Touch action recognition method and device, storage medium and electronic equipment

Abstract

The invention provides a touch action recognition method and device, a storage medium and electronic equipment, and belongs to the technical field of human-computer interaction. The method comprises the following steps: acquiring multi-frame queue data; calculating an effective contact number of the current frame data in the multi-frame queue data, and determining an action type to which the touch action of the current frame belongs based on the effective contact number; if the action type is a point contact type, calculating a corresponding dispersion degree according to the current frame data and the previous frame data, and determining whether the touch action of the current frame is one of a light touch action or a sliding action based on the dispersion degree; if the action type is a beating type or a palm type, calculating the number of continuous frames which are judged to be the same action type in the multi-frame queue data; and determining whether the touch action of the corresponding frame is one of a slow shooting action, a snapshot action, a long press action, a continuous touch action or a single touch action based on the number of the continuous frames. According to the invention, the recognition accuracy of complex gestures can be improved.

Description

Technical Field

[0001] This application relates to the field of human-computer interaction technology, and in particular to a touch action recognition method, device, storage medium and electronic device. Background Technology

[0002] With the widespread adoption of smart terminals, wearable devices, and IoT embedded devices, touch interaction, as an intuitive and convenient human-computer interaction method, has been widely applied in consumer electronics, smart homes, medical devices, and other fields. Users' demands for touch interaction are no longer limited to simple clicks and swipes, but are evolving towards more complex compound gestures (such as long presses, continuous taps, and combinations of light touches and swipes), adaptability to multiple scenarios (such as operation while wearing gloves or in humid environments), and low-power operation.

[0003] Currently, mainstream touch action recognition technologies are mainly divided into two categories: one is based on capacitive touch screen detection technology, which determines the touch position by detecting changes in capacitance during the touch process. However, this technology can only recognize simple clicks and swipes on a two-dimensional plane and cannot effectively distinguish complex gestures that rely on pressure features or timing features, such as light touches and continuous taps. The other is based on acceleration threshold detection technology, which recognizes specific actions (such as tapping) by setting a fixed acceleration threshold. However, this technology still has difficulty distinguishing complex gesture sequences.

[0004] Therefore, how to overcome the limitations of existing technologies and provide a touch detection technology that can accurately recognize a variety of complex gestures has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] The purpose of this application is to provide a touch action recognition method, device, storage medium, and electronic device to solve at least one of the above-mentioned technical problems.

[0006] In a first aspect, this application provides a touch action recognition method, the method comprising: Acquire multi-frame queue data, wherein each frame of the multi-frame queue data includes the effective channel values ​​of each channel collected by the multi-channel capacitive sensor at a corresponding moment; Using one frame of data from a multi-frame queue as the current frame data, the number of valid touch points in the current frame data is calculated, and the action type of the touch action in the current frame is determined based on the number of valid touch points. The action type includes point touch, slap, and palm touch. If the action type of the current frame is a touch type, calculate the corresponding dispersion based on the data of the current frame and the data of the previous frame, and determine whether the touch action of the current frame is a light touch or a swipe action based on the dispersion. If the action type of the current frame is a slapping type or a palm type, calculate the number of consecutive frames in the multi-frame queue data that are determined to have the same action type. Based on the number of consecutive frames, determine whether the touch action of the corresponding frame is one of the following: slow tap, fast tap, long press, continuous stroking, or single stroking.

[0007] Optionally, calculating the number of valid touch points in the current frame data includes: counting the number of valid channel values ​​in the current frame data to obtain the number of valid touch points.

[0008] Optionally, determining the action type of the touch action in the current frame based on the number of valid touch points includes: If 0 < number of valid touch points in the current frame ≤ first touch point threshold, the action type of the current frame is determined to be a point touch type. If the second touch threshold ≤ the number of valid touches in the current frame ≤ the third touch threshold, the action type of the current frame is determined to be a slapping type; If the number of valid touch points in the current frame is greater than the third touch point threshold, the action type of the current frame is determined to be the palm type.

[0009] Optionally, calculating the corresponding dispersion based on the current frame data and the previous frame data includes: Based on the effective channel value and effective channel position of each channel in each frame of data, calculate the pressure center, total channel value, and distance between each effective channel position and the pressure center of the corresponding frame of data. The dispersion of the corresponding frame is calculated based on the effective channel value, the distance between each effective channel position and the pressure center, and the total channel value.

[0010] Optionally, determining whether the touch action in the current frame is a tap or a swipe based on the dispersion includes: If the dispersion is less than or equal to the dispersion threshold, the center offset between the current frame and the previous frame is calculated based on the pressure center of the current frame and the pressure center of the previous frame. The corresponding channel change value is calculated based on the effective channel values ​​in the current frame and the previous frame that are in the same channel position. If the center offset is greater than or equal to the offset threshold, the touch action in the current frame is determined to be a swipe action; If the channel change value is greater than or equal to the channel change threshold, the touch action in the current frame is determined to be a light touch action.

[0011] Optionally, determining whether the touch action of the corresponding frame is a slow-motion action, a fast-motion action, a long-press action, a continuous stroking action, or a single stroking action based on the number of consecutive frames includes: The duration of the corresponding touch action is determined based on the number of consecutive frames. When the same action type is tapping, if the duration is less than or equal to the first duration threshold, the corresponding touch action is determined to be a fast tapping action; if the first duration threshold is less than the duration and less than or equal to the second duration threshold, the corresponding touch action is determined to be a slow tapping action. When the same action type is the palm type, if the duration is less than or equal to the third duration threshold, the corresponding touch action is determined to be a single stroking action; if the third duration threshold is less than the duration and less than or equal to the fourth duration threshold, the corresponding touch action is determined to be a continuous stroking action; if the duration is greater than the fourth duration threshold, the corresponding touch action is determined to be a long press action.

[0012] Optionally, the method further includes: It responds to the identified touch actions in a matching manner; Get the cooldown duration corresponding to the selected touch action; Ignore the recognition and / or response to touch actions in each frame of data within the cooling time.

[0013] Optionally, before determining the action type of the touch action in the current frame based on the number of valid touch points, the method further includes: If the number of valid touch points in the current frame is 0, the data of the current frame will be considered an invalid frame. The number of invalid frames with consecutive invalid frames in the multi-frame queue data is counted. If the number of invalid frames is greater than or equal to the interruption threshold, it is determined that no valid touch action has occurred. If the number of invalid frames is less than the interruption threshold, the action type of the touch action in the current frame is determined based on the number of valid touch points.

[0014] A second aspect of this application provides a touch action recognition device, the device comprising: The touch action acquisition module is used to acquire multi-frame queue data, wherein each frame of the multi-frame queue data includes the effective channel values ​​of each channel collected by the multi-channel capacitive sensor at a corresponding moment. The action type recognition module is used to take one frame of data in a multi-frame queue as the current frame data, calculate the number of valid touch points of the current frame data, and determine the action type of the touch action of the current frame based on the number of valid touch points. The action type includes point touch type, slapping type, and palm type. The touch type analysis module is used to calculate the corresponding dispersion based on the current frame data and the previous frame data if the action type of the current frame is touch type, and determine whether the touch action of the current frame is a light touch action or a swipe action based on the dispersion. The patting or palm type analysis module is used to calculate the number of consecutive frames in the multi-frame queue data that are determined to be of the same action type if the action type of the current frame is patting or palm type; and to determine whether the touch action of the corresponding frame is one of slow patting action, fast patting action, long press action, continuous stroking action or single stroking action based on the number of consecutive frames.

[0015] In a third aspect, this application provides a computer-readable storage medium storing executable instructions that, when executed by a processor, cause the processor to perform the method described in any embodiment of this application.

[0016] In a fourth aspect, this application provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to perform the method as described in any embodiment of this application.

[0017] The touch action recognition method, device, storage medium, and electronic device in this application acquire effective channel values ​​using a multi-channel capacitive sensor at a preset acquisition frequency. For each moment, the effective channel values ​​of each channel are collected to form a corresponding frame of data. A preset number of consecutively acquired frames are then combined to form a multi-frame queue of data. The number of touch points in this multi-frame queue is then detected to determine the corresponding action type. Based on the determined action type, corresponding detection is performed. If it is a touch type, dispersion detection is performed to identify whether it is a light touch or a swipe action. If it is a tapping or palm strike type, the number of consecutive frames of the same action type is identified. Based on this number, it is determined whether the touch action is a slow tap, fast tap, long press, continuous stroking, or single stroking action. Therefore, this application can accurately distinguish whether the corresponding touch action is one of seven types of touch actions: light touch, swipe, slow tap, fast tap, long press, continuous stroking, and single stroking, thus improving the accuracy of complex gesture recognition. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation on the scope of this application.

[0019] Figure 1 This is a flowchart illustrating a touch action recognition method in one embodiment; Figure 2 This is a partial flowchart of a touch action recognition method in one embodiment; Figure 3 This is a structural block diagram of a touch action recognition device in one embodiment; Figure 4 This is a structural block diagram of a touch action recognition device in another embodiment; Figure 5 This is a schematic diagram of the structure of an electronic device in one embodiment. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] All terms used in this application (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0022] For example, the terms "first," "second," etc., used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0023] For example, the terms "comprising" or "including" used in this application indicate the presence of features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0024] like Figure 1 As shown, this application provides a touch action recognition method, which includes: Step 110: Obtain multi-frame queue data.

[0025] In this embodiment, the electronic device is equipped with a touch device, which can be a touchscreen. The touch device further integrates a multi-channel capacitive sensor, which can acquire the user's touch actions on the touch device. The multi-channel capacitive sensor is a sensor array composed of multiple capacitive sensing units, sensing touch actions by detecting changes in capacitance. For example, the multi-channel capacitive sensor can be a 14-channel capacitive sensor or any other suitable number of channels. Each channel in the multi-channel capacitive sensor acquires a touch action in a corresponding area of ​​the touch device, obtaining a corresponding channel value. This channel value can be a voltage value or a capacitance value; the magnitude of the channel value reflects the force of the touch action, and the location of the capacitive channel reflects the position where the touch action occurred.

[0026] A "frame" represents the data set collected by the sensor at a specific moment, serving as the basic time unit for time-series gesture recognition. The collected data set can be the collection of channel values ​​acquired by each channel of the sensor. Multi-frame queue data is a collection of multiple frames arranged in chronological order, forming a sliding window for time-series analysis. During a sliding window, a preset number of frames can be collected, such as 20 frames or any other suitable number. A multi-frame queue data set is the collection of frames acquired within a sliding window. Each frame in the multi-frame queue data set includes the valid channel values ​​of each channel acquired by the multi-channel capacitive sensor at a corresponding moment. These valid channel values ​​are the values ​​retained after data preprocessing.

[0027] Optionally, each frame in the multi-frame queue data is the frame data acquired within the most recent sliding window. The frame data in the multi-frame queue data is updated in real time, retaining only the most recent number of frames, such as retaining the most recent 20 frames in real time.

[0028] In one embodiment, noise filtering is performed on the original channel values ​​of each channel in each frame of data, and the remaining original channel values ​​are used as the valid channel values ​​for the corresponding channels.

[0029] The raw channel values ​​are unprocessed channel values ​​directly acquired by the multi-channel capacitive sensor (e.g., capacitance data). These raw channel values ​​contain environmental noise (such as electromagnetic interference and sensor drift) and valid touch signals. The electronic device performs preprocessing on the raw channel values, such as noise reduction, to remove environmental noise or channel values ​​resulting from accidental touches, retaining the remaining values ​​as valid channel values.

[0030] Specifically, noise filtering can include numerical filtering and / or filter filtering. A corresponding noise threshold can be set to remove data in the original channel values ​​below the threshold as noise; for example, the noise threshold can be set to 300. Taking the original channel value as capacitance data, its range is typically 0~60000 (unit: capacitance units). For example, channel 3 might collect a channel value of 50 when there is no touch (this data is noise data), and a value of 58000 when touched (this is a valid signal). Optionally, the noise threshold can be pre-calibrated based on experimental results and adaptively determined according to one or more factors such as the environment and usage scenario of the electronic device, and usage habits. For example, when the current environment is identified as dry, the noise threshold is determined to be 250; if the current environment is humid, the noise threshold is adaptively adjusted to 300.

[0031] In addition to noise threshold filtering, a secondary filtering is performed using relevant filtering algorithms, and the channel values ​​retained after the secondary filtering are used as valid channel values.

[0032] Data values ​​below a noise threshold in the original channel values ​​of each frame are treated as noise and removed. The remaining channel values ​​after noise threshold removal are considered potential effective channel values. These potential effective channel values ​​are then subjected to a secondary filtering algorithm, such as a hybrid filtering algorithm or a wavelet transform filtering algorithm. The potential effective channel values ​​after this secondary filtering are then used as the effective channel values ​​in the corresponding frame data. This secondary filtering improves the accuracy of the frame data.

[0033] Step 120: Take one frame of data from the multi-frame queue as the current frame data, calculate the number of valid touch points in the current frame data, and determine the action type of the touch action in the current frame based on the number of valid touch points.

[0034] In this embodiment, the action types include touch type, patting type, and palm type. Specifically, touch type can include light touch and swipe action; patting type can include slow patting and fast patting action; palm type can include long press action, continuous stroking action, and single stroking action.

[0035] The number of valid touch points indicates the number of valid channel values ​​in the corresponding frame of data, that is, how many channel sensing units have collected valid channel values. The number of valid touch points reflects the area affected by the user's touch action, and the channel position corresponding to each valid channel value reflects the location of the touch action. The channel position can be represented by the channel number of the corresponding channel.

[0036] Specifically, based on the numerical range of the number of valid touch points, it can be divided into corresponding action types. This enables the identification of touch action types.

[0037] In one embodiment, calculating the number of valid touch points in the current frame data includes: counting the number of valid channel values ​​in the current frame data to obtain the number of valid touch points.

[0038] For each frame of data, a corresponding structured touch data object can be formed. This structured touch data object contains the timestamp of the corresponding frame, a touch point distribution mapping table, and the number of valid touch points. It can also further include one or more of the following: the timing index of the corresponding frame, the average channel value, and the maximum channel value. The touch point distribution mapping table (touchTable) is represented by a key-value pair data structure. The "key" (sensor_index) represents the physical channel number of the multi-channel capacitive sensor (i.e., the sensor's location identifier); the "value" (pressure_value) represents the valid channel value of the corresponding channel. For example, with a timing index (index) = 1 and a channel value of pressure value, the data structure of an exemplary structured touch data object (TouchClass) is as follows: TouchClass = { index = 1, -- Time-series index (increments from 1) time = 168723456789, -- timestamp touchTable = {-- Touch point distribution mapping table [3] = 58000, -- Channel 3 pressure value [4] = 42000, -- Channel 4 pressure value }, activeCount = 2, -- Number of valid touch points avgPressure = 50000, -- Average pressure value (dynamically calculated) maxPressure = 58000 -- Maximum pressure value (dynamically extracted) } This example shows that the frame data was acquired at time (timestamp) 168723456789, with effective channel values ​​of 58000 for the 3rd channel and 42000 for the 4th channel. The number of effective touch points (activeCount) is 2. The average channel value (avgPressure) of the effective channel values ​​in this frame data is 50000, and the maximum channel value (maxPressure) is 58000.

[0039] Step 130: If the action type of the current frame is a touch type, calculate the corresponding dispersion based on the current frame data and the previous frame data, and determine whether the touch action of the current frame is a tap action or a swipe action based on the dispersion.

[0040] In this embodiment, dispersion detection is performed on point-type touch actions. Dispersion is used to quantify the concentration of touch points (touch positions). The smaller the dispersion, the more concentrated the pressure distribution, and the greater the probability that the corresponding touch action is a light touch or a swipe action.

[0041] By analyzing frame data from two adjacent frames, changes in the user's touch position during a touch process can be identified. This positional change is quantified as dispersion, and based on this dispersion, it can be determined whether a touch action is a tap or a swipe. The dispersion is calculated based on the effective channel values ​​and corresponding channel positions in the frame data of two adjacent frames.

[0042] Step 140: If the action type of the current frame is a slapping type or a palm type, calculate the number of consecutive frames in the multi-frame queue data that are determined to be of the same action type.

[0043] For actions identified as slapping or hand gestures, multi-frame recognition can be performed to detect whether multiple consecutive frames contain the same action type, and the number of such consecutive frames is counted. For example, there might be 5 consecutive slapping frames or 8 consecutive hand gesture frames.

[0044] Step 150: Determine whether the touch action of the corresponding frame is one of the following based on the number of consecutive frames: slow tap, fast tap, long press, continuous stroking, or single stroking.

[0045] Since the durations corresponding to hand gestures and slapping gestures are different, the duration of the corresponding touch action can be calculated based on the number of consecutive frames obtained. Furthermore, if multiple consecutive frames are identified as the same action type, the timestamps of the first and last frames can be extracted, and the duration can be calculated based on the difference between these timestamps.

[0046] Based on the mapping relationship between the duration and the corresponding specific touch action, it is possible to determine which type of touch action corresponds to the multi-frame queue data, namely slow tap, fast tap, long press, continuous stroking, or single stroking.

[0047] The touch action recognition method in this application uses a multi-channel capacitive sensor to collect effective channel values ​​at a preset sampling frequency. For each moment, the effective channel values ​​of each channel are collected to form a corresponding frame of data. A preset number of consecutively collected frames are then formed into a multi-frame queue of data. The number of touch points in this multi-frame queue is then detected to determine the corresponding action type. Based on the determined action type, corresponding detection is performed. If it is a touch type, dispersion detection is performed to identify whether it is a light touch or a swipe action. If it is a tapping or palm strike type, the number of consecutive frames of the same action type is identified. Based on this number, it is determined whether the touch action is a slow tap, fast tap, long press, continuous stroking, or single stroking action. Based on this, this application can accurately distinguish whether the corresponding touch action is a light touch, swipe, slow tap, fast tap, long press, continuous stroking, or single stroking action—a total of seven touch actions—improving the accuracy of complex gesture recognition.

[0048] In one embodiment, determining the action type of the touch action in the current frame based on the number of valid touch points includes: if 0 < number of valid touch points in the current frame ≤ first touch point threshold, the action type of the current frame is determined to be a point touch type; if the second touch point threshold ≤ number of valid touch points in the current frame ≤ third touch point threshold, the action type of the current frame is determined to be a patting type; if the number of valid touch points in the current frame > third touch point threshold, the action type of the current frame is determined to be a palm type.

[0049] In this embodiment, the first contact threshold c1, the second contact threshold c2, and the third contact threshold c3 can be preset appropriate values. These values ​​can be determined based on one or more factors, such as the number of channels in the multi-channel capacitive sensor and the user's usage scenario and habits. Specifically, the first contact threshold c1 ≤ the second contact threshold c2 ≤ the third contact threshold c3. The first contact threshold c1 is a critical value used to distinguish point touch types from other types; the second contact threshold is a critical value used to distinguish the lower limit of the tapping type; and the third contact threshold is a critical value used to distinguish the tapping type from the palm type.

[0050] Touch actions of the "tap" type are generated by small-area contact such as fingertips, with a small number of effective contact points, such as lightly touching the screen with a finger. Based on this, c1=3 can be set. Touch actions of the "slap" type are generated by medium-area contact such as the edge of the palm or knuckles, with a medium number of effective contact points, such as slapping the device with the side of the palm. Touch actions of the "palm" type are generated by large-area contact of the entire palm, with a large number of effective contact points, such as pressing or stroking the surface of the device. Based on this, c2=5 and c3=7 can be set. Understandably, the values ​​of c1, c2, and c3 can also be set to any other suitable values, such as c1=3, c2=6, c3=8, or c1=5, c2=8, c3=10, etc.

[0051] Taking a 14-channel capacitive sensor as an example, c1=3, c2=5, c3=7 can be set. If 0 < the number of effective contact points (≤3), the action type of the corresponding frame is determined to be a point touch type. This action type could be a sliding action or one of the sliding actions, or it could be noise such as a mis-touch. If 5 ≤ the number of effective contact points (≤7), the action type of the corresponding frame is determined to be a tapping type. This action type could be a fast tapping action or a slow tapping action, or it could be noise such as a mis-touch. If the number of effective contact points > 7, the action type of the corresponding frame is determined to be a palm strike type. This action type could be a single stroking action, a continuous stroking action, or a long press action, or it could be noise such as a mis-touch.

[0052] Furthermore, if the number of valid touch points in the current frame is greater than the first touch point threshold, the number of valid touch points in the current frame is corrected, and the corrected number of valid touch points replaces the previous number of valid touch points in the current frame. For the corrected number of valid touch points in the current frame, its numerical range is detected. If the second touch point threshold ≤ the number of valid touch points in the current frame (i.e., the corrected number of valid touch points in the current frame) ≤ the third touch point threshold, the action type of the current frame is determined to be a slapping type; if the number of valid touch points in the current frame (i.e., the corrected number of valid touch points in the current frame) > the third touch point threshold, the action type of the current frame is determined to be a palm type.

[0053] Specifically, the electronic device also sets a correction threshold, which is greater than the aforementioned noise threshold, for example, a correction threshold of 800. For each valid channel value, it can be further checked whether it is greater than or equal to the correction threshold. The number of valid channel values ​​greater than or equal to the correction threshold is counted, and this number is taken as the corrected number of valid touch points in the current frame. The numerical range in which the corrected number of valid touch points falls is compared, and the action type of the current frame is determined to be either a slapping type or a hand gesture type based on the numerical range in which it falls.

[0054] In one embodiment, the dispersion is calculated based on the current frame data and the previous frame data, including: calculating the pressure center, total channel value, and distance between each effective channel position and the pressure center of the corresponding frame data based on the effective channel value and effective channel position of each channel in each frame data; and calculating the dispersion of the corresponding frame based on the effective channel value, the distance between each effective channel position and the pressure center, and the total channel value.

[0055] In this embodiment, the pressure center refers to the (weighted) center position of all valid touch points in the corresponding frame data, reflecting the concentration point of touch pressure, which can be calculated by the valid channel value and its corresponding channel position; the total channel value is the sum of all valid channel values ​​in the corresponding frame data, such as the total pressure value.

[0056] Total channel value The pressure center of the corresponding frame It can be calculated using the following formula: Dispersion .

[0057] Where m represents the number of valid touch points in the corresponding frame. This represents the value of the i-th valid channel in the corresponding frame data. This represents the channel position of the i-th valid channel value, which can be represented by the corresponding channel number. Indicates the channel position of the i-th valid channel value. Its pressure center The distance between them , It can be understood as The weights are calculated by summing the distances of each channel location to its pressure center, and the corresponding dispersion d can be obtained.

[0058] For example, the data structure of the touch point distribution mapping table in the current frame data is as follows: touchTable = {-- Touch point distribution mapping table [3] = 58000, -- Channel 3 pressure value [4] = 42000, -- Channel 4 pressure value If the number of valid touch points in the current frame is m=2, then the total channel value is... 58000 + 42000 = 100000; Center of pressure (58000×3+42000×4) / 100000=3.42. The channel position of the i-th valid channel value. Its pressure center Distance between =| |, for example, channel 3 ( The distance between its pressure center and its pressure center =|3-3.42|=0.42, Channel 4 ( The distance between its pressure center and its pressure center =|4-3.42|=0.58, the dispersion is obtained by weighted summation of the distances between each channel position and its pressure center in the current frame. =0.42×0.58+0.58×0.42=0.49.

[0059] The data structure of the touch point distribution mapping table in the frame data of the previous frame of the current frame is as follows: touchTable = {-- Touch point distribution mapping table [3] = 52000, -- Channel 3 pressure value [4] = 43000, -- Channel 4 pressure value [5] = 48000, -- Channel 5 pressure value If the frame data of the previous frame is m=3, then the frame data of the previous frame is m=3. =143000; =3.97.

[0060] By weighted summing the distances between each channel position and its pressure center in the current frame, the dispersion d = 0.69 is obtained.

[0061] In one embodiment, determining whether a touch action in the current frame is a tap or a swipe based on dispersion includes: if dispersion ≤ dispersion threshold, calculating the center offset between the current frame and the previous frame based on the pressure center of the current frame and the pressure center of the previous frame, and calculating the corresponding channel change value based on the effective channel values ​​at the same channel position in the current frame and the previous frame; if the center offset ≥ offset threshold, determining the touch action in the current frame as a swipe; if the channel change value ≥ channel change threshold, determining the touch action in the current frame as a tap.

[0062] Similar to the first contact threshold, electronic devices also preset corresponding dispersion thresholds, offset thresholds, and channel change thresholds. These thresholds can be fixed values, or they can be determined based on one or more factors such as the number of channels in the multi-channel capacitive sensor and the user's usage scenario and habits. For example, the dispersion threshold could be set to 3, the offset threshold to 0.5, and the channel change threshold to 1500.

[0063] Optionally, after calculating the dispersion of each frame's data, the numerical range of the dispersion of each frame is identified. When the dispersion of multiple consecutive frames (such as the current frame and its previous frame) is less than or equal to a preset dispersion threshold, further calculations of center offset and channel change values ​​are performed. Based on the center offset and channel change values, it is determined whether the touch action in those multiple consecutive frames is a swipe action or a tap action. If, in the current frame and its previous frame, only one frame has a dispersion of less than or equal to the dispersion threshold, it is determined not to be a tap type.

[0064] Among them, the center offset is the distance between the pressure centers of two adjacent frames (such as the current frame and the previous frame), reflecting the movement range of the touch position; the same channel position means that there are channels with valid channel values ​​in both frames (the current frame and the previous frame) (such as channel 3 having valid channel values ​​in both frames); the channel change threshold is a critical value used to judge whether the pressure has changed significantly. For example, if it is set to 150, a channel change value ≥ 150 indicates that the pressure change is obvious.

[0065] The channel variation value y is the average of the absolute values ​​of the effective channel differences at each corresponding channel position in two frames of data. The channel variation value y can be calculated using the formula... The result is obtained through calculation. Here, k represents the total number of identical channel positions in the two frames of data (the number of valid comparison points). This represents the valid channel value at the j-th common channel position in the current frame among two frames of data; This represents the valid channel value at the j-th same channel position in the previous frame among two frames of data.

[0066] Continuing with the example of the current frame data and its previous frame data, the center offset between the current frame and the previous frame... That is, the absolute value of the difference between the two pressure centers, i.e. |5.99-5.53|=0.46. The total number of the same channel positions in the current frame data and the previous frame data is k=2 (that is, both have valid channel values ​​of channel 3 and channel 4). =58000, =42000; =52000, =43000, based on which the channel change value y=3500 is calculated.

[0067] If the preset offset threshold is 0.4 and the preset channel change threshold is 1500, then the channel change value is greater than or equal to the channel change threshold and the center offset is less than the offset threshold, and it is determined to be a light touch action.

[0068] If the center offset in two adjacent frames of data < the offset threshold, and the channel change value < the channel change threshold, then further analyze the touch action, or determine that the touch action corresponding to this frame of data is an incorrect operation (i.e., no valid touch action occurs), and does not belong to any of the 7 touch actions.

[0069] In this embodiment, through the triple judgment of the dispersion degree, channel change value, and center offset of two adjacent frames, the confusion problem of similar point-touch actions (such as light touch and small sliding) is solved, and the sliding action and light touch action can be accurately identified.

[0070] In one embodiment, based on the number of consecutive frames, it is determined whether the touch action corresponding to the corresponding frame is one of the slow-tap action, fast-tap action, long-press action, continuous stroke action, or single stroke action, including: determining the duration of the corresponding touch action based on the number of consecutive frames; when the same action type is the tap type, if the duration ≤ the first duration threshold, determine that the corresponding touch action is a fast-tap action, if the first duration threshold < the duration ≤ the second duration threshold, determine that the corresponding touch action is a slow-tap action; when the same action type is the palm type, if the duration ≤ the third duration threshold, determine that the corresponding touch action is a single stroke action, if the third duration threshold < the duration ≤ the fourth duration threshold, determine that the corresponding touch action is a continuous stroke action, if the duration > the fourth duration threshold, determine that the corresponding touch action is a long-press action.

[0071] In this embodiment, the magnitude of the duration is related to the number of consecutive frames and the acquisition frequency. Based on the number of consecutive frames and the corresponding acquisition frequency, the corresponding duration can be calculated.

[0072] The magnitudes of the first duration threshold, the second duration threshold, the third duration threshold, and the fourth duration threshold can be calibrated according to experiments. For example, the first duration threshold t1 = 200 ms, the second duration threshold t2 = 500 ms, the third duration threshold t3 = 700 ms, and the fourth duration threshold t4 = 2000 ms can be set. It can be understood that the magnitudes of the 4 duration thresholds can also be set to other any suitable values.

[0073] For the obtained consecutive multiple frames that are all of the tap type, calculate the corresponding duration t0. If t0 ≤ t1, determine that the touch action is a fast-tap action. If t1 < t0 ≤ t2, determine that it is a slow-tap action. If it does not belong to the above two situations, further analyze the corresponding touch action, or determine that the touch action corresponding to this frame of data is an incorrect operation and does not belong to any of the 7 touch actions.

[0074] For the consecutive multiple frames that are all of the palm type, the corresponding duration t0 of the consecutive multiple frames is also calculated. If t0 ≤ t3, it is determined that the touch action is a single stroke action. If t3 < t0 ≤ t4, it is determined that it is a continuous stroke action. If t0 > t4, it is determined that the corresponding touch action is a long press action. If it does not belong to any of the above situations, the corresponding touch action is further analyzed, or it is determined that the touch action corresponding to the frame data is an incorrect operation (that is, no valid touch action occurs) and does not belong to any of the 7 touch actions.

[0075] In this embodiment, by comparing the duration with multiple groups of thresholds, the sub-actions of the patting type (quick pat / slow pat) and the palm type (single stroke / continuous stroke / long press) are clearly divided, solving the recognition confusion of similar composite gestures (such as the confusion between quick patting and slow patting, the confusion between short-time stroking and long pressing), and can accurately identify the quick pat action, slow pat action, single stroke action, continuous stroke action, and long press action.

[0076] In one embodiment, as Figure 2 shown, the above method further includes: Step 210, making a matching response to the recognized touch action.

[0077] In this embodiment, after the corresponding touch action is recognized, a corresponding touch control instruction is generated according to the touch action and the corresponding position where it acts, and the touch control instruction is executed. For example, if the recognized touch action is a tap action and its acting position is the flash control, the device responds to the touch action to turn on / off the flash; if the recognized touch action is a swipe action and its swipe trajectory is that the fingertip swipes from left to right, the device responds to the touch action to switch the display page.

[0078] Step 220, obtaining the cooling duration corresponding to the recognized touch action.

[0079] Step 230, ignoring the recognition and / or response of the touch actions of each frame data within the cooling duration.

[0080] In this embodiment, the cooling duration represents a "refractory period" preset for a specific touch action. During this period, the device suppresses the recognition and / or response of the same type or related touch actions to avoid repeated triggering in a short time. The cooling duration is set according to the action characteristics. For example, the cooling duration of quick repeated actions (such as patting, tapping) is relatively short, and the cooling duration of continuous actions (such as stroking, swiping) is relatively long.

[0081] For example, a cooldown time of 1800ms is set for a swipe action; 1000ms is set for a tap, a quick pat, and a slow pat; and 2000ms is set for a single swipe, a continuous swipe, and a long press. Understandably, the cooldown time for each touch action can also be set to any other suitable value.

[0082] The start time of the cooldown period can be one of several times, such as the response time to the touch action, the acquisition time of the first frame corresponding to the touch action, or the acquisition time of the last frame corresponding to the touch action. During the cooldown period, the electronic device will still acquire touch action data at a preset acquisition frequency and form corresponding frame data. However, for this frame data, the electronic device will no longer perform touch action recognition and analysis within the cooldown period, or will not respond to the identified touch actions. For example, after detecting a light touch action and responding to the action command, if a swipe action is identified within the cooldown period (e.g., within 1000ms), the electronic device will ignore the swipe action, refuse to respond to it, or issue a prompt to the user indicating frequent operation.

[0083] Ignoring only the response is suitable for scenarios where data records need to be retained, such as recognizing actions but not performing operations during the cooling-off period (cooling-off duration). Only the historical queue data is updated, and user operation habits can be analyzed based on this historical queue data. Based on these operation habits, the cooling-off duration corresponding to each touch action and various thresholds in this application (such as the first touch threshold, the second touch threshold, the third touch threshold, the dispersion threshold, the offset threshold, the channel change threshold, the first duration threshold, the second duration threshold, the third duration threshold, the fourth duration threshold, and the interruption threshold) can be adaptively adjusted.

[0084] Completely ignoring identification is suitable for low-power scenarios, such as skipping steps like calculating the number of effective contacts and dispersion analysis during the cooling period to reduce CPU usage (e.g., wearable devices in deep sleep). Electronic devices adaptively determine, based on the current scenario, whether to adopt a strategy of simply ignoring responses or completely ignoring identification for frame data during the cooling period.

[0085] In this embodiment, different cooling durations are set for different action types, balancing operational flexibility and system stability. Cooling durations suppress repetitive responses within a short period, such as preventing repeated screen activation due to user hand tremors or frequent page switching due to rapid swiping, thus improving interaction stability. Furthermore, the cooling mechanism filters out environmental interference (such as sensor false triggers caused by vibration) or unintentional repetitive user actions (such as not releasing the touch promptly), ensuring that responses are only for valid user actions. During the cooling period, some recognition steps can be selectively skipped (e.g., completely ignoring recognition), reducing the computational load on the embedded device, lowering power consumption, and extending battery life.

[0086] In one embodiment, before determining the action type of the touch action in the current frame based on the number of valid touch points, the method further includes: if the number of valid touch points in the current frame is 0, treating the current frame data as an invalid frame; counting the number of invalid frames in the multi-frame queue data that have consecutive invalid frames; if the number of invalid frames is greater than or equal to an interruption threshold, determining that no valid touch action has occurred; if the number of invalid frames is less than the interruption threshold, then performing the determination of the action type of the touch action in the current frame based on the number of valid touch points.

[0087] In this embodiment, the size of the interruption threshold can also be adaptively determined based on one or more factors such as the device's environment, usage scenario, and usage habits. For example, the interruption threshold can be set to any suitable value such as 2, 3, or 5. For a frame of data, if its set of valid channel values ​​is empty (e.g., all channel values ​​are < noise threshold), then the number of valid touch points = 0, and the frame is determined to be an invalid frame; if the set of valid channel values ​​is not empty (e.g., the channel values ​​of channels 3 and 4 are ≥ noise threshold), then the number of valid touch points is ≥ 1, and the frame is a valid frame, proceeding to the subsequent action type determination process.

[0088] For example, if the interruption threshold is set to 2, and frames 8-12 in the multi-frame queue are all invalid frames, then the number of invalid frames is 5, the touch recognition is determined to be interrupted, and the device stops the current action recognition process; if frames 1, 8, and 13 in the multi-frame queue are all invalid, then the number of invalid frames is 1, and the device continues to recognize the current action.

[0089] When a touch interruption is detected, the electronic device may perform one or more of the following operations: 1. Clear the action state and reset the currently recognized action type (e.g., clear the "long press" state); 2. Release resources and stop unnecessary calculations (e.g., pause dispersion calculation); 3. Enter standby mode, with the sensor reducing the sampling frequency or some circuits entering a sleep state.

[0090] For example, if a user presses and holds the device for 3 seconds and then releases their hand, and the sensor detects 6 consecutive invalid frames (invalid frame count = 6 ≥ 5), the electronic device determines that the touch has been interrupted, stops the recording function (which is a response to the long press action), and reduces the sensor sampling frequency from 100Hz to 10Hz to save power.

[0091] In this embodiment, by statistically analyzing consecutive invalid frames, firstly, environmental noise (such as isolated invalid frames caused by electrostatic interference) can be effectively filtered out, avoiding misjudgment as touch interruptions. For example, brief signal fluctuations may generate 1-2 invalid frames, but because the interruption threshold (such as 5) is not reached, the system still maintains the current action recognition process. Secondly, power consumption management can be optimized and subsequent processing can be simplified. By timely identifying genuine touch interruptions (such as when the user leaves the device), the system can enter a low-power state, extending battery life. Furthermore, complex action type determination steps can be skipped directly (such as not needing to calculate the number of valid touch points or dispersion), reducing computational resource consumption. For example, a smartwatch automatically reduces the operating frequency of the touch sensor 50ms after the user stops touching (corresponding to 5 invalid frames). Thirdly, the continuity of action recognition is enhanced. For example, it can avoid interrupting ongoing action recognition due to accidental invalid frames (such as a finger slightly lifted but not completely removed). For example, if the user briefly lifts their finger during a swipe (generating 3 invalid frames), because the number of invalid frames is less than 5, the system will still determine the subsequent frames as a continuation of the swipe action.

[0092] In one embodiment, such as Figure 3 As shown, a touch action recognition device is provided, the device comprising: The touch action acquisition module 310 is used to acquire multi-frame queue data. Each frame of the multi-frame queue data includes the effective channel values ​​of each channel collected by the multi-channel capacitive sensor at a corresponding moment.

[0093] The action type recognition module 320 is used to take one frame of data in the multi-frame queue as the current frame data, calculate the number of valid touch points of the current frame data, and determine the action type of the touch action of the current frame based on the number of valid touch points. The action types include point touch type, slap type, and palm type.

[0094] The touch type analysis module 330 is used to calculate the corresponding dispersion based on the current frame data and the previous frame data if the action type of the current frame is touch type, and determine whether the touch action of the current frame is a light touch action or a swipe action based on the dispersion.

[0095] The patting or palm type analysis module 340 is used to calculate the number of consecutive frames in the multi-frame queue data that are determined to be of the same action type if the action type of the current frame is patting or palm type; and to determine whether the touch action of the corresponding frame is one of slow patting action, fast patting action, long press action, continuous stroking action or single stroking action based on the number of consecutive frames.

[0096] In one embodiment, the action type recognition module 320 is further used to count the number of effective channel values ​​in the current frame data to obtain the number of effective touch points; if 0 < the number of effective touch points in the current frame ≤ the first touch point threshold, it is determined that the action type of the current frame belongs to the point touch type; if the second touch point threshold ≤ the number of effective touch points in the current frame ≤ the third touch point threshold, it is determined that the action type of the current frame belongs to the slapping type; if the number of effective touch points in the current frame > the third touch point threshold, it is determined that the action type of the current frame belongs to the palm type.

[0097] In one embodiment, the touch type analysis module 330 is further configured to calculate the pressure center, total channel value, and distance between each effective channel position and the pressure center of the corresponding frame data based on the effective channel value and effective channel position of each channel in each frame data; and to calculate the dispersion of the corresponding frame based on the effective channel value, the distance between each effective channel position and the pressure center, and the total channel value.

[0098] In one embodiment, the touch type analysis module 330 is further configured to: if the dispersion is less than or equal to the dispersion threshold, calculate the center offset between the current frame and the previous frame based on the pressure center of the current frame and the pressure center of the previous frame; calculate the corresponding channel change value based on the effective channel values ​​in the current frame and the previous frame that are in the same channel position; if the center offset is greater than or equal to the offset threshold, determine that the touch action of the current frame is a sliding action; if the channel change value is greater than or equal to the channel change threshold, determine that the touch action of the current frame is a light touch.

[0099] In one embodiment, the patting or palm type analysis module 340 is further configured to determine the duration of the corresponding touch action based on the number of consecutive frames; when the same action type is patting, if the duration is ≤ a first duration threshold, the corresponding touch action is determined to be a fast patting action; if the first duration threshold < duration ≤ a second duration threshold, the corresponding touch action is determined to be a slow patting action; when the same action type is palm, if the duration is ≤ a third duration threshold, the corresponding touch action is determined to be a single stroking action; if the third duration threshold < duration ≤ a fourth duration threshold, the corresponding touch action is determined to be a continuous stroking action; if the duration > a fourth duration threshold, the corresponding touch action is determined to be a long press action.

[0100] In one embodiment, such as Figure 4 As shown, the aforementioned touch action recognition device also includes: The response execution module 350 is used to make a matching response to the recognized touch action.

[0101] The cooling processing module 360 ​​is used to obtain the cooling duration corresponding to the identified touch action; and ignores the recognition and / or response of touch actions in each frame of data within the cooling duration.

[0102] In one embodiment, the action type identification module 320 is further configured to: if the number of valid touch points in the current frame is 0, treat the current frame data as an invalid frame; count the number of invalid frames in the multi-frame queue data that have consecutive invalid frames; if the number of invalid frames is greater than or equal to the interruption threshold, determine that no valid touch action has occurred; if the number of invalid frames is less than the interruption threshold, determine the action type to which the touch action of the current frame belongs based on the number of valid touch points.

[0103] In one embodiment, a computer-readable storage medium is provided having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the steps in the above method embodiments.

[0104] In one embodiment, an electronic device is also provided, including one or more processors; and a memory storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform the steps in the above method embodiments. The electronic device may be a device that deploys the above-described delivery control system, such as a backend server that communicates with the overhead crane, controlling its operation or sending travel paths to it.

[0105] In one embodiment, such as Figure 5 The diagram illustrates the structure of an electronic device used to implement an embodiment of this application. The electronic device includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes based on a program stored in a read-only memory (ROM) 502 or a program loaded from a storage portion 508 into a random access memory (RAM) 503. The RAM 503 also stores various programs and data required for the operation of the electronic device 500. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0106] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 510 as needed so that computer programs read from it can be installed into storage section 508 as needed.

[0107] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer-readable medium carrying instructions that, in such embodiments, can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the instructions are executed by central processing unit (CPU) 501, the various method steps described in this application are performed.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0109] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of this application and form different embodiments. For example, all the embodiments above can be used in any combination. The information disclosed in this background section is intended only to enhance the understanding of the general background of this application and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art.

Claims

1. A touch action recognition method, characterized in that, The method includes: Acquire multi-frame queue data, wherein each frame of the multi-frame queue data includes the effective channel values ​​of each channel collected by the multi-channel capacitive sensor at a corresponding moment; Using one frame of data from a multi-frame queue as the current frame data, the number of valid touch points in the current frame data is calculated. Based on the number of valid touch points, the action type of the touch action in the current frame is determined. The action type includes point touch, slap, and palm touch. If the action type of the current frame is a touch type, calculate the corresponding dispersion based on the current frame data and the previous frame data, and determine whether the touch action of the current frame is a light touch or a swipe action based on the dispersion. If the action type of the current frame is a slapping type or a palm type, calculate the number of consecutive frames in the multi-frame queue data that are determined to have the same action type. Based on the number of consecutive frames, determine whether the touch action of the corresponding frame is one of the following: slow tap, fast tap, long press, continuous stroking, or single stroking.

2. The touch action recognition method according to claim 1, characterized in that, The calculation of the number of valid touch points in the current frame data includes: The number of valid channel values ​​in the current frame data is counted to obtain the number of valid touch points; The process of determining the action type of the touch action in the current frame based on the number of valid touch points includes: If 0 < number of valid touch points in the current frame ≤ first touch point threshold, the action type of the current frame is determined to be a point touch type. If the second touch threshold ≤ the number of valid touches in the current frame ≤ the third touch threshold, the action type of the current frame is determined to be a slapping type; If the number of valid touch points in the current frame is greater than the third touch point threshold, the action type of the current frame is determined to be the palm type.

3. The touch action recognition method according to claim 1, characterized in that, The step of calculating the corresponding dispersion based on the current frame data and the previous frame data includes: Based on the effective channel value and effective channel position of each channel in each frame of data, calculate the pressure center, total channel value, and distance between each effective channel position and the pressure center of the corresponding frame of data. The dispersion of the corresponding frame is calculated based on the effective channel value, the distance between each effective channel position and the pressure center, and the total channel value.

4. The touch action recognition method according to claim 1, characterized in that, Determining whether the touch action in the current frame is a tap or a swipe based on the dispersion includes: If the dispersion is less than or equal to the dispersion threshold, the center offset between the current frame and the previous frame is calculated based on the pressure center of the current frame and the pressure center of the previous frame. The corresponding channel change value is calculated based on the effective channel values ​​in the current frame and the previous frame that are in the same channel position. If the center offset is greater than or equal to the offset threshold, the touch action in the current frame is determined to be a swipe action; If the channel change value is greater than or equal to the channel change threshold, the touch action in the current frame is determined to be a light touch action.

5. The touch action recognition method according to claim 1, characterized in that, The step of determining whether the touch action of the corresponding frame is a slow-motion action, a fast-motion action, a long-press action, a continuous stroking action, or a single stroking action based on the number of consecutive frames includes: The duration of the corresponding touch action is determined based on the number of consecutive frames. When the same action type is tapping, if the duration is less than or equal to the first duration threshold, the corresponding touch action is determined to be a fast tapping action; if the first duration threshold is less than the duration and less than or equal to the second duration threshold, the corresponding touch action is determined to be a slow tapping action. When the same action type is the palm type, if the duration is less than or equal to the third duration threshold, the corresponding touch action is determined to be a single stroking action; if the third duration threshold is less than the duration and less than or equal to the fourth duration threshold, the corresponding touch action is determined to be a continuous stroking action; if the duration is greater than the fourth duration threshold, the corresponding touch action is determined to be a long press action.

6. The touch action recognition method according to any one of claims 1 to 5, characterized in that, The method further includes: It responds to the identified touch actions in a matching manner; Get the cooldown duration corresponding to the selected touch action; Ignore the recognition and / or response to touch actions in each frame of data within the cooling time.

7. The touch action recognition method according to any one of claims 1 to 5, characterized in that, Before determining the action type of the touch action in the current frame based on the number of valid touch points, the method further includes: If the number of valid touch points in the current frame is 0, the data of the current frame will be considered an invalid frame. The number of invalid frames with consecutive invalid frames in the multi-frame queue data is counted. If the number of invalid frames is greater than or equal to the interruption threshold, it is determined that no valid touch action has occurred. If the number of invalid frames is less than the interruption threshold, the action type of the touch action in the current frame is determined based on the number of valid touch points.

8. A touch action recognition device, characterized in that, The device includes: The touch action acquisition module is used to acquire multi-frame queue data, wherein each frame of the multi-frame queue data includes the effective channel values ​​of each channel collected by the multi-channel capacitive sensor at a corresponding moment. The action type recognition module is used to take one frame of data in a multi-frame queue as the current frame data, calculate the number of valid touch points of the current frame data, and determine the action type of the touch action of the current frame based on the number of valid touch points. The action type includes point touch type, slapping type, and palm type. The touch type analysis module is used to calculate the corresponding dispersion based on the current frame data and the previous frame data if the action type of the current frame is touch type, and determine whether the touch action of the current frame is a light touch action or a swipe action based on the dispersion. The patting or palm type analysis module is used to calculate the number of consecutive frames in the multi-frame queue data that are determined to be of the same action type if the action type of the current frame is patting or palm type; and to determine whether the touch action of the corresponding frame is one of slow patting action, fast patting action, long press action, continuous stroking action or single stroking action based on the number of consecutive frames.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores executable instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: One or more processors; A memory for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to perform the method as described in any one of claims 1 to 7.

Images