CN111160165A - Self-adaptive posture error correction detection method and device - Google Patents
Self-adaptive posture error correction detection method and device Download PDFInfo
- Publication number
- CN111160165A CN111160165A CN201911310507.6A CN201911310507A CN111160165A CN 111160165 A CN111160165 A CN 111160165A CN 201911310507 A CN201911310507 A CN 201911310507A CN 111160165 A CN111160165 A CN 111160165A
- Authority
- CN
- China
- Prior art keywords
- attitude
- detection
- posture
- adaptive
- error correction
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 147
- 238000012937 correction Methods 0.000 title claims abstract description 40
- 230000002159 abnormal effect Effects 0.000 claims abstract description 4
- 230000003044 adaptive effect Effects 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 11
- 238000013145 classification model Methods 0.000 claims description 9
- 238000004891 communication Methods 0.000 claims description 9
- 230000036541 health Effects 0.000 claims description 5
- 230000003993 interaction Effects 0.000 claims description 5
- 238000012423 maintenance Methods 0.000 claims description 4
- 239000000853 adhesive Substances 0.000 claims description 3
- 230000001070 adhesive effect Effects 0.000 claims description 3
- 241000190070 Sarracenia purpurea Species 0.000 claims description 2
- 238000000034 method Methods 0.000 abstract description 14
- 210000003128 head Anatomy 0.000 description 30
- 238000005516 engineering process Methods 0.000 description 7
- 238000009434 installation Methods 0.000 description 7
- 230000005540 biological transmission Effects 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000006378 damage Effects 0.000 description 2
- 230000004379 myopia Effects 0.000 description 2
- 208000001491 myopia Diseases 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241001456553 Chanodichthys dabryi Species 0.000 description 1
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 235000019646 color tone Nutrition 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000005059 dormancy Effects 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Social Psychology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a self-adaptive posture error correction detection method and a self-adaptive posture error correction detection device, wherein the method comprises the following steps: s1: fixing the position of a camera, and configuring configuration information of a current attitude detection state; s2: selecting a posture detection model according to the configuration information; s3: the camera collects the current attitude image, and the selected attitude detection model detects the attitude image: if the detected posture is abnormal, an alarm is sent out, otherwise, the step S3 is repeated, and the device comprises: the device comprises a shell, a fixed connecting piece, an attitude detection module, a warning module and a control module; the gesture detection module comprises a camera and a gesture judgment unit; the control module is respectively in signal connection with the attitude determination unit and the warning module and is used for controlling the warning module to warn according to the regulation and control signal, configuring configuration information of the current attitude detection state and selecting an attitude detection model of the attitude determination unit according to the configuration information. The invention has the technical characteristics of strong self-adaptive capacity, wide application range and good posture detection performance.
Description
Technical Field
The invention belongs to the technical field of intelligent detection, and particularly relates to a self-adaptive posture error correction detection method and device.
Background
The wrong posture can cause symptoms such as myopia, humpback, head bending, high and low shoulders and the like over time, and can induce diseases of cervical vertebra, lumbar vertebra, eyes and other parts to cause irreversible damage to the body.
The existing gesture detection technology includes a method for judging gestures by using ultrasonic ranging or infrared ranging, and the method can solve a part of problems of wrong gesture detection in the actual use process, but under the condition that other objects are shielded or the non-trunk part of a human body is shielded, false detection alarm is easily generated, the specific condition of wrong gestures cannot be analyzed in a targeted manner, and the method has great limitation in the use process.
The other part forces the user to be in a specified posture state by using a rigid structure, and the technical scheme has certain practicability in the aspect of forcibly correcting the wrong posture, but limits the activity space of people to a certain extent and causes certain inconvenience.
In addition, different crowds, different environments and gesture detection standards have certain differences, so that the limitation of the current gesture detection technology is caused, meanwhile, the use of the current gesture detection technology is limited to a specific application scene, self-adaption cannot be performed on the different application scene detection technologies, and the practicability of the technology is greatly reduced.
Disclosure of Invention
The invention aims to provide a self-adaptive posture error correction detection method and a self-adaptive posture error correction detection device, which have the technical characteristics of strong self-adaptive capacity, wide application range and good posture detection performance.
In order to solve the problems, the technical scheme of the invention is as follows:
an adaptive attitude error correction detection method comprises the following steps:
s1: fixing the position of a camera, and configuring configuration information of a current attitude detection state;
s2: selecting a posture detection model according to the configuration information;
s3: the camera collects the current attitude image, and the selected attitude detection model detects the attitude image: if the detected posture is abnormal, an alarm is sent out, otherwise, the step S3 is repeated.
According to an embodiment of the present invention, the configuration information includes user information, usage environment information, and a gesture image library, and before step S1, the method further includes the following steps:
a1: acquiring configuration information under different attitude detection states;
a2: and training through different configuration information to obtain a plurality of different posture detection models.
According to an embodiment of the present invention, the pose detection model is a classification model, and in step a2, the images required by the classification model training need to be labeled with pose categories.
According to an embodiment of the present invention, the pose detection model is a regression model, and in step a2, pose parameters are labeled on the images required for training the regression model.
According to an embodiment of the invention, the posture parameters include the distance of the eyes from the table top, the height of the left and right shoulders, the head lowering angle, the head turning angle, and the head leaning angle.
According to an embodiment of the present invention, in step S1, the usage environment information includes a placement position and an angle of the camera, wherein the placement position and the angle are configured through manual input or automatic identification;
in step S2, the specific step of selecting the posture detection model according to the configuration information is: and selecting a posture detection model according to the placing position and the angle.
According to an embodiment of the present invention, the configuration of the placement position and the angle through automatic identification specifically includes the following steps:
b1: establishing a camera detection model, and collecting or inputting a standard attitude image, wherein the camera detection model is a regression model;
b2: regressing the standard attitude image through a camera detection model to obtain angles of head lowering, head turning and head deviation;
b3: and reversely deducing to obtain the placing position and the angle of the camera according to the angles of the lower head, the rotating head and the offset head.
The invention also provides a self-adaptive posture error correction detection device, which is used for posture detection by applying the self-adaptive posture error correction detection method, and comprises the following steps: the device comprises a shell, a fixed connecting piece arranged on the shell, and a posture detection module, a warning module and a control module which are arranged in the shell;
the gesture detection module comprises a camera for collecting gesture images and a gesture judgment unit for detecting the gesture images and generating regulation and control signals;
the control module is respectively in signal connection with the attitude determination unit and the warning module and is used for controlling the warning module to warn according to the regulation and control signal, configuring configuration information of the current attitude detection state and selecting the attitude detection model of the attitude determination unit according to the configuration information.
According to an embodiment of the present invention, the fixing connector is a magnetic member, a locking member, an adhesive member, a clamping member, or a locking member.
According to an embodiment of the invention, the portable electronic device further comprises a carrier, the shell is internally fixed in the carrier or externally fixed on the carrier through a fixed connecting piece, and the carrier is a table lamp, a sound box, a display, a water cup, a clock, a desk or a bookshelf.
According to an embodiment of the present invention, the attitude correction device further includes a communication module electrically connected to the control module for feeding back the working status and the attitude health report of the adaptive attitude correction detection device.
According to an embodiment of the invention, the control module is further configured to count the maintenance time of the normal posture, and control the warning module to remind fatigue according to the maintenance time.
According to an embodiment of the present invention, the device further comprises a speaker and a microphone for voice interaction and signal connection with the control module, wherein the speaker and the microphone are further used for acoustic ranging between the device and a user to assist in adjusting the fixed position of the device.
Compared with the prior art, the invention has the following advantages and positive effects:
1) the method can fix the fixed camera at any position, and the gesture detection model selected by the configuration information of the current gesture detection state is adapted to the current gesture detection condition, and specifically, before detection, a plurality of different gesture monitoring models are obtained by acquiring configuration training of different gesture detection states and pre-training for selection during actual gesture detection, wherein the gesture detection condition training is carried out by configuring a gesture image library, user information and use environment information, so that the self-adapting capability is improved, and the application range is expanded;
2) in the method of the invention, the placing position and the angle of the camera can be set in various ways, and because the placing position and the angle of the camera have great relation with the judgment of the posture detection, different placing positions and angles correspond to different posture detection models, therefore, for the configuration information, the configuration can be performed by manual input under the condition of known installation information or automatic identification and configuration under the condition of unknown installation information, specifically, regression detection is performed on the standard attitude image through a camera detection model of a regression model to obtain the angles of the head lowering, the head turning and the head deviation, and the placing position and the angle of the camera can be obtained by reverse thrust, so that the required posture detection model can be automatically selected, the self-adaption of the fixed position of the camera can be realized, so that the self-adaption capability is improved, and the application range is expanded;
3) the device can fix the self-adaptive posture correction detection device at any position in the environment through various fixed connecting pieces, and can also be combined with any article in the environment, so that the self-adaptive capacity is improved, and the application range is expanded;
4) the device provided by the invention is convenient for man-machine interaction by arranging the loudspeaker and the microphone, has a sound wave ranging function in the fixing process of the self-adaptive posture correction detection device, and can assist in installation of the device, so that the self-adaptive capacity is improved, and the application range is expanded.
Drawings
FIG. 1 is a main flow chart of an adaptive attitude error correction detection method according to the present invention;
FIG. 2 is a flow chart of automatic identification of the placement position and angle of a camera according to the adaptive posture error correction detection method of the present invention;
FIG. 3 is a schematic structural diagram of an adaptive attitude error correction detection apparatus according to the present invention.
Description of reference numerals:
1-fixing the connecting piece; 2-a communication module; 3-a shell; 4-warning buzzer; 5-a key module; 6-push button; 7-warning light in the right position; 8-wrong posture warning light; 9- -left position warning light; 10-a camera; 11-a data transmission port; 12-charging interface.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description will be made with reference to the accompanying drawings. It is obvious that the drawings in the following description are only some examples of the invention, and that for a person skilled in the art, other drawings and embodiments can be derived from them without inventive effort.
For the sake of simplicity, the drawings only schematically show the parts relevant to the present invention, and they do not represent the actual structure as a product. In addition, in order to make the drawings concise and understandable, components having the same structure or function in some of the drawings are only schematically illustrated or only labeled. In this document, "one" means not only "only one" but also a case of "more than one".
The following describes a method and an apparatus for adaptive posture error correction provided by the present invention in further detail with reference to the accompanying drawings and specific embodiments.
Example 1
Referring to fig. 1, the present application provides an adaptive attitude error correction detection method, including the following steps:
s1: fixing the position of a camera, and configuring configuration information of a current attitude detection state;
s2: selecting a posture of the posture detection model according to the configuration information;
s3: the camera collects the current attitude image, and the selected attitude detection model detects the attitude image: if the detected posture is abnormal, an alarm is sent out, otherwise, the step S3 is repeated.
The present embodiment will now be specifically explained:
in step S1 of this embodiment, the camera fixing position has multiple forms, which is specifically limited according to the placement position and angle of the camera, and can be fixed anywhere in the bookshelf, the desk lamp, the wall, the ground, the clock, etc. where the posture can be photographed, the configuration information includes the user information, the usage environment information, the posture image library, the usage environment information can include the height and the inclination angle of the desk, the position and the angle of the camera, the height of the chair, etc., the user information can include the height, the weight, the sex, the degree of myopia, the distance between the upper half body length and the two eyes, the distance between the eyes and the nose tip, whether to wear glasses, the height of the eyes from the desktop when viewing the front, the height of the shoulders from the desktop, the distance between the chest and the desk, etc., the posture image library can be a plurality of different specific posture images for different users under different collecting environments, the configuration mode of the configuration information can be manual input or automatic identification measurement, for example, the automatic identification measurement can be performed through technologies such as image identification, infrared distance measurement, acoustic distance measurement and the like.
Referring to fig. 1, the present embodiment provides a plurality of posture detection models to be selected to adapt to different posture detection states, and before step S1, the method further includes the following steps: a1: acquiring configuration information under different attitude detection states; a2: and training through different configuration information to obtain a plurality of different posture detection models. Specifically, in step a1, different user information, usage environment information, and configuration information of the pose image library may be collected, and in step a2, a plurality of different pose detection models are obtained by training the deep convolutional network model using different configuration information.
The posture detection model of this embodiment may adopt a regression model or a classification model: the posture detection model is a classification model, and in a corresponding step A2, the images required by the classification model training need to be labeled with posture categories; the pose detection model is a regression model, and in step a2, pose parameters are labeled for the images required by the classification model training, and preferably, the pose parameters include the distance between the eyes and the desktop, the height of the left and right shoulders, the head lowering angle, the head turning angle, the head leaning angle, the face position, and the eye and nose position. The human face position and the eye and nose position can be marked or not marked, the whole algorithm is not influenced, the regression precision can be improved, and meanwhile, a large amount of marking cost is increased.
The arrangement positions and angles of the cameras in the configuration information are greatly related to posture detection, certain differences exist among different arrangement positions and angles and corresponding posture detection models, the multiple posture detection models are provided for selection, and different posture detection models correspond to different arrangement positions and angles of the cameras. In step S2 of this embodiment, a corresponding posture detection model may be selected according to the camera placement position and angle obtained through manual input or automatic recognition, so as to adapt to posture detection in the current detection environment.
Referring specifically to fig. 2, the configuration of the placement position and angle by automatic identification specifically includes the following steps:
b1: establishing a camera detection model, and collecting or inputting a standard attitude image, wherein the camera detection model is a regression model;
b2: regressing the standard attitude image through a camera detection model to obtain angles of head lowering, head turning and head deviation;
b3: and reversely deducing to obtain the placing position and the angle of the camera according to the angles of the lower head, the rotating head and the offset head.
In the method of the embodiment, the placing position and the angle of the camera can be set in various ways, and because the placing position and the angle of the camera have great relation with the judgment of the posture detection, different placing positions and angles correspond to different posture detection models, therefore, for the configuration information, the configuration can be performed by manual input under the condition of known installation information or automatic identification and configuration under the condition of unknown installation information, specifically, regression detection is performed on the standard attitude image through a camera detection model of a regression model to obtain the angles of the head lowering, the head turning and the head deviation, and the placing position and the angle of the camera can be obtained by reverse thrust, so that the required posture detection model can be automatically selected, the self-adaption of the fixed position of the camera can be realized, so that the self-adaption capability is improved, and the application range is expanded.
Preferably, in this embodiment S3, the gesture correctness criterion corresponding to the gesture detection model may include: the distance between eyes and the desktop cannot be less than 30cm, the height difference between two shoulders and the desktop cannot exceed 2cm, the height difference between the eyes and the desktop when the front is seen in visual mode and the height difference between the eyes and the desktop when the user sits straight and the front is seen in visual mode does not exceed 5cm, the head lowering angle cannot exceed 45 degrees, the head turning angle does not exceed 30 degrees, the head deviation angle does not exceed 15 degrees, when all conditions are met, the posture is correct, and otherwise, the posture is wrong. Wherein, the height of eyes apart from the desktop when the user visually observes the place ahead is not more than 5cm from the height when sitting straight and visually observes the place ahead, and whether the user bends over at present can be judged to this standard. If the posture detection model is adopted as the regression model, an image of a user sitting straight and looking ahead needs to be input in advance, the distance between eyes of the user and the desktop is regressed, and whether the user bends or not is indirectly judged according to the current angle of lowering the head and the position of the eyes away from the desktop.
Preferably, the requirements for the gesture criteria may be inconsistent when people are in different situations, such as reading and writing, e.g. people tend to move head to the left when writing with the right hand, and some adjustment of the gesture criteria is required. If a classification model is used in a2, multiple models need to be trained according to different criteria; if a regression model is used in a2, the judgment criteria need only be modified accordingly.
Preferably, before step S3 of the present embodiment is executed, a determination of whether gesture detection is required, that is, whether there is a person in the screen and whether the person is in a proper position may be added, specifically, when no person is in the screen or the person is not in a proper position, a standby is performed, and otherwise, step S3 is executed.
Preferably, in step S3 of this embodiment, when the user gesture is wrong, an alarm may be given in real time, or an alarm may be given after a certain time according to the user gesture error.
Example 2
Referring to fig. 3, the present application provides an adaptive posture-error-correction detection apparatus according to embodiment 1, which applies the adaptive posture-error-correction detection method described in any one of the above to perform posture detection, including: the device comprises a shell 3, a fixed connecting piece 1 arranged on the shell 3, and a posture detection module, a warning module and a control module which are arranged in the shell 3;
the gesture detection module comprises a camera 10 for collecting gesture images and a gesture judgment unit for detecting the gesture images and generating regulation and control signals;
the control module is respectively in signal connection with the attitude determination unit and the warning module and is used for controlling the warning module to warn according to the regulation and control signal, configuring configuration information of the current attitude detection state and selecting the attitude detection model of the attitude determination unit according to the configuration information.
The present embodiment will now be described in detail:
the fixing connector 1 of the embodiment is a magnetic connector or a buckle or an adhesive or a clamping or locking member, and supports two installation modes of being externally worn and being internally installed in other products, and can be fixed on various external interfaces such as but not limited to desks, bookshelves, walls, the ground, table lamps, clocks and watches.
The device of the embodiment can fix the self-adaptive posture correction detection device at any position in the environment through various fixed connecting pieces 1, and can also be combined with any article in the environment, so that the self-adaptive capacity is improved, and the application range is expanded.
The camera 10 of the present embodiment preferably uses a camera 10 of 100 to 300 ten thousand pixels with a viewing angle of 75 ° to 90 °. Preferably, the camera 10 of the present embodiment is disposed at a perpendicular or approximately perpendicular angle to the horizontal plane, so as to facilitate the detection of the posture, and further preferably, the placement position of the camera 10 uses the central axis of the human body of the user as an axis, uses the normal line of the front of the human body as a 0 ° reference line, has a placement angle of 0 °, has a left-right deviation of 5 °, rotates 40 ° to 45 ° counterclockwise, rotates 40 ° to 45 ° along the pointer, has a height of 15cm to 35cm from the desktop, and has a distance of 40cm to 70cm from the human body. Similarly, the camera 10 of the present embodiment is not limited to the above setting, and may be fixed at other placement positions and angles, and may be adapted to posture detection by selecting an internal posture detection model.
Referring to fig. 3, the warning module of this embodiment includes a right position warning light 7, an error posture warning light 8, a left position warning light 9, and a warning buzzer 4, and adopts various warning modes, such as flashing frequency of the lights, sound of the buzzer, and the like, so as to present different warning effects of different posture errors.
The control module of the embodiment can count the maintaining time of the normal posture and control the warning module to remind fatigue according to the maintaining time so as to prevent a user from maintaining a posture for too long time. The control module can also judge the attitude once every a period of preset time so as to reduce the frequency of attitude detection and prolong the endurance time of the device.
Referring to fig. 3, the present embodiment further includes a communication module 2, a charging interface 12, a data transmission interface 11, and a key module 5, wherein the communication module 2 is electrically connected to the control module for feeding back the working status and the posture health report of the adaptive posture correction detection device, the charging interface 12 is used for charging or supplying power to the adaptive posture correction detection device, the data transmission interface 11 is used for performing data transmission with the outside, such as posture detection data, and the key module 5 is used for controlling the relevant operations of the module.
Specifically, the communication module 2 may communicate with an external client or device based on communication technologies such as WIFI, bluetooth, 2G/3G/4G, and the like, so as to upload posture-related data, and preferably, the control unit may send a posture detection condition for a period of time to the mobile client via the communication module 2 to generate a posture health report for the user to refer to.
The key module 5 of the present embodiment is provided with a button 6, and the button 6 may be a physical button 6 or a virtual button 6, and specifically may perform diversified operations such as alarm cancellation, a posture detection switch, mode selection, device reset, and the like.
Preferably, the gesture determination unit is further configured to perform a user identity recognition function, and when a new user feature is recognized, prompt is performed or corresponding information is transmitted to the mobile client through the communication unit, so that the user is timely reminded of the currently detected identity information. The gesture determination unit can also be used for an environment scene recognition function, tool information used by the current user is collected through the camera 10, recognizable use tools include but are not limited to books, pens, mobile phones, tablets and the like, the work scene of the current user is determined according to the recognized use tool information, and on the basis, the control module is used for adjusting work parameters such as brightness and color tones of external products to protect the eyesight of the user.
Preferably, the present embodiment further comprises a speaker and a microphone for voice interaction, in signal connection with the control module, wherein the speaker and the microphone are also used for acoustic ranging between the device and the user to assist in adjusting the fixed position of the device. The device of this embodiment is through setting up loudspeaker and microphone, not only conveniently carries out human-computer interaction to play the function of sound wave range finding at the fixed process of self-adaptation formula gesture error correction detection device, but auxiliary device's installation when human distance device is too near or too far away, suggestion user adjusting device's position, so, improved self-adaptability, enlarged application range extensively. Similarly, the warning module can also carry out voice warning through the loudspeaker.
Preferably, this embodiment configures an infrared sensor for the device, detects whether someone enters the effective detection range, and the device realizes automatic opening when detecting someone is in the effective detection range, and realizes automatic dormancy when the user leaves the effective detection range.
The embodiment provides a pair of self-adaptation formula gesture error correction detection device, can carry out the gesture detection of pertinence according to different crowds, indicate the current gesture of user to be in wrong state when detecting the gesture mistake, stop the suggestion when the user gets back to the correct gesture, avoid the user to be in the gesture state to the human injury for a long time, help the user to form good gesture custom under the prerequisite that does not restrict the user action, for healthy health protection drives and protects the navigation, product simple structure, easily with other article assembly combination, have very big practicality in the gesture detection effect.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, it is still within the scope of the present invention if they fall within the scope of the claims of the present invention and their equivalents.
Claims (13)
1. An adaptive attitude error correction detection method is characterized by comprising the following steps:
s1: fixing the position of a camera, and configuring configuration information of a current attitude detection state;
s2: selecting a posture detection model according to the configuration information;
s3: the camera collects a current attitude image, and the selected attitude detection model detects the attitude image: if the detected posture is abnormal, an alarm is sent out, otherwise, the step S3 is repeated.
2. The adaptive pose error correction detection method according to claim 1, wherein the configuration information comprises user information, usage environment information, and a pose image library, and wherein the step S1 is preceded by the steps of:
a1: collecting the configuration information under different attitude detection states;
a2: and training through different configuration information to obtain a plurality of different posture detection models.
3. The adaptive pose error correction detection method according to claim 2, wherein the pose detection model is a classification model, and in step a2, the images required by the classification model are labeled with pose categories.
4. The adaptive pose error correction detection method according to claim 2, wherein the pose detection model is a regression model, and in step a2, the images required by the regression model are labeled with pose parameters.
5. The adaptive pose error correction detection method of claim 4, wherein the pose parameters comprise eye distance from table top, left and right shoulder height, head lowering angle, head turning angle, head tip angle.
6. The adaptive posture error correction detection method according to any one of claims 2-4, wherein in step S1, the usage environment information includes a pose position and an angle of the camera, wherein the pose position and the angle are configured by manual input or automatic recognition;
in step S2, the specific step of selecting the posture detection model according to the configuration information is: and selecting the posture detection model according to the placing position and the angle.
7. The adaptive attitude error correction detection method according to claim 6, wherein the configuring of the placement position and the angle by automatic recognition specifically comprises the steps of:
b1: establishing a camera detection model, and collecting or inputting a standard attitude image, wherein the camera detection model is a regression model;
b2: regressing the standard attitude image through the camera detection model to obtain angles of head lowering, head turning and head deviation;
b3: and reversely deducing to obtain the placing position and the angle of the camera according to the angles of the low head, the rotating head and the offset head.
8. An adaptive attitude error correction detection apparatus, characterized in that the attitude detection is performed by applying the adaptive attitude error correction detection method according to any one of claims 1 to 7, comprising: the device comprises a shell, a fixed connecting piece arranged on the shell, and a posture detection module, a warning module and a control module which are arranged in the shell;
the gesture detection module comprises a camera for collecting gesture images and a gesture judgment unit for detecting the gesture images and generating regulation and control signals;
the control module is respectively in signal connection with the attitude determination unit and the warning module and is used for controlling the warning module to warn according to the regulation and control signal, configuring configuration information of a current attitude detection state and selecting an attitude detection model of the attitude determination unit according to the configuration information.
9. The apparatus according to claim 8, wherein the fixing member is a magnetic member, a locking member, an adhesive member, a clamping member, or a locking member.
10. The adaptive posture error correction detection device of claim 8, further comprising a carrier, wherein the housing is internally fixed in the carrier or externally fixed on the carrier through the fixed connection piece, and the carrier is a table lamp, a sound box, a display, a water cup, a clock, a desk or a bookshelf.
11. The adaptive attitude correction detection device according to claim 8, further comprising a communication module electrically connected to the control module for feeding back an operating status and an attitude health report of the adaptive attitude correction detection device.
12. The adaptive posture error correction detection device according to any one of claims 8-11, wherein the control module is further configured to count a maintenance time for a normal posture, and control the warning module to perform a fatigue prompt according to the maintenance time.
13. The adaptive attitude correction detection device according to any one of claims 8-11, further comprising a speaker and a microphone in signal connection with the control module for voice interaction, wherein the speaker and the microphone are further used for acoustic ranging between the device and a user to assist in adjusting the fixed position of the device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911310507.6A CN111160165B (en) | 2019-12-18 | 2019-12-18 | Self-adaptive posture error correction detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911310507.6A CN111160165B (en) | 2019-12-18 | 2019-12-18 | Self-adaptive posture error correction detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111160165A true CN111160165A (en) | 2020-05-15 |
| CN111160165B CN111160165B (en) | 2023-12-19 |
Family
ID=70557842
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911310507.6A Active CN111160165B (en) | 2019-12-18 | 2019-12-18 | Self-adaptive posture error correction detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111160165B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112998699A (en) * | 2021-04-02 | 2021-06-22 | 四川写正智能科技有限公司 | Radar sensor system for detecting user motion state and detection method |
| CN113836965A (en) * | 2020-06-08 | 2021-12-24 | 北京字节跳动网络技术有限公司 | Method, device, terminal and storage medium for processing information |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006004043A1 (en) * | 2004-07-07 | 2006-01-12 | Nec Corporation | Wide field-of-view image input method and device |
| CN103034859A (en) * | 2012-12-13 | 2013-04-10 | 华为技术有限公司 | Method and device for obtaining posture model |
| CN103955272A (en) * | 2014-04-16 | 2014-07-30 | 北京尚德智产投资管理有限公司 | Terminal equipment user posture detecting system |
| CN206674138U (en) * | 2017-04-01 | 2017-11-24 | 连云港艾加加信息科技有限公司 | Video-assisted equipment and set top box with posture error correction |
| CN109948509A (en) * | 2019-03-11 | 2019-06-28 | 成都旷视金智科技有限公司 | Obj State monitoring method, device and electronic equipment |
| CN110321786A (en) * | 2019-05-10 | 2019-10-11 | 北京邮电大学 | A kind of human body sitting posture based on deep learning monitors method and system in real time |
| CN110414419A (en) * | 2019-07-25 | 2019-11-05 | 四川长虹电器股份有限公司 | A kind of posture detecting system and method based on mobile terminal viewer |
-
2019
- 2019-12-18 CN CN201911310507.6A patent/CN111160165B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006004043A1 (en) * | 2004-07-07 | 2006-01-12 | Nec Corporation | Wide field-of-view image input method and device |
| CN103034859A (en) * | 2012-12-13 | 2013-04-10 | 华为技术有限公司 | Method and device for obtaining posture model |
| CN103955272A (en) * | 2014-04-16 | 2014-07-30 | 北京尚德智产投资管理有限公司 | Terminal equipment user posture detecting system |
| CN206674138U (en) * | 2017-04-01 | 2017-11-24 | 连云港艾加加信息科技有限公司 | Video-assisted equipment and set top box with posture error correction |
| CN109948509A (en) * | 2019-03-11 | 2019-06-28 | 成都旷视金智科技有限公司 | Obj State monitoring method, device and electronic equipment |
| CN110321786A (en) * | 2019-05-10 | 2019-10-11 | 北京邮电大学 | A kind of human body sitting posture based on deep learning monitors method and system in real time |
| CN110414419A (en) * | 2019-07-25 | 2019-11-05 | 四川长虹电器股份有限公司 | A kind of posture detecting system and method based on mobile terminal viewer |
Non-Patent Citations (1)
| Title |
|---|
| 韩晓明;冉春风;侯雪峰;关然;潘巍;: "基于视频的学生坐姿检测与纠正***研究" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113836965A (en) * | 2020-06-08 | 2021-12-24 | 北京字节跳动网络技术有限公司 | Method, device, terminal and storage medium for processing information |
| CN112998699A (en) * | 2021-04-02 | 2021-06-22 | 四川写正智能科技有限公司 | Radar sensor system for detecting user motion state and detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111160165B (en) | 2023-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104239860B (en) | Sitting posture detection and reminding method and device during use of intelligent terminal | |
| CN204631388U (en) | A kind of new-type intelligent glasses structure | |
| WO2014094486A1 (en) | Device for preventing myopia and cervical spondylosis | |
| CN204303137U (en) | A kind of body posture monitoring terminal and system | |
| CN105118247B (en) | A kind of spectacle-frame type Intelligent eyesight protection instrument | |
| CN204901502U (en) | Desk lamp with position of sitting monitor function | |
| CN111160165A (en) | Self-adaptive posture error correction detection method and device | |
| CN103908065A (en) | Intelligent desk with sitting posture correcting function and correcting method implemented by intelligent desk | |
| CN110632773A (en) | Intelligent myopia prevention and control glasses | |
| CN106569612A (en) | Method for protecting eyes, display device for protecting eyes and terminal | |
| CN103976733A (en) | Multi-passage brain wave control glasses | |
| US20130265169A1 (en) | Eyewear Device Configured To Track Head Movement | |
| CN104503537A (en) | Computer display with function of preventing myopia and lumbar vertebra and vertebral column diseases | |
| CN109164918A (en) | Intelligent sitting posture tracking and method of adjustment, device, intelligent elevated table and storage medium | |
| CN109255314A (en) | Information cuing method, device, intelligent glasses and storage medium | |
| CN207946657U (en) | A kind of electronic watch based on eyesight protection system | |
| CN103529563A (en) | Eyewear device capable of monitoring and preventing cervical vertebra diseases in real time | |
| CN107730834A (en) | A kind of antifatigue Intelligent worn device being combined based on attitude detection with image recognition | |
| CN202171700U (en) | Spectacle capable of correcting sitting posture | |
| CN104914591A (en) | Eyeglasses capable of monitoring distance from display screen | |
| CN113160260A (en) | Head-eye double-channel intelligent man-machine interaction system and operation method | |
| CN206532049U (en) | The anti-fatigue eyeglasses of novel intelligent | |
| CN204463383U (en) | A kind of based on monolithic processor controlled Intelligent eyesight protection instrument | |
| CN209278917U (en) | A kind of stereoscopic vision sitting posture monitoring attention gatter | |
| CN212183797U (en) | Luminous environment adjusting device suitable for public health safety |
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 |