WO2017201992A1 - 一种压力屏校准方法和装置 - Google Patents

一种压力屏校准方法和装置 Download PDF

Info

Publication number
WO2017201992A1
WO2017201992A1 PCT/CN2016/106932 CN2016106932W WO2017201992A1 WO 2017201992 A1 WO2017201992 A1 WO 2017201992A1 CN 2016106932 W CN2016106932 W CN 2016106932W WO 2017201992 A1 WO2017201992 A1 WO 2017201992A1
Authority
WO
WIPO (PCT)
Prior art keywords
pressure
neural network
network model
artificial neural
pressure value
Prior art date
Application number
PCT/CN2016/106932
Other languages
English (en)
French (fr)
Inventor
孟龙龙
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2017201992A1 publication Critical patent/WO2017201992A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
    • G06F3/041—Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
    • G06F3/041—Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means
    • G06F3/0414—Digitisers, e.g. for touch screens or touch pads, characterised by the transducing means using force sensing means to determine a position

Definitions

  • the main principle of terminal pressure sensing is to arrange several pressure sensors according to certain rules under the pressure screen.
  • each pressure sensor will read a pressure value, and finally use the preset algorithm according to each
  • the pressure values of the pressure sensors determine a final value as the pressure value at which the user presses the pressure screen. Due to the influence of the structure, the layout of the pressure sensor and the characteristics of each pressure sensor, the pressure screen of the terminal needs to be calibrated at the factory. Otherwise, the pressure sensing performance of different parts of the pressure screen varies greatly, resulting in the user's experience. good.
  • the calibration method of the terminal pressure screen mainly uses the instrument to pressurize different positions of the pressure screen, and then find the calibration coefficient of different positions.
  • the obtained pressure value is multiplied by the calibration coefficient of the position as the user. Press the pressure value of the screen.
  • the pressure screen after the calibration of the method still has some problems that the sensor pressure performance of the user is not good, and the pressure sensitivity of the entire pressure screen is poor.
  • embodiments of the present invention are expected to provide a pressure screen calibration method and apparatus, which can effectively adapt to the number and arrangement of different pressure sensors, and can provide excellent calibration accuracy and improve user experience.
  • Embodiments of the present invention provide a pressure screen calibration method, including:
  • the pressure value sensed by each pressure sensor is used as an input of an artificial neural network model, and the actual applied pressure value is used as an output of the artificial neural network model to determine the artificial neural network model.
  • the artificial neural network model for determining the weight parameter is configured to determine a pressure value received by the pressure screen according to a pressure value sensed by each pressure sensor.
  • the artificial neural network model including: using a back propagation (BP) model;
  • the method further includes: preset a training error, and determining a weight parameter of the BP neural network model when the preset training error requirement is reached.
  • the method further includes: performing data return on the acquired pressure values of the pressure sensors.
  • One-time treatment One-time treatment
  • the method further includes: performing data recovery on the acquired pressure value actually applied at each of the preset pressure receiving positions One treatment.
  • the obtaining the pressure value actually applied at each of the preset pressure receiving positions comprises:
  • the pressure value of each preset pressure receiving position is measured by an external device.
  • the artificial neural network model based on determining the weight parameter determines the pressure value received by the pressure screen according to the pressure value sensed by each pressure sensor, including:
  • the method further includes: normalizing the pressure values of the current pressure sensors before inputting the pressure values of the current pressure sensors as the artificial neural network model for determining the weight parameters;
  • the output value is inverse normalized to determine the pressure value to which the current pressure screen is subjected.
  • the embodiment of the invention further provides a pressure screen calibration device, the device comprising: an acquisition module and a determination module; wherein
  • the acquiring module is configured to acquire a pressure value sensed when each pressure sensor applies pressure to each preset pressure receiving position of the pressure screen, and obtain a pressure value actually applied at each of the preset pressure receiving positions;
  • the determining module is configured to adopt an artificial neural network model, and the pressure value sensed by each pressure sensor is used as an input of an artificial neural network model, and the actually applied pressure value is used as an output of the artificial neural network model. Determining a weight parameter of the artificial neural network model;
  • the artificial neural network model for determining the weight parameter is configured to determine a pressure value received by the pressure screen according to a pressure value sensed by each pressure sensor.
  • the determining module the artificial neural network model comprises: a BP neural network model
  • the determining module is further configured to: preset a training error, and determine a weight parameter of the BP neural network model when the preset training error requirement is reached.
  • the acquiring module is configured to acquire an external device to measure a pressure value of each preset pressure receiving position under pressure.
  • the determining module is further configured to:
  • the obtained pressure values sensed by the pressure sensors are normalized by data
  • the pressure screen calibration method and device acquire the pressure values sensed when the pressure sensors are pressed at the preset pressure receiving positions of the pressure screen, and obtain the actual pressure applied at the preset pressure receiving positions. a pressure value; using an artificial neural network model, the pressure value sensed by each pressure sensor is used as an input of an artificial neural network model, and the actual applied pressure value is used as an output of the artificial neural network model to determine the artificial The weight parameter of the neural network model; when the user actually uses the pressure screen, the artificial neural network model for determining the weight parameter is used, and the pressure value of the pressure position of the pressure screen is determined according to the pressure value sensed by each pressure sensor; For the number and arrangement of different pressure sensors, only the number of neuron nodes of the artificial neural network model can be adjusted to adapt; thus, it can effectively adapt to the number and arrangement of different pressure sensors, and can provide excellent calibration accuracy and improve user experience. .
  • FIG. 1 is a schematic flow chart of a pressure screen calibration method according to an embodiment of the present invention.
  • FIG. 2 is a schematic diagram showing the arrangement of a terminal pressure sensor according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of a BP neural network model according to an embodiment of the present invention.
  • FIG. 4 is a schematic structural diagram of a pressure screen calibration apparatus according to an embodiment of the present invention.
  • the pressure values sensed when the pressure sensors are pressed at the preset pressure receiving positions of the pressure screen are obtained, and the pressure values actually applied at the preset pressure receiving positions are obtained; the artificial neural network model is adopted. And using the pressure value sensed by each pressure sensor as an input of an artificial neural network model, and using the actually applied pressure value as the artificial neural network Outputting a model, determining a weight parameter of the artificial neural network model according to a preset training error; wherein the artificial neural network model for determining the weight parameter is set to be determined according to a pressure value sensed by each pressure sensor The pressure value to which the pressure screen is subjected.
  • the main purpose of the pressure screen calibration method provided by the embodiment of the present invention is to establish an artificial neural network model, which can be set to represent the pressure value induced on each pressure sensor on the terminal and the pressure of the actual pressure receiving position. Relationship between values; in subsequent use, according to the pressure value induced on each pressure sensor, the artificial neural network model can accurately determine the pressure value received by the actual pressure position; as shown in FIG. 1, the method includes:
  • Step 101 Acquire a pressure value sensed when each pressure sensor applies pressure at each preset pressure receiving position of the pressure screen, and obtain a pressure value actually applied at each of the preset pressure receiving positions;
  • a plurality of pressure sensors are disposed under the terminal pressure screen, and are set to sense a pressure signal received by the pressure screen, and the pressure sensor signal transmits the pressure signal to the pressure information processing chip; 9 pressures are arranged under the terminal pressure screen as shown in FIG. 2
  • the sensors are numbered from the upper left corner to the lower right corner: No. 1, No. 2, ... No. 9, and the pressure information processing chip of the pressure sensor signal is in the lower right corner of the screen, and the pressure information processing chip quantizes the pressure signal sent by the pressure sensor into The pressure value is sent to the terminal processor.
  • a plurality of pressure receiving positions may be set in advance on the terminal pressure screen, and pressure pressing instruments are used to sequentially pressurize the respective pressure receiving positions; a wired or wireless connection may be established between the terminal and the pressure calibration instrument through the terminal pressure Initialize the calibration instrument and set the pressure of the calibration instrument. Generally, the same pressure can be used for calibration of each pressure position;
  • the pressure calibration instrument starts to apply pressure to the pressure receiving position on the pressure screen according to the set pressure value, and applies pressure to each pressure applying position, and the terminal reads the pressure value of each pressure sensor through the pressure information processing chip, and the pressure.
  • the pressure value actually calibrated by the instrument to the screen; the pressure value actually hit by the calibration instrument on the screen can be measured and transmitted to the terminal through the feedback system of the pressure calibration instrument; the terminal can store the data until the pressure calibration instrument is The pressure position measurement pressure is all over.
  • Step 102 Using an artificial neural network model, using the pressure value sensed by each pressure sensor as an input of an artificial neural network model, and determining the artificial force by using the actually applied pressure value as an output of the artificial neural network model.
  • the weight parameter of the neural network model is a parameter that determines the artificial force by using the actually applied pressure value as an output of the artificial neural network model.
  • an artificial neural network is used to process the pressure value of each pressure sensor read by the terminal and the pressure value actually pressed by the pressure calibration instrument to the screen; here, as shown in FIG. 3, a BP neural network model may be adopted;
  • the terminal reads the pressure value of each pressure sensor as an input to the BP neural network model;
  • the pressure value actually applied to the screen by the pressure calibration instrument is used as the output of the BP neural network model.
  • the empirical formula can be used for the number of neurons in the hidden layer.
  • the empirical formula for the number of neurons in the hidden layer can be expressed by the expression (1):
  • n i is the number of neurons in the input layer
  • m is the number of neurons in the output layer
  • ⁇ is a constant between [1, 10]; according to the expression (1)
  • the number of neurons is calculated to be between 5 and 14; the number of neurons in the hidden layer can be selected as 6; in this case, the training times and training errors of the BP neural network model can be preset to determine the preset.
  • the weight parameter of the BP neural network model is required when the training error is required; the BP neural network model can use epochs to indicate the number of trainings of the BP network, and the goal can be used to represent the training error of the BP network;
  • the training data of the BP neural network model can be expressed by the expression (2) and the expression (3):
  • the expressions (2) and (3) respectively represent the pressure sensor induced pressure data corresponding to all the pressed positions stored in the database and the pressure values actually hitting the screen.
  • the calibrated neural network cannot adapt to different user pressing forces; it can be obtained by normalization processing.
  • the pressure in this way, whether the pressure used in the calibration or the actual pressing force of the user, can fall into a uniform range after normalization; enable the neural network to adapt to various pressures; here, the expression can be
  • the data of the expression (3) is normalized.
  • the data of expression (2) and expression (3) can be normalized by expression (4) and expression (5), respectively:
  • Each column vector in Expression (4) is input to the BP neural network model, and each value in Expression (5) corresponds to a column vector of (4), and as its desired output.
  • the S-type tangent function tansig can be selected as the excitation function of the hidden layer neurons, and the pure linear function purelin is selected as the output layer function;
  • the S-type tangent function can be expressed by the expression (6)
  • the purely linear function can be represented by the expression (7):
  • x represents the input of the BP neural network model and y represents the output of the BP neural network model.
  • the pressure value of each pressure sensor is read and normalized, and the normalized data is input as a trained BP neural network model.
  • a normalized output value is obtained by performing an inverse normalization process on the output value to obtain a pressure value of the user pressing the terminal screen.
  • the main purpose of the pressure screen calibration apparatus provided by the embodiment of the present invention is to establish an artificial neural network model, which can be set to represent the pressure value induced on each pressure sensor on the terminal and the pressure received by the actual pressure receiving position. Relationship between values; in subsequent use, according to the pressure value induced on each pressure sensor, the artificial neural network model can accurately determine the pressure value received by the actual pressure receiving position; as shown in FIG. 4, the device includes: obtaining Module 41 and determining module 42; wherein
  • the obtaining module 41 is configured to acquire each preset pressure position of each pressure sensor on the pressure screen The pressure value sensed when the pressure is applied, and the pressure value actually applied at each of the preset pressure receiving positions is obtained;
  • a plurality of pressure sensors are disposed under the terminal pressure screen, and are set to sense a pressure signal received by the pressure screen, and the pressure sensor signal transmits the pressure signal to the pressure information processing chip; 9 pressures are arranged under the terminal pressure screen as shown in FIG. 2
  • the sensors are numbered from the upper left corner to the lower right corner: No. 1, No. 2, ... No. 9, and the pressure information processing chip of the pressure sensor signal is in the lower right corner of the screen, and the pressure information processing chip quantizes the pressure signal sent by the pressure sensor into The pressure value is sent to the terminal processor.
  • a plurality of pressure receiving positions may be set in advance on the terminal pressure screen, and pressure pressing instruments are used to sequentially pressurize the respective pressure receiving positions; a wired or wireless connection may be established between the terminal and the pressure calibration instrument through the terminal pressure Initialize the calibration instrument and set the pressure of the calibration instrument. Normally, the same pressure can be used to calibrate each pressurized position.
  • the pressure calibration instrument starts to apply pressure to the pressure receiving position on the pressure screen according to the set pressure value, and applies pressure to each pressure applying position, and the terminal reads the pressure value of each pressure sensor through the pressure information processing chip, and the pressure.
  • the pressure value actually calibrated by the instrument to the screen; the pressure value actually hit by the calibration instrument on the screen can be measured and transmitted to the terminal through the feedback system of the pressure calibration instrument; the terminal can store the data until the pressure calibration instrument is The pressure position measurement pressure is all over.
  • the determining module 42 is configured to adopt an artificial neural network model, and the pressure value sensed by each pressure sensor is used as an input of an artificial neural network model, and the actually applied pressure value is used as an output of the artificial neural network model. Determining a weight parameter of the artificial neural network model;
  • an artificial neural network is used to process the pressure value of each pressure sensor read by the terminal and the pressure value that the pressure calibration instrument actually hits the screen;
  • a BP neural network model can be used; the terminal reads each pressure The pressure value of the sensor is used as an input to the BP neural network model; the pressure value actually applied to the screen by the pressure calibration instrument is used as the output of the BP neural network model.
  • the training data of the BP neural network model can be expressed by the expression (2) and the expression (3); wherein, the expression (2) and the expression (3) ) respectively represents the pressure sensor sensing pressure data corresponding to all the pressed positions stored in the database and the pressure value actually hitting the screen.
  • the calibrated neural network cannot adapt to different user pressing forces; it can be obtained by normalization processing.
  • the pressure in this way, whether the pressure used in the calibration or the actual pressing force of the user, can fall into a uniform range after normalization; enable the neural network to adapt to various pressures; here, the expression can be
  • the data of the expression (3) is normalized.
  • the data of expression (2) and expression (3) can be normalized by expression (4) and expression (5) respectively; each column vector in expression (4) is used as BP neural network.
  • each value in the expression (5) corresponds to a column vector of (4) and serves as its desired output.
  • the S-type tangent function tansig can be selected as the excitation function of the hidden layer neurons, and the pure linear function purelin is selected as the output layer function;
  • the S-type tangent function can be expressed by the expression (6)
  • the purely linear function can be represented by the expression (7); wherein x represents the input of the BP neural network model and y represents the output of the BP neural network model;
  • the pressure value of each pressure sensor is read and normalized, and the normalized data is input as a trained BP neural network model.
  • a normalized output value is obtained by performing an inverse normalization process on the output value to obtain a pressure value of the user pressing the terminal screen.
  • the obtaining module 41 and the determining module 42 are combined by a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA) of the terminal. achieve.
  • CPU central processing unit
  • MPU microprocessor
  • DSP digital signal processor
  • FPGA field programmable gate array
  • Embodiments of the present invention also provide a storage medium.
  • the foregoing storage medium stores an execution instruction, where the execution instruction is used to perform one or a combination of the steps in the foregoing method embodiments.
  • the foregoing storage medium may include, but is not limited to, a USB flash drive, a Read-Only Memory (ROM), and a Random Access Memory (RAM).
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • the pressure screen calibration method and apparatus have the following beneficial effects: obtaining the pressure value sensed by each pressure sensor when pressure is applied to each preset pressure position of the pressure screen, and obtaining the actual position. Depressing the pressure value applied by each preset pressure position; using an artificial neural network model, using the pressure value sensed by each pressure sensor as an input of an artificial neural network model, using the actually applied pressure value as the artificial nerve The output of the network model determines the weight parameter of the artificial neural network model; when the user actually uses the pressure screen, the artificial neural network model for determining the weight parameter is used, and the pressure is determined according to the pressure value sensed by each pressure sensor.
  • the pressure value of the pressure position of the screen further, for the number and arrangement of different pressure sensors, only the number of neuron nodes of the artificial neural network model can be adjusted to adapt; thus, the number and arrangement of different pressure sensors can be effectively adapted, and Provides excellent calibration accuracy and enhances the user experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Measuring Fluid Pressure (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种压力屏校准方法和装置，获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值（101）；采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数（102）；用户实际使用压力屏时，采用所述确定权值参数的人工神经网络模型，根据各压力传感器感应的压力值，确定所述压力屏受压位置的压力值。

Description

一种压力屏校准方法和装置 技术领域
本发明涉及终端压力屏技术，尤其涉及一种压力屏校准方法和装置。
背景技术
压力感应是目前终端新增的一种用户体验方式，其应用场景主要包括基于压力感应的称重、游戏、图片预览及应用图标右键菜单控制等。
终端压力感应的主要原理是在其压力屏下方按照一定的规则布置数个压力传感器，当用户手指按压压力屏时，每个压力传感器都会读取到一个压力值，最后利用预设的算法根据每个压力传感器的压力值求出一个最终值，作为用户按压压力屏的压力值。由于结构、压力传感器的布局、每个压力传感器特性差异等因素的影响，终端的压力屏在出厂时需要做校准测试，否则压力屏不同部位的压力感应性能差异很大，造成用户的体验性不佳。
目前，终端压力屏的校准方法主要是利用仪器对压力屏不同位置加压，然后求出不同位置的校准系数，当用户按压不同的位置时，获取的压力值乘以该位置的校准系数作为用户按压屏幕的压力值。该方法校准后的压力屏仍然存在某些区域感知用户手指压力性能不佳的问题，整个压力屏压力感应的一致性较差。
因此，如何适应不同压力传感器的数量及布置方式，并能提供优良的校准精度，提升用户体验，是亟待解决的问题。
发明内容
有鉴于此，本发明实施例期望提供一种压力屏校准方法和装置，能有效适应不同压力传感器数量及布置方式，并能提供优良的校准精度，提升用户体验。
为达到上述目的，本发明的技术方案是这样实现的：
本发明实施例提供了一种压力屏校准方法，所述方法包括：
获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；
采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；
所述确定权值参数的人工神经网络模型，设置为根据各压力传感器感应到的压力值确定所述压力屏受到的压力值。
上述方案中，所述采用人工神经网络模型，包括：采用反向传播(BP，Back Propagation)神经网络模型；
所述方法还包括：预设训练误差，确定达到所述预设训练误差要求时所述BP神经网络模型的权值参数。
上述方案中，所述方法还包括：选择S型正切函数作为所述BP神经网络模型的隐含层神经元的激励函数，选择纯线性函数作为所述BP神经网络模型的输出层函数。
上述方案中，所述将所述各压力传感器感应到的压力值作为人工神经网络模型的输入之前，所述方法还包括：对所述获取的所述各压力传感器感应到的压力值进行数据归一化处理；
所述在将所述实际施加的压力值作为所述人工神经网络模型的输出之前，所述方法还包括：对所述获取的实际在所述各预设受压位置施加的压力值进行数据归一化处理。
上述方案中，所述获取实际在所述各预设受压位置施加的压力值，包括：
采用外部装置测量各预设受压位置受压的压力值。
上述方案中，所述基于确定权值参数的人工神经网络模型，根据各压力传感器感应到的压力值确定所述压力屏受到的压力值，包括：
读取当前各压力传感器的压力值，并作为所述确定权值参数的人工神经网络模型输入，将获取输出值确定为当前压力屏受到的压力值；
所述方法还包括：所述将当前各压力传感器的压力值作为所述确定权值参数的人工神经网络模型输入之前，对所述当前各压力传感器的压力值作归一化处理；对所述输出值作反归一化处理，确定为当前压力屏受到的压力值。
本发明实施例还提供了一种压力屏校准装置，所述装置包括：获取模块和确定模块；其中，
所述获取模块，设置为获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；
所述确定模块，设置为采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；
所述确定权值参数的人工神经网络模型，设置为根据各压力传感器感应到的压力值确定所述压力屏受到的压力值。
上述方案中，所述确定模块，所述人工神经网络模型包括：BP神经网络模型；
所述确定模块，还设置为：预设训练误差，确定达到所述预设训练误差要求时所述BP神经网络模型的权值参数。
上述方案中，所述获取模块，设置为获取外部装置测量各预设受压位置受压的压力值。
上述方案中，所述确定模块，还设置为：
在将所述各压力传感器感应到的压力值作为人工神经网络模型的输入之前，对所述获取的所述各压力传感器感应到的压力值进行数据归一化处理；
在将所述实际施加的压力值作为所述人工神经网络模型的输出之前，对所述获取的实际在所述各预设受压位置施加的压力值进行数据归一化处理。
本发明另一实施例提供了一种计算机存储介质，所述计算机存储介质存储有执行指令，所述执行指令用于执行上述方法实施例中的步骤之一或其组合。
本发明实施例所提供的压力屏校准方法和装置，获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；用户实际使用压力屏时，采用所述确定权值参数的人工神经网络模型，根据各压力传感器感应的压力值，确定所述压力屏受压位置的压力值；进一步，对于不同压力传感器数量及布置方式，只需要调整人工神经网络模型的神经元节点数即可适应；如此，能有效适应不同压力传感器数量及布置方式，并能提供优良的校准精度，提升用户体验。
附图说明
图1为本发明实施例压力屏校准方法的流程示意图；
图2为本发明实施例终端压力感应器布置示意图；
图3为本发明实施例BP神经网络模型结构示意图；
图4为本发明实施例压力屏校准装置的组成结构示意图。
具体实施方式
本发明实施例中，获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络 模型的输出，根据预设的训练误差，确定所述人工神经网络模型的权值参数；其中，所述确定权值参数的人工神经网络模型，设置为根据各压力传感器感应到的压力值，确定所述压力屏受到的压力值。
下面结合实施例对本发明再作进一步详细的说明。
本发明实施例提供的压力屏校准方法，主要目的是要建立一个人工神经网络模型，所述人工神经网络模型可以设置为表示终端上各压力感应器上感应的压力值和实际受压位置承受压力值的关系；后续使用中，根据各压力感应器上感应的压力值，通过所述人工神经网络模型能够准确确定实际受压位置承受的压力值；如图1所示，所述方法包括：
步骤101：获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；
通常，终端压力屏下设置有多个压力传感器，设置为感应压力屏受到的压力信号，压力传感器信号将压力信号传送到压力信息处理芯片；如图2所示的终端压力屏下方布置9个压力传感器，从左上角至右下角依次编号为：1号、2号、……9号，屏幕右下角为压力传感器信号的压力信息处理芯片，该压力信息处理芯片将压力传感器发送压力信号转量化为压力值，并发送给终端处理器。
这里，可以预先在终端压力屏上设置多个受压位置，并采用压力校准仪器对各个受压位置依次施压；终端和压力校准仪器之间可以建立有线或无线的连接，通过所述终端压力初始化校准仪器，设置校准仪器的压力。通常，对各受压位置可以采用相同的压力进行校准；
压力校准仪器根据设置的压力值开始向压力屏上的受压位置依次施压，每向一个施压位置施压，终端通过所述压力信息处理芯片读取每个压力传感器的压力值，以及压力校准仪器实际打到屏幕上的压力值；所述校准仪器实际打到屏幕上的压力值可以通过压力校准仪器的反馈系统测量并传送到终端；终端可以存储这些数据，直至压力校准仪器对所有受压位置测施压全部结束。
步骤102：采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；
可选地，采用人工神经网络来处理终端读取的每个压力传感器的压力值以及压力校准仪器实际打到屏幕上的压力值；这里，如图3所示，可以采用BP神经网络模型；将终端读取每个压力传感器的压力值作为BP神经网络模型的输入；将压力校准仪器实际打到屏幕上的压力值作为BP神经网络模型的输出。在图2所示的终端上设置有9个压力传感器，因此，可以设置输入层为9个节点，9个压力传感器的压力输出值分别作为每个节点的输入；以N表示压力屏校准仪器对各受压位置施压的次数，
Figure PCTCN2016106932-appb-000001
表示第一个压力传感器在第n次施压时感知的压力值，
Figure PCTCN2016106932-appb-000002
表示第二个压力传感器在第n次施压时感知的压力值，以此类推，
Figure PCTCN2016106932-appb-000003
表示第九个压力传感器在第n次施压时感知的压力值；其中，n∈[1,N]；输出层包含1个节点，以pn表示第n次打点时压力校准仪器实际打到屏幕上的压力值，该值即为BP神经网络模型第n组向量的期望输出；
隐含层中神经元数目可以采用经验公式，隐含层神经元个数的经验公式可以用表达式(1)表示：
Figure PCTCN2016106932-appb-000004
其中，l为隐含层神经元个数，ni为输入层神经元个数，m为输出层神经元个数，α为[1，10]之间的常数；根据表达式(1)可以计算出神经元个数为5-14个之间；计算中可以选择隐含层神经元个数为6；这里，可以预设BP神经网络模型的训练次数和训练误差，确定达到所述预设训练误差要求时所述BP神经网络模型的权值参数；BP神经网络模型可以用epochs表示BP网络的训练次数、可以用goal表示BP网络的训练误差；
根据图2的终端，进行N次的压力屏施压后，BP神经网络模型的训练数据可以用表达式(2)和表达式(3)表示：
Figure PCTCN2016106932-appb-000005
[p1 p2 p3 … pn … pN]              (3)
其中，表达式(2)和表达式(3)分别表示数据库中存储的所有受压位置对应的压力传感器感应压力数据及实际打到屏幕上的压力值。
可选地，因为用户手指按压压力屏的力度是不确定的，和校准中采用的压力可能不在一个范围内，这样校准的神经网络就无法适应不同的用户按压力度；可以采用归一化处理获取的压力，这样，无论是校准中使用的压力还是用户实际的按压力度，在归一化后都能落到统一的范围中；使神经网络能够适应各种压力；这里，可以将表达式(2)和表达式(3)的数据进行归一化处理。表达式(2)和表达式(3)的数据进行归一化处理后可以分别用表达式(4)和表达式(5)表示：
Figure PCTCN2016106932-appb-000006
Figure PCTCN2016106932-appb-000007
表达式(4)中的每个列向量作为BP神经网络模型的输入，表达式(5)中的每个值分别对应式表达(4)的一个列向量，并且作为其期望输出。
在选择BP神经网络模型的训练中，可以选择S型正切函数tansig作为隐含层神经元的激励函数，选取纯线性函数purelin作为输出层函数；所述S型正切函数可以用表达式(6)表示，所述纯线性函数可以用表达式(7)表示：
y＝tansig(x)＝2/(1+exp(-2*x))-1           (6)
y＝purelin(x)＝x          (7)
其中，x表示BP神经网络模型的输入，y表示BP神经网络模型的输出。
可以根据设定：ni＝9、m＝1，确定l＝6，并取epochs＝50，goal＝0.00001；利用表达式(4)和表达式(5)中的归一化数据开始训练建立的BP神经网络模型，其中，表达式(4)作为人工神经网络模型的输入，表达式(5)作为BP神经网络模型的期望输出；当训练误差达到goal以下时停止训练，记录BP神经网络模型的所有权值参数。权值参数确定后，所述BP神经网络模型即为已经完成训练的BP神经网络模型；表明压力屏的校准已经完成。
后续压力屏的使用过程中，当用户按压终端压力屏时，读取每个压力传感器的压力值并做归一化处理，将该归一化数据作为已训练的BP神经网络模型输入，能得到一个归一化输出值，对该输出值做反归一化处理即可得到用户按压终端屏幕的压力值。
实际应用中，针对不同压力屏中不同数量的压力传感器，校准过程中，只需调整BP神经网络模型的输入节点的个数，并且BP神经网络模型几乎不受压力传感器布置方式的影响；通过在训练BP神经网络模型时goal的调整，能够对校准的精度进行调整，获取用户需求的精度，提升了用户体验。
本发明实施例提供的压力屏校准装置，主要目的是要建立一个人工神经网络模型，所述人工神经网络模型可以设置为表示终端上各压力感应器上感应的压力值和实际受压位置承受压力值的关系；后续使用中，根据各压力感应器上感应的压力值，通过所述人工神经网络模型能够准确确定实际受压位置承受的压力值；如图4所示，所述装置包括：获取模块41和确定模块42；其中，
所述获取模块41，设置为获取各压力传感器在压力屏各预设受压位 置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；
通常，终端压力屏下设置有多个压力传感器，设置为感应压力屏受到的压力信号，压力传感器信号将压力信号传送到压力信息处理芯片；如图2所示的终端压力屏下方布置9个压力传感器，从左上角至右下角依次编号为：1号、2号、……9号，屏幕右下角为压力传感器信号的压力信息处理芯片，该压力信息处理芯片将压力传感器发送压力信号转量化为压力值，并发送给终端处理器。
这里，可以预先在终端压力屏上设置多个受压位置，并采用压力校准仪器对各个受压位置依次施压；终端和压力校准仪器之间可以建立有线或无线的连接，通过所述终端压力初始化校准仪器，设置校准仪器的压力。通常，对各受压位置可以采用相同的压力进行校准。
压力校准仪器根据设置的压力值开始向压力屏上的受压位置依次施压，每向一个施压位置施压，终端通过所述压力信息处理芯片读取每个压力传感器的压力值，以及压力校准仪器实际打到屏幕上的压力值；所述校准仪器实际打到屏幕上的压力值可以通过压力校准仪器的反馈系统测量并传送到终端；终端可以存储这些数据，直至压力校准仪器对所有受压位置测施压全部结束。
所述确定模块42，设置为采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；
可选地，采用人工神经网络来处理终端读取的每个压力传感器的压力值以及压力校准仪器实际打到屏幕上的压力值；这里，可以采用BP神经网络模型；将终端读取每个压力传感器的压力值作为BP神经网络模型的输入；将压力校准仪器实际打到屏幕上的压力值作为BP神经网络模型的输出。在图2所示的终端上设置有9个压力传感器，因此，可以设置输 入层为9个节点，9个压力传感器的压力输出值分别作为每个节点的输入；以N表示压力屏校准仪器对各受压位置施压的次数，
Figure PCTCN2016106932-appb-000008
表示第一个压力传感器在第n次施压时感知的压力值，
Figure PCTCN2016106932-appb-000009
表示第二个压力传感器在第n次施压时感知的压力值，以此类推，
Figure PCTCN2016106932-appb-000010
表示第九个压力传感器在第n次施压时感知的压力值；其中，n∈[1,N]；输出层包含1个节点，以pn表示第n次打点时压力校准仪器实际打到屏幕上的压力值，该值即为BP神经网络模型第n组向量的期望输出。
隐含层中神经元数目可以采用经验公式，隐含层神经元个数的经验公式可以用表达式(1)表示；其中，l为隐含层神经元个数，ni为输入层神经元个数，m为输出层神经元个数，α为[1，10]之间的常数；根据表达式(1)可以计算出神经元个数为5-14个之间；计算中可以选择隐含层神经元个数为6；这里，可以预设BP神经网络模型的训练次数和训练误差确定达到所述预设训练误差要求时所述BP神经网络模型的权值参数。；BP神经网络模型可以用epochs表示BP网络的训练次数、可以用goal表示BP网络的训练误差。
根据图2的终端，进行N次的压力屏施压后，BP神经网络模型的训练数据可以用表达式(2)和表达式(3)表示；其中，表达式(2)和表达式(3)分别表示数据库中存储的所有受压位置对应的压力传感器感应压力数据及实际打到屏幕上的压力值。
可选地，因为用户手指按压压力屏的力度是不确定的，和校准中采用的压力可能不在一个范围内，这样校准的神经网络就无法适应不同的用户按压力度；可以采用归一化处理获取的压力，这样，无论是校准中使用的压力还是用户实际的按压力度，在归一化后都能落到统一的范围中；使神经网络能够适应各种压力；这里，可以将表达式(2)和表达式(3)的数据进行归一化处理。表达式(2)和表达式(3)的数据进行归一化处理后可以分别用表达式(4)和表达式(5)表示；表达式(4)中的每个列向量作为BP神经网络模型的输入，表达式(5)中的每个值分别对应式表达(4)的一个列向量，并且作为其期望输出。
在选择BP神经网络模型的训练中，可以选择S型正切函数tansig作为隐含层神经元的激励函数，选取纯线性函数purelin作为输出层函数；所述S型正切函数可以用表达式(6)表示，所述纯线性函数可以用表达式(7)表示；其中，x表示BP神经网络模型的输入，y表示BP神经网络模型的输出；
可以根据设定：ni＝9、m＝1，确定l＝6，并取epochs＝50，goal＝0.00001；利用表达式(4)和表达式(5)中的归一化数据开始训练建立的BP神经网络模型，其中，表达式(4)作为人工神经网络模型的输入，表达式(5)作为BP神经网络模型的期望输出；当训练误差达到goal以下时停止训练，记录BP神经网络模型的所有权值参数。权值参数确定后，所述BP神经网络模型即为已经完成训练的BP神经网络模型；表明压力屏的校准已经完成。
后续压力屏的使用过程中，当用户按压终端压力屏时，读取每个压力传感器的压力值并做归一化处理，将该归一化数据作为已训练的BP神经网络模型输入，能得到一个归一化输出值，对该输出值做反归一化处理即可得到用户按压终端屏幕的压力值。
实际应用中，针对不同压力屏中不同数量的压力传感器，校准过程中，只需调整BP神经网络模型的输入节点的个数，并且BP神经网络模型几乎不受压力传感器布置方式的影响；通过在训练BP神经网络模型时goal的调整，能够对校准的精度进行调整，获取用户需求的精度，提升了用户体验。
在实际应用中，所述获取模块41和确定模块42由终端的中央处理器(CPU)、微处理器(MPU)、数字信号处理器(DSP)、或现场可编程门阵列(FPGA)等组合实现。
本发明的实施例还提供了一种存储介质。可选地，在本实施例中，上述存储介质中存储有执行指令，该执行指令用于执行上述方法实施例中的步骤之一或其组合。
可选地，在本实施例中，上述存储介质可以包括但不限于：U盘、只读存储器(Read-Only Memory，简称为ROM)、随机存取存储器(Random Access Memory，简称为RAM)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本发明的最佳实施例而已，并非用于限定本发明的保护范围，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。
工业实用性
如上所述，本发明实施例提供的一种压力屏校准方法和装置具有以下有益效果：获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；用户实际使用压力屏时，采用所述确定权值参数的人工神经网络模型，根据各压力传感器感应的压力值，确定所述压力屏受压位置的压力值；进一步，对于不同压力传感器数量及布置方式，只需要调整人工神经网络模型的神经元节点数即可适应；如此，能有效适应不同压力传感器数量及布置方式，并能提供优良的校准精度，提升用户体验。

Claims (10)

  1. 一种压力屏校准方法，所述方法包括：
    获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；
    采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；
    所述确定权值参数的人工神经网络模型，设置为根据各压力传感器感应到的压力值确定所述压力屏受到的压力值。
  2. 根据权利要求1所述的方法，其中，所述采用人工神经网络模型，包括：采用反向传播BP神经网络模型；
    所述方法还包括：预设训练误差，确定达到所述预设训练误差要求时所述BP神经网络模型的权值参数。
  3. 根据权利要求2所述的方法，其中，所述方法还包括：选择S型正切函数作为所述BP神经网络模型的隐含层神经元的激励函数，选择纯线性函数作为所述BP神经网络模型的输出层函数。
  4. 根据权利要求1至3任一项所述的方法，其中，所述将所述各压力传感器感应到的压力值作为人工神经网络模型的输入之前，所述方法还包括：对所述获取的所述各压力传感器感应到的压力值进行数据归一化处理；
    所述在将所述实际施加的压力值作为所述人工神经网络模型的输出之前，所述方法还包括：对所述获取的实际在所述各预设受压位置施加的压力值进行数据归一化处理。
  5. 根据权利要求1至3任一项所述的方法，其中，所述获取实际在所述各预设受压位置施加的压力值，包括：
    采用外部装置测量各预设受压位置受压的压力值。
  6. 根据权利要求4所述的方法，其中，所述基于确定权值参数的 人工神经网络模型，根据各压力传感器感应到的压力值确定所述压力屏受到的压力值，包括：
    读取当前各压力传感器的压力值，并作为所述确定权值参数的人工神经网络模型输入，将获取输出值确定为当前压力屏受到的压力值；
    所述方法还包括：所述将当前各压力传感器的压力值作为所述确定权值参数的人工神经网络模型输入之前，对所述当前各压力传感器的压力值作归一化处理；对所述输出值作反归一化处理，确定为当前压力屏受到的压力值。
  7. 一种压力屏校准装置，所述装置包括：获取模块和确定模块；其中，
    所述获取模块，设置为获取各压力传感器在压力屏各预设受压位置施压时感应到的压力值，并获取实际在所述各预设受压位置施加的压力值；
    所述确定模块，设置为采用人工神经网络模型，将所述各压力传感器感应到的压力值作为人工神经网络模型的输入，将所述实际施加的压力值作为所述人工神经网络模型的输出，确定所述人工神经网络模型的权值参数；
    所述确定权值参数的人工神经网络模型，设置为根据各压力传感器感应到的压力值确定所述压力屏受到的压力值。
  8. 根据权利要求7所述的装置，其中，所述确定模块，所述人工神经网络模型包括：BP神经网络模型；
    所述确定模块，还设置为：预设训练误差，确定达到所述预设训练误差要求时所述BP神经网络模型的权值参数。
  9. 根据权利要求7或8所述的装置，其中，所述获取模块，设置为获取外部装置测量各预设受压位置受压的压力值。
  10. 根据权利要求7或8所述的装置，其中，所述确定模块，还 设置为：
    在将所述各压力传感器感应到的压力值作为人工神经网络模型的输入之前，对所述获取的所述各压力传感器感应到的压力值进行数据归一化处理；
    在将所述实际施加的压力值作为所述人工神经网络模型的输出之前，对所述获取的实际在所述各预设受压位置施加的压力值进行数据归一化处理。
PCT/CN2016/106932 2016-05-23 2016-11-23 一种压力屏校准方法和装置 WO2017201992A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610344997.1A CN107422891A (zh) 2016-05-23 2016-05-23 一种压力屏校准方法和装置
CN201610344997.1 2016-05-23

Publications (1)

Publication Number Publication Date
WO2017201992A1 true WO2017201992A1 (zh) 2017-11-30

Family

ID=60411052

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/106932 WO2017201992A1 (zh) 2016-05-23 2016-11-23 一种压力屏校准方法和装置

Country Status (2)

Country Link
CN (1) CN107422891A (zh)
WO (1) WO2017201992A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111176473A (zh) * 2018-11-12 2020-05-19 北京钛方科技有限责任公司 触控板的按压力识别方法及系统
WO2021037189A1 (zh) * 2019-08-28 2021-03-04 北京钛方科技有限责任公司 校准方法、装置及存储介质
CN113419646A (zh) * 2021-06-03 2021-09-21 维沃移动通信有限公司 压力感应通道的校准方法、装置及电子设备
CN111176473B (zh) * 2018-11-12 2024-05-24 北京钛方科技有限责任公司 触控板的按压力识别方法及系统

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110554792B (zh) * 2018-06-04 2021-10-19 北京钛方科技有限责任公司 传感器校准调节装置及方法
CN110442262A (zh) * 2019-07-22 2019-11-12 珠海格力电器股份有限公司 基于触碰压力的工控触摸屏实时校准方法、装置及设备
WO2022126468A1 (zh) * 2020-12-17 2022-06-23 深圳市汇顶科技股份有限公司 压力校准方法、测试机、触控芯片和触摸板
CN112558791B (zh) * 2020-12-17 2022-11-01 深圳市汇顶科技股份有限公司 压力校准方法、测试机、触控芯片和触摸板
CN114224321B (zh) * 2021-11-01 2024-05-10 福建世新机器人科技有限公司 一种基于电容性压力传感阵列的足底压力测量方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090066663A1 (en) * 2007-09-11 2009-03-12 Samsung Electronics Co., Ltd. Apparatus and method for controlling operation of mobile terminal
CN102521611A (zh) * 2011-12-13 2012-06-27 广东威创视讯科技股份有限公司 一种基于触摸屏的触摸物识别方法
CN105278846A (zh) * 2015-05-14 2016-01-27 维沃移动通信有限公司 控制移动终端屏幕方向的方法及其移动终端
CN105279405A (zh) * 2015-10-28 2016-01-27 同济大学 触屏用户按键行为模式构建与分析系统及其身份识别方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102141455B (zh) * 2010-12-10 2014-06-25 北京航空航天大学 一种非介入式压力测量方法
CN102759430B (zh) * 2012-06-28 2014-08-20 北京自动化控制设备研究所 基于bp神经网络的谐振筒压力传感器高精度校试方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090066663A1 (en) * 2007-09-11 2009-03-12 Samsung Electronics Co., Ltd. Apparatus and method for controlling operation of mobile terminal
CN102521611A (zh) * 2011-12-13 2012-06-27 广东威创视讯科技股份有限公司 一种基于触摸屏的触摸物识别方法
CN105278846A (zh) * 2015-05-14 2016-01-27 维沃移动通信有限公司 控制移动终端屏幕方向的方法及其移动终端
CN105279405A (zh) * 2015-10-28 2016-01-27 同济大学 触屏用户按键行为模式构建与分析系统及其身份识别方法

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111176473A (zh) * 2018-11-12 2020-05-19 北京钛方科技有限责任公司 触控板的按压力识别方法及系统
CN111176473B (zh) * 2018-11-12 2024-05-24 北京钛方科技有限责任公司 触控板的按压力识别方法及系统
WO2021037189A1 (zh) * 2019-08-28 2021-03-04 北京钛方科技有限责任公司 校准方法、装置及存储介质
CN112445357A (zh) * 2019-08-28 2021-03-05 北京钛方科技有限责任公司 校准方法、装置及存储介质
CN112445357B (zh) * 2019-08-28 2022-06-14 北京钛方科技有限责任公司 校准方法、装置及存储介质
CN113419646A (zh) * 2021-06-03 2021-09-21 维沃移动通信有限公司 压力感应通道的校准方法、装置及电子设备
CN113419646B (zh) * 2021-06-03 2023-05-16 维沃移动通信有限公司 压力感应通道的校准方法、装置及电子设备

Also Published As

Publication number Publication date
CN107422891A (zh) 2017-12-01

Similar Documents

Publication Publication Date Title
WO2017201992A1 (zh) 一种压力屏校准方法和装置
US9446287B2 (en) Sensor-based athletic activity measurements
RU2510071C2 (ru) Устройство для тестирования личности
JP6586818B2 (ja) 診療支援装置、診療支援方法、診療支援プログラム、生体情報測定装置
KR102043687B1 (ko) 음성 인식 기반의 열화상 카메라를 이용하는 스마트 미러 장치 및 그 방법
JP2009531709A (ja) 感圧センサアレイの制御回路及び関連する方法
Wainwright et al. The validity and reliability of a sample of 10 Wattbike cycle ergometers
AU2019320700A1 (en) Method and device for self-measurement of intra-ocular pressure
TW201514905A (zh) 肥胖基因檢測評估系統及資料處理方法
US20180126219A1 (en) Methods and apparatuses for handgrip strength assessment using pressure-sensitive elements
CN107890342A (zh) 执行受检者测量
KR102460802B1 (ko) 센싱 장치, 센싱데이터의 처리 방법 및 프로그램
US20220092300A1 (en) Display apparatus and method for controlling thereof
WO2021047376A1 (zh) 数据处理方法、数据处理装置及相关设备
Lemerle et al. Application of pressure mapping techniques to measure push and gripping forces with precision
US10463915B2 (en) System, method and program product for monitoring the use of an exercise ball by a user and exercise ball which can be used in such a system
Kumar et al. Accuracy of oscillometric-based blood pressure monitoring devices: impact of pulse volume, arrhythmia, and respiratory artifact
US11125635B2 (en) Sensor array for consolidated force measurement
KR102390599B1 (ko) 내면 집중력 훈련 방법 및 장치
WO2020008380A1 (en) Method for carrying out oxymetric tests and medical apparatus with an improved oxymetric device with a reflective sensor
TWI580404B (zh) 肌肉張力感測方法及系統
WO2022042200A1 (zh) 人体成分检测的方法、系统、电子设备及存储介质
JP2018149081A (ja) 情報処理装置、情報処理方法及びプログラム
US10753784B2 (en) Fluid-filled personal scale responsive to temperature and pressure
CN108426628A (zh) 一种基于站立方向的智能体重秤

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16902976

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 16902976

Country of ref document: EP

Kind code of ref document: A1