WO2016138728A1 - 一种可随手势而自动旋转界面的智能手表及其控制方法 - Google Patents
一种可随手势而自动旋转界面的智能手表及其控制方法 Download PDFInfo
- Publication number
- WO2016138728A1 WO2016138728A1 PCT/CN2015/085063 CN2015085063W WO2016138728A1 WO 2016138728 A1 WO2016138728 A1 WO 2016138728A1 CN 2015085063 W CN2015085063 W CN 2015085063W WO 2016138728 A1 WO2016138728 A1 WO 2016138728A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- smart watch
- interface
- motion data
- sensor
- control method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G04—HOROLOGY
- G04G—ELECTRONIC TIME-PIECES
- G04G21/00—Input or output devices integrated in time-pieces
Definitions
- the present invention relates to the field of smart wearable devices, and more particularly to a smart watch capable of automatically rotating an interface with a gesture and a control method thereof.
- the object of the present invention is to provide a smart watch capable of automatically rotating an interface with a gesture and a control method thereof, aiming at solving the problem that the interface of the existing smart watch cannot automatically rotate with the gesture.
- a smart watch control method capable of automatically rotating an interface with a gesture, comprising the steps of:
- the pre-processing includes correcting and filtering motion data, denoising, filtering, and time domain segmentation, thereby extracting pre-processed motion data;
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the sensor comprises a magnetic sensor, a gravity acceleration sensor, a direction sensor, and a gyro sensor.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the motion data is subjected to time domain segmentation processing to form a feature value vector, and the feature value vector is used as an input of algorithm modeling, and an action is obtained through feature analysis. model.
- the smart watch control method that automatically rotates an interface with a gesture, wherein the step of forming the feature value vector after the motion data is subjected to time domain segmentation processing includes:
- a feature value vector is formed.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the angle includes: an angle for indicating that the top of the smart watch faces north, for indicating that the top or the tail of the smart watch is tilted The angle and the angle used to indicate the left or right side of the smart watch.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the magnetic induction information comprises: a magnetic field induction intensity.
- a smart watch control method capable of automatically rotating an interface with a gesture, comprising the steps of:
- A. Obtain motion data of the smart watch through a sensor disposed in the smart watch, and preprocess the motion data;
- the upper layer corresponds to the rotating graphical interface by the rotation angle of the output.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the sensor comprises a magnetic sensor, a gravity acceleration sensor, a direction sensor, and a gyro sensor.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein in the step A, the data pre-processing includes correcting and filtering motion data, denoising, filtering, and time domain segmentation, thereby extracting pre-processing After the exercise data.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the motion data includes acceleration, angular velocity, angle, and magnetic field information in a three-dimensional direction of the smart watch.
- the smart watch control method capable of automatically rotating an interface with a gesture, wherein the motion data is subjected to time domain segmentation processing to form a feature value vector, and the feature value vector is used as an input of the algorithm modeling, and the action model is obtained through feature analysis.
- a smart watch that automatically rotates the interface with gestures including:
- a preprocessing module configured to acquire motion data of the smart watch through a sensor disposed in the smart watch, and preprocess the motion data
- a model establishing module configured to perform data mining on the preprocessed motion data to establish an action model, identify a rotation angle of the smart watch, and output the same;
- the rotation module is used for the upper layer to correspond to the rotation graphical interface through the rotation angle of the output.
- the smart watch can automatically rotate an interface with a gesture, wherein the sensor comprises a magnetic sensor, a gravity acceleration sensor, a direction sensor, and a gyro sensor.
- the smart watch capable of automatically rotating an interface with a gesture, wherein in the pre-processing module, data pre-processing includes correcting and filtering motion data, denoising, filtering, and time domain segmentation, thereby extracting pre-processed Sports data.
- the smart watch can automatically rotate an interface with a gesture, wherein the motion data includes acceleration, angular velocity, angle and magnetic field information in a three-dimensional direction of the smart watch.
- the smart watch can automatically rotate the interface according to the gesture, wherein the motion data is subjected to time domain segmentation processing to form a feature value vector, and the feature value vector is used as an input of the algorithm modeling, and the action model is obtained through feature analysis.
- the invention acquires the movement data change information of the smart watch through the sensor disposed in the smart watch, outputs the rotation angle through the bottom layer driving and the algorithm processing, and the upper layer APP performs the rotation of the graphic interface according to the selection angle.
- FIG. 1 is a flow chart of a preferred embodiment of a smart watch control method that automatically rotates an interface with a gesture according to the present invention.
- FIG. 2 is a schematic structural view of the smart watch of the present invention before rotation.
- FIG. 3 is a schematic structural view of the smart watch of the present invention after being rotated.
- FIG. 4 is a structural block diagram of a preferred embodiment of a smart watch capable of automatically rotating an interface with a gesture according to the present invention.
- the present invention provides a smart watch that can automatically rotate an interface with a gesture and a control method thereof.
- a smart watch that can automatically rotate an interface with a gesture and a control method thereof.
- FIG. 1 is a flow chart of a preferred embodiment of a smart watch control method for automatically rotating an interface with a gesture according to the present invention. As shown in the figure, the method includes the following steps:
- S102 Perform data mining on the motion data to establish an action model, identify a rotation angle of the smart watch, and output the same;
- the upper layer corresponds to the rotation graphic interface by the rotation angle of the output.
- the senor includes a magnetic sensor, a gravity acceleration sensor, and a gyro sensor.
- the rotating action of the smart watch needs to be collected by various sensors integrated in the remote controller, such as magnetic sensor, acceleration sensor, direction sensor, gyro sensor, etc.
- the collected information includes the magnetic field in the three-dimensional direction of the sensor and the acceleration in the three-dimensional direction.
- Motion data such as the angle in the three-dimensional direction and the angular velocity in the three-dimensional direction.
- the data pre-processing includes correcting and filtering the motion data, denoising, filtering, and time domain segmentation, thereby extracting the pre-processed motion data.
- the above data correction and screening are processed separately for different data types, such as raw data collected by magnetic sensors, direction sensors and gyro sensors.
- the linear and non-linear correction functions are needed to adjust the angular range of each axis. Between 0-360 degrees, the original data that is difficult to analyze has the positional meaning corresponding to the real three-dimensional space.
- the watch is kept horizontal and 90° perpendicular to the eye. Baseline direction.
- the raw data collected by the acceleration sensor needs to be based on the acceleration of gravity, and the acceleration in each direction is corrected accordingly, so that the original data has a corresponding physical meaning.
- the corrected acceleration sensor output is processed.
- the raw data should only have a gravitational acceleration value of about 9.8 m/s2 in the direction of the center of the earth.
- the direction of the X-axis is from left to right along the horizontal direction of its screen. If the smart watch is not square or rectangular, the shorter sides need to be placed horizontally, and the longer sides need to be placed vertically to determine the X-axis direction.
- the direction of the Y-axis is from the lower left corner of the screen to the top of the screen along the vertical direction of the screen.
- Event.values[0] represents the acceleration in the X-axis direction.
- Event.values[1] represents the acceleration in the Y-axis direction.
- Event.values[2] represents the acceleration in the Z-axis direction.
- the result is a float value so that the coordinate system can be created.
- the magnetic sensor is used to determine the position and angle of the smart watch at any time.
- the angular difference can be determined by calculating the azimuth difference.
- a three-axis magnetic sensor IC can obtain the magnetic field sensing in the X, Y and Z directions in the current environment. Intensity, for the Android middle layer is to read the three values measured by the magnetic sensor.
- the angle at which the smart watch UI needs to be deflected or horizontally deflected to the left or right can be obtained;
- the sensor determines the angular velocity of the watch at any time; the direction sensor can return to three angles, which determine the placement of the smart watch.
- the first angle indicates the angle between the top of the smart watch and the north.
- the angle value changes. For example, when the angle is 0°, it indicates that the top of the smart watch is facing true north; when the angle is 90°, it means that the top of the smart watch is facing the east; when the angle is 180°, it means that the top of the smart watch is facing south; the intelligent angle At 270°, it means the top of the watch is facing west.
- the second angle indicates the angle at which the top or tail of the smart watch is tilted.
- the angle value changes.
- the angle ranges from -180° to 180°.
- the smart watch screen is placed horizontally on the table. If the table is completely horizontal, the angle should be 0°. If you lift from the top of the smart watch until the smart watch is rotated 180° along the X axis (the screen is placed horizontally down on the table), the angle will change from 0° to -180° during this rotation.
- the angle value when lifting from the top of the smart watch, the angle value will gradually decrease until it is equal to -180°; if it is lifted from the bottom of the smart watch until the smart watch is rotated 180° along the X axis (the screen is level down) Placed on the table), the angle value will change from 0° to 180°. That is to say, when lifting from the top of the smart watch, the angle value will gradually increase until it is equal to 180°.
- the third angle indicates the angle at which the left or right side of the smart watch is tilted.
- the angle value changes.
- the angle ranges from -90° to 90°.
- the smart watch screen is placed horizontally on the table. If the table is completely horizontal, the angle should be 0°.
- the left side of the smart watch is gradually raised, knowing to rotate the smart watch 90° along the Y axis (the watch is perpendicular to the desktop). During this rotation, the angle value will change from 0° to -90°.
- the angle value when lifting from the left side of the smart watch, the angle value will gradually decrease until it is equal to -90°; if the right side of the smart watch is gradually raised, it is known to rotate the watch 90° along the Y axis (watch and desktop) Vertical), the angle value will change from 0 to 90 during this rotation. That is to say, when lifting from the right side of the smart watch, the angle value will gradually increase until it is equal to 90°.
- a specific example is that when the user chooses to use the function of the present invention, the reference value of the smart watch is set and recorded once. Then, through the change of the Z-axis acceleration of the gyro sensor, it is judged whether the user performs the action of lifting the smart watch, and then triggers the bright screen. This process also determines the horizontal deflection angle of the user with respect to the reference direction through the X-axis of the acceleration sensor and The judgment of the deflection to the left and the right is combined with the X-axis of the magnetic sensor and the rotation angle of the Y-axis to read the pitch angle and the roll angle based on the reference direction. Thereby triggering the graphical interface to make stereoscopic angular changes including horizontal and vertical directions. All in all, the combination of the three sensors can determine the gestures of the person and determine whether it is necessary to rotate the UI to suit the position of the human eye for a more comfortable look.
- the preprocessing also includes denoising, filtering, time domain segmentation, etc.
- the denoising, filtering and time domain segmentation preprocessing in the present invention all process data according to an algorithm. For example, for the user to habitually view the habit of the smart watch for user analysis, it is assumed that the viewing action is a periodic repetitive action, and the periodic waveform is reflected in the received motion data for data analysis. Then, the filtering process will filter out some abnormal signals whose periodic features are not obvious, such as signals with excessive sloshing amplitude. This process is filtering preprocessing; the viewing action of the smart watch should be within a certain frequency range. Then, the remaining periodic signals that do not meet this frequency range are determined as noise.
- time domain segmentation will be static for the smart watch.
- the motion data and the periodic signal with amplitude are segmented accordingly to find out the periodic signals that meet the user's habits, thereby segmenting the useful data, which is the time domain segmentation preprocessing.
- the time domain segmentation preprocessing of the present invention is implemented based on a data definition system.
- the data definition system classifies the actions of the user to view the smart watch according to the requirements of the algorithm. For example, the action of the user viewing the smart watch is divided into: the user lifts up Start the smart watch, flip the action of the smart watch, etc., and then define these actions separately; then establish a time domain segmentation window for data segmentation, and according to these defined actions, the motion data in the time domain segmentation window is passed through the feature value. After processing, the eigenvalue vector is formed. The eigenvalue vector will be used as the input of the algorithm modeling.
- the next new motion data enters the time domain segmentation window again, and the eigenvalue processing is performed to input the data identification. Entrance.
- the above process makes the actions that are segmented when the algorithm is modeled and the data is identified within the defined range of motion, so that the data modeling entry and the data recognition entry are in a substantially consistent state, ensuring data input. accuracy.
- step S103 the pre-processed motion data is used as a basis for establishing an action model obtained by feature analysis.
- a neural network system based on motion data is established through a random forest decision algorithm. Its specific data mining modeling process, including steps:
- S201 randomly extracting K samples from the original training set, and randomly extracting J variables from the eigenvalue variable set, and forming the local training set by the K samples and the J variables;
- the smart watch of the present invention looks at the graphical interface at any different angles, and the content displayed thereon is always 90 degrees vertical to the user, and the display angle is adjusted correspondingly as the angle of the hand changes. .
- the UI rotates at a corresponding angle.
- the smart watch graphic interface is always 90 degrees perpendicular to the body directly above, and the smart watch experience is not seen because the hand is far away from the body, and the figure is not clear. Clocks that change with gestures, 12 points are always in line with the line of sight for easy viewing time.
- the present invention further provides a preferred embodiment of a smart watch that can automatically rotate an interface with a gesture.
- the method includes:
- the pre-processing module 100 is configured to acquire motion data of the smart watch by using a sensor disposed in the smart watch, and pre-process the motion data;
- the model establishing module 200 is configured to perform data mining on the preprocessed motion data to establish an action model, identify a rotation angle of the smart watch, and output the same;
- the rotation module 300 is configured to correspond to the rotation graphic interface by the rotation angle of the output.
- the senor includes a magnetic sensor, a gravity acceleration sensor, and a gyro sensor.
- the data pre-processing includes correction and screening of motion data, denoising, filtering, and time domain segmentation, thereby extracting pre-processed motion data.
- the motion data includes acceleration, angular velocity, angle, and magnetic field information in a three-dimensional direction of the smart watch.
- the motion data is subjected to time domain segmentation processing to form a feature value vector, and the feature value vector is used as an input of the algorithm modeling, and the action model is obtained through feature analysis.
- the present invention acquires the motion data change information of the smart watch through the sensor disposed in the smart watch, and outputs the rotation angle through the underlying driving and algorithm processing, and the upper layer APP performs the rotation of the graphic interface according to the selection angle.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
本发明公开一种可随手势而自动旋转界面的智能手表及其控制方法,其中方法包括通过设置在智能手表中的传感器来获取智能手表的运动数据,对运动数据进行预处理;对经预处理的运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;上层通过输出的旋转角度来对应旋转图形界面。本发明可以使智能手表旋转界面。
Description
本发明涉及智能穿戴设备领域,尤其涉及一种可随手势而自动旋转界面的智能手表及其控制方法。
随着智能手表的越来越普及,智能手表的图形界面用户体验也变得越来越重要,但由于智能手表屏幕较小,在原来手机平板的图形界面设计明显不在适合在手表上运用,这个问题在圆形表盘上尤其突出。这就迫切需要一套贴合用户和圆形表盘的图形界面设计。目前可见的圆形表盘的界面都是静止的,界面不会随着手势的不同位置而旋转,用户在观看手表和使用时,需要转头才能读取信息,使用起来不方便。
因此,现有技术还有待于改进和发展。
本发明的目的在于提供一种可随手势而自动旋转界面的智能手表及其控制方法,旨在解决现有的智能手表其界面不能随手势自动旋转的问题。
一种可随手势而自动旋转界面的智能手表控制方法,其包括步骤:
A、通过设置在智能手表中的传感器来获取所述智能手表的运动数据,并对所述运动数据进行预处理,其中,所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息,所述预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据;
B、对经预处理的所述运动数据进行数据挖掘建立动作模型,识别所述智能手表的旋转角度并输出;以及
C、通过输出的所述旋转角度来对应旋转图形界面。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述传感器包括磁感应器、重力加速度传感器、方向传感器以及陀螺仪传感器。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述运动数据经过时域分割处理后形成特征值向量,所述特征值向量作为算法建模的输入,通过特征分析得到动作模型。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述运动数据经过时域分割处理后形成特征值向量的步骤包括:
根据算法的需求将用户查看所述智能手表的动作归类;
建立用于数据分割的时域分割窗口;以及
根据所述动作归类,将所述时域分割窗口中的运动数据经过特征值处理之后,形成特征值向量。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述角度包括:用于表示所述智能手表顶部朝向正北方的夹角,用于表示所述智能手表顶部或尾部翘起的角度以及用于表示所述智能手表左侧或右侧翘起的角度。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述磁感应信息包括:磁场感应强度。
一种可随手势而自动旋转界面的智能手表控制方法,其中,包括步骤:
A、通过设置在智能手表中的传感器来获取智能手表的运动数据,并对运动数据进行预处理;
B、对经预处理的所述运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;
C、上层通过输出的旋转角度来对应旋转图形界面。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述传感器包括磁感应器、重力加速度传感器、方向传感器以及陀螺仪传感器。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述步骤A中,数据预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息。
所述的可随手势而自动旋转界面的智能手表控制方法,其中,所述运动数据经过时域分割处理后形成特征值向量,特征值向量作为算法建模的输入,通过特征分析得到动作模型。
一种可随手势而自动旋转界面的智能手表,其中,包括:
预处理模块,用于通过设置在智能手表中的传感器来获取智能手表的运动数据,并对运动数据进行预处理;
模型建立模块,用于对经预处理的所述运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;
旋转模块,用于上层通过输出的旋转角度来对应旋转图形界面。
所述的可随手势而自动旋转界面的智能手表,其中,所述传感器包括磁感应器、重力加速度传感器、方向传感器以及陀螺仪传感器。
所述的可随手势而自动旋转界面的智能手表,其中,所述预处理模块中,数据预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据。
所述的可随手势而自动旋转界面的智能手表,其中,所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息。
所述的可随手势而自动旋转界面的智能手表,其中,所述运动数据经过时域分割处理后形成特征值向量,特征值向量作为算法建模的输入,通过特征分析得到动作模型。
本发明通过设置在智能手表中的传感器,来获取智能手表运动数据变化信息,通过底层驱动和算法处理,输出旋转角度,上层APP根据选择角度进行图形界面的旋转。
图1为本发明一种可随手势而自动旋转界面的智能手表控制方法较佳实施例的流程图。
图2为本发明的智能手表旋转前的结构示意图。
图3为本发明的智能手表旋转后的结构示意图。
图4为本发明一种可随手势而自动旋转界面的智能手表较佳实施例的结构框图。
本发明提供一种可随手势而自动旋转界面的智能手表及其控制方法,为使本发明的目的、技术方案及效果更加清楚、明确,以下对本发明进一步详细说明。应当理解,此处所描述的具体实施例仅仅用以解释本发明,并不用于限定本发明。
请参阅图1,图1为本发明一种可随手势而自动旋转界面的智能手表控制方法较佳实施例的流程图,如图所示,其包括步骤:
S101、通过设置在智能手表中的传感器来获取智能手表的运动数据,并对运动数据进行预处理;
S102、对所述运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;
S103、上层通过输出的旋转角度来对应旋转图形界面。
具体来说,所述传感器包括磁感应器、重力加速度传感器以及陀螺仪传感器。
智能手表的旋转动作需要通过集成在遥控器中的各种传感器来采集,例如磁感应器、加速度传感器、方向传感器、陀螺仪传感器等等,所采集的信息包括传感器三维方向的磁场、三维方向的加速度、三维方向的角度及三维方向的角速度等动作数据。
所述步骤S101中,数据预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据。
上述数据校正及筛选是针对数据类型的不同分别进行处理的,例如磁感应器、方向传感器和陀螺仪传感器采集的原始数据,需要通过线性及非线性的校正函数,将每一轴的方向角度范围调整在0-360度之间,使得原本难以分析的原始数据具有了与真实三维空间相对应的位置含义,使用时先做一次校正,以抬手看表,手表保持水平,与眼睛垂直90°为基准方向。又例如,加速度传感器采集的原始数据需要以重力加速度为定基,对各个方向上的加速度做相对应的校正,使得原始数据具有相应的物理含义,当智能手表静止时,校正处理后的加速度传感器输出的原始数据应该只有向地心方向的9.8m/s²左右的重力加速度值。
如果智能手表是矩形或正方形手表来说,X轴的方向是沿着其屏幕的水平方向从左向右。如果智能手表不是正方形或矩形的话,较短的边需要水平放置,较长的边需要垂直放置,从而确定X轴方向。
Y轴的方向是从屏幕的左下角开始沿着屏幕的垂直方向指向屏幕的顶端。
将智能手表平放在水平面上,Z轴的方向是从手表里指向天空。
在方法:
@Override
public void onSensorChanged(SensorEvent event)
{
// TODO Auto-generated method stub
}
中,event.values[0]表示X轴方向上的加速度。
event.values[1]表示Y轴方向上的加速度。
event.values[2]表示Z轴方向上的加速度。
获得的是float(浮动)值,这样就可以建立坐标系了。通过三轴的加速度差值可以比较准确的智能手表是否被抬起来,如果被抬起来就触发中断,唤醒智能手表屏幕。通过磁感应器确定智能手表任意时刻所处的位置和角度,通过计算方位差可以确定运动过程中的角度变化;一个三轴的磁感应器IC可以得到当前环境下X、Y和Z方向上的磁场感应强度,对于Android中间层来说就是读取该磁感应器测量到的这3个值。当抬手看表时,通过与前文记录的基准方向对比前面相应的3个值之间的方位差值,即可获取智能手表UI需要向左还是向右偏转及水平偏转的角度;通过陀螺仪传感器可确定任意时刻手表的角速度;方向传感器可以返回三个角度,这三个角度即可确定智能手表的摆放状态。
关于方向传感器返回的三个角度的说明如下:
第一个角度:表示智能手表顶部朝向正北方的夹角。当智能手表绕着Z轴旋转时,该角度值发生改变。例如当该角度为0°时,表明智能手表顶部朝向正北;该角度为90°时,代表智能手表顶部朝向正东;该角度为180°时,代表智能手表顶部朝向正南;该智能角度为270°时,代表手表顶部朝向正西。
第二个角度:表示智能手表顶部或尾部翘起的角度。当智能手表绕着X轴倾斜时,该角度值发生改变。该角度的取值范围是-180°~180°。假设将智能手表屏幕朝上水平放在桌面上,如果桌面是完全水平的,该角度值应该是0°。假如从智能手表顶部开始抬起,直到将智能手表沿X轴旋转180°(屏幕向下水平放在桌面上),在这个旋转过程中,该角度值会从0°变化为-180°。也就是说,从智能手表顶部抬起时,该角度值会逐渐减小,直到等于-180°;如果从智能手表底部开始抬起,直到将智能手表沿X轴旋转180°(屏幕向下水平放在桌面上),该角度值会从0°变化为180°。也就是说,从智能手表顶部抬起时,该角度值会逐渐增大,直到等于180°。
第三个角度:表示智能手表左侧或右侧翘起的角度。当智能手表绕着Y轴倾斜时,该角度值发生改变。该角度的取值范围是-90°~90°。假设将智能手表屏幕朝上水平放在桌面上,如果桌面是完全水平的,该角度值应该是0°。假设将智能手表左侧逐渐抬起,知道将智能手表沿Y轴旋转90°(手表与桌面垂直)。在这个旋转过程中,该角度值会从0°变化为-90°。也就是说,从智能手表左侧抬起时,该角度值会逐渐减小,直到等于-90°;如果将智能手表右侧逐渐抬起,知道将手表沿Y轴旋转90°(手表与桌面垂直),在这个旋转过程中,该角度值会从0变化为90°。也就是说,从智能手表右侧抬起时,该角度值会逐渐增大,直到等于90°。
一个具体的实例是:当用户选择使用本发明的功能后,会对智能手表的方位做一次基准值的设定和记录。然后通过陀螺仪传感器的Z轴加速度的变化判断用户是否做抬手智能看表的动作,然后触发亮屏,这个过程还通过加速度传感器的X轴,判断出用户相对于基准方向的水平偏转角度及往左往右偏转的判断,再结合磁传感器的X轴,Y轴的旋转角度,判读出基于基准方向的仰俯角度和滚转角度。从而触发图形界面做立体的角度变化包括水平方向的,垂直方向的。总而言之,三种传感器结合起来可以确定人的手势动作,判断是否需要旋转UI,以适应人眼位置,达到更舒适看表的效果。
预处理还包括去噪、滤波、时域分割等,本发明中的去噪、滤波及时域分割预处理都是根据算法来对数据进行处理的。例如,对于用户习惯性查看智能手表这一习性来做用户分析,假定这种查看动作是一种周期性的反复动作,体现在接收的运动数据上对数据分析有作用的是周期性的波形,那么滤波处理的过程会将周期性特征不明显的一些异常信号,例如晃动幅度过大的信号滤除掉,此过程即为滤波预处理;智能手表的查看动作应该是在一定的频率范围内的,那么将其余的不符合这个频率范围的周期信号判定为噪声,比如环境因素可能造成噪声,所以需要将该噪声过滤掉,此过程即为去噪预处理;时域分割会对智能手表静止的运动数据、有振幅的周期性信号进行相应的分割,以将符合用户习惯的周期性信号寻找出来,从而分割出有用的数据,此即为时域分割预处理。
本发明的时域分割预处理是基于数据定义系统来实现的,数据定义系统是根据算法的需求将用户查看智能手表的动作归类,例如,将用户查看智能手表的动作分为:用户向上抬起智能手表、翻转智能手表的动作等,然后分别对这些动作进行定义;然后建立用于数据分割的时域分割窗口,根据这些定义好的动作,将时域分割窗口中的运动数据经过特征值处理之后,形成特征值向量,特征值向量将作为算法建模的输入,构建模型,建模基本完成之后,下一个新的运动数据再次进入时域分割窗口,经过特征值处理,输入到数据识别入口。上述过程使得算法建模时和数据识别时分割出来的动作都是处在定义好的动作范围之内,从而使数据建模入口与数据识别入口处于基本一致的状态之下,确保了数据输入的准确性。
在步骤S103中,经过预处理后的运动数据将作为建立动作模型的基础,所述动作模型为通过特征分析得到。经过模式化的数据挖掘建模过程,通过随机森林决策算法,建立起基于运动数据的神经网络系统。其具体的数据挖掘建模过程,包括步骤:
S201、从原始训练集中随机有放回地抽出K个样本,从特征值变量集随机抽出J个变量,由所述K个样本及J个变量构成局部训练集;
S202、创建节点;
S203、判断所述局部训练集是否为空,当所述局部训练集为空时,标记节点为错误,并返回步骤S202,重新创建节点;
S204、当所述局部训练集不为空时,若局部训练集中的所有记录都属于同一个类别,则以该类别标记节点;若局部训练集中的属性集为空,则标记节点N为最普通的类;
S205、对局部训练集中连续的特征值变量进行离散化,并从所述特征值变量中选取具有最高信息增益的属性D;
S206、根据所述属性D构成有X棵树的随机森林模型;关于随机森林模型的建模过程的更多细节也可参考现有技术,当然本发明也可选用其他建模方式来建立模型,同样也可实现本发明的目的。
这样请参见图2及图3,本发明的智能手表在任何不同角度去看图形界面,其上显示的内容对于用户一直是90度垂直的,并且随着手的角度变化,显示角度也会相应调整。对于图中所示的圆形表盘,随着智能手表的角度变化,UI做相应角度旋转。
通过本发明,智能手表图形界面正上方始终与身体保持垂直90度,不会因为手离身体较远,而看不清楚图形,更舒适的智能手表体验。随手势而变化的时钟,12点始终与视线保持在一条直线上,方便查看时间。
基于上述方法,本发明还提供一种可随手势而自动旋转界面的智能手表较佳实施例,如图4所示,其包括:
预处理模块100,用于通过设置在智能手表中的传感器来获取智能手表的运动数据,并对运动数据进行预处理;
模型建立模块200,用于对经预处理的所述运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;
旋转模块300,用于上层通过输出的旋转角度来对应旋转图形界面。
进一步,所述传感器包括磁感应器、重力加速度传感器以及陀螺仪传感器。
进一步,所述预处理模块100中,数据预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据。
进一步,所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息。
进一步,所述运动数据经过时域分割处理后形成特征值向量,特征值向量作为算法建模的输入,通过特征分析得到动作模型。
关于上述模块单元的技术细节在前面的方法中已有详述,故不再赘述。
综上所述,本发明通过设置在智能手表中的传感器,来获取智能手表运动数据变化信息,通过底层驱动和算法处理,输出旋转角度,上层APP根据选择角度进行图形界面的旋转。
应当理解的是,本发明的应用不限于上述的举例,对本领域普通技术人员来说,可以根据上述说明加以改进或变换,所有这些改进和变换都应属于本发明所附权利要求的保护范围。
Claims (16)
- 一种可随手势而自动旋转界面的智能手表控制方法,其中包括步骤:A、通过设置在智能手表中的传感器来获取所述智能手表的运动数据,并对所述运动数据进行预处理,其中,所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息,所述预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据;B、对经预处理的所述运动数据进行数据挖掘建立动作模型,识别所述智能手表的旋转角度并输出;以及C、通过输出的所述旋转角度来对应旋转图形界面。
- 根据权利要求1所述的可随手势而自动旋转界面的智能手表控制方法,其中所述传感器包括磁感应器、重力加速度传感器、方向传感器以及陀螺仪传感器。
- 根据权利要求1所述的可随手势而自动旋转界面的智能手表控制方法,其中所述运动数据经过时域分割处理后形成特征值向量,所述特征值向量作为算法建模的输入,通过特征分析得到动作模型。
- 根据权利要求3所述的可随手势而自动旋转界面的智能手表控制方法,其中所述运动数据经过时域分割处理后形成特征值向量的步骤包括:根据算法的需求将用户查看所述智能手表的动作归类;建立用于数据分割的时域分割窗口;以及根据所述动作归类,将所述时域分割窗口中的运动数据经过特征值处理之后,形成特征值向量。
- 根据权利要求1所述的可随手势而自动旋转界面的智能手表控制方法,其中所述角度包括:用于表示所述智能手表顶部朝向正北方的夹角,用于表示所述智能手表顶部或尾部翘起的角度以及用于表示所述智能手表左侧或右侧翘起的角度。
- 根据权利要求1所述的可随手势而自动旋转界面的智能手表控制方法,其中所述磁感应信息包括:磁场感应强度。
- 一种可随手势而自动旋转界面的智能手表控制方法,其中包括步骤:A、通过设置在智能手表中的传感器来获取智能手表的运动数据,并对运动数据进行预处理;B、对经预处理的所述运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;以及C、通过输出的旋转角度来对应旋转图形界面。
- 根据权利要求7所述的可随手势而自动旋转界面的智能手表控制方法,其中所述传感器包括磁感应器、重力加速度传感器、方向传感器以及陀螺仪传感器。
- 根据权利要求8所述的可随手势而自动旋转界面的智能手表控制方法,其中所述步骤A中,数据预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的运动数据。
- 根据权利要求8所述的可随手势而自动旋转界面的智能手表控制方法,其中所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息。
- 根据权利要求9 所述的可随手势而自动旋转界面的智能手表控制方法,其中所述运动数据经过时域分割处理后形成特征值向量,特征值向量作为算法建模的输入,通过特征分析得到动作模型。
- 一种可随手势而自动旋转界面的智能手表,其中包括:预处理模块,用于通过设置在智能手表中的传感器来获取智能手表的运动数据,并对运动数据进行预处理;模型建立模块,用于对经预处理的所述运动数据进行数据挖掘建立动作模型,识别智能手表的旋转角度并输出;旋转模块,用于上层通过输出的旋转角度来对应旋转图形界面,使图形界面相对人体视线保持不变。
- 根据权利要求12所述的可随手势而自动旋转界面的智能手表,其中所述传感器包括磁感应器、重力加速度传感器、方向传感器以及陀螺仪传感器。
- 根据权利要求13所述的可随手势而自动旋转界面的智能手表,其中所述预处理模块中,数据预处理包括对运动数据的校正及筛选、去噪、滤波以及时域分割,从而提取出预处理后的数据。
- 根据权利要求12所述的可随手势而自动旋转界面的智能手表,其中所述运动数据包括智能手表三维方向的加速度、角速度、角度及磁场信息。
- 根据权利要求12所述的可随手势而自动旋转界面的智能手表,其中所述运动数据经过时域分割处理后形成特征值向量,特征值向量作为算法建模的输入,通过特征分析得到动作模型。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510093987.0A CN104698831B (zh) | 2015-03-03 | 2015-03-03 | 一种可随手势而自动旋转界面的智能手表及其控制方法 |
| CN201510093987.0 | 2015-03-03 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2016138728A1 true WO2016138728A1 (zh) | 2016-09-09 |
Family
ID=53346076
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2015/085063 Ceased WO2016138728A1 (zh) | 2015-03-03 | 2015-07-24 | 一种可随手势而自动旋转界面的智能手表及其控制方法 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104698831B (zh) |
| WO (1) | WO2016138728A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12135585B2 (en) | 2021-09-28 | 2024-11-05 | International Business Machines Corporation | Automatic alignment of wrist-mounted display with another display device |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104698831B (zh) * | 2015-03-03 | 2017-05-10 | 惠州Tcl移动通信有限公司 | 一种可随手势而自动旋转界面的智能手表及其控制方法 |
| KR102373491B1 (ko) * | 2015-07-15 | 2022-03-11 | 삼성전자주식회사 | 회전체의 회전 인식 방법 및 그 방법을 처리하는 전자 장치 |
| CN105117119B (zh) * | 2015-07-28 | 2018-12-11 | 广东欧珀移动通信有限公司 | 一种屏幕画面旋转的方法及智能手表 |
| CN105575317B (zh) * | 2016-03-14 | 2021-06-15 | 联想(北京)有限公司 | 电子设备和显示处理方法 |
| US10890981B2 (en) | 2016-10-24 | 2021-01-12 | Ford Global Technologies, Llc | Gesture-based vehicle control |
| CN106527911A (zh) * | 2016-10-26 | 2017-03-22 | 广东小天才科技有限公司 | 一种终端播放器控制方法及装置 |
| US10684693B2 (en) | 2017-03-02 | 2020-06-16 | Samsung Electronics Co., Ltd. | Method for recognizing a gesture and an electronic device thereof |
| CN108508974A (zh) * | 2018-03-30 | 2018-09-07 | 四川斐讯信息技术有限公司 | 可穿戴设备处理器及其数据处理方法、及可穿戴设备 |
| CN108803896B (zh) * | 2018-05-28 | 2021-07-16 | Oppo(重庆)智能科技有限公司 | 控制屏幕的方法、装置、终端及存储介质 |
| CN110187759B (zh) | 2019-05-08 | 2023-01-17 | 安徽华米信息科技有限公司 | 显示方法、装置、智能穿戴设备及存储介质 |
| CN110162177A (zh) * | 2019-05-21 | 2019-08-23 | 出门问问信息科技有限公司 | 智能手表的显示方法、装置、智能手表及可读存储介质 |
| CN111708273A (zh) * | 2020-06-01 | 2020-09-25 | 广东乐芯智能科技有限公司 | 一种手表的控制方法 |
| CN113382472A (zh) * | 2021-06-09 | 2021-09-10 | 合肥市云联鸿达信息技术有限公司 | 一种儿童手表定位系统 |
| CN114995094B (zh) * | 2022-06-08 | 2023-07-11 | 深圳市万如夏科技有限公司 | 一种抬手亮屏的智能手表及抬手亮屏方法 |
| CN116483241B (zh) * | 2023-06-21 | 2024-03-29 | 深圳市微克科技股份有限公司 | 一种智能设备床头钟自适应方向调整方法、系统及介质 |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101174178A (zh) * | 2006-10-31 | 2008-05-07 | 佛山市顺德区顺达电脑厂有限公司 | 观看图案的方法和装置 |
| CN102323854A (zh) * | 2011-03-11 | 2012-01-18 | 中国科学院研究生院 | 人体运动捕获装置 |
| WO2013081861A1 (en) * | 2011-11-30 | 2013-06-06 | Qualcomm Mems Technologies, Inc. | Gesture-responsive user interface for an electronic device |
| CN104035321A (zh) * | 2014-05-16 | 2014-09-10 | 华为技术有限公司 | 智能手表的屏幕控制方法和智能手表 |
| US20150015502A1 (en) * | 2013-07-11 | 2015-01-15 | Khalid Al-Nasser | Smart watch |
| CN104698831A (zh) * | 2015-03-03 | 2015-06-10 | 惠州Tcl移动通信有限公司 | 一种可随手势而自动旋转界面的智能手表及其控制方法 |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101989126B (zh) * | 2009-08-07 | 2015-02-25 | 深圳富泰宏精密工业有限公司 | 手持式电子装置及其屏幕画面自动旋转方法 |
| CN102809927A (zh) * | 2011-05-30 | 2012-12-05 | 昆达电脑科技(昆山)有限公司 | 可调整信息显示角度的电子表 |
| KR101357292B1 (ko) * | 2012-03-12 | 2014-01-28 | 삼성전자주식회사 | 휴대용 단말기의 정보 표시 장치 및 방법 |
| CN103514240B (zh) * | 2012-11-29 | 2016-10-12 | Tcl集团股份有限公司 | 一种基于遥控器的家庭成员关系挖掘方法及系统 |
| CN103024579B (zh) * | 2012-12-24 | 2016-09-21 | 深圳Tcl新技术有限公司 | 显示设备画面自动旋转的方法及装置 |
| CN103970291B (zh) * | 2013-01-31 | 2018-08-14 | 索尼公司 | 移动终端 |
| CN103455256B (zh) * | 2013-08-21 | 2017-09-15 | 小米科技有限责任公司 | 旋转屏幕显示画面的方法和终端 |
| CN103472993A (zh) * | 2013-09-02 | 2013-12-25 | 小米科技有限责任公司 | 旋转显示方法、装置及终端设备 |
| CN103645804A (zh) * | 2013-12-18 | 2014-03-19 | 三星电子(中国)研发中心 | 识别人体手势的方法、装置以及使用其的手表 |
| CN103885584B (zh) * | 2014-01-06 | 2016-11-09 | 北京奇虎科技有限公司 | 智能手表 |
| CN103760982B (zh) * | 2014-01-22 | 2017-09-22 | 深圳市金立通信设备有限公司 | 一种控制终端屏幕状态的方法及终端 |
| CN103793075B (zh) * | 2014-02-14 | 2017-02-15 | 北京君正集成电路股份有限公司 | 一种应用在智能手表上的识别方法 |
| CN204129441U (zh) * | 2014-09-05 | 2015-01-28 | 青岛歌尔声学科技有限公司 | 一种智能系统和智能手表 |
| CN104375648B (zh) * | 2014-11-26 | 2017-09-08 | 三星电子(中国)研发中心 | 一种腕式设备及其使用方法 |
-
2015
- 2015-03-03 CN CN201510093987.0A patent/CN104698831B/zh active Active
- 2015-07-24 WO PCT/CN2015/085063 patent/WO2016138728A1/zh not_active Ceased
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101174178A (zh) * | 2006-10-31 | 2008-05-07 | 佛山市顺德区顺达电脑厂有限公司 | 观看图案的方法和装置 |
| CN102323854A (zh) * | 2011-03-11 | 2012-01-18 | 中国科学院研究生院 | 人体运动捕获装置 |
| WO2013081861A1 (en) * | 2011-11-30 | 2013-06-06 | Qualcomm Mems Technologies, Inc. | Gesture-responsive user interface for an electronic device |
| US20150015502A1 (en) * | 2013-07-11 | 2015-01-15 | Khalid Al-Nasser | Smart watch |
| CN104035321A (zh) * | 2014-05-16 | 2014-09-10 | 华为技术有限公司 | 智能手表的屏幕控制方法和智能手表 |
| CN104698831A (zh) * | 2015-03-03 | 2015-06-10 | 惠州Tcl移动通信有限公司 | 一种可随手势而自动旋转界面的智能手表及其控制方法 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12135585B2 (en) | 2021-09-28 | 2024-11-05 | International Business Machines Corporation | Automatic alignment of wrist-mounted display with another display device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104698831B (zh) | 2017-05-10 |
| CN104698831A (zh) | 2015-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016138728A1 (zh) | 一种可随手势而自动旋转界面的智能手表及其控制方法 | |
| WO2012036327A1 (ko) | 이동 통신 단말기에서의 스케쥴 표시 방법 및 장치 | |
| CN105025440B (zh) | 室内外场景检测方法及设备 | |
| CN103354582B (zh) | 一种通过传感器智能关联应用的方法及系统 | |
| CN110926473A (zh) | 识别楼层的方法、装置、电子设备及存储介质 | |
| CN104640184A (zh) | 一种屏幕唤醒方法及装置 | |
| AU2020301254B2 (en) | Sticker generating method and apparatus, and medium and electronic device | |
| WO2016088981A1 (ko) | 사용자 인터페이스를 제공하기 위한 방법, 디바이스, 시스템 및 비일시성의 컴퓨터 판독 가능한 기록 매체 | |
| CN107092882B (zh) | 一种基于子动作感知的行为识别系统及其工作方法 | |
| CN114581998A (zh) | 一种基于目标对象关联特征融合的布控方法、系统、设备和介质 | |
| JP2025041608A (ja) | 電子デバイスの現在位置を表示及び改善するためのユーザインタフェース | |
| CN111510553A (zh) | 一种运动轨迹展示方法、装置和可读存储介质 | |
| CN108693982A (zh) | 一种将手机内置传感器采集数据从手机坐标系转换到参考坐标系的处理方法 | |
| CN110647881A (zh) | 确定图像对应的卡片类型的方法、装置、设备及存储介质 | |
| CN113138665A (zh) | 一种基于智能手表的智能设备指向操控方法及系统 | |
| CN109567814B (zh) | 刷牙动作的分类识别方法、计算设备、系统及存储介质 | |
| CN114708641B (zh) | 睡眠检测方法、装置、计算机可读存储介质及终端设备 | |
| WO2018093197A2 (ko) | 영상 처리 방법 및 이를 지원하는 전자 장치 | |
| CN111077570B (zh) | 确定油藏中辉绿岩的侵入位置的方法、装置和存储介质 | |
| CN119225538B (zh) | 一种掌上平板电脑的数据交互方法和系统 | |
| CN211928663U (zh) | 一种基于树莓派的人脸识别智能门禁系统 | |
| CN113920222A (zh) | 获取地图建图数据的方法、装置、设备及可读存储介质 | |
| TW201908922A (zh) | 具有防近視功能的電子裝置及方法 | |
| CN114022570B (zh) | 相机间外参的标定方法及电子设备 | |
| CN116414495A (zh) | 显示卡片的方法和电子设备 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15883781 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 15883781 Country of ref document: EP Kind code of ref document: A1 |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 15883781 Country of ref document: EP Kind code of ref document: A1 |