CN116578227B - Screen control method, device and equipment of intelligent watch and storage medium - Google Patents

Screen control method, device and equipment of intelligent watch and storage medium Download PDF

Info

Publication number
CN116578227B
CN116578227B CN202310833555.3A CN202310833555A CN116578227B CN 116578227 B CN116578227 B CN 116578227B CN 202310833555 A CN202310833555 A CN 202310833555A CN 116578227 B CN116578227 B CN 116578227B
Authority
CN
China
Prior art keywords
subset
screen control
control model
value
preset
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.)
Active
Application number
CN202310833555.3A
Other languages
Chinese (zh)
Other versions
CN116578227A (en
Inventor
黄艳锋
简胜奇
陈国平
鲍亮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Yisai Communication Technology Co ltd
Original Assignee
Shenzhen Yisai Communication Technology Co ltd
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 Shenzhen Yisai Communication Technology Co ltd filed Critical Shenzhen Yisai Communication Technology Co ltd
Priority to CN202310833555.3A priority Critical patent/CN116578227B/en
Publication of CN116578227A publication Critical patent/CN116578227A/en
Application granted granted Critical
Publication of CN116578227B publication Critical patent/CN116578227B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0487—Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
    • G—PHYSICS
    • G04—HOROLOGY
    • G04G—ELECTRONIC TIME-PIECES
    • G04G9/00—Visual time or date indication 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
    • G06F3/033—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor
    • G06F3/0346—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor with detection of the device orientation or free movement in a 3D space, e.g. 3D mice, 6-DOF [six degrees of freedom] pointers using gyroscopes, accelerometers or tilt-sensors
    • 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
    • G06F3/048—Interaction techniques based on graphical user interfaces [GUI]
    • G06F3/0484—Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Abstract

The method comprises the steps of obtaining a plurality of third sampling values collected by the triaxial acceleration sensor in a second preset time period when the state of a wrist of a user in the first moment is inconsistent with the state in the first preset time period, inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model to judge whether the screen of the intelligent watch needs to be lightened at the current moment, and realizing intelligent control on the screen of the intelligent watch so as to improve the use experience of the user; the first preset time period is a time period connected with and before the first time, the second preset time period is a time period between the first time and a second time, and the second time is a current time.

Description

Screen control method, device and equipment of intelligent watch and storage medium
Technical Field
The application relates to the field of screen control of intelligent watches, in particular to a screen control method, device and equipment of an intelligent watch and a storage medium.
Background
Watches are common wear in people's daily lives and are typically worn on the wrist for timing. With the development of mobile device technology, the functions of the watch are also becoming diversified, and a product of a smart watch is generated. Compared with a common watch with only a timing function, the intelligent watch also has one or more functions of reminding, navigation, calibration, heart rate detection, interaction, connection with a mobile phone, step counting and the like, and has an electronic touch display screen.
Due to the increase of functions, the energy consumption of the smart watch is also increased more than that of the traditional watch. In order to save power and avoid false touches, the smart watch is often set to automatically close the screen without operating the watch, and when the user wants to watch the screen each time, the user needs to touch or press a key to wake up the screen, which is inconvenient in some situations. Thus, a method is needed to solve this problem.
Disclosure of Invention
The application provides a screen control method, device, equipment and storage medium of an intelligent watch, and aims to realize intelligent control on a screen of the intelligent watch and improve the use experience of a user.
In a first aspect, the present application provides a screen control method of a smart watch, including:
When the screen state of the intelligent watch is in an extinction state, acquiring a plurality of first sampling values acquired by the triaxial acceleration sensor in a first preset time period and second sampling values acquired at a first moment, and inputting the plurality of first sampling values into a preset state detection model to acquire the state of the wrist of a user in the first preset time period; the wrist of the user wears the intelligent watch, and the first preset time period is a time period which is connected with the first moment and is before the first moment;
judging whether the state of the wrist of the user at the first moment is consistent with the state in the first preset time period or not according to the plurality of first sampling values and the second sampling values;
if the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period, acquiring a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period; the second preset time period is a duration between the first time and a second time, and the second time is a current time;
inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model to judge whether the screen of the intelligent watch needs to be lightened at the current moment;
And if the screen of the intelligent watch needs to be lightened at the current moment, the screen of the intelligent watch is lightened.
Further, the determining, according to the plurality of first sampling values and the plurality of second sampling values, whether the state of the wrist of the user at the first moment is consistent with the state in the first preset time period includes:
calculating a first standard deviation, a second standard deviation and a third standard deviation according to the plurality of first sampling values and the second sampling values; the first standard deviation is a standard deviation of acceleration values in a first direction in the plurality of first sampling values and the second sampling values, the second standard deviation is a standard deviation of acceleration values in a second direction in the plurality of first sampling values and the second sampling values, and the third standard deviation is a standard deviation of acceleration values in a third direction in the plurality of first sampling values and the second sampling values;
acquiring a first preset standard deviation, a second preset standard deviation and a third preset standard deviation according to the state of the wrist of the user in the first preset time period; wherein the first preset standard deviation is a maximum value of standard deviations of acceleration values in the first direction that match a state of the wrist of the user within the first preset time period, the second preset standard deviation is a maximum value of standard deviations of acceleration values in the second direction that match a state of the wrist of the user within the first preset time period, and the third preset standard deviation is a maximum value of standard deviations of acceleration values in the third direction that match a state of the wrist of the user within the first preset time period;
Comparing the first standard deviation with the first preset standard deviation, comparing the second standard deviation with the second preset standard deviation, and comparing the third standard deviation with the third preset standard deviation;
if at least one of the first standard deviation is greater than the first preset standard deviation, the second standard deviation is greater than the second preset standard deviation and the third standard deviation is greater than the third preset standard deviation is met, determining that the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period.
Further, the training method of the preset screen control model comprises the following steps:
acquiring a training sample set, wherein the training sample set comprises a plurality of mapping relations, and the mapping relations are mapping relations between a state of the wrist of an experimenter and a triaxial acceleration value of the wrist of the experimenter, which is acquired when the wrist of the experimenter deviates from the state of the wrist of the experimenter;
labeling a plurality of mapping relations in the training sample set respectively to distinguish positive samples and negative samples, wherein the positive samples are the reasons for the state that the wrist of the experimenter deviates from the wrist of the experimenter are that the mapping relations matched with the screen of the intelligent watch are checked;
Constructing a convolutional neural network model, wherein the convolutional neural network model sequentially comprises four parallel branches, a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first full-connection layer, a tanh activation layer, a second full-connection layer and a sigmoid activation layer;
and training the convolutional neural network model based on the marked training sample set to obtain the screen control model.
Further, the training the convolutional neural network model based on the labeled training sample set to obtain the screen control model includes:
randomly dividing the training sample set into a first subset, a second subset, a third subset, a fourth subset and a fifth subset in a hierarchical sampling mode;
training the convolutional neural network model by using the first subset, the second subset, the third subset and the fourth subset to obtain a first initial screen control model, and verifying the first initial screen control model by using the fifth subset to obtain a first accuracy rate;
training the first initial screen control model by using the first subset, the second subset, the third subset and the fifth subset to obtain a second initial screen control model, and verifying the second initial screen control model by using the fourth subset to obtain a second accuracy rate;
Training the second initial screen control model by using the first subset, the second subset, the fourth subset and the fifth subset to obtain a third initial screen control model, and verifying the third initial screen control model by using the third subset to obtain a third accuracy rate;
training the third initial screen control model by using the first subset, the third subset, the fourth subset and the fifth subset to obtain a fourth initial screen control model, and verifying the fourth initial screen control model by using the second subset to obtain a fourth accuracy rate;
training the fourth initial screen control model by using the second subset, the third subset, the fourth subset and the fifth subset to obtain a fifth initial screen control model, and verifying the fifth initial screen control model by using the first subset to obtain a fifth accuracy rate;
calculating average values of the first accuracy rate, the second accuracy rate, the third accuracy rate, the fourth accuracy rate and the fifth accuracy rate, and comparing the average values with preset average values;
If the average value is not smaller than the preset average value, respectively calculating average values of corresponding parameters in the first initial screen control model, the second initial screen control model, the third initial screen control model, the fourth initial screen control model and the fifth initial screen control model;
generating a screen control model to be optimized by using all the average values;
constructing a two-class cross entropy loss function;
and optimizing the screen control model to be optimized by using the two-classification cross entropy loss function to obtain the screen control model.
Further, the method further comprises:
the screen of the intelligent watch is lightened, a fourth sampling value acquired by the triaxial acceleration sensor is acquired at the same time, and a fifth sampling value acquired by the acceleration sensor is acquired after the fourth sampling value is acquired;
calculating a first absolute value, a second absolute value and a third absolute value by using the fourth sampling value and the fifth sampling value; wherein the first absolute value is an absolute value of a difference between an acceleration value in a first direction in the fourth sample value and an acceleration value in a first direction in the fifth sample value, the second absolute value is an absolute value of a difference between an acceleration value in a second direction in the fourth sample value and an acceleration value in a second direction in the fifth sample value, and the third absolute value is an absolute value of a difference between an acceleration value in a third direction in the fourth sample value and an acceleration value in a third direction in the fifth sample value;
Comparing the first absolute value, the second absolute value and the third absolute value with preset absolute values respectively;
and if at least one of the first absolute value, the second absolute value and the third absolute value is larger than the preset absolute value, extinguishing the screen of the intelligent watch.
In a second aspect, the present application provides a screen control device of a smart watch, the smart watch being provided with a triaxial acceleration sensor, the screen control device comprising:
the first acquisition module is used for acquiring a plurality of first sampling values acquired by the triaxial acceleration sensor in a first preset time period and second sampling values acquired at the current moment when the screen state of the intelligent watch is in an off state, and inputting the plurality of first sampling values into a preset state detection model so as to acquire the state of the wrist of a user in the first preset time period; the wrist of the user wears the intelligent watch, and the first preset time period is a time period which is connected with the current moment and is before the current moment;
the judging module is used for judging whether the current state of the wrist of the user is consistent with the state in the first preset time period or not according to the plurality of first sampling values and the second sampling values;
The second acquisition module is used for acquiring a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period if the current state of the wrist of the user is inconsistent with the state of the wrist in the first preset time period; the second preset time period is a time period which is connected with the current time and is after the current time;
the input module is used for inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model so as to judge whether the screen of the intelligent watch needs to be lightened;
and the execution module is used for lighting the screen of the intelligent watch if the screen of the intelligent watch is required to be lighted.
In a third aspect, the present application provides a terminal device comprising a processor, a memory and a computer program stored on the memory and executable by the processor, wherein the computer program, when executed by the processor, implements a screen control method of any one of the smart watches as described above.
In a fourth aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a screen control method of any of the smartwatches described above.
The application discloses a screen control method, a device, equipment and a storage medium of an intelligent watch, wherein the screen control method of the intelligent watch obtains a plurality of third sampling values collected by a triaxial acceleration sensor in a second preset time period when the state of a wrist of a user in the first moment is inconsistent with the state in the first preset time period, inputs the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model so as to judge whether the screen of the intelligent watch needs to be lightened at the current moment, and lightens the screen of the intelligent watch if the screen of the intelligent watch needs to be lightened at the current moment, so that intelligent control on the screen of the intelligent watch is realized to improve the use experience of the user.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a screen control method of a smart watch according to an embodiment of the present application;
fig. 2 is a schematic block diagram of a screen control device of a smart watch according to an embodiment of the present application;
fig. 3 is a schematic block diagram of a structure of a terminal device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The flow diagrams depicted in the figures are merely illustrative and not necessarily all of the elements and operations/steps are included or performed in the order described. For example, some operations/steps may be further divided, combined, or partially combined, so that the order of actual execution may be changed according to actual situations.
It is also to be understood that the terminology used in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Watches are common wear in people's daily lives and are typically worn on the wrist for timing. With the development of mobile device technology, the functions of the watch are also becoming diversified, and a product of a smart watch is generated. Compared with a common watch with only a timing function, the intelligent watch also has one or more functions of reminding, navigation, calibration, heart rate detection, interaction, connection with a mobile phone, step counting and the like, and has an electronic touch display screen.
Due to the increase of functions, the energy consumption of the smart watch is also increased more than that of the traditional watch. In order to save power and avoid false touches, the smart watch is often set to automatically close the screen without operating the watch, and when the user wants to watch the screen each time, the user needs to touch or press a key to wake up the screen, which is inconvenient in some situations. Therefore, the embodiment of the application provides a screen control method, device and equipment of a smart watch and a storage medium, so as to solve the problems.
Some embodiments of the present application are described in detail below with reference to the accompanying drawings. The following embodiments and features of the embodiments may be combined with each other without conflict.
Referring to fig. 1, fig. 1 is a flowchart of a screen control method of a smart watch according to an embodiment of the present application, and as shown in fig. 1, the flowchart of the screen control method of a smart watch according to an embodiment of the present application includes steps S100 to S500.
Step S100, when the screen state of the intelligent watch is in an extinction state, acquiring a plurality of first sampling values acquired by the triaxial acceleration sensor in a first preset time period and a second sampling value acquired at a first moment, and inputting the plurality of first sampling values into a preset state detection model to acquire the state of the wrist of a user in the first preset time period; the smart watch is worn on the wrist of the user, and the first preset time period is a time period which is connected with the first time and is before the first time.
It will be appreciated that the state of the user's wrist may take on a plurality of states in a short time, and that to improve the accuracy of the method, the first preset time period is set to a short time period, preferably not less than 2 seconds.
The state of the wrist of the user in the first preset time period comprises the state that the wrist is placed on a table, the wrist is placed on a thigh when the user sits, the wrist is backed by the waist when the user types, the state that the wrist is placed vertically when the user stands, the state of the wrist when the user walks, the state of the wrist when the user runs, the state that the wrist is placed when the user sleeps, the state of the wrist when the user sleeps and turns over, and the like.
The preset state detection model is obtained through convolutional neural network training.
And step 200, judging whether the state of the wrist of the user at the first moment is consistent with the state in the first preset time period or not according to the plurality of first sampling values and the second sampling values.
Step S300, if the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period, acquiring a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period; the second preset time period is a duration between the first time and a second time, and the second time is a current time.
It will be appreciated that if the state of the user's wrist at the first time does not coincide with the state of the wrist in the first preset time period, it is stated that, with respect to the first preset time period, the state of the wrist has suddenly changed at the first time period, which may be caused by the user's need to view the screen of the smart watch, or may be caused by the wrist changing from one state to another state, and therefore, it is necessary to acquire the triaxial acceleration of the wrist in the second preset time period after the first time period to determine whether the sudden change occurs due to the wrist needing to view the screen of the smart watch, and that, regardless of the state of the wrist, the time required for the user to move the wrist from the first time period to the time when the user can view the screen is generally greater than 0.5 seconds and less than 1.5 seconds, and therefore, the second preset time period may be set to between 0.5 seconds and 1.5 seconds.
Step 400, inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model to judge whether the screen of the intelligent watch needs to be lightened at the current moment.
It is understood that, when the states of the wrist within the first preset period are different, the plurality of third sampling values acquired by the triaxial acceleration are different even if the reason why the abrupt change of the wrist occurs is due to the user viewing the screen.
And S500, if the screen of the intelligent watch needs to be lightened at the current moment, the screen of the intelligent watch is lightened.
In step S500, only if the acquired plurality of third sampling values match with actual sampling values of the wrist in the second preset period when the user views the screen in the state of the wrist in the first preset period, the screen of the smart watch can be lightened.
According to the screen control method of the intelligent watch, when the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period, a plurality of third sampling values collected by the triaxial acceleration sensor in the second preset time period are obtained, the state of the user in the first preset time period and the plurality of third sampling values are input into a preset screen control model, so that whether the screen of the intelligent watch needs to be lightened at the current moment or not is judged, and if the screen of the intelligent watch needs to be lightened at the current moment, the screen of the intelligent watch is lightened, intelligent control on the screen of the intelligent watch is achieved, and the use experience of the user is improved.
In some embodiments, after step S500, the method further comprises steps S600 to S900.
And S600, acquiring a fourth sampling value acquired by the triaxial acceleration sensor while lighting the screen of the intelligent watch, and acquiring a fifth sampling value acquired by the acceleration sensor after acquiring the fourth sampling value.
It can be understood that the fourth sampling value is a triaxial acceleration value acquired by the triaxial acceleration sensor when the user views the screen.
Step S700, calculating a first absolute value, a second absolute value and a third absolute value by using the fourth sampling value and the fifth sampling value; wherein the first absolute value is an absolute value of a difference between an acceleration value in a first direction in the fourth sampling value and an acceleration value in a first direction in the fifth sampling value, the second absolute value is an absolute value of a difference between an acceleration value in a second direction in the fourth sampling value and an acceleration value in a second direction in the fifth sampling value, and the third absolute value is an absolute value of a difference between an acceleration value in a third direction in the fourth sampling value and an acceleration value in a third direction in the fifth sampling value.
Step S800, comparing the first absolute value, the second absolute value and the third absolute value with preset absolute values respectively.
And step 900, extinguishing the screen of the smart watch if at least one of the first absolute value, the second absolute value and the third absolute value is larger than the preset absolute value.
It will be appreciated that, after the third sampling value is acquired by the three-axis acceleration sensor, there may be a case where the fifth sampling value is acquired multiple times, and each time the third sampling value is acquired by the three-axis acceleration sensor, steps S700 to S800 are performed once, and only when the fifth sampling value satisfies the condition in step S900, step S900 is performed once. It may also be appreciated that when the fifth sampling value satisfies the condition in step S900 at a time, it indicates that the user does not view the screen of the smart watch any more, and in order to save the power consumption of the smart watch, when the user does not view the screen of the smart watch any more, the screen of the smart watch is extinguished.
In some embodiments, step S200 includes steps S201 to S204.
Step S201, calculating a first standard deviation, a second standard deviation and a third standard deviation according to a plurality of first sampling values and second sampling values; the first standard deviation is a standard deviation of acceleration values in a first direction in the plurality of first sampling values and the second sampling values, the second standard deviation is a standard deviation of acceleration values in a second direction in the plurality of first sampling values and the second sampling values, and the third standard deviation is a standard deviation of acceleration values in a third direction in the plurality of first sampling values and the second sampling values.
The calculation method of the standard deviation is a prior art, and is not described herein.
Step S202, acquiring a first preset standard deviation, a second preset standard deviation and a third preset standard deviation according to the state of the wrist of the user in the first preset time period; the first preset standard deviation is the maximum value of the standard deviation of the acceleration value in the first direction, which is matched with the state of the wrist of the user in the first preset time period, the second preset standard deviation is the maximum value of the standard deviation of the acceleration value in the second direction, which is matched with the state of the wrist of the user in the first preset time period, and the third preset standard deviation is the maximum value of the standard deviation of the acceleration value in the third direction, which is matched with the state of the wrist of the user in the first preset time period.
It will be appreciated that the first, second and third preset standard deviations depend on the state of the user's wrist within the first preset time period, that is, the first, second and third preset standard deviations are different when the state of the user's wrist within the first preset time period is different. For example, when the user places the wrist on the table in the first preset time period, the first preset standard deviation is 1m/s, the second preset standard deviation is 2m/s, the third preset standard deviation is 0.5m/s, and when the user places the wrist behind the waist in the first preset time period, the first preset standard deviation is 0.5m/s, the second preset standard deviation is 1.5m/s, and the third preset standard deviation is 0.1 m/s.
Step S203, comparing the first standard deviation with the first preset standard deviation, comparing the second standard deviation with the second preset standard deviation, and comparing the third standard deviation with the third preset standard deviation.
Step S204, if at least one of the first standard deviation is greater than the first preset standard deviation, the second standard deviation is greater than the second preset standard deviation, and the third standard deviation is greater than the third preset standard deviation is met, determining that the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period.
It is understood that if at least one of the first standard deviation is greater than the first preset standard deviation, the second standard deviation is greater than the second preset standard deviation, and the third standard deviation is greater than the third preset standard deviation, it is indicated that the state of the wrist is mutated at the first time point compared to the first preset time period.
In this embodiment, by comparing the first standard deviation with the first preset standard deviation, comparing the second standard deviation with the second preset standard deviation, and comparing the third standard deviation with the third preset standard deviation, and determining whether the state of the wrist of the user at the first time is inconsistent with the state in the first preset time period when at least one of the first standard deviation is greater than the first preset standard deviation, the second standard deviation is greater than the second preset standard deviation, and the third standard deviation is greater than the third preset standard deviation is established, it is possible to more accurately determine whether the state of the wrist of the user at the first time is inconsistent with the state in the first preset time period.
In some embodiments, the training method of the preset screen control model includes steps S1000 to S4000.
Step S1000, obtaining a training sample set, wherein the training sample set comprises a plurality of mapping relations, and the mapping relations are mapping relations between a state of the wrist of an experimenter and a triaxial acceleration value of the wrist of the experimenter, which is acquired when the wrist of the experimenter deviates from the state of the wrist of the experimenter.
It will be appreciated that the reason why the experimenter 'S wrist deviates from the state in which the experimenter' S wrist is located may be due to the experimenter 'S viewing of the screen of the smart watch during data collection, or may be due to the transition of the experimenter' S wrist from one state to another during data collection, for each state in which the experimenter 'S wrist is located, a plurality of triaxial acceleration values within a second preset time period when the experimenter' S wrist deviates from the state in which the experimenter 'S wrist is located due to the two reasons described above need to be collected, and it is noted that the second preset time period in step S1000 and the second preset time period in step S300 tend to coincide in time length, but the second preset time period in step S1000 is a time period set based on the length of time required for the experimenter' S wrist to move to the screen of the smart watch to be able to view. The data in the training sample set is collected before step S1000, and the method for collecting the data may be that a triaxial acceleration sensor is worn on the wrist of the experimenter, so that the triaxial acceleration sensor collects the data in the second preset time period in step S1000 when the wrist of the experimenter deviates from a certain state, and records the data, where multiple sets of different data need to be collected for any state where the wrist of the experimenter is located, where the data must be collected when the experimenter makes an action of viewing a screen of the smart watch.
And S2000, marking a plurality of mapping relations in the training sample set respectively to distinguish positive samples and negative samples, wherein the positive samples are the reasons that the wrist of the experimenter deviates from the state of the wrist of the experimenter are the mapping relations matched by looking up the screen of the intelligent watch.
The negative sample is in a state that the wrist of the experimenter deviates from the wrist of the experimenter, and the reason is not to look at the mapping relation matched with the screen of the intelligent watch.
Step S3000, constructing a convolutional neural network model, wherein the convolutional neural network model sequentially comprises four parallel branches, a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first full-connection layer, a tanh activation layer, a second full-connection layer and a sigmoid activation layer.
And step 4000, training the convolutional neural network model based on the marked training sample set to obtain the screen control model.
According to the embodiment, the training sample set is marked to distinguish the positive sample from the negative sample, and training is performed on the convolutional neural network model based on the marked training sample set, so that the training effect of the preset screen control model can be improved, and the control effect of the preset screen control model is better.
In some embodiments, step S4000 includes steps S4001 to S4011.
Step S4001, randomly dividing the training sample set into a first subset, a second subset, a third subset, a fourth subset and a fifth subset by adopting a hierarchical sampling mode.
Step S4002, training the convolutional neural network model by using the first subset, the second subset, the third subset, and the fourth subset to obtain a first initial screen control model, and verifying the first initial screen control model by using the fifth subset to obtain a first accuracy.
Step S4003, training the first initial screen control model by using the first subset, the second subset, the third subset, and the fifth subset to obtain a second initial screen control model, and verifying the second initial screen control model by using the fourth subset to obtain a second accuracy rate.
Step S4004, training the second initial screen control model by using the first subset, the second subset, the fourth subset, and the fifth subset to obtain a third initial screen control model, and verifying the third initial screen control model by using the third subset to obtain a third accuracy rate.
Step S4005, training the third initial screen control model by using the first subset, the third subset, the fourth subset, and the fifth subset to obtain a fourth initial screen control model, and verifying the fourth initial screen control model by using the second subset to obtain a fourth accuracy rate.
Step S4006, training the fourth initial screen control model by using the second subset, the third subset, the fourth subset and the fifth subset to obtain a fifth initial screen control model, and verifying the fifth initial screen control model by using the first subset to obtain a fifth accuracy rate.
Step S4007, calculating an average value of the first accuracy, the second accuracy, the third accuracy, the fourth accuracy, and the fifth accuracy, and comparing the average value with a preset average value.
Step S4008, if the average value is not less than the preset average value, calculating average values of corresponding parameters in the first initial screen control model, the second initial screen control model, the third initial screen control model, the fourth initial screen control model and the fifth initial screen control model, respectively.
Step S4009, generating a screen control model to be optimized by using all the average values.
Step S4010, constructing a two-class cross entropy loss function.
The two-classification cross entropy loss function is constructed based on the labeled training sample set.
And step S4011, optimizing the screen control model to be optimized by using the two-classification cross entropy loss function to obtain the screen control model.
It can be understood that, since the screen control model is a two-class model, the control effect of the screen control model obtained by optimizing the screen control model to be optimized through the two-class cross entropy loss function is better.
Referring to fig. 2, fig. 2 is a schematic block diagram of a screen control device 100 of a smart watch according to an embodiment of the present application, where the smart watch is provided with a tri-axial acceleration sensor, and as shown in fig. 2, the screen control device 100 of the smart watch includes:
the first obtaining module 110 is configured to obtain, when the screen state of the smart watch is in an off state, a plurality of first sampling values collected by the triaxial acceleration sensor in a first preset time period and a second sampling value collected at a current time, and input the plurality of first sampling values into a preset state detection model, so as to obtain a state of a wrist of a user in the first preset time period; the wrist of the user wears the intelligent watch, and the first preset time period is a time period which is connected with the current moment and is before the current moment;
The judging module 120 is configured to judge, according to the plurality of first sampling values and the second sampling values, whether a current state of the wrist of the user is consistent with a state in the first preset time period;
a second obtaining module 130, configured to obtain a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period if the current state of the wrist of the user is inconsistent with the state of the wrist in the first preset time period; the second preset time period is a time period which is connected with the current time and is after the current time;
the input module 140 is configured to input a state of the user in the first preset time period and a plurality of third sampling values into a preset screen control model, so as to determine whether to light a screen of the smart watch;
and the execution module 150 is configured to light the screen of the smart watch if the screen of the smart watch needs to be lightened.
It should be noted that, for convenience and brevity of description, specific working processes of the above-described apparatus and each module may refer to corresponding processes in the foregoing embodiment of the screen control method of the smart watch, and will not be described herein again.
The press control device 100 provided in the above-described embodiment may be implemented in the form of a computer program that can be run on the terminal apparatus 200 as shown in fig. 3.
Referring to fig. 3, fig. 3 is a schematic block diagram of a structure of a terminal device 200 according to an embodiment of the present application, where the terminal device 200 includes a processor 201 and a memory 202, and the processor 201 and the memory 202 are connected through a system bus 203, and the memory 202 may include a nonvolatile storage medium and an internal memory.
The non-volatile storage medium may store a computer program. The computer program comprises program instructions which, when executed by the processor 201, cause the processor 201 to perform any of the press control methods described above.
The processor 201 is used to provide computing and control capabilities supporting the operation of the overall terminal device 200.
The internal memory provides an environment for the execution of a computer program in a non-volatile storage medium that, when executed by the processor 201, causes the processor 201 to perform any of the smart watch screen control methods described above.
It will be appreciated by those skilled in the art that the structure shown in fig. 3 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation of the terminal device 200 related to the present application, and that a specific terminal device 200 may include more or less components than those shown in the drawings, or may combine some components, or have a different arrangement of components.
It should be appreciated that the processor 201 may be a central processing unit (Central Processing Unit, CPU), and the processor 201 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
In some embodiments, the processor 201 is configured to execute a computer program stored in the memory to implement the following steps:
when the screen state of the intelligent watch is in an extinction state, acquiring a plurality of first sampling values acquired by the triaxial acceleration sensor in a first preset time period and second sampling values acquired at a first moment, and inputting the plurality of first sampling values into a preset state detection model to acquire the state of the wrist of a user in the first preset time period; the wrist of the user wears the intelligent watch, and the first preset time period is a time period which is connected with the first moment and is before the first moment;
Judging whether the state of the wrist of the user at the first moment is consistent with the state in the first preset time period or not according to the plurality of first sampling values and the second sampling values;
if the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period, acquiring a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period; the second preset time period is a duration between the first time and a second time, and the second time is a current time;
inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model to judge whether the screen of the intelligent watch needs to be lightened at the current moment;
and if the screen of the intelligent watch needs to be lightened at the current moment, the screen of the intelligent watch is lightened.
In some embodiments, when implementing the determining, according to the plurality of first sampling values and the second sampling values, whether the state of the wrist of the user at the first moment is consistent with the state within the first preset period of time, the processor 201 is configured to implement:
Calculating a first standard deviation, a second standard deviation and a third standard deviation according to the plurality of first sampling values and the second sampling values; the first standard deviation is a standard deviation of acceleration values in a first direction in the plurality of first sampling values and the second sampling values, the second standard deviation is a standard deviation of acceleration values in a second direction in the plurality of first sampling values and the second sampling values, and the third standard deviation is a standard deviation of acceleration values in a third direction in the plurality of first sampling values and the second sampling values;
acquiring a first preset standard deviation, a second preset standard deviation and a third preset standard deviation according to the state of the wrist of the user in the first preset time period; wherein the first preset standard deviation is a maximum value of standard deviations of acceleration values in the first direction that match a state of the wrist of the user within the first preset time period, the second preset standard deviation is a maximum value of standard deviations of acceleration values in the second direction that match a state of the wrist of the user within the first preset time period, and the third preset standard deviation is a maximum value of standard deviations of acceleration values in the third direction that match a state of the wrist of the user within the first preset time period;
Comparing the first standard deviation with the first preset standard deviation, comparing the second standard deviation with the second preset standard deviation, and comparing the third standard deviation with the third preset standard deviation;
if at least one of the first standard deviation is greater than the first preset standard deviation, the second standard deviation is greater than the second preset standard deviation and the third standard deviation is greater than the third preset standard deviation is met, determining that the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period.
In some embodiments, the processor 201 is further configured to implement:
acquiring a training sample set, wherein the training sample set comprises a plurality of mapping relations, and the mapping relations are mapping relations between a state of the wrist of an experimenter and a triaxial acceleration value of the wrist of the experimenter, which is acquired when the wrist of the experimenter deviates from the state of the wrist of the experimenter;
labeling a plurality of mapping relations in the training sample set respectively to distinguish positive samples and negative samples, wherein the positive samples are the reasons for the state that the wrist of the experimenter deviates from the wrist of the experimenter are that the mapping relations matched with the screen of the intelligent watch are checked;
Constructing a convolutional neural network model, wherein the convolutional neural network model sequentially comprises four parallel branches, a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first full-connection layer, a tanh activation layer, a second full-connection layer and a sigmoid activation layer;
and training the convolutional neural network model based on the marked training sample set to obtain the screen control model.
In some embodiments, when implementing the training sample set after labeling to train the convolutional neural network model to obtain the screen control model, the processor 201 is configured to implement:
randomly dividing the training sample set into a first subset, a second subset, a third subset, a fourth subset and a fifth subset in a hierarchical sampling mode;
training the convolutional neural network model by using the first subset, the second subset, the third subset and the fourth subset to obtain a first initial screen control model, and verifying the first initial screen control model by using the fifth subset to obtain a first accuracy rate;
training the first initial screen control model by using the first subset, the second subset, the third subset and the fifth subset to obtain a second initial screen control model, and verifying the second initial screen control model by using the fourth subset to obtain a second accuracy rate;
Training the second initial screen control model by using the first subset, the second subset, the fourth subset and the fifth subset to obtain a third initial screen control model, and verifying the third initial screen control model by using the third subset to obtain a third accuracy rate;
training the third initial screen control model by using the first subset, the third subset, the fourth subset and the fifth subset to obtain a fourth initial screen control model, and verifying the fourth initial screen control model by using the second subset to obtain a fourth accuracy rate;
training the fourth initial screen control model by using the second subset, the third subset, the fourth subset and the fifth subset to obtain a fifth initial screen control model, and verifying the fifth initial screen control model by using the first subset to obtain a fifth accuracy rate;
calculating average values of the first accuracy rate, the second accuracy rate, the third accuracy rate, the fourth accuracy rate and the fifth accuracy rate, and comparing the average values with preset average values;
If the average value is not smaller than the preset average value, respectively calculating average values of corresponding parameters in the first initial screen control model, the second initial screen control model, the third initial screen control model, the fourth initial screen control model and the fifth initial screen control model;
generating a screen control model to be optimized by using all the average values;
constructing a two-class cross entropy loss function;
and optimizing the screen control model to be optimized by using the two-classification cross entropy loss function to obtain the screen control model.
In some embodiments, the processor 201 is further configured to implement:
the screen of the intelligent watch is lightened, a fourth sampling value acquired by the triaxial acceleration sensor is acquired at the same time, and a fifth sampling value acquired by the acceleration sensor is acquired after the fourth sampling value is acquired;
calculating a first absolute value, a second absolute value and a third absolute value by using the fourth sampling value and the fifth sampling value; wherein the first absolute value is an absolute value of a difference between an acceleration value in a first direction in the fourth sample value and an acceleration value in a first direction in the fifth sample value, the second absolute value is an absolute value of a difference between an acceleration value in a second direction in the fourth sample value and an acceleration value in a second direction in the fifth sample value, and the third absolute value is an absolute value of a difference between an acceleration value in a third direction in the fourth sample value and an acceleration value in a third direction in the fifth sample value;
Comparing the first absolute value, the second absolute value and the third absolute value with preset absolute values respectively;
and if at least one of the first absolute value, the second absolute value and the third absolute value is larger than the preset absolute value, extinguishing the screen of the intelligent watch.
It should be noted that, for convenience and brevity of description, specific working processes of the terminal device 200 described above may refer to corresponding processes of the screen control method of the smart watch, and are not described herein.
Embodiments of the present application also provide a computer readable storage medium storing a computer program, where the computer program is executed by one or more processors to cause the one or more processors to implement a screen control method of a smart watch as provided in the embodiments of the present application.
The computer readable storage medium may be an internal storage unit of the terminal device 200 of the foregoing embodiment, for example, a hard disk or a memory of the terminal device 200. The computer readable storage medium may also be an external storage device of the terminal device 200, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which the terminal device 200 is equipped with.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (5)

1. A screen control method of a smart watch, wherein the smart watch is provided with a triaxial acceleration sensor, the method comprising:
when the screen state of the intelligent watch is in an extinction state, acquiring a plurality of first sampling values acquired by the triaxial acceleration sensor in a first preset time period and second sampling values acquired at a first moment, and inputting the plurality of first sampling values into a preset state detection model to acquire the state of the wrist of a user in the first preset time period; the wrist of the user wears the intelligent watch, and the first preset time period is a time period which is connected with the first moment and is before the first moment;
judging whether the state of the wrist of the user at the first moment is consistent with the state in the first preset time period or not according to the plurality of first sampling values and the second sampling values;
If the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period, acquiring a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period; the second preset time period is a duration between the first time and a second time, and the second time is a current time;
inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model to judge whether the screen of the intelligent watch needs to be lightened at the current moment;
if the screen of the intelligent watch needs to be lightened at the current moment, the screen of the intelligent watch is lightened;
the screen of the intelligent watch is lightened, a fourth sampling value acquired by the triaxial acceleration sensor is acquired at the same time, and a fifth sampling value acquired by the acceleration sensor is acquired after the fourth sampling value is acquired;
calculating a first absolute value, a second absolute value and a third absolute value by using the fourth sampling value and the fifth sampling value; wherein the first absolute value is an absolute value of a difference between an acceleration value in a first direction in the fourth sample value and an acceleration value in a first direction in the fifth sample value, the second absolute value is an absolute value of a difference between an acceleration value in a second direction in the fourth sample value and an acceleration value in a second direction in the fifth sample value, and the third absolute value is an absolute value of a difference between an acceleration value in a third direction in the fourth sample value and an acceleration value in a third direction in the fifth sample value;
Comparing the first absolute value, the second absolute value and the third absolute value with preset absolute values respectively;
if at least one of the first absolute value, the second absolute value and the third absolute value is larger than the preset absolute value, extinguishing a screen of the intelligent watch;
the training method of the preset screen control model comprises the following steps:
acquiring a training sample set, wherein the training sample set comprises a plurality of mapping relations, and the mapping relations are mapping relations between a state of the wrist of an experimenter and a triaxial acceleration value of the wrist of the experimenter, which is acquired when the wrist of the experimenter deviates from the state of the wrist of the experimenter;
labeling a plurality of mapping relations in the training sample set respectively to distinguish positive samples and negative samples, wherein the positive samples are the reasons for the state that the wrist of the experimenter deviates from the wrist of the experimenter are that the mapping relations matched with the screen of the intelligent watch are checked;
constructing a convolutional neural network model, wherein the convolutional neural network model sequentially comprises four parallel branches, a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first full-connection layer, a tanh activation layer, a second full-connection layer and a sigmoid activation layer;
Training the convolutional neural network model based on the marked training sample set to obtain the screen control model;
the training the convolutional neural network model based on the labeled training sample set to obtain the screen control model comprises the following steps:
randomly dividing the training sample set into a first subset, a second subset, a third subset, a fourth subset and a fifth subset in a hierarchical sampling mode;
training the convolutional neural network model by using the first subset, the second subset, the third subset and the fourth subset to obtain a first initial screen control model, and verifying the first initial screen control model by using the fifth subset to obtain a first accuracy rate;
training the first initial screen control model by using the first subset, the second subset, the third subset and the fifth subset to obtain a second initial screen control model, and verifying the second initial screen control model by using the fourth subset to obtain a second accuracy rate;
training the second initial screen control model by using the first subset, the second subset, the fourth subset and the fifth subset to obtain a third initial screen control model, and verifying the third initial screen control model by using the third subset to obtain a third accuracy rate;
Training the third initial screen control model by using the first subset, the third subset, the fourth subset and the fifth subset to obtain a fourth initial screen control model, and verifying the fourth initial screen control model by using the second subset to obtain a fourth accuracy rate;
training the fourth initial screen control model by using the second subset, the third subset, the fourth subset and the fifth subset to obtain a fifth initial screen control model, and verifying the fifth initial screen control model by using the first subset to obtain a fifth accuracy rate;
calculating average values of the first accuracy rate, the second accuracy rate, the third accuracy rate, the fourth accuracy rate and the fifth accuracy rate, and comparing the average values with preset average values;
if the average value is not smaller than the preset average value, respectively calculating average values of corresponding parameters in the first initial screen control model, the second initial screen control model, the third initial screen control model, the fourth initial screen control model and the fifth initial screen control model;
Generating a screen control model to be optimized by using all the average values;
constructing a two-class cross entropy loss function;
and optimizing the screen control model to be optimized by using the two-classification cross entropy loss function to obtain the screen control model.
2. The screen control method of the smart watch according to claim 1, wherein the determining whether the state of the wrist of the user at the first time is consistent with the state in the first preset period of time according to the plurality of first sampling values and the second sampling values includes:
calculating a first standard deviation, a second standard deviation and a third standard deviation according to the plurality of first sampling values and the second sampling values; the first standard deviation is a standard deviation of acceleration values in a first direction in the plurality of first sampling values and the second sampling values, the second standard deviation is a standard deviation of acceleration values in a second direction in the plurality of first sampling values and the second sampling values, and the third standard deviation is a standard deviation of acceleration values in a third direction in the plurality of first sampling values and the second sampling values;
acquiring a first preset standard deviation, a second preset standard deviation and a third preset standard deviation according to the state of the wrist of the user in the first preset time period; wherein the first preset standard deviation is a maximum value of standard deviations of acceleration values in the first direction that match a state of the wrist of the user within the first preset time period, the second preset standard deviation is a maximum value of standard deviations of acceleration values in the second direction that match a state of the wrist of the user within the first preset time period, and the third preset standard deviation is a maximum value of standard deviations of acceleration values in the third direction that match a state of the wrist of the user within the first preset time period;
Comparing the first standard deviation with the first preset standard deviation, comparing the second standard deviation with the second preset standard deviation, and comparing the third standard deviation with the third preset standard deviation;
if at least one of the first standard deviation is greater than the first preset standard deviation, the second standard deviation is greater than the second preset standard deviation and the third standard deviation is greater than the third preset standard deviation is met, determining that the state of the wrist of the user at the first moment is inconsistent with the state in the first preset time period.
3. The utility model provides a screen control device of intelligent wrist-watch, its characterized in that, intelligent wrist-watch is equipped with triaxial acceleration sensor, screen control device includes:
the first acquisition module is used for acquiring a plurality of first sampling values acquired by the triaxial acceleration sensor in a first preset time period and second sampling values acquired at the current moment when the screen state of the intelligent watch is in an off state, and inputting the plurality of first sampling values into a preset state detection model so as to acquire the state of the wrist of a user in the first preset time period; the wrist of the user wears the intelligent watch, and the first preset time period is a time period which is connected with the current moment and is before the current moment;
The judging module is used for judging whether the current state of the wrist of the user is consistent with the state in the first preset time period or not according to the plurality of first sampling values and the second sampling values;
the second acquisition module is used for acquiring a plurality of third sampling values acquired by the triaxial acceleration sensor in a second preset time period if the current state of the wrist of the user is inconsistent with the state of the wrist in the first preset time period; the second preset time period is a time period which is connected with the current time and is after the current time;
the input module is used for inputting the state of the user in the first preset time period and the plurality of third sampling values into a preset screen control model so as to judge whether the screen of the intelligent watch needs to be lightened;
the execution module is used for lighting the screen of the intelligent watch if the screen of the intelligent watch needs to be lighted;
the screen of the intelligent watch is lightened, a fourth sampling value acquired by the triaxial acceleration sensor is acquired at the same time, and a fifth sampling value acquired by the acceleration sensor is acquired after the fourth sampling value is acquired;
Calculating a first absolute value, a second absolute value and a third absolute value by using the fourth sampling value and the fifth sampling value; wherein the first absolute value is an absolute value of a difference between an acceleration value in a first direction in the fourth sample value and an acceleration value in a first direction in the fifth sample value, the second absolute value is an absolute value of a difference between an acceleration value in a second direction in the fourth sample value and an acceleration value in a second direction in the fifth sample value, and the third absolute value is an absolute value of a difference between an acceleration value in a third direction in the fourth sample value and an acceleration value in a third direction in the fifth sample value;
comparing the first absolute value, the second absolute value and the third absolute value with preset absolute values respectively;
if at least one of the first absolute value, the second absolute value and the third absolute value is larger than the preset absolute value, extinguishing a screen of the intelligent watch;
the training method of the preset screen control model comprises the following steps:
acquiring a training sample set, wherein the training sample set comprises a plurality of mapping relations, and the mapping relations are mapping relations between a state of the wrist of an experimenter and a triaxial acceleration value of the wrist of the experimenter, which is acquired when the wrist of the experimenter deviates from the state of the wrist of the experimenter;
Labeling a plurality of mapping relations in the training sample set respectively to distinguish positive samples and negative samples, wherein the positive samples are the reasons for the state that the wrist of the experimenter deviates from the wrist of the experimenter are that the mapping relations matched with the screen of the intelligent watch are checked;
constructing a convolutional neural network model, wherein the convolutional neural network model sequentially comprises four parallel branches, a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first full-connection layer, a tanh activation layer, a second full-connection layer and a sigmoid activation layer;
training the convolutional neural network model based on the marked training sample set to obtain the screen control model;
the training the convolutional neural network model based on the labeled training sample set to obtain the screen control model comprises the following steps:
randomly dividing the training sample set into a first subset, a second subset, a third subset, a fourth subset and a fifth subset in a hierarchical sampling mode;
training the convolutional neural network model by using the first subset, the second subset, the third subset and the fourth subset to obtain a first initial screen control model, and verifying the first initial screen control model by using the fifth subset to obtain a first accuracy rate;
Training the first initial screen control model by using the first subset, the second subset, the third subset and the fifth subset to obtain a second initial screen control model, and verifying the second initial screen control model by using the fourth subset to obtain a second accuracy rate;
training the second initial screen control model by using the first subset, the second subset, the fourth subset and the fifth subset to obtain a third initial screen control model, and verifying the third initial screen control model by using the third subset to obtain a third accuracy rate;
training the third initial screen control model by using the first subset, the third subset, the fourth subset and the fifth subset to obtain a fourth initial screen control model, and verifying the fourth initial screen control model by using the second subset to obtain a fourth accuracy rate;
training the fourth initial screen control model by using the second subset, the third subset, the fourth subset and the fifth subset to obtain a fifth initial screen control model, and verifying the fifth initial screen control model by using the first subset to obtain a fifth accuracy rate;
Calculating average values of the first accuracy rate, the second accuracy rate, the third accuracy rate, the fourth accuracy rate and the fifth accuracy rate, and comparing the average values with preset average values;
if the average value is not smaller than the preset average value, respectively calculating average values of corresponding parameters in the first initial screen control model, the second initial screen control model, the third initial screen control model, the fourth initial screen control model and the fifth initial screen control model;
generating a screen control model to be optimized by using all the average values;
constructing a two-class cross entropy loss function;
and optimizing the screen control model to be optimized by using the two-classification cross entropy loss function to obtain the screen control model.
4. A terminal device, characterized in that the terminal device comprises a processor, a memory and a computer program stored on the memory and executable by the processor, wherein the computer program, when executed by the processor, implements the screen control method of a smart watch according to any one of claims 1 to 2.
5. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program, wherein the computer program, when executed by a processor, implements the screen control method of a smart watch according to any one of claims 1 to 2.
CN202310833555.3A 2023-07-10 2023-07-10 Screen control method, device and equipment of intelligent watch and storage medium Active CN116578227B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310833555.3A CN116578227B (en) 2023-07-10 2023-07-10 Screen control method, device and equipment of intelligent watch and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310833555.3A CN116578227B (en) 2023-07-10 2023-07-10 Screen control method, device and equipment of intelligent watch and storage medium

Publications (2)

Publication Number Publication Date
CN116578227A CN116578227A (en) 2023-08-11
CN116578227B true CN116578227B (en) 2024-01-16

Family

ID=87541627

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310833555.3A Active CN116578227B (en) 2023-07-10 2023-07-10 Screen control method, device and equipment of intelligent watch and storage medium

Country Status (1)

Country Link
CN (1) CN116578227B (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106020670A (en) * 2016-05-24 2016-10-12 北京搜狗科技发展有限公司 Screen lightening control method, device and electronic equipment
CN107102717A (en) * 2017-04-26 2017-08-29 广东小天才科技有限公司 Screen control method and device
CN108347758A (en) * 2018-02-05 2018-07-31 广东欧珀移动通信有限公司 screen awakening method and device, terminal, computer readable storage medium
CN108418953A (en) * 2018-02-05 2018-08-17 广东欧珀移动通信有限公司 The screen control method and device of terminal, readable storage medium storing program for executing, terminal
CN108427578A (en) * 2018-02-08 2018-08-21 上海惠芽信息技术有限公司 A kind of screen control method, computer readable storage medium and a kind of intelligent terminal
CN108803896A (en) * 2018-05-28 2018-11-13 Oppo（重庆）智能科技有限公司 Control method, apparatus, terminal and the storage medium of screen
CN112067015A (en) * 2020-09-03 2020-12-11 青岛歌尔智能传感器有限公司 Step counting method and device based on convolutional neural network and readable storage medium
CN116225227A (en) * 2023-03-10 2023-06-06 东莞市猎声电子科技有限公司 Method for judging bright screen based on finite state machine

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016039587A1 (en) * 2014-09-11 2016-03-17 Samsung Electronics Co., Ltd. Wearable device

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106020670A (en) * 2016-05-24 2016-10-12 北京搜狗科技发展有限公司 Screen lightening control method, device and electronic equipment
CN107102717A (en) * 2017-04-26 2017-08-29 广东小天才科技有限公司 Screen control method and device
CN108347758A (en) * 2018-02-05 2018-07-31 广东欧珀移动通信有限公司 screen awakening method and device, terminal, computer readable storage medium
CN108418953A (en) * 2018-02-05 2018-08-17 广东欧珀移动通信有限公司 The screen control method and device of terminal, readable storage medium storing program for executing, terminal
CN108427578A (en) * 2018-02-08 2018-08-21 上海惠芽信息技术有限公司 A kind of screen control method, computer readable storage medium and a kind of intelligent terminal
CN108803896A (en) * 2018-05-28 2018-11-13 Oppo（重庆）智能科技有限公司 Control method, apparatus, terminal and the storage medium of screen
CN112067015A (en) * 2020-09-03 2020-12-11 青岛歌尔智能传感器有限公司 Step counting method and device based on convolutional neural network and readable storage medium
CN116225227A (en) * 2023-03-10 2023-06-06 东莞市猎声电子科技有限公司 Method for judging bright screen based on finite state machine

Also Published As

Publication number Publication date
CN116578227A (en) 2023-08-11

Similar Documents

Publication Publication Date Title
CN104995581B (en) The gestures detection management of electronic equipment
US20130018837A1 (en) Emotion recognition apparatus and method
US9329694B2 (en) Preemptive machine learning-based gesture recognition
Dobbins et al. Detecting physical activity within lifelogs towards preventing obesity and aiding ambient assisted living
US9804679B2 (en) Touchless user interface navigation using gestures
CN110262947B (en) Threshold warning method and device, computer equipment and storage medium
CN103561634B (en) Analyte testing method and system with height analyte trend notice
CN106030446A (en) Adjusting a power mode of a wearable computing device based on motion data
CN111225603B (en) Electronic device and method for providing stress index corresponding to user activity
CN110476141A (en) Sight tracing and user terminal for executing this method
EP3315914B1 (en) Step counting method, device and terminal
CN103729056A (en) System and method for controlling electronic equipment by knock
CN109753777A (en) A kind of personal identification method, device, storage medium and mobile terminal
US20230359541A1 (en) Electronic device for identifying force touch and method for operating same
US20200319768A1 (en) Gesture Recognition on Watch Bezel Using Strain Gauges
CN110476180A (en) User terminal for providing the method for the reward type advertising service based on text reading and for carrying out this method
CN105426961A (en) Method for capturing user intentions by utilizing intelligent bracelet and intelligent mobile phone
CN116578227B (en) Screen control method, device and equipment of intelligent watch and storage medium
CN111931075B (en) Content recommendation method and device, computer equipment and storage medium
CN112163095A (en) Data processing method, device, equipment and storage medium
CN111797856B (en) Modeling method and device, storage medium and electronic equipment
CN106073791B (en) Calorie computing method and device based on Intelligent bracelet
CN113222771A (en) Method and device for determining target group based on knowledge graph and electronic equipment
CN113127272B (en) Screen detection method and screen detection electronic equipment
CN111813639B (en) Method and device for evaluating equipment operation level, storage medium and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant