CN113313917A - Method for solving false alarm generated when no target exists in front of detector in sitting posture detection - Google Patents
Method for solving false alarm generated when no target exists in front of detector in sitting posture detection Download PDFInfo
- Publication number
- CN113313917A CN113313917A CN202010118356.0A CN202010118356A CN113313917A CN 113313917 A CN113313917 A CN 113313917A CN 202010118356 A CN202010118356 A CN 202010118356A CN 113313917 A CN113313917 A CN 113313917A
- Authority
- CN
- China
- Prior art keywords
- sitting posture
- face
- detection
- standard
- data
- 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
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
Abstract
The invention provides a method for solving the problem of false alarm generated when no target exists in front of a detector in sitting posture detection, which comprises the following steps: s1 processing the sitting posture calculation module for the first time; s2 registering a standard sitting posture; s3 judging whether the registration is successful, if not, proceeding to S2; if yes, go to S4; s4 again processing the sitting posture calculation module; s5 comparing the real-time sitting posture data with the standard sitting posture data; s6, judging whether the deviation exceeds the threshold value, if not, entering S4; if yes, go to S7; s7 judging whether the state duration time exceeds the designated period, if not, entering S4, if yes, carrying out S8; s8 judges whether or not a target is present: judging by a face counting accumulator, if the face counting accumulator is not more than 1, detecting no face in the detection period, judging that no target person is in the detection range of the detector, and carrying out S9; s9 alarm of abnormal sitting posture; s10 again processes the sitting posture calculation module; s11 comparing the real-time sitting posture data with the standard sitting posture data again; s12, judging whether the sitting posture meets the standard, if not, entering S9; if yes, the alarm is cancelled and the process goes to S4.
Description
Technical Field
The invention relates to the field of face recognition, in particular to a method for solving the problem of false alarm generated when no target exists in front of a detector in sitting posture detection.
Background
With the continuous development of science and technology, particularly the development of computer vision technology, the face recognition technology is widely applied to various fields of information security, electronic authentication and the like, and the image feature extraction method has good recognition performance. Among them, sitting posture detection is also an important content in the field of recognition. In the prior art, the method comprises the steps of 1, measuring the distance between a person and a sensor based on ultrasonic waves, laser lines and the like, and judging whether the sitting posture is standard or not according to the distance. 2. Based on a common image sensor, feature data such as a face angle, a human figure posture and the like are obtained and are matched with preset standard posture feature data so as to judge whether the sitting posture is abnormal or not.
However, in the prior art, the following disadvantages exist:
1. the method has the advantages that the distance between a person and the detection device is detected based on laser rays and ultrasonic waves, whether the sitting posture of the person is correct or not is judged according to the detection result, the method has a good detection effect on large-amplitude posture change, but the method cannot accurately judge local unhealthy sitting postures such as a side face and a head bending position, and the ultrasonic waves and the laser rays used by the methods are harmful to the health of the person in the radioactive environments for a long time.
2. Patent CN102096801A is based on the information of image, through the angle of inclination of acquireing the people's face, glasses area difference, head and shoulder curve difference to it is the position of sitting anomaly to do with the initial standard data who obtains to be poor and exceed the difference of preset threshold value, because of only solving in this method and waiting to detect the difference that the target is in the position of sitting normal and unusual in detection scope, do not solve and wait to detect the condition that the target leaves the detection scope, can lead to the detector to judge that the unusual position of sitting is reported to the police this moment, influence and use experience.
3. The patent CN110334631A is based on face detection and sitting posture detection of binary operation, and uses the adaboost face detection algorithm to position the head position of the standard sitting posture, and sets a tolerance as a standard, and detects whether the target head position exceeds the standard tolerance as a sitting posture abnormality judgment standard. Because of shallow layer face detection algorithm or degree of depth CNN face detection algorithm based on such as the adboost, have certain limit to face detection ability, too big or the application environment complicacy of face angle can lead to face detection algorithm can't detect the face, influence follow-up judgement, and the same unable differentiation of no face detection output condition when this kind of condition is normally left the detection range with the target of waiting to detect consequently can lead to the wrong report to produce, influences use experience.
Common terms in the prior art include:
1. the face detection module: an image containing a complete face is input, and the face detection module returns the bounding box (bounding box) coordinates of the face in the image.
2. Face space angle: the human face three-angle face three-dimensional face angles, wherein the pitch rotates around the y axis; yaw is rotation about the z-axis; the roll is rotated about the x-axis. The explanation for pitch, yaw, roll is shown in FIG. 3.
3. Face angle detection module: inputting a gray image of ROI data of a human face, and acquiring a spatial angle of the human face, wherein the ROI (region of interest) is an interested region. In machine vision and image processing, a region to be processed, called a region of interest, ROI, is delineated from a processed image in the form of a box, circle, ellipse, irregular polygon, or the like.
4. A face counting accumulator: and when the face detection module detects a face, the face counting accumulator performs an addition 1 operation.
5. Undetected face count accumulator: and when the face detection module does not detect the face, the undetected face counting accumulator is added with 1 for operation.
6. Sitting posture information: the system comprises a face space angle, face space position information, a face counting accumulator and an undetected face counting accumulator.
7. Standard sitting posture information: a person has one or more sets of sitting posture information collected by the present invention while sitting at a table or other location in a standard posture.
8. IOU: the IOU is called Intersection over Union (Intersection over Union), and the IOU calculates the ratio of the Intersection and Union of two bounding boxes, as shown in FIG. 4, and is called Intersection/Union. IOU: one concept used in target detection is the overlap ratio of the generated candidate frame (candidate frame) and the original labeled frame (ground truth frame), i.e. the ratio of their intersection to union.
9. Face spatial position offset value: and calculating the IOU of the bounding box as the face position deviation value according to the bounding box detected in the step 1.
Disclosure of Invention
In order to solve the above problems, the present invention is directed to: a new method is provided for solving the problem that when a common image sensor is used, the extraction result of the human face and human figure posture features under the condition that no target person exists and the result of the human face and human figure features which cannot be normally extracted are not distinguished in a sitting posture detection scheme based on a human face and human figure posture feature extraction algorithm.
Specifically, the invention provides a method for solving the problem of false alarm generated when no target exists in front of a detector in sitting posture detection, which comprises the following steps:
s1, a first-time processing sitting posture calculation module for obtaining sitting posture information, comprising: the sitting posture detecting device comprises a face detecting module, a face angle detecting module, a sitting posture information calculating module, a face counting accumulator and an undetected face counting accumulator;
s2, registering a standard sitting posture;
s3, determine whether the registration was successful? If not, re-enter step S2; if yes, carrying out the next step;
s4, processing the sitting posture calculation module again to obtain sitting posture information;
s5, comparing the real-time sitting posture data with the standard sitting posture data;
s6, determine if the deviation exceeds a threshold? If not, re-enter step S4; if yes, carrying out the next step;
s7, determine if the state duration exceeds a specified period? If not, the step S4 is re-entered, and if yes, the next step is carried out;
s8, determining the presence or absence of a target:
judging the face counting accumulator, if the face counting accumulator is not more than 1, determining that no face is detected in the detection period, and determining that no target person is in the detection range of the detector, and performing the next step S9;
s9, alarming for abnormal sitting posture, and assigning Alarm to 1;
s10, processing the sitting posture calculation module again to obtain sitting posture information;
s11, comparing the data of the real-time sitting posture and the standard sitting posture again;
s12, determine whether the sitting posture meets the standard? If not, re-enter step S9; if yes, the Alarm is cancelled, Alarm is reset to 0, and the process re-enters step S4.
Thus, the present application has the advantages that: by adopting the method, the sitting posture can be detected only by using a common image sensor, the abnormal sitting posture alarm can be realized, the influence of similar ultrasonic waves and laser rays on the human health can be avoided, the product cost can be reduced, and particularly, the method well solves the problem that the extraction result of the human face and human-shaped posture characteristics under the condition without a target figure and the extraction result of the human face and human-shaped characteristics which are not normally extracted by the target figure cannot be distinguished in the prior art.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention.
Fig. 1 is a schematic block diagram of a flow chart of a method to which the present invention relates.
FIG. 2 is a schematic block diagram of a sitting posture calculation module in steps of the method of the present invention.
FIG. 3 is a schematic view of the sitting posture detection of the present invention.
Fig. 4 is a schematic diagram of the IOU ratio to which the present invention relates.
Detailed Description
In order that the technical contents and advantages of the present invention can be more clearly understood, the present invention will now be described in further detail with reference to the accompanying drawings.
As shown in FIG. 1, the present invention relates to a method for solving the false alarm generated when there is no target in front of the detector in the sitting posture detection, which comprises the following steps:
s1, a first-time processing sitting posture calculation module for obtaining sitting posture information, comprising: the sitting posture detecting device comprises a face detecting module, a face angle detecting module, a sitting posture information calculating module, a face counting accumulator and an undetected face counting accumulator;
s2, registering a standard sitting posture;
s3, determine whether the registration was successful? If not, re-enter step S2; if yes, carrying out the next step;
s4, processing the sitting posture calculation module again to obtain sitting posture information;
s5, comparing the real-time sitting posture data with the standard sitting posture data;
s6, determine if the deviation exceeds a threshold? If not, re-enter step S4; if yes, carrying out the next step;
s7, determine if the state duration exceeds a specified period? If not, the step S4 is re-entered, and if yes, the next step is carried out;
s8, determining the presence or absence of a target:
judging the face counting accumulator, if the face counting accumulator is not more than 1, determining that no face is detected in the detection period, and determining that no target person is in the detection range of the detector, and performing the next step S9;
s9, alarming for abnormal sitting posture, and assigning Alarm to 1;
s10, processing the sitting posture calculation module again to obtain sitting posture information;
s11, comparing the data of the real-time sitting posture and the standard sitting posture again;
s12, determine whether the sitting posture meets the standard? If not, re-enter step S9; if yes, the Alarm is cancelled, Alarm is reset to 0, and the process re-enters step S4.
As shown in fig. 2, the processing seating posture calculating module of step S1, S4 or S10 further includes:
a, the face detection module obtains a face bounding box from image data;
b, deducting human face ROI data according to the human face boundary frame and transmitting the human face ROI data to a human face angle detection module, acquiring a human face space angle, adding 1 to a human face counting accumulator, wherein when the human face detection module cannot acquire the human face boundary frame from image data, the human face counting accumulator does not process, and the human face counting accumulator is not detected and is added with 1;
and c, integrating the face bounding box to be obtained and the face space angle together by the sitting posture information calculating module, and updating the face counting accumulator and the undetected face counting accumulator for other modules to calculate.
The step S2 is registered in a standard sitting posture, and further includes: the sitting posture information is obtained through step S1, and a preset number n of sitting posture information results are continuously obtained as a result of the sitting posture information obtained from one frame of data, where n is a positive integer greater than or equal to 3.
In step S2, the method further includes: the condition for judging the success of registering the standard sitting posture is that,
1) at least 3 maximum values of the angle deviation values in the sitting posture information result are smaller than a preset first threshold value;
2) the spatial position deviation value meets a preset second threshold value;
3) the above two conditions 1) and 2) must be satisfied simultaneously, then the standard sitting posture registration is successful.
Comparing the real-time sitting posture with the standard sitting posture deviation in the step S5 or S11, further comprising:
A. in the detection, acquiring sitting posture information of the current frame and performing deviation calculation on the registered sitting posture information;
B. and calculating a face space angle deviation value and a face space position deviation value.
In step S6, whether the deviation exceeds a threshold value further includes: if one or both of the calculated deviation values of the face space angle and the face space position deviation value exceed the preset threshold, or if the face count accumulator is not detected to exceed the preset threshold, the current sitting posture is considered to be abnormal, and the next step S7 is performed.
In step S7, the determining whether the state duration exceeding the threshold exceeds a predetermined period further includes: when the deviation exceeds the threshold value, starting to accumulate the number of the abnormal sitting posture frames and simultaneously starting to time when the sitting posture is abnormal, and when the abnormal sitting posture count exceeds the preset times in the specified period, confirming that the abnormal sitting posture is established, giving an alarm signal, and performing the next step S8; otherwise, the information is considered to be false information, the step S4 of the sitting posture calculation module is carried out again, and the sitting posture data information is continuously monitored.
The preset times are positive integers larger than 1.
The Alarm signal is given, and Alarm is equal to 1.
In step S8, if it is determined that no target person is within the detector detection range, the Alarm signal is cancelled and Alarm is set to 0.
The specified period is a preset time period, for example, 1 minute or 3 minutes, and is preset according to the needs of the customer.
The preset times are positive integers larger than 1, such as 1 time, 3 times or 5 times, 10 times and the like, and are set according to the needs of customers.
The methods for calculating the face space angle deviation value, the face space position deviation value, and the preset threshold are already relatively mature prior art, for example, refer to the schemes in patents CN102096801A and CN110334631A, and will not be described herein again.
In addition, the technical solutions specifically referred to in the present application can be further explained as follows:
1. a sitting posture calculation module:
the face detection module obtains a face bounding box from the image data, deducts face ROI data according to the face bounding box and transmits the face ROI data to the face angle detection module, obtains a face space angle, and adds 1 to a face counting accumulator, when the face detection module cannot obtain the face bounding box from the image data, the face counting accumulator does not process, and the sum of 1 is added to the undetected face counting accumulator. And calculating sitting posture information, namely integrating the obtained face bounding box and the face space angle, and updating the face counting accumulator and the undetected face counting accumulator for other modules to calculate.
2. Registering a standard sitting posture:
the sitting posture information is acquired through the sitting posture detection module 1, the sitting posture information acquired by one frame of data is used as a result, and the preset number of sitting posture results are continuously acquired. The condition for successfully verifying the sitting posture registration comprises 1) that the maximum value of three angle deviations in the sitting posture result is smaller than a preset threshold value and 2) that the spatial position deviation value meets the preset threshold value. The two conditions are simultaneously met, namely the standard sitting posture registration is successful.
3. Comparing the deviation between the real-time sitting posture and the standard sitting posture:
and in the detection, the sitting posture information of the current frame and the registered sitting posture information are obtained to carry out deviation calculation, and a face space angle deviation value and a face space position deviation value are calculated.
4. Deviation exceeds a threshold value:
and if neither deviation value obtained by calculation in the step 3 exceeds a preset threshold value, the currently collected sitting posture is considered to be normal, the sitting posture information is continuously monitored, and if one deviation value exceeds the preset threshold value or both deviation values exceed the preset threshold value or the undetected face count accumulator exceeds the preset threshold value, the currently collected sitting posture is considered to be abnormal, and the next judgment is carried out.
5. The state duration period:
and when the sitting posture is judged to be abnormal in step 4, accumulating the number of the abnormal sitting posture frames and starting timing at the same time, when the abnormal sitting posture count exceeds a preset threshold value in a preset time period, determining that the abnormal sitting posture is established, giving an Alarm signal if Alarm is 1, otherwise, considering the abnormal sitting posture count as false information, and continuing calculating by the sitting posture calculation module.
6. Judging whether a target exists or not:
and judging a face counting accumulator, if the face counting accumulator is not more than 1, determining that no face is detected in the detection period, determining that no target person is in the detection range of the detector, cancelling the Alarm signal when Alarm is 0, and starting to carry out steps 1, 3 and 4.
7. Whether the sitting posture is standard or not:
and when the Alarm signal of Alarm 1 is triggered in the step 5, starting the steps 1, 3 and 4, and if the step 4 judges that the sitting posture is normal, canceling the Alarm signal of Alarm 1. And continuing the steps 1, 3 and 4 to monitor the sitting posture information.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for solving the problem of false alarm generated when no target exists in front of a detector in sitting posture detection is characterized by comprising the following steps:
s1, a first-time processing sitting posture calculation module for obtaining sitting posture information, comprising: the sitting posture detecting device comprises a face detecting module, a face angle detecting module, a sitting posture information calculating module, a face counting accumulator and an undetected face counting accumulator;
s2, registering a standard sitting posture;
s3, determine whether the registration was successful? If not, re-enter step S2; if yes, carrying out the next step;
s4, processing the sitting posture calculation module again to obtain sitting posture information;
s5, comparing the real-time sitting posture data with the standard sitting posture data;
s6, determine if the deviation exceeds a threshold? If not, re-enter step S4; if so, proceed to the next step S7;
s7, determine if the state duration exceeds a specified period? If not, re-entering the step S4, if yes, proceeding to the next step S8;
s8, determining the presence or absence of a target:
judging the face counting accumulator, if the face counting accumulator is not more than 1, determining that no face is detected in the detection period, and determining that no target person is in the detection range of the detector, and performing the next step S9;
s9, alarming for abnormal sitting posture, and assigning Alarm to 1;
s10, processing the sitting posture calculation module again to obtain sitting posture information;
s11, comparing the data of the real-time sitting posture and the standard sitting posture again;
s12, determine whether the sitting posture meets the standard? If not, re-enter step S9; if yes, the Alarm is cancelled, Alarm is reset to 0, and the process re-enters step S4.
2. The method of claim 1, wherein the sitting posture calculating module of step S1, S4 or S10 further comprises:
a, the face detection module obtains a face bounding box from image data;
b, deducting human face ROI data according to the human face boundary frame and transmitting the human face ROI data to a human face angle detection module, acquiring a human face space angle, adding 1 to a human face counting accumulator, wherein when the human face detection module cannot acquire the human face boundary frame from image data, the human face counting accumulator does not process, and the human face counting accumulator is not detected and is added with 1;
and c, integrating the face bounding box to be obtained and the face space angle together by the sitting posture information calculating module, and updating the face counting accumulator and the undetected face counting accumulator for other modules to calculate.
3. The method as claimed in claim 1, wherein the step S2 of registering in a standard sitting posture further comprises: the sitting posture information data is obtained through step S1, and a preset number n of sitting posture information data results are successively obtained as a result of the sitting posture information data obtained from one frame of data, where n is a positive integer greater than or equal to 3.
4. The method as claimed in claim 3, wherein the step S2 further comprises: the condition for judging the success of registering the standard sitting posture is that,
1) at least 3 maximum values of the angle deviation values in the sitting posture information result are smaller than a preset first threshold value;
2) the spatial position deviation value meets a preset second threshold value;
3) the above two conditions 1) and 2) must be satisfied simultaneously, then the standard sitting posture registration is successful.
5. The method of claim 1, wherein comparing the real-time sitting posture with the standard sitting posture deviation in step S5 or S11 further comprises:
A. in the detection, acquiring sitting posture information of the current frame and performing deviation calculation on the registered sitting posture information;
B. and calculating a face space angle deviation value and a face space position deviation value.
6. The method as claimed in claim 1, wherein the step S6, if the deviation exceeds the threshold, further comprises: if one or both of the calculated deviation values of the face space angle and the face space position deviation value exceed the preset threshold, or if the face count accumulator is not detected to exceed the preset threshold, the current sitting posture is considered to be abnormal, and the next step S7 is performed.
7. The method of claim 1, wherein the step S7 of determining whether the duration of the state exceeding the threshold exceeds a specified period further comprises: when the deviation exceeds the threshold value, starting to accumulate the number of the abnormal sitting posture frames and simultaneously starting to time when the sitting posture is abnormal, and when the abnormal sitting posture count exceeds the preset times in the specified period, confirming that the abnormal sitting posture is established, giving an alarm signal, and performing the next step S8; otherwise, the information is considered to be false information, the step S4 is carried out again to process the sitting posture calculation module, and the sitting posture data information is continuously monitored.
8. The method of claim 7, wherein the predetermined number is a positive integer greater than 1.
9. The method for solving the problem of false Alarm generated when there is no target in front of the detector in sitting posture detection as claimed in claim 7, wherein said Alarm signal is given and Alarm is set to 1.
10. The method according to claim 7, wherein in step S8, if it is determined that there is no target person in the detection range of the detector, the Alarm signal is cancelled and Alarm is set to 0; if the face count accumulator is greater than 1, it is determined that a face is detected in the detection period, and it is determined that a target person is within the detection range of the detector, and step S4 is performed again.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010118356.0A CN113313917B (en) | 2020-02-26 | 2020-02-26 | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010118356.0A CN113313917B (en) | 2020-02-26 | 2020-02-26 | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113313917A true CN113313917A (en) | 2021-08-27 |
| CN113313917B CN113313917B (en) | 2022-12-16 |
Family
ID=77370110
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010118356.0A Active CN113313917B (en) | 2020-02-26 | 2020-02-26 | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113313917B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096801A (en) * | 2009-12-14 | 2011-06-15 | 北京中星微电子有限公司 | Sitting posture detecting method and device |
| CN105788185A (en) * | 2014-12-24 | 2016-07-20 | 刁宇童 | Method and device for monitoring sitting posture |
| CN107491751A (en) * | 2017-08-14 | 2017-12-19 | 成都伞森科技有限公司 | Sitting posture analysis method and device |
| WO2018113582A1 (en) * | 2016-12-22 | 2018-06-28 | 欧普照明股份有限公司 | Sitting posture recognition system and sitting posture recognition method |
| CN109523755A (en) * | 2018-12-17 | 2019-03-26 | 石家庄爱赛科技有限公司 | Stereoscopic vision sitting posture reminder and based reminding method |
| CN109872359A (en) * | 2019-01-27 | 2019-06-11 | 武汉星巡智能科技有限公司 | Sitting posture detecting method, device and computer readable storage medium |
| CN110334631A (en) * | 2019-06-27 | 2019-10-15 | 西安工程大学 | A kind of sitting posture detecting method based on Face datection and Binary Operation |
-
2020
- 2020-02-26 CN CN202010118356.0A patent/CN113313917B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096801A (en) * | 2009-12-14 | 2011-06-15 | 北京中星微电子有限公司 | Sitting posture detecting method and device |
| CN105788185A (en) * | 2014-12-24 | 2016-07-20 | 刁宇童 | Method and device for monitoring sitting posture |
| WO2018113582A1 (en) * | 2016-12-22 | 2018-06-28 | 欧普照明股份有限公司 | Sitting posture recognition system and sitting posture recognition method |
| CN107491751A (en) * | 2017-08-14 | 2017-12-19 | 成都伞森科技有限公司 | Sitting posture analysis method and device |
| CN109523755A (en) * | 2018-12-17 | 2019-03-26 | 石家庄爱赛科技有限公司 | Stereoscopic vision sitting posture reminder and based reminding method |
| CN109872359A (en) * | 2019-01-27 | 2019-06-11 | 武汉星巡智能科技有限公司 | Sitting posture detecting method, device and computer readable storage medium |
| CN110334631A (en) * | 2019-06-27 | 2019-10-15 | 西安工程大学 | A kind of sitting posture detecting method based on Face datection and Binary Operation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113313917B (en) | 2022-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108629946B (en) | Human body falling detection method based on RGBD sensor | |
| Dubois et al. | Human activities recognition with RGB-Depth camera using HMM | |
| CN107219925B (en) | Posture detection method and device and server | |
| CN108875533B (en) | Face recognition method, device, system and computer storage medium | |
| CN109830078B (en) | Intelligent behavior analysis method and intelligent behavior analysis equipment suitable for narrow space | |
| CN106919254B (en) | Head-mounted virtual reality equipment and safety protection method | |
| CN111539253A (en) | Face detection method, device, terminal and storage medium | |
| CN112861564A (en) | Realization device for sitting posture detection | |
| US20220095959A1 (en) | Feigned Injury Detection Systems And Methods | |
| US20190370996A1 (en) | Image processing apparatus | |
| Wang et al. | Fall detection algorithm for the elderly based on human characteristic matrix and SVM | |
| CN113313917B (en) | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection | |
| WO2020217812A1 (en) | Image processing device that recognizes state of subject and method for same | |
| CN113312938A (en) | Method and system for preventing false alarm generated when no target exists in front of detector in sitting posture detection | |
| CN113312939A (en) | System for preventing false alarm generated when no target exists in front of detector in sitting posture detection | |
| CN112861562A (en) | Detection method and system for detecting sitting posture abnormality | |
| CN112861563A (en) | Sitting posture detection method and system | |
| US20220284790A1 (en) | Posture discrimination device | |
| CN112800885B (en) | Data processing system and method based on big data | |
| Kong et al. | Three-states-transition method for fall detection algorithm using depth image | |
| US20210019508A1 (en) | System and Method for Detecting Potentially Dangerous Situations Based on Video Data | |
| CN113739819A (en) | Verification method and device, electronic equipment, storage medium and chip | |
| JPH0991432A (en) | Method for extracting doubtful person | |
| Minar et al. | Intruder detection for automated access control systems with Kinect device | |
| CN113129374A (en) | Method and device for assisting reading and writing posture judgment by laser point cloud face recognition |
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 |