CN113255523B - Method, system, device and storage medium for improving gesture recognition precision and recognition - Google Patents
Method, system, device and storage medium for improving gesture recognition precision and recognition Download PDFInfo
- Publication number
- CN113255523B CN113255523B CN202110580963.3A CN202110580963A CN113255523B CN 113255523 B CN113255523 B CN 113255523B CN 202110580963 A CN202110580963 A CN 202110580963A CN 113255523 B CN113255523 B CN 113255523B
- Authority
- CN
- China
- Prior art keywords
- gesture
- sensor
- user
- induction
- sensing
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The application relates to a method, a system, a device and a storage medium for improving gesture recognition precision, relates to the technical field of automobile skylight gesture recognition, and is based on a plurality of sensing modules; establishing an incidence relation between the comparison result signal parameter and a gesture database and judging whether a user has a requirement for controlling the skylight; and if the user has the requirement for controlling the skylight, dynamically changing the sensing sensitivity and the sensing threshold of each sensing module according to the initial gesture action of the user. The gesture recognition device has the effects of high gesture recognition precision and small interference of the sensing module.
Description
Technical Field
The present application relates to the field of automotive electronics, and in particular, to a method, system, device and storage medium for improving gesture recognition accuracy.
Background
With the development of society, scientific technology is also continuously advancing. The common switches in life are gradually changed from traditional mechanical switches into simple touch switches and non-contact gesture sensing switches. For example, in a car sunroof, a non-contact gesture-sensing switch is currently implemented.
In the related technology, the gesture induction opening technology part of the automobile skylight captures gestures by using a camera, and controls the skylight to act based on different captured gestures, and a switch using the technology is generally expensive; in addition, a plurality of sensors are combined to detect gestures by using a simple infrared or capacitance sensing technology, and then the skylight action is controlled based on different detected gestures, so that the switch price by using the technology is relatively low. However, in practical operation, the scheme of performing gesture recognition by using the capacitive sensing technology often has the problem that gestures cannot be successfully recognized due to the fact that the gestures of an operator are not standard.
Therefore, the applicant believes that the gesture recognition by using the capacitive sensing technology has the defect of low gesture recognition accuracy, and needs to be improved.
Disclosure of Invention
In order to solve the problem that the recognition precision is low when a gesture is performed by utilizing a capacitance sensing technology, the application provides a method, a system, a device and a storage medium for improving the gesture recognition precision.
In a first aspect, the present application provides a method for improving gesture recognition accuracy, which adopts the following technical scheme:
a method for improving gesture recognition accuracy is based on a plurality of induction modules and comprises the following steps:
acquiring a signal characteristic parameter corresponding to the gesture action of the current user, comparing the signal characteristic parameter with a preset effective induction range parameter, and obtaining and storing a comparison result parameter;
establishing an incidence relation between the comparison result signal parameter and a gesture database and judging whether a user has a requirement for controlling the skylight;
and if the user has the requirement for controlling the skylight, dynamically changing the sensing sensitivity and the sensing threshold of each sensing module according to the initial gesture action of the user.
By adopting the technical scheme, whether the initial gesture of the user is in the preset effective sensing range or not is judged firstly after the sensing module senses the gesture action of the user, and if the initial gesture of the user is in the preset effective sensing range, the gesture judgment on the next step is carried out, and when the gesture posture of the user is incorrect or the gesture moving speed is high, the sensing sensitivity and the sensing threshold value of the sensing module are dynamically changed, so that the mutual interference condition of a plurality of sensing modules is reduced, and the gesture recognition precision is improved.
Preferably, the method includes the steps of obtaining a signal characteristic parameter corresponding to the gesture motion of the current user, comparing the signal characteristic parameter with a preset effective induction range parameter, and obtaining and storing a comparison result parameter, and specifically includes the following steps:
the sensing module acquires and acquires initial gesture actions of a user, converts the initial gesture actions of the user into gesture signals, converts the gesture signals into gesture signal characteristic parameters based on the chip and stores the gesture signal characteristic parameters;
calculating a preset effective induction range parameter based on the induction sensitivity and the induction threshold parameter of the induction module, and comparing the signal characteristic parameter with a preset effective induction range value to obtain a comparison result parameter;
and writing the comparison result parameter into a memory, and recording and storing.
By adopting the technical scheme, the gesture action is firstly acquired, the gesture action is converted into the gesture signal, then the gesture signal is processed through the chip, the processed gesture signal is compared with the preset effective induction range value, and the comparison result parameter obtained after comparison is stored and recorded, so that the requirement of the initial gesture of the user is determined.
Preferably, the dynamically changing the sensing sensitivity and the sensing threshold of the sensing module according to the initial gesture of the user specifically includes:
respectively arranging the sensing modules in a first sensing area, a second sensing area and a third sensing area, wherein the first sensing area and the third sensing area are used for acquiring and acquiring initial or final gesture actions of a user, and the second sensing area is used for acquiring intermediate gesture actions of the user;
if the induction module in the first induction area or the second induction area is triggered, dynamically adjusting the induction sensitivity and the induction threshold of the induction module in the third induction area;
if the induction module in the second induction area or the third induction area is triggered, dynamically adjusting the induction sensitivity and the induction threshold of the induction module in the first induction area;
if the gesture moving speed of the user is too fast, after the sensing module in the first sensing area or the third sensing area is triggered, the sensing sensitivity and the sensing threshold value of the sensing module in the second sensing area are dynamically adjusted, and the time for detecting the gesture action by the sensing module in the second sensing area is changed.
By adopting the technical scheme, after the induction modules in different induction areas are triggered, the induction threshold values and the induction sensitivity of other induction modules are dynamically adjusted, so that the minimum induction interference is realized on the premise of not influencing the induction precision of the induction modules, and the effect of improving the gesture recognition precision is further achieved.
In a second aspect, the present application provides an identification method for improving gesture identification accuracy, which adopts the following technical scheme:
the recognition method for improving the gesture recognition precision is based on the requirement of a user for controlling an automobile skylight, wherein the control requirement comprises an opening requirement, a closing requirement and a pause requirement, and is characterized by comprising the following steps of:
adjusting parameters of the induction module, including initializing an induction range of the induction module;
the sensing module converts the initial gesture or the subsequent gesture into gesture parameters based on a related algorithm after acquiring the initial gesture and the subsequent gesture;
matching the gesture parameters with parameters prestored in a gesture database based on a correlation algorithm to obtain result parameters;
and the sensing modules dynamically adjust the sensing sensitivity and the sensing threshold of each sensing module based on the result parameters.
By adopting the technical scheme, the parameters of each induction module are initialized, the gesture parameters are matched with the parameters in the gesture recognition database, and the induction sensitivity and the induction threshold of each induction module are dynamically adjusted, so that the interference of the induction range of each induction module is reduced, and the effect of improving the gesture recognition precision is achieved.
In a third aspect, the present application provides a gesture recognition control system, which adopts the following technical scheme:
a control system for improving gesture recognition accuracy, comprising:
the sensing module comprises a first sensor, a second sensor and a third sensor and is used for acquiring and identifying gesture actions of a user;
the gesture judgment module is in signal connection with the switch module, and the gesture judgment comprises opening gesture judgment, closing gesture judgment and pause gesture judgment and is used for judging the gesture of a user to control the switch module;
and the threshold adjusting module is used for adjusting the threshold and the induction sensitivity of the second sensor.
The zero clearing module is used for controlling the switch module according to the gesture action if the gesture action is determined to be effective; if the gesture action is determined to be invalid, clearing to zero and re-identifying the current gesture action;
by adopting the technical scheme, in actual use, the initial gesture action of the user is firstly identified through the sensing module, and whether the user needs to control the skylight is determined; when the user is judged to have the demand, the gesture judgment module judges the gesture of the user and sends the judgment result to the switch module; if the gesture control requirement is judged to be absent, the zero clearing module judges that the gesture is invalid and zero cleared, and recognizes the gesture again; after the gesture is judged, the sensing threshold value and the sensing sensitivity of the sensing module are adjusted through the threshold value adjusting module, so that the interference of the sensing range among the sensing modules is reduced, and the gesture recognition precision is improved.
Preferably, if the user performs a gesture operation on the sunroof, the opening operation specifically includes the following steps:
when the first sensor collects the user stopping gesture, the first sensor enters an opening awakening state, and meanwhile, the threshold parameter and the sensing sensitivity of the third sensor are changed;
when the first sensor and the second sensor acquire user gestures and the third sensor does not acquire the user gestures, entering an opening preparation state;
when the second sensor and the third sensor both acquire the user gesture and the first sensor does not acquire the user gesture or the first sensor and the second sensor both do not acquire the user gesture and the third sensor acquires the user gesture, opening the skylight;
the closing operation specifically comprises the following steps:
when the third sensor collects the user stopping gesture, the state of closing and awakening is entered, and meanwhile, the threshold value and the sensing sensitivity of the first sensor are changed;
when the second sensor and the third sensor acquire the user gestures and the first sensor does not acquire the user gestures, entering a closing preparation state;
and when the first sensor and the second sensor both collect the user gesture and the third sensor does not collect the user gesture or when the second sensor and the third sensor both do not collect the user gesture and the first sensor collects the user gesture, closing the skylight.
The pause operation specifically includes that if the first sensor, the second sensor and the third sensor acquire the user stop gesture, the skylight enters a pause state.
By adopting the technical scheme, the gesture action is decomposed into three stages by means of the three sensors, the gestures in different stages are respectively collected by means of the three sensors, the induction sensitivity and the induction threshold value of the sensors are dynamically changed, the interference condition of the induction ranges of the three sensors is reduced, and the accuracy of gesture recognition is improved.
Preferably, the first sensor, the second sensor and the third sensor are all capacitance sensors.
By adopting the technical scheme, the capacitive sensor has the advantages of simple structure, high sensitivity, zero magnetic hysteresis, vacuum compatibility, strong overload capacity, good dynamic response characteristic, strong adaptability to severe conditions such as high temperature, radiation, strong vibration and the like, and the stability of sensor induction is improved.
In a fourth aspect, the application provides a device for improving gesture recognition accuracy, which adopts the following technical scheme:
a device for improving gesture recognition accuracy comprises an automobile skylight for executing the system.
By adopting the technical scheme, the automobile skylight is convenient to realize gesture recognition non-contact control, and the accuracy of gesture recognition control is improved.
In a fifth aspect, the storage medium for improving gesture recognition accuracy provided by the application adopts the following technical scheme:
a storage medium for improving gesture recognition accuracy comprises a computer program which can be loaded by a processor and a chip and executes the method.
By adopting the technical scheme, the operation of the whole method is controlled by a computer program, and the stability and the accuracy of the operation of the whole method are improved.
In summary, the present application includes at least one of the following beneficial technical effects:
1. whether the initial gesture of the user is in the preset effective sensing range or not is judged firstly after the sensing module senses the gesture of the user, if so, the gesture judgment of the next step is carried out, and when the gesture posture of the user is incorrect or the gesture moving speed is high, the sensing sensitivity and the sensing threshold value of the sensing module are dynamically changed, so that the mutual interference condition of a plurality of sensing modules is reduced, and the gesture recognition precision is improved.
2. The gesture motion is divided into three stages, the gestures in different stages are collected by means of the three sensors respectively, the sensing sensitivity and the sensing threshold value of the sensors are changed dynamically, the interference condition of the sensing range among the three sensors is reduced, and the accuracy of gesture recognition is improved.
Drawings
Fig. 1 is a schematic block diagram of a method for improving gesture recognition accuracy, which is mainly embodied in the embodiment of the present application;
FIG. 2 is a schematic block diagram of a sub-step structure mainly embodying step S1 according to an embodiment of the present application;
fig. 3 is a schematic block diagram of an identification method for improving gesture recognition accuracy according to an embodiment of the present disclosure;
fig. 4 is a schematic diagram of a system structure mainly embodying gesture recognition control in the embodiment of the present application;
fig. 5 is a schematic diagram mainly embodying a gesture recognition apparatus according to an embodiment of the present disclosure.
Reference numerals: 1. a sensing module; 11. a first sensor; 12. a second sensor; 13. a third sensor; 2. a first sensing region; 3. a second sensing region; 4. a third sensing region; 5. a gesture judgment module; 6. a threshold adjustment module; 7. and clearing the module.
Detailed Description
The present application is described in further detail below with reference to figures 1-5.
The embodiment of the application discloses a method for improving gesture recognition precision, which can be applied to action induction switches in different fields. The embodiment of the application is described in detail with reference to the control applied to the sunroof as an example.
A method for improving gesture recognition accuracy comprises the following steps:
s1, a sensing module 1 acquires a signal characteristic parameter corresponding to the current user gesture action, compares the signal characteristic parameter with a preset effective sensing range parameter, and obtains and stores a comparison result parameter;
s2, establishing an incidence relation between the comparison result signal parameter and the gesture database and judging whether the user has the requirement for controlling the skylight;
and S3, if the user has the requirement for controlling the skylight, dynamically changing the sensing sensitivity and the sensing threshold of each sensing module according to the initial gesture action of the user.
In the embodiment of the application, signal characteristic parameters corresponding to the gesture actions of the current user are acquired through a plurality of sensing modules 1 arranged in a sensing area on the automobile skylight. The sensing module 1 can be a capacitive proximity sensor, the capacitive proximity sensor can convert detected capacitance into digital data depending on a chip, the digital data can obtain a corresponding signal value after being processed by the chip, when a user hand approaches the capacitive proximity sensor, the capacitance value which can be detected by the capacitive proximity sensor can be increased, and correspondingly, the signal value can also be increased. Here, the signal characteristic parameter acquired in the step S1 corresponding to the current user gesture is a signal value obtained from the current user gesture.
The preset effective sensing range parameter is a threshold value capable of triggering the sensing module 1 to a signal value capable of starting to output, namely when the collected signal characteristic parameter corresponding to the gesture action of the current user reaches the threshold value, the sensing module 1 can be triggered to be in an ON state, otherwise, the sensing module is in an OFF state. The larger the effective sensing range is, the smaller the threshold of the corresponding signal value is, that is, the easier the sensing module 1 is to be triggered.
Referring to fig. 2, the S1 step includes the following sub-steps:
s101, an induction module 1 collects and acquires initial gesture actions of a user, converts the initial gesture actions of the user into gesture signals, and converts the gesture signals into gesture signal characteristic parameters based on a chip and stores the gesture signal characteristic parameters. It should be added that the conversion of the gesture motion into the gesture signal may be performed by converting the gesture signal into the gesture signal characteristic parameter through a chip of the seilance company, and storing the converted gesture signal characteristic parameter.
S102, calculating a preset effective induction range parameter based on the induction sensitivity and the induction threshold parameter of the induction module 1, and comparing the signal characteristic parameter with a preset effective induction range value to obtain a comparison result parameter;
and S103, writing the comparison result parameters into a memory, and recording and storing.
And S2, establishing an incidence relation between the comparison result signal parameter and the gesture database and judging whether the user has the requirement of controlling the skylight.
The signal characteristic parameters corresponding to the gesture actions of the current user are acquired through the plurality of sensing modules 1 arranged in the sensing area ON the automobile skylight, when one or more sensing modules 1 are triggered to be in an ON state, the requirement that the user controls the skylight can be judged, and the initial action of the user can be judged through the position of the sensing module 1 which is triggered firstly, so that the specific control requirement (such as opening, closing or pausing of the skylight) of the user can be judged.
Therefore, by establishing the association relationship between the comparison result signal parameter in S2 and the control command in the gesture database, the control requirement of the current user can be determined.
In practice, the spatial distance is difficult for the operator to grasp, and the requirements for the hand operation angle and speed also reduce the use experience of the operator. Practice shows that when an operator does not exercise or carelessly make a gesture or the gesture operation is too fast, the success rate of recognizing the control command set above is greatly reduced.
For this purpose, step S3: if the user is judged to have the requirement of controlling the skylight, according to the initial gesture action of the user, the sensing sensitivity and the sensing threshold value of each sensing module are dynamically changed, and the method specifically comprises the following steps:
respectively arranging the sensing modules in a first sensing area, a second sensing area and a third sensing area, wherein the first sensing area and the third sensing area are used for acquiring and acquiring initial or final gesture actions of a user, and the second sensing area is used for acquiring intermediate gesture actions of the user;
if the induction module in the first induction area or the second induction area is triggered, dynamically adjusting the induction sensitivity and the induction threshold of the induction module in the third induction area;
if the induction module in the second induction area or the third induction area is triggered, dynamically adjusting the induction sensitivity and the induction threshold of the induction module in the first induction area;
if the gesture moving speed of the user is too fast, after the sensing module in the first sensing area or the third sensing area is triggered, the sensing sensitivity and the sensing threshold value of the sensing module in the second sensing area are dynamically adjusted, and the time for detecting the gesture action by the sensing module in the second sensing area is changed.
In S3, the initial gesture action of the user is judged by identifying the position of the induction module 1 which is triggered firstly, and the requirement of the user for controlling the skylight in advance can be judged by identifying the initial gesture of the user; then, the sensing threshold of each sensing module 1 is dynamically changed to improve the gesture recognition effect, that is, the effective sensing range parameter of the sensing module 1 which is not required to be triggered under the current gesture action is reduced (that is, the signal threshold of the sensing module 1 which is not required to be triggered is increased) to reduce the possibility that the sensing module 1 which is not required to be triggered under the current gesture action is triggered, so that the gesture recognition effect is improved.
The embodiment of the application provides an identification method for improving gesture identification precision, which comprises the following steps:
D1. based on the requirement of a user for controlling the automobile skylight, the control requirement comprises an opening requirement, a closing requirement and a pause requirement:
D2. adjusting parameters of the induction module, including initializing an induction range of the induction module;
D3. the sensing module converts the initial gesture or the subsequent gesture into gesture parameters based on a related algorithm after acquiring the initial gesture and the subsequent gesture;
D4. matching the gesture parameters with parameters stored in a gesture database in advance based on a correlation algorithm to obtain result parameters;
D5. and the sensing modules dynamically adjust the sensing sensitivity and the sensing threshold of each sensing module based on the result parameters.
Through initializing the parameters of each induction module, matching the gesture parameters with the parameters in the gesture recognition database, and dynamically adjusting the induction sensitivity and the induction threshold of each induction module, the interference of the induction range of each induction module is reduced, and the effect of improving the gesture recognition precision is achieved.
The embodiment of the application also discloses a gesture recognition control system:
referring to fig. 4, the device includes a sensing module 1, a gesture determining module 5, a threshold adjusting module 6, and a zero clearing module 7. The sensing module 1 comprises a first sensor 11, a second sensor 12 and a third sensor 13, which are used for collecting and recognizing gesture actions of a user. The first sensor 11, the second sensor 12 and the third sensor 13 are all capacitance proximity sensors, and the capacitance proximity sensors have the advantages of simple structure, high sensitivity, zero magnetic hysteresis, vacuum compatibility, strong overload capacity, good dynamic response characteristic, strong adaptability to severe conditions such as high temperature, radiation, strong vibration and the like. Of course, the first sensor 11, the second sensor 12 and the third sensor 13 may also be other proximity sensors, such as infrared sensors, etc., and the principle thereof may be similar to that of the above capacitive proximity sensor.
The threshold adjustment module 6 is mainly used to adjust the threshold parameters and the sensing sensitivity of the first sensor 11, the second sensor 12 and the third sensor 13, and is usually controlled by a computer. The zero clearing module 7 is used for zero clearing the control system, and if the gesture action is determined to be effective, the switch module is controlled according to the gesture action; and if the gesture action is determined to be invalid, resetting through a zero clearing reset algorithm, and re-identifying the current gesture action.
The sensing module 1 is mainly realized through a sensor, specifically, when a user performs an operation of opening the skylight, a user gesture swings along the direction from the first sensor 11 to the third sensor 13, and when the first sensor 11 acquires that the user stops the gesture and triggers to a state ON to enter an open awakening state, a threshold parameter and sensing sensitivity of the third sensor 13 are changed;
when the first sensor 11 and the second sensor 12 both collect the user gesture and trigger to the state ON and the third sensor 13 does not collect the user gesture and trigger to the state OFF, entering an opening preparation state;
when the second sensor 12 and the third sensor 13 both collect the gesture of the user and trigger to the state ON and the first sensor 11 does not collect the gesture of the user and trigger to the state OFF, or the first sensor 11 and the second sensor 12 do not collect the gesture of the user and trigger to the state OFF and the third sensor 13 collects the gesture of the user and trigger to the state ON, the skylight is opened.
The closing operation specifically comprises the steps that a user gesture waves along the direction from the third sensor 13 to the first sensor 11, when the third sensor 13 collects a user stopping gesture and triggers the user to be in a state ON, the user enters a closing and awakening state, and meanwhile, the threshold parameter and the induction sensitivity of the first sensor 11 are changed;
when the second sensor 12 and the third sensor 13 both collect the user gesture and trigger to the state ON and the first sensor 11 does not collect the user gesture and trigger to the state OFF, entering a closing preparation state;
when the first sensor 11 and the second sensor 12 both collect the gesture of the user and trigger to the state ON and the third sensor 13 does not collect the gesture of the user and trigger to the state OFF, or when the second sensor 12 and the third sensor 13 both do not collect the gesture of the user and trigger to the state OFF and the first sensor 11 collects the gesture of the user and trigger to the state ON, the skylight is closed.
The pause operation specifically comprises that the palm of the user is parallel to the automobile skylight, if the first sensor 11, the second sensor 12 and the third sensor 13 collect the user stop gesture and trigger the user stop gesture to be in the state ON, the skylight enters the pause state to be opened, and the time of the skylight being in the state ON is not less than the preset time threshold.
The gesture motion is divided into three stages, the gestures in different stages are collected by means of the three sensors respectively, the sensing sensitivity and the sensing threshold value of the sensors are changed dynamically, the interference condition of the sensing ranges of the three sensors is reduced, and the accuracy of gesture recognition is improved.
Referring to fig. 5, the embodiment of the application further discloses a device for improving gesture recognition accuracy, which comprises an automobile skylight, wherein an induction area is formed in the middle of the automobile skylight along the length direction of the automobile skylight, and the induction module 1 is installed in the induction area.
The embodiment of the application also discloses a storage medium for improving the gesture recognition precision, and a computer program which can be loaded by the processor and used for executing the gesture recognition control method is stored. A random access memory, also called main memory, may be used, which is an internal memory that exchanges data directly with the CPU. It can be read and written at any time, and is quick in speed, and can be extensively used as temporary data storage medium of operation system or other running program.
The above embodiments are preferred embodiments of the present application, and the protection scope of the present application is not limited by the above embodiments, so: all equivalent changes made according to the structure, shape and principle of the present application shall be covered by the protection scope of the present application.
Claims (8)
1. A method for improving gesture recognition accuracy is characterized by comprising the following steps based on a plurality of induction modules (1):
acquiring a signal characteristic parameter corresponding to the gesture action of the current user, comparing the signal characteristic parameter with a preset effective induction range parameter, and obtaining and storing a comparison result parameter;
establishing an incidence relation between the comparison result signal parameters and the gesture database and judging whether the user has a requirement for controlling the skylight;
if the user has the requirement for controlling the skylight, dynamically changing the sensing sensitivity and the sensing threshold of each sensing module (1) according to the initial gesture action of the user;
the step of dynamically changing the sensing sensitivity and the sensing threshold of the sensing module (1) according to the initial gesture of the user specifically comprises the following steps:
respectively arranging each induction module (1) in a first induction area (2), a second induction area (3) and a third induction area (4), wherein the first induction area (2) and the third induction area (4) are used for collecting and obtaining initial or final gesture actions of a user, and the second induction area (3) is used for collecting middle gesture actions of the user;
if the induction module (1) in the first induction area (2) or the second induction area (3) is triggered, dynamically adjusting the induction sensitivity and the induction threshold of the induction module (1) in the third induction area (4);
if the induction module (1) in the second induction area (3) or the third induction area (4) is triggered, dynamically adjusting the induction sensitivity and the induction threshold of the induction module (1) in the first induction area (2);
if the gesture moving speed of the user is too fast, after the sensing module (1) in the first sensing area (2) or the third sensing area (4) is triggered, the sensing sensitivity and the sensing threshold of the sensing module (1) in the second sensing area (3) are dynamically adjusted, and the gesture motion detection time of the sensing module (1) in the second sensing area (3) is changed.
2. The method for improving gesture recognition accuracy according to claim 1, wherein the method comprises the following steps: the method comprises the following steps of obtaining signal characteristic parameters corresponding to the gesture actions of a current user, comparing the signal characteristic parameters with preset effective induction range parameters, and obtaining and storing comparison result parameters, wherein the method specifically comprises the following steps:
the sensing module (1) collects and acquires initial gesture actions of a user, converts the initial gesture actions of the user into gesture signals, and converts the gesture signals into gesture signal characteristic parameters based on a chip and stores the gesture signal characteristic parameters;
calculating a preset effective induction range parameter based on the induction sensitivity and the induction threshold parameter of the induction module (1), and comparing the signal characteristic parameter with a preset effective induction range value to obtain a comparison result parameter;
and writing the comparison result parameter into a memory, and recording and storing.
3. An identification method for improving gesture identification accuracy based on the requirement of a user to control an automobile skylight according to claim 1, wherein the control requirement comprises an opening requirement, a closing requirement and a pause requirement, and is characterized by comprising the following steps of:
adjusting parameters of the induction module (1), including initializing an induction range of the induction module (1);
the sensing module (1) converts an initial gesture or a subsequent gesture into gesture parameters based on a related algorithm after acquiring the initial gesture and the subsequent gesture;
matching the gesture parameters with parameters stored in a gesture database in advance based on a correlation algorithm to obtain result parameters;
the sensing modules (1) dynamically adjust the sensing sensitivity and the sensing threshold of each sensing module (1) based on the result parameters.
4. A gesture recognition control system applied to the method according to any one of claims 1 to 2, comprising:
the sensing module (1) comprises a first sensor (11), a second sensor (12) and a third sensor (13) and is used for collecting and recognizing gesture actions of a user;
the gesture judgment module (5) is in signal connection with the switch module, and the gesture judgment comprises opening gesture judgment, closing gesture judgment and pause gesture judgment and is used for judging the gesture of a user to control the switch module;
the threshold adjusting module (6) is used for adjusting the threshold and the induction sensitivity of the first sensor (11), the second sensor (12) and the third sensor (13);
a clear module (7) for controlling the switch module according to the gesture action if the gesture action is determined to be valid; and if the gesture action is determined to be invalid, clearing to zero and re-identifying the current gesture action.
5. The gesture recognition control system of claim 4, wherein if the user performs gesture operation on the skylight,
the opening operation specifically comprises the following steps:
when the first sensor (11) collects the user stopping gesture, the user enters an opening awakening state, and meanwhile, the threshold parameter and the induction sensitivity of the third sensor (13) are changed;
when the first sensor (11) and the second sensor (12) both acquire user gestures and the third sensor (13) does not acquire the user gestures, entering an opening preparation state;
when the second sensor (12) and the third sensor (13) both acquire the user gesture and the first sensor (11) does not acquire the user gesture or the first sensor (11) and the second sensor (12) both acquire the user gesture and the third sensor (13) acquires the user gesture, opening the skylight;
the closing operation specifically comprises the following steps:
when the third sensor (13) collects a user stopping gesture, the state of closing and awakening is entered, and meanwhile, the threshold value and the induction sensitivity of the first sensor (11) are changed;
when the second sensor (12) and the third sensor (13) acquire user gestures and the first sensor (11) does not acquire the user gestures, entering a closing preparation state;
when the first sensor (11) and the second sensor (12) both collect the user gesture and the third sensor (13) does not collect the user gesture or when the second sensor (12) and the third sensor (13) both collect the user gesture and the first sensor (11) collects the user gesture, closing the skylight;
the pause operation specifically comprises that if the first sensor (11), the second sensor (12) and the third sensor (13) collect the user stop gesture, the skylight enters a pause state.
6. A gesture recognition control system according to claim 4, characterized in that the first sensor (11), the second sensor (12) and the third sensor (13) are all capacitive proximity sensors.
7. An apparatus for improving gesture recognition accuracy, comprising a sunroof for performing the system of any one of claims 4 to 6.
8. A storage medium for improving gesture recognition accuracy, comprising a computer program capable of being loaded by a processor, a chip and executing the method according to any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110580963.3A CN113255523B (en) | 2021-05-26 | 2021-05-26 | Method, system, device and storage medium for improving gesture recognition precision and recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110580963.3A CN113255523B (en) | 2021-05-26 | 2021-05-26 | Method, system, device and storage medium for improving gesture recognition precision and recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113255523A CN113255523A (en) | 2021-08-13 |
| CN113255523B true CN113255523B (en) | 2022-07-12 |
Family
ID=77184855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110580963.3A Active CN113255523B (en) | 2021-05-26 | 2021-05-26 | Method, system, device and storage medium for improving gesture recognition precision and recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113255523B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104278908A (en) * | 2013-07-10 | 2015-01-14 | 麦格纳覆盖件有限公司 | Method and system for hands-free operation of a power window of a vehicle |
| CN112488958A (en) * | 2020-12-15 | 2021-03-12 | 西安交通大学 | Image contrast enhancement method based on scale space |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014172323A1 (en) * | 2013-04-15 | 2014-10-23 | Flextronics Ap, Llc | Driver facts behavior information storage system |
| US9818246B2 (en) * | 2015-07-29 | 2017-11-14 | Ford Global Technologies, Llc | System and method for gesture-based control of a vehicle door |
| US11104229B2 (en) * | 2016-07-11 | 2021-08-31 | Shanghai Yanfeng Jinqiao Automotive Trim Systems Co. Ltd. | Vehicle interior component |
| US11209908B2 (en) * | 2017-01-12 | 2021-12-28 | Sony Corporation | Information processing apparatus and information processing method |
| WO2018147254A1 (en) * | 2017-02-10 | 2018-08-16 | パナソニックＩｐマネジメント株式会社 | Vehicular input apparatus |
| CN111325877B (en) * | 2018-12-14 | 2021-09-28 | 云丁网络技术(北京)有限公司 | Electronic equipment control method and device and electronic equipment |
-
2021
- 2021-05-26 CN CN202110580963.3A patent/CN113255523B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104278908A (en) * | 2013-07-10 | 2015-01-14 | 麦格纳覆盖件有限公司 | Method and system for hands-free operation of a power window of a vehicle |
| CN112488958A (en) * | 2020-12-15 | 2021-03-12 | 西安交通大学 | Image contrast enhancement method based on scale space |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113255523A (en) | 2021-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102055836B (en) | Mobile terminal with action recognition function and action recognition method thereof | |
| EP0973123A1 (en) | Finger sensor operating technique | |
| CN109074819B (en) | Operation-sound based preferred control method for multi-mode command and electronic device using the same | |
| US10140499B1 (en) | Systems and methods for touch and press detection using a sensor | |
| KR100941418B1 (en) | A localization method of moving robot | |
| JP2005129048A (en) | Sensor for detecting input operation and for detecting fingerprint | |
| WO2008049288A1 (en) | Biometric characteristic identifying device for accessing control | |
| US10782821B2 (en) | Method of classifying a finger touch in respect of finger pressure and fingerprint sensing system | |
| CN105303172A (en) | Calibration parameter obtaining method and device of fingerprint sensor and mobile terminal | |
| CN102981671B (en) | Multi-class object touch point detection method | |
| CN106249311A (en) | Proximity sensor and detection method for proximity sensor | |
| CN105447506B (en) | A kind of gesture identification method based on section distribution probability feature | |
| CN101806174A (en) | Touch control type anti-theft device | |
| CN113255523B (en) | Method, system, device and storage medium for improving gesture recognition precision and recognition | |
| CN113585912B (en) | Automatic door control method based on image processing technology and automatic door | |
| CN101620723A (en) | Offline intelligent image information processing method and intelligent image information processing system | |
| CN105975044B (en) | Method and device for automatically controlling touch screen wet hand mode through fingerprint detection | |
| WO2020224149A1 (en) | Electronic device, fingerprint recognition method therefor, and apparatus thereof | |
| WO2022142204A1 (en) | Vehicle lock control method and device, vehicle and storage medium | |
| KR101269107B1 (en) | Method for recognizing hand gesture using camera and thereof apparatus | |
| TWM562435U (en) | A living human face detection device | |
| CN100385947C (en) | System of identifying image of human bodies | |
| CN107358173A (en) | The fingerprint identification method and identification device of a kind of intelligent terminal | |
| CN100515080C (en) | Device with setting of image monitoring area and method thereof | |
| CN110255313B (en) | Elevator monitoring method and device based on infrared detection device |
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 |