving direction is identified. The specific method includes:

[0026] Step 5.1: Collect multiple sliding touch tracks, including the sliding touch coordinates (x t ,y t ), calculated tangential sliding rate Slide touch direction vector And normalize all feature values, save the sliding trajectory direction label, and construct the sliding touch data set;

[0027] Step 5.2: Use the self-organizing map network SOM as a classifier of the sliding touch movement direction, use the constructed sliding touch data set to pre-train the self-organizing map network SOM model classification, and obtain an optimized sliding trajectory classifier;

[0028] Step 5.3: Input the touch coordinates of each point of the sliding touch trajectory collected in real time by the user and the calculated sliding speed into the optimized sliding trajectory classifier to identify the moving direction of the user's sliding touch.

[0029] A further technical solution is that in step 5, the tangential sliding rate of each touch point is Slide touch direction vector The calculation formula is as follows:

[0030]

[0031] In addition, the present invention also solves the above-mentioned problem through the following technical solutions:

[0032] A sliding touch behavior recognition device based on a TV remote control includes a memory and a processor, wherein:

[0033] The memory is used to store computer programs;

[0034] The processor is used to read the computer program in the memory and execute the steps of the sliding touch behavior recognition method as described above.

[0035] And another improvement of the present invention:

[0036] A computer-readable storage medium stores a readable computer program, which, when executed by a processor, implements the steps of the sliding touch behavior recognition method as described above.

[0037] Another improvement of the present invention:

[0038] A sliding touch behavior recognition device based on a TV remote control comprises:

[0039] The detection module is used to obtain the usage status of the touch panel on the TV remote control in real time; and when the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained;

[0040] A judgment processing module is used to judge whether it is an accidental touch behavior of the user according to the touch coordinate data and touch event status obtained by the detection module; and when the judgment result is a normal sliding touch behavior of the user, the touch coordinate data is processed by Kalman filtering to remove the noise influence existing in the data collection and achieve a trajectory smoothing effect;

[0041] The recognition module is used to recognize the user's sliding touch intention and the user's sliding touch movement direction.

[0042] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0043] The recognition method of the present invention avoids the problems of false touches of the TV remote control, noise errors in the sliding touch trajectory, and difficulty in identifying the direction of complex sliding touch movements; a pre-trained self-organizing map network (Self-organizing Maps, SOM) is used to identify the user's complex sliding touch movement direction, thereby improving the recognition accuracy of the sliding touch behavior of the TV remote control, allowing the user's sliding touch behavior feedback to be accurate, and improving the overall user experience of the TV device. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The figure is a flow chart of a method for identifying sliding touch behavior based on a TV remote controller according to the present invention.

[0045] Figure 2 The present invention is a structural block diagram of a sliding touch behavior recognition device based on a TV remote controller. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below in conjunction with the drawings in the preferred embodiments of the present application. In the drawings, the same or similar reference numerals throughout represent the same or similar parts or parts with the same or similar functions. The described embodiments are part of the embodiments of the present application, not all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limitations on the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0047] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0048] In the description of the present application, it should be noted that, unless expressly provided and limited otherwise, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or display comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or displays.

[0049] Example 1

[0050] Combined with Figure 1 As shown, the sliding touch behavior recognition method based on the TV remote control includes:

[0051] Step 1: Get the usage status of the touchpad on the TV remote control in real time.

[0052] When the touch is not pressed, the touch event information collection is not triggered, and the touch panel does not return the monitoring end usage status flag for the time being.

[0053] In this embodiment, the use status of the touch pad on the TV remote control is obtained in real time. The use status flag of the touch pad on the remote control is 0 by default. When it is detected that the user is using the touch pad, the use status flag is changed to 1.

[0054] Step 2: When the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained.

[0055] When the touch state is pressed, the touchpad immediately returns the touch point coordinates and the use status mark "0" to the monitoring end, that is, the touch starting point information; during the sliding process, the touch point coordinates and the use status mark "1" are collected at time intervals Δt; when the touch state is lifted, the touch point coordinates and the use status mark "2" are immediately returned to the monitoring end, that is, the touch end point information; a complete touch event and its related touch information are collected.

[0056] Specifically, in this embodiment, when the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained, that is, when the use status flag of the touchpad on the TV remote control is "1", the touch coordinates and touch event status are collected at a fixed frequency at a time interval Δt. A complete touch event includes three states: press, slide, and lift, and the corresponding flags are "0", "1", and "2" respectively. Among them, the press and lift states will only appear once in a complete touch behavior, which are the first moment when the finger touches the touchpad and the first moment when the finger leaves the touchpad. Triggering the press and lift states will immediately record the current touch information.

[0057] Step 3: Determine whether it is an accidental touch by the user based on the acquired touch coordinate data and touch event status;

[0058] First, determine whether the touch point coordinates of the collected data have changed. If not, it is considered that this touch event is an accidental touch behavior that does not generate a sliding touch trajectory. If the touch point coordinates of the collected data have changed, that is, a sliding touch trajectory is generated, then calculate the distance d between each touch point and the starting point of the touch trajectory. If they are all less than the threshold h, it is considered that this sliding touch event is an accidental touch behavior of the user, otherwise it is considered to be a normal sliding touch behavior of the user.

[0059] In this embodiment, the method for determining whether the user's touch behavior is an accidental touch includes the following steps:

[0060] Step 3.1: When the touch event is detected as being in the pressed state, that is, when the touch event flag is 0, the monitoring end starts to record the initial touch coordinates (x_start, y_start) at this time;

[0061] Step 3.2: When the touch event is detected as a sliding state, the sliding touch coordinates (x t ,y t ) until the touch event is lifted, and record the final touch coordinates (x_final, y_final);

[0062] Step 3.3: If the initial and final touch coordinates are at the same coordinates as the sliding touch coordinates, and the sliding state and the lifting state do not change the touch coordinates, then:

[0063] (x t ,y t )=(x_start,y_start);

[0064] (x_final,y_final)=(x_start,y_start);

[0065] The touch event is determined to be an accidental touch behavior that does not generate a sliding touch track.

[0066] Step 3.4: If the sliding state changes the touch coordinates in a single touch event, that is, a sliding touch track is generated, it is regarded as a user sliding operation, and the sliding touch coordinates (x t ,y t ) and the distance d between the starting point of the touch track, that is, the initial touch coordinates (x_start, y_start), and the distance d between the starting point of the sliding touch, that is, the initial touch coordinates (x_start, y_start) and the end point, that is, the final touch coordinates (x_final, y_final) c ;

[0067] Furthermore, the distance d and the distance d c The calculation formula is:

[0068]

[0069] Step 3.5: Set distance d and distance d c The thresholds are τ1 and τ2 respectively; if the distance d is less than the threshold τ1 until the lifting state, the distance d c If both are less than the threshold τ2, the current sliding touch event is considered to be an accidental sliding touch by the user, and the process returns to step 1; otherwise, the process goes to step 4;

[0070] Step 4: If the user's sliding touch behavior is normal, Kalman filtering is used to process the touch coordinate data to remove the noise that may exist during data collection and achieve a smooth trajectory effect;

[0071] The touch data of the user's normal sliding touch behavior is processed by Kalman filtering, the state vector and covariance matrix of the Kalman filter are initialized, and the state estimation value of the touch point at the next moment t+Δt is predicted using the Kalman filter prediction equation, combined with the actual collected touch coordinates (x t+Δt ,y t+Δt ) and the current touch coordinates (x t ,y t ), update equations to compare predicted values ​​with actual collected values, use Kalman update equations to update state estimation and covariance matrix, and completely predict and update the touch trajectory to obtain a smoothed sliding touch trajectory. Remove the influence of noise errors that may exist during data collection to make the sliding touch trajectory smooth.

[0072] In this embodiment, step 4 specifically includes the following steps:

[0073] Step 4.1: Collect complete sliding touch data, including the initial touch coordinates (x_start, y_start) and final touch coordinates (x_final, y_final) when the touch event is pressed and lifted, as well as the sliding touch coordinates (x_final, y_final) collected at time intervals Δt during the sliding process. t ,y t );

[0074] Step 4.2: Initialize the state vector and covariance matrix of the Kalman filter. The state vector contains the coordinates of each touch point in the touch trajectory and the vertical and horizontal velocities. The covariance matrix is ​​used to represent the uncertainty of the state estimation.

[0075] Step 4.3: Use the Kalman filter prediction equation to predict the state estimate of the touch point at the next moment t+Δt, combined with the actual collected sliding touch coordinates (x t+Δt ,y t+Δt ), the update equation compares the predicted value with the actual collected value, and uses the update equation to update the state estimation and covariance matrix, completely predicting and updating the touch trajectory to obtain a smoothed sliding touch trajectory.

[0076] Step 5: Determine the user's sliding touch intention and identify the user's sliding touch movement direction.

[0077] Before completing the sliding direction classification of the user's sliding touch trajectory data, multiple sliding touch trajectories are collected first, including vertical sliding, horizontal sliding, oblique sliding and other sliding trajectories, and the coordinates of each touch point; according to the time interval Δt and touch coordinates (x t ,y t ) Calculate the tangential sliding rate of each touch point The coordinates of each touch point in the complete touch process (xt ,y t ) and the touch point coordinates at the next moment (x t+Δt ,y t+Δt ) can calculate the sliding touch direction vector of each touch point Each sliding touch trajectory data contains the trajectory touch point coordinates, touch point tangential sliding rate, touch point sliding touch direction vector and real sliding touch direction label, constructing the sliding touch data set. The value of each sliding touch data is normalized to make the dimension of the input model of all data consistent. Initialize the SOM weight, use Euclidean distance calculation to quantify the similarity between the input sample and the neurons in the network, and combine the real sliding touch direction label of the trajectory data to perform training and tuning in a supervised learning mode to obtain the optimized sliding trajectory classifier.

[0078] The above user sliding touch trajectory data is used to calculate the tangential sliding rate (x t ,y t ) and the sliding touch direction vector Then, data processing is performed according to the input dimension of the pre-trained sliding trajectory classifier. If the number of touch points in the user's touch trajectory exceeds the maximum number of collected touch points n and the dimension is too long, the sampling is truncated. If the number of touch points is small and the dimension is too small, 0 is used to fill the dimension at the end. The processed user sliding touch trajectory data is input into the classifier to obtain the classification result of the user sliding touch movement direction.

[0079] In this embodiment, the user's sliding touch intention is determined, and the user's sliding touch movement direction is identified. The specific method includes the following steps:

[0080] Step 5.1: Collect multiple sliding touch tracks, including vertical sliding, horizontal sliding, and diagonal sliding, etc. The sliding touch coordinates (x t ,y t ), calculated tangential sliding rate Slide touch direction vector Special gesture tracks can be expanded.

[0081] Tangential sliding rate of each touch point Slide touch direction vector The calculation formula is as follows:

[0082]

[0083] The length and direction of each sliding touch trajectory data are different, the trajectory curve is smooth, and all feature values ​​are normalized. By setting the maximum detectable sliding touch trajectory distance, the maximum number of collected touch points n can be calculated. In order to ensure the consistency of the feature dimensions of the collected sliding touch trajectory data, if the number of touch points in all collected trajectories is less than n, the touch point coordinates, tangential sliding rate, and sliding touch direction vector after the final touch point lifted are padded with 0; if the number of touch points in the collected trajectory exceeds n, the trajectory of n touch points is sampled and intercepted from the starting point. Each sliding touch trajectory data contains the trajectory touch point coordinates, touch point tangential sliding rate, touch point sliding touch direction vector, and real sliding touch direction label to construct a sliding touch data set.

[0084] Step 5.2: Use the self-organizing map network SOM as a classifier for the sliding touch movement direction, use the constructed sliding touch data set to pre-train the self-organizing map network SOM model classification, and optimize the parameters according to the classification effect to obtain the optimized sliding trajectory classifier;

[0085] Step 5.3: The touch coordinates of each point in the sliding touch track collected by the user in real time and the calculated tangential sliding rate and the sliding touch direction vector Data preprocessing is performed according to the input dimension of the pre-trained sliding trajectory classifier, and the processed user sliding touch trajectory data is input into the optimized sliding trajectory classifier to identify the user sliding touch movement direction result.

[0086] Example 2

[0087] Refer to the attached Figure 2 , a sliding touch behavior recognition device based on a TV remote control, comprising:

[0088] The detection module is used to obtain the usage status of the touch panel on the TV remote control in real time; and when the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained;

[0089] A judgment processing module is used to judge whether it is an accidental touch behavior of the user according to the touch coordinate data and touch event status obtained by the detection module; and when the judgment result is a normal sliding touch behavior of the user, the touch coordinate data is processed by Kalman filtering to remove the noise influence existing in the data collection and achieve a trajectory smoothing effect;

[0090] The recognition module is used to recognize the user's sliding touch intention and the user's sliding touch movement direction.

[0091] Example 3

[0092] A sliding touch behavior recognition device based on a TV remote controller includes a memory and a processor, wherein:

[0093] The memory is used to store computer programs;

[0094] The processor is used to read the computer program in the memory and execute the steps of the sliding touch behavior recognition method as described in Example 1.

[0095] In an optional embodiment, a computer-readable storage medium is further provided, on which a readable computer program is stored. When the program is executed by a processor, the steps of the sliding touch behavior recognition method described in Example 1 are implemented.

[0096] Although the present invention is described herein with reference to the illustrative embodiments of the present invention, the above embodiments are only preferred embodiments of the present invention, and the embodiments of the present invention are not limited to the above embodiments. It should be understood that those skilled in the art can design many other modifications and embodiments, which will fall within the scope and spirit of the principles disclosed in this application.

Claims

1. A sliding touch behavior recognition method based on a TV remote control, characterized in that: include: Step 1: Get the usage status of the touchpad on the TV remote control in real time; Step 2: When the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained; Step 3: judging whether it is an accidental touch according to the acquired touch coordinate data and touch event status, and returning to step 1 if it is an accidental touch; Step 4: If the user's sliding touch behavior is normal, Kalman filtering is used to process the touch coordinate data to remove the noise that may exist during data collection and achieve a smooth trajectory effect; Step 5: Determine the user's sliding touch intention and identify the user's sliding touch movement direction.

2. The identification method according to claim 1, characterized in that: The method for determining whether the user's touch behavior is an accidental touch includes the following steps: Step 3.1: When the touch event is detected as a pressed state, the monitoring end starts to record the initial touch coordinates (x_start, y_start) at this time; Step 3.2: When the touch event is detected as a sliding state, the sliding touch coordinates (x t ,y t ) until the touch event is lifted, and record the final touch coordinates (x_final, y_final); Step 3.3: If the initial and final touch coordinates are at the same coordinates as the sliding touch coordinates, and the sliding state and the lifting state do not change the touch coordinates, then the touch event is determined to be an error touch behavior that does not generate a sliding touch track.

3. The identification method according to claim 2, characterized in that: The method for determining whether the user's touch behavior is an accidental touch also includes: Step 3.4: If the sliding state changes the touch coordinates in a single touch event, that is, a sliding touch track is generated, it is regarded as a user sliding operation, and the sliding touch coordinates (x t ,y t ) and the distance d between the starting point of the touch track, that is, the initial touch coordinates (x_start, y_start), and the distance d between the starting point of the sliding touch, that is, the initial touch coordinates (x_start, y_start) and the end point, that is, the final touch coordinates (x_final, y_final) c ; Step 3.5: If the distance d is less than the threshold τ1 until the lifting state, the distance d c If both are less than the threshold τ2, the current sliding touch event is considered to be an accidental sliding touch by the user.

4. The identification method according to claim 3, characterized in that: The distance d and the distance d c The calculation formula is:

5. The identification method according to claim 1, characterized in that: In step 4, the specific steps include: Step 4.1: Collect complete sliding touch data, including the initial touch coordinates (x_start, y_start) and final touch coordinates (x_final, y_final) when the touch event is pressed and lifted, as well as the sliding touch coordinates (x_final, y_final) collected at time intervals Δt during the sliding process. t ,y t ); Step 4.2: Initialize the state vector and covariance matrix of the Kalman filter. The state vector contains the coordinates of each touch point in the touch trajectory and the vertical and horizontal velocities. The covariance matrix is ​​used to represent the uncertainty of the state estimation. Step 4.3: Use the Kalman filter prediction equation to predict the state estimate of the touch point at the next moment t+Δt, combined with the actual collected sliding touch coordinates (x t+Δt ,y t+Δt ), the update equation compares the predicted value with the actual collected value, and uses the update equation to update the state estimation and covariance matrix, completely predicting and updating the touch trajectory to obtain a smoothed sliding touch trajectory.

6. The identification method according to claim 1, characterized in that: In step 5, the user's sliding touch intention is determined, and the user's sliding touch movement direction is identified. The specific method includes: Step 5.1: Collect multiple sliding touch tracks, including the sliding touch coordinates (x t ,y t ), calculated tangential sliding rate Slide touch direction vector And normalize all feature values, save the sliding trajectory direction label, and construct the sliding touch data set; Step 5.2: Use the self-organizing map network SOM as a classifier of the sliding touch movement direction, use the constructed sliding touch data set to pre-train the self-organizing map network SOM model classification, and obtain an optimized sliding trajectory classifier; Step 5.3: Input the touch coordinates of each point of the sliding touch trajectory collected in real time by the user and the calculated sliding speed into the optimized sliding trajectory classifier to identify the moving direction of the user's sliding touch.

7. The identification method according to claim 6, characterized in that: In step 5, the tangential sliding rate of each touch point Slide touch direction vector The calculation formula is as follows:

8. A sliding touch behavior recognition device based on a TV remote control, characterized in that: comprising a memory and a processor, wherein: The memory is used to store computer programs; The processor is used to read the computer program in the memory and execute the steps of the sliding touch behavior recognition method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: A readable computer program is stored thereon, and when the program is executed by a processor, the steps of the sliding touch behavior recognition method as described in any one of claims 1 to 7 are implemented.

10. A sliding touch behavior recognition device based on a TV remote control, characterized in that: include: A detection module is used to obtain the usage status of the touch panel on the TV remote control in real time; When the user's sliding touch behavior is detected, the touch coordinates and touch event status are obtained; A judgment processing module is used to judge whether it is an accidental touch behavior of the user according to the touch coordinate data and touch event status obtained by the detection module; and when the judgment result is a normal sliding touch behavior of the user, the touch coordinate data is processed by Kalman filtering to remove the noise influence existing in the data collection and achieve a trajectory smoothing effect; The recognition module is used to recognize the user's sliding touch intention and the user's sliding touch movement direction.

Images