CN110458010A - A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment - Google Patents
A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment Download PDFInfo
- Publication number
- CN110458010A CN110458010A CN201910601482.9A CN201910601482A CN110458010A CN 110458010 A CN110458010 A CN 110458010A CN 201910601482 A CN201910601482 A CN 201910601482A CN 110458010 A CN110458010 A CN 110458010A
- Authority
- CN
- China
- Prior art keywords
- head
- user
- shoulder
- sitting posture
- human body
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 208000001491 myopia Diseases 0.000 claims abstract description 37
- 238000004891 communication Methods 0.000 claims description 25
- 230000008859 change Effects 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 12
- 241001282135 Poromitra oscitans Species 0.000 claims description 8
- 206010048232 Yawning Diseases 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 3
- 238000000465 moulding Methods 0.000 claims 1
- 210000003128 head Anatomy 0.000 abstract description 70
- 230000004379 myopia Effects 0.000 abstract description 16
- 238000001514 detection method Methods 0.000 abstract description 14
- 230000002265 prevention Effects 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 5
- NJPPVKZQTLUDBO-UHFFFAOYSA-N novaluron Chemical compound C1=C(Cl)C(OC(F)(F)C(OC(F)(F)F)F)=CC=C1NC(=O)NC(=O)C1=C(F)C=CC=C1F NJPPVKZQTLUDBO-UHFFFAOYSA-N 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000004438 eyesight Effects 0.000 description 3
- 239000004519 grease Substances 0.000 description 3
- 238000001914 filtration Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004397 blinking Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 210000000744 eyelid Anatomy 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000035764 nutrition Effects 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000002618 waking effect Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F21—LIGHTING
- F21S—NON-PORTABLE LIGHTING DEVICES; SYSTEMS THEREOF; VEHICLE LIGHTING DEVICES SPECIALLY ADAPTED FOR VEHICLE EXTERIORS
- F21S6/00—Lighting devices intended to be free-standing
- F21S6/002—Table lamps, e.g. for ambient lighting
- F21S6/003—Table lamps, e.g. for ambient lighting for task lighting, e.g. for reading or desk work, e.g. angle poise lamps
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F21—LIGHTING
- F21V—FUNCTIONAL FEATURES OR DETAILS OF LIGHTING DEVICES OR SYSTEMS THEREOF; STRUCTURAL COMBINATIONS OF LIGHTING DEVICES WITH OTHER ARTICLES, NOT OTHERWISE PROVIDED FOR
- F21V33/00—Structural combinations of lighting devices with other articles, not otherwise provided for
- F21V33/0064—Health, life-saving or fire-fighting equipment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Ophthalmology & Optometry (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of intelligent desk lamps, including table lamp body, camera, control module, reminding module, wherein: control module is used to receive the video image of camera acquisition, user's human body contour outline figure is generated by preset rules, judge head and shoulder position, the difference for calculating head and shoulder position judges near-sighted sitting posture occurred and controls reminding module output prompting message if difference meets preset threshold.Intelligent desk lamp of the invention calculates the relative position of human body head and shoulder by control module to judge whether user's sitting posture is accurate, when overcoming the prior art and directly acquiring the specific features such as eyes of user or mouth, because individual difference it is inconsistent to camera requirement for height caused by sitting posture detect error situation, have many advantages, such as that sitting posture detection accuracy is high, practical, safe and reliable, the pre- myopia prevention of user can be helped well.The invention also discloses a kind of sitting posture detecting method and a kind of electronic equipment.
Description
Technical field
The present invention relates to intelligent electric appliance technical field, it is a kind of to relate generally to a kind of intelligent desk lamp, a kind of sitting posture detecting method
Electronic equipment.
Background technique
There is news report to claim, adolescents in China rate of myopia height ranks first in the world: the rate of myopia of Chinese high school student and university student
More than seventy percent, and increase year by year, pupil's rate of myopia is also close to 40%.In contrast, U.S. students in middle and primary schools rate of myopia is only
It is 10%.In inducing the big factor of myopia three (heredity, environment and nutrition), scientists think, hair of the environmental factor to myopia
Raw role is bigger, wherein abnormal sitting posture is really (improper including Writing distance) key factor for influencing user's eyesight, is needed
User's correction is reminded using suitable technological means, to protect user's eyesight.
Desk lamp is the schooling apparatus of student's indispensability, since current school work competitive pressure is very big, is fought by torchlight in also becoming
The homely food of state student, this is but also importance of the desk lamp in the daily life of student is higher and higher.Current desk lamp is poor
The opposite sex is mainly reflected in the power of light, different using different light sources, such as the light of incandescent lamp, LED light, fluorescent lamp.
Occur in the market in recent years many " myopia-proof intelligent lamp ", mainly carries out face using multiple sensors or binocular camera
Calibration, to judge the Writing distance of student.For example eyes of user is observed, if the eyes of user cannot be captured by camera
It arrives, then judges that the height of writing of user judges use if such case duration exceeds predetermined value lower than normal level value
Family was write closely, to eyes nocuousness.Although this method is simple, since user has individual difference, accurate setting sensing
Itself is a difficult points with local features such as enough capture eyes for device or camera shooting grease head highness, if sensor or camera shooting grease head highness setting
It is improper, judge that the result of sitting posture is also inaccurate.In addition, sensor radiation is strong, it is larger to the actual bodily harm of user.
Summary of the invention
In view of this, user's sitting posture detection function, and structure can be accurately realized it is really necessary to propose a kind of intelligent desk lamp
Simply, securely and reliably, cost performance it is high.
A kind of intelligent desk lamp, including table lamp body, camera, control module, reminding module, in which:
Camera is fixed on table lamp body, for acquiring the user video image of predeterminable area, and by collected video
Image is sent to control module, and the predeterminable area includes user's head, neck, shoulder;
Control module generates user's human body contour outline figure by preset rules, sentences for receiving the video image of camera acquisition
Broken end portion and shoulder position, the difference for calculating head and shoulder position are judged if difference meets preset threshold
Show near-sighted sitting posture and controls reminding module output prompting message.
It further, include: to utilize background subtraction or interframe by the method that preset rules generate user's human body contour outline figure
Calculus of finite differences obtains human body contour outline.
Further, the method that control module calculates the difference of head and shoulder position includes: to user's human body wheel
Exterior feature figure carries out gradient calculating, and using second largest position of change of gradient in human body contour outline figure as head position, change of gradient is maximum
Position as shoulder position, calculate the difference of head position and shoulder position.
Further, it includes: to calculate from bottom to top that control module, which calculates head and the method for the difference of shoulder position,
The width of adjacent position in human body contour outline figure, if current width is than upper position reduced width presupposition multiple, judgement upper one
Position is shoulder position;If current width increases presupposition multiple than upper position width, judge current location for head position
It sets, calculates the difference of head position and shoulder position.
Further, the control module is also used to video image identification human eye based on the received, and calculates preset time
Number of winks in section inner video image judges fatigue state occurred if number of winks is more than preset threshold, and controls and mention
Module of waking up exports prompting message.
Further, the control module is also used to video image identification mouth based on the received, and calculates preset time
Number of yawning in section inner video image, if number of yawning judges fatigue state occurred, and control more than preset threshold
Reminding module processed exports prompting message.
Further, the intelligent desk lamp further includes user characteristics input module, described for obtaining active user's feature
Control module is provided with preset threshold corresponding with user characteristics, and the control module is obtained according to user characteristics input module
User characteristics call corresponding preset threshold.
Further, the method for control module identification human eye or mouth includes: the view that control module receives camera acquisition
Image recognition is carried out using the library function of preset OpenCV after frequency image.
Further, the reminding module is alarm lamp or sound and light alarm device or voice alarming device.
Further, the desk lamp further includes communication module, and user is had myopia by communication module and sat by control module
The information of appearance or fatigue state is sent to exterior terminal equipment.
Further, the camera is monocular cam.
The invention also discloses a kind of sitting posture detecting methods, can improve the accuracy of sitting posture testing result, and have certain
Versatility can be applicable in wider scene to help the pre- myopia prevention of user.
The present invention includes the following contents:
A kind of sitting posture detecting method, comprising the following steps:
User's human body contour outline figure is generated by preset rules according to the user video image of the predeterminable area of acquisition, it is described default
Region includes user's head, neck, shoulder;
Judge head and shoulder position, calculate the difference on head and shoulder position, if difference meets default threshold
Value, then judge near-sighted sitting posture occurred.
The third aspect, the invention also discloses a kind of electronic equipment, including processor, communication interface, memory and communication
Bus, in which:
Processor, communication interface, memory complete mutual communication by communication bus;
Memory, for storing computer program;
Processor, when for executing the program stored on memory, the user including the predeterminable area according to acquisition is regarded
Frequency image generates user's human body contour outline figure by preset rules, and the predeterminable area includes user's head, neck, shoulder;
Judge head and shoulder position, calculate the difference on head and shoulder position, if difference meets default threshold
Value, then judge near-sighted sitting posture occurred.
The beneficial effect of the present invention compared with prior art is: intelligent desk lamp of the invention is acquired default using camera
Region inner video image calculates the relative position of human body head and shoulder by control module to judge whether user's sitting posture is quasi-
Really, when overcoming the prior art and directly acquiring the specific features such as eyes of user or mouth, because individual difference wants camera shooting grease head highness
Sitting posture detects error situation caused by asking inconsistent.And camera is also radiationless, will not damage user's body.Of the invention
Intelligent desk lamp has many advantages, such as that sitting posture detection accuracy is high, practical, safe and reliable, can help user's prevention close well
Depending on.
Detailed description of the invention
Fig. 1 is a kind of intelligent desk lamp structural schematic diagram in example 1.
Fig. 2-1 is the schematic diagram of the correct sitting posture of user in example 1.
Fig. 2-2 is the schematic diagram of the correct sitting posture of user in example 1.
Fig. 2-3 is the schematic diagram of the near-sighted sitting posture of user in example 1.
Fig. 3-1 is that in example 2, eyes open the structural schematic diagram of state when blink detection.
Fig. 3-2 is that in example 2, eyes close the structural schematic diagram of state when blink detection.
Fig. 4 is a kind of flow chart of sitting posture detecting method in the third embodiment.
Fig. 5 is the structural schematic diagram of a kind of electronic equipment in some embodiments.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Embodiment one
As shown in Figure 1, a kind of intelligent desk lamp, including table lamp body (not shown), camera 10, control module 20,
Reminding module 30.Table lamp body include at least pedestal, be fixed on the base bracket, be set on bracket light bulb, be set to
The basic structures such as the power supply in pedestal consider that light bulb can be LED bulb from Energy Angle.The table lamp body can be general
Desk lamp, the present embodiment to its structure with no restrictions.
This intelligent desk lamp further includes camera 10, and can be fixed on bracket also may be mounted on pedestal, for acquiring
Predeterminable area inner video image, and collected video image is sent to control module 20.Predeterminable area include user's head,
Neck, shoulder etc., in order to acquire the upper part of the body image of different users, camera 10 is preferably mounted on Desk lamp feet, tiltedly
Upwardly facing the upper part of the body of user.In addition, camera 10 further preferably uses the lesser camera lens of focal length to widen the vision.Camera 10
It can be binocular camera 10 or monocular cam 10, but binocular camera 10 is computationally intensive, hardware configuration is also more complicated, no
Conducive to causing desk lamp miniaturization difficulty to increase, hardware cost is also high.Due to generally using desk lamp in fixed and quiet usage scenario,
Therefore the present embodiment preferably uses monocular cam 10, and monocular cam 10 is small in size, does not also need constantly to update and safeguard,
Be conducive to the miniaturization for meeting intelligent desk lamp and low at demand.
Control module 20, can be set in Desk lamp feet, is built-in with the units such as memory, central processing unit, has distinguished
At video image storage and processing function.After receiving the video image that monocular cam 10 is sent, the center of control module 20
Processor will mainly complete following functions:
1. generating user's human body contour outline figure by preset rules generates human body contour outline by preset rules.
The embodiment of the present invention can use background subtraction or frame differential method etc. and obtain human body contour outline.To use frame-to-frame differences
It is done for point-score as described below:
(1) median filtering pretreatment first is carried out to image, removes the random noise in image.
(2) background image Bx (x, y) is chosen from video image, making it only includes background image.
(3) continuous two field pictures are chosen from video image, previous frame image is denoted as Pk-1 (x, y), and present frame is denoted as
Pk(x,y).Since desk lamp user does not have quick movement generally, the two field pictures of larger time difference should be selected, are avoided
Since selection of time is too short, user is almost overlapped in two frame of front and back.
(4) it calculates present frame and the poor of background frames Bx (x, y) obtains FD (x, y), complete target is extracted from image.
(5) it calculates present frame and the poor of former frame obtains FG (x, y), obtain the variable quantity of target.
(6) intersection of frame difference FD (x, y) and FG (x, y) is asked to obtain the coarse moving region image of target to get use is arrived
The human body contour outline of the user of desk lamp.
2. judging head and shoulder position, the difference on head and shoulder position is calculated.
In fact, the position on head is a region, but for ease of calculation, in the present embodiment, head position can be with
It is considered as the junction on neck and head.Fig. 2-1, Fig. 2-2, Fig. 2-3 left side show camera 10 be fixed on desk lamp
Pedestal, obliquely towards user when multiple user's human body contour outline figures for generating, including head 100, neck 200, shoulder 300 are
Facilitate understanding, lateral plan when user corresponding with human body contour outline figure writes is also shown on the right side of every pair figure.
In some embodiments, according to a large amount of statistical data it can be found that normal person neck is close in human body contour outline figure
For rectangle, there is change of gradient in neck and head and the junction of shoulder, and the change of gradient of neck and shoulder is maximum, neck
The change of gradient on portion and head is second largest, and the position of human body head and shoulder can be judged with this principle.Specifically, knot
Fig. 2-1, Fig. 2-2, Fig. 2-3 are closed, gradient calculating is carried out to user's human body contour outline figure, change of gradient in human body contour outline figure is second largest
Position as head position (position of T2 in such as figure), the maximum position of change of gradient is as shoulder position (such as T1 in figure
Position), calculate the two difference.This method realizes that process is simple and accuracy is high.
It in other embodiments, also found according to a large amount of statistical data, in human body contour outline figure, due to the shoulder of normal person
Width is generally at least twice or more neck width, the width at head part and neck contacts generally also greater than neck width, because
This can also judge human head location and shoulder position by the width of each section in detection human body contour outline figure.Specifically,
It can use shape function or mat function or remaining effective algorithm calculate adjacent position in human body contour outline figure from bottom to top
Width judge current location if current width such as reduces one times or more than upper position reduced width presupposition multiple
For shoulder position;If current width increases presupposition multiple than upper position width, judge current location for head position, most
Calculate the difference of head position Yu shoulder position again afterwards.
The position of head and shoulder can certainly be judged using remaining feasible method, the embodiment of the present invention does not limit this
System.
3. by obtained head with the difference of shoulder position compared with preset threshold, if difference meets preset threshold,
Then judge near-sighted sitting posture occurred and controls the output prompting message of reminding module 30.
Fig. 2-1, Fig. 2-2, Fig. 2-3 show the alternate position spike situation on head and shoulder in user's difference sitting posture, ordinary circumstance
Correct sitting posture when writing be upper body is straight and even, two shoulders flush, head just and can be slightly canted.The present embodiment is to user's human body contour outline
It is illustrated for the method for figure progress gradient calculating, passes through and count great amount of samples data, it is assumed that the position of head position and shoulder
Difference is set less than after 0.5cm, i.e., it is believed that near-sighted sitting posture of having gone on a journey, then Fig. 2-1, Fig. 2-2 are correct sitting posture, Fig. 2-3 is myopia
Sitting posture.At this point, control module 20, which controls reminding module 30, exports prompting message.
It in some embodiments, can also be according to user crowd's when determining preset threshold by counting great amount of samples data
Individual features determine different threshold values, such as using the age as user characteristics, user can be divided into 14 years old or less children, 14 years old-
18 years old teenagers, 18 years old or more adult etc., it is possible to understand that the neck of 14 years old or less children be generally shorter than 18 years old or more and grow up
People, therefore the relative distance of head and shoulder can be closer, in order to improve the accuracy of sitting posture detection, is preferably set according to age characteristics
Set different threshold values.
Specifically, intelligent desk lamp further includes the user characteristics input module (Fig. 1 is not shown) connecting with control module, it is used for
It obtains active user's feature (such as age), control module is provided with preset threshold corresponding with user characteristics, control module root
Corresponding preset threshold is called according to the user characteristics that user characteristics input module obtains.Such as user be 10 years old children, then with
After the age of user that family feature input module will acquire is sent to control module, control module is called 14 years old or less children's is default
Whether threshold value is compared with obtained head with the difference of shoulder position, judge user's sitting posture correctly to improve sitting posture inspection
The accuracy of survey.
Correspondingly, control module 20 is equipped with the feature input submodule for inputting individual features for user, future judges user
Obtained head can be compared with the difference of shoulder position with the threshold value that feature input submodule obtains when sitting posture,
Reminding module 30 can be set on Desk lamp feet or be arranged on bracket, can be alarm lamp, audible and visual alarm
Device or voice alarming device, the present embodiment are illustrated by taking voice alarming device as an example, which includes audio
The structures such as decoder module, overpower amplifier, loudspeaker, when control module 20 judges that near-sighted sitting posture occurs in user, just starting is mentioned
Awake voice alarming device, reminds user to correct sitting posture, pre- myopia prevention in time.
In some embodiments, in order to realize the monitoring situation to intelligent desk lamp user, which further includes communication
The information that user has near-sighted sitting posture is sent to exterior terminal equipment by communication module by module, control module 20.Such as family
Long to wish the long-range study sitting posture for understanding child to help the pre- myopia prevention of child, then intelligent desk lamp is when there is near-sighted sitting posture in child
Corresponding warning information is sent to the mobile phone of parent.The communication module can be the mobile communication modules such as 3G, 4G, 5G, can also
Think the communication modules such as WiFi.
Intelligent desk lamp of the invention acquires predeterminable area inner video image using camera 10, is calculated by control module 20
Whether the relative position of human body head and shoulder is accurate to judge user's sitting posture, overcomes the prior art and directly acquires eyes of user
Or when the specific features such as mouth, because sitting posture detects wrong feelings caused by individual difference is inconsistent to 10 requirement for height of camera
Condition.And camera 10 is also radiationless, will not damage user's body.Intelligent desk lamp of the invention has sitting posture detection accuracy
The advantages that high, practical, safe and reliable, can help the pre- myopia prevention of user well.In addition, the intelligent desk lamp of the present embodiment is adopted
With monocular cam 10, structure is simple, is conducive to the miniaturization of desk lamp, and cost is also less expensive.
Embodiment two
Compared with embodiment one, the intelligent desk lamp of the present embodiment is not only able to achieve near-sighted sitting posture alarm, moreover it is possible to realize fatigue
The function that state is reminded.Can judge whether user is tired by detection blink situation or situation of yawning, in human fatigue
User's rest is reminded in time, can further help the pre- myopia prevention of user.
In some embodiments, control module 20 is also used to video image identification human eye based on the received, and calculates default
Number of winks in period inner video image judges fatigue state occurred, and control if number of winks is more than preset threshold
Reminding module 30 processed exports prompting message.
Specifically, utilizing the library function of preset OpenCV after the video image that the reception camera 10 of control module 20 acquires
Image recognition is carried out, identifies position of human eye.
Then blink detection is carried out.In existing algorithm, typically only have chosen eyes left eye angle and right eye angle this
Blink situation has occurred to judge the shape and detecting whether of eyes in two points, but due to the size of the eyes of different people with
Shape is different, and due to eyes be it is nonlinear, this also results in only by two points the standard for obtaining eyes
True shape is unrealistic, therefore will result in the number of winks detection inaccuracy to eyes, cannot really react the use of user
Eye shape condition.The present embodiment preferably obtains four points on eye contour, the profile of eyes can be accurately admitted to, Er Qieshi
Eye condition for all age brackets.
Specifically, as shown in connection with fig. 3, left side is that eyes open state, and right side is closed-eye state, by each eye with four
Then coordinate representation can show clockwise since the left comer of eyes around the rest part of eyes region, by this four
A point is respectively labeled as P1~P4.It is the horizontal length of eyes between P1 and P3, is the vertical length of eyes between P2 and P4.Meter
Calculate the aspect ratio (EAR) of eyes:
EAR=| P3-P1 |/| P2-P4 |
By numerous studies number it has been found that eyes open when, the length-width ratio of eyes be about it is constant, but blink when
Zero can be quickly fallen to.Can determine in this way the number of blink and whether occur not batting an eyelid for a long time and for a long time do not open eyes or
The excessively high problem of person's frequency of wink.
When blinking statistics, it is also necessary to which an eye closing threshold value is set, when the transverse and longitudinal of eyes ratio is less than the threshold value, so that it may
It is judged as eye closing, however it is not possible that everyone same threshold value of selection, somebody's eyes are big, and somebody's eyes are small, at this moment
Threshold value should be with people difference without stopping changing, can all be judged by accident when threshold value is excessive and too small, it is therefore desirable to
One algorithm for having more generalization ability, the blink threshold value of adaptive calculating different people are set.In the present embodiment, can will make
The maximum transverse and longitudinal ratio for using desk lamp to occur in for a period of time takes its half as threshold value as the upper limit.It in this way can be according to different people
Eyes ratio intelligence setting threshold value.
When control module 20 calculates user's number of winks more than preset threshold, judge that fatigue state occurs in user, and
It controls reminding module 30 and exports prompting message.
In further embodiments, control module 20 is also used to video image identification mouth based on the received, and calculates pre-
If the number of yawning in period inner video image, if number of yawning judges tired shape occurred more than preset threshold
State, and control reminding module 30 and export prompting message.
Image knowledge can be carried out using the library function of preset OpenCV after receiving the video image that camera 10 acquires
Not, it identifies mouth position, then detects whether the case where being yawned by calculating the aspect ratio of mouth, can also make
With the size of dehiscing (i.e. the mouth opening of mouth) etc. for calculating mouth, specific calculation method can refer to existing literature.But by
Not necessarily it is exactly fatigue in yawning, judges whether user is fatigue state in order to more acurrate, in the present embodiment, only one
When the number yawned in a preset time period reaches preset threshold, just judge user for fatigue state.
In the present embodiment, reminding module 30 can be alarm lamp or sound and light alarm device or voice alarming device, Ke Yiyou
It is selected as voice alarming device, when control module 20 judges that near-sighted sitting posture occurs in user, just voice alarming device is reminded in starting, and
When remind user rest, pre- myopia prevention.The intelligent desk lamp also further includes communication module, and control module 20 will be used by communication module
The information that family has fatigue is sent to exterior terminal equipment, for example the study sitting posture of remotely understanding child is wished by such as parent
To help the pre- myopia prevention of child, then corresponding warning information is sent to parent's when child has near-sighted sitting posture by intelligent desk lamp
Mobile phone.
Embodiment three
The embodiment of the invention also discloses a kind of sitting posture detecting methods, comprising the following steps:
S01 generates user's human body contour outline figure by preset rules according to the user video image of the predeterminable area of acquisition, described
Predeterminable area includes user's head, neck, shoulder.
The embodiment of the present invention can use background subtraction or frame differential method etc. and obtain human body contour outline.To use frame-to-frame differences
It is done for point-score as described below:
(1) median filtering pretreatment first is carried out to image, removes the random noise in image.
(2) background image Bx (x, y) is chosen from video image, making it only includes background image.
(3) continuous two field pictures are chosen from video image, previous frame image is denoted as Pk-1 (x, y), and present frame is denoted as
Pk(x,y).Since desk lamp user does not have quick movement generally, the two field pictures of larger time difference should be selected, are avoided
Since selection of time is too short, user is almost overlapped in two frame of front and back.
(4) it calculates present frame and the poor of background frames Bx (x, y) obtains FD (x, y), complete target is extracted from image.
(5) it calculates present frame and the poor of former frame obtains FG (x, y), obtain the variable quantity of target.
(6) intersection of frame difference FD (x, y) and FG (x, y) is asked to obtain the coarse moving region image of target to get use is arrived
The human body contour outline of the user of desk lamp.
S02 judges head and shoulder position, calculates the difference on head and shoulder position.
The position of true head portion is a region, but for ease of calculation, head position be considered neck with
The junction on head.
In some embodiments, it is possible to understand that, in human body contour outline figure, normal person neck is close to rectangle, neck with
There is change of gradient in head and the junction of shoulder, and the change of gradient of neck and shoulder is maximum, the ladder of neck and head
Degree variation is second largest, and the position of human body head and shoulder can be judged with this principle.Specifically, to user's human body contour outline
Figure carries out gradient calculating, and using second largest position of change of gradient in human body contour outline figure as head position, change of gradient is maximum
Position calculates the two difference as shoulder position.This method realizes that process is simple and accuracy is high.
In other embodiments, in human body contour outline figure, since the shoulder width of normal person is generally at least twice or more
Width at neck width, head part and neck contacts can also pass through detection human body wheel generally also greater than neck width
The width of each section judges human head location and shoulder position in wide figure.Specifically, can use shape function or
Mat function calculates the width for calculating adjacent position in human body contour outline figure from bottom to top, if current width contracts than upper position width
Small presupposition multiple, such as reduces one times or more, then judges current location for shoulder position;If current width is wider than a upper position
Degree increases presupposition multiple, then judges that current location for head position, finally calculates the difference of head position Yu shoulder position again.
S03 judges near-sighted sitting posture occurred if difference meets preset threshold.
The present embodiment is a large amount of by counting to be illustrated for carrying out the method for gradient calculating to user's human body contour outline figure
After the position difference of sample data, head position and shoulder is less than 0.5cm, i.e., it is believed that near-sighted sitting posture of having gone on a journey.Therefore, Fig. 2-
1, Fig. 2-2 is correct sitting posture, and Fig. 2-3 is near-sighted sitting posture.
It in some embodiments, can also be according to user crowd's when determining preset threshold by counting great amount of samples data
Individual features determine different threshold values, such as using the age as user characteristics, user can be divided into 14 years old or less children, 14 years old-
18 years old teenagers, 18 years old or more adult etc., it is possible to understand that the neck of 14 years old or less children be generally shorter than 18 years old or more and grow up
People, therefore the relative distance of head and shoulder can be closer, in order to improve the accuracy of sitting posture detection, is preferably set according to age characteristics
Set different threshold values.
The embodiment of the present invention judges whether user's sitting posture is accurate by calculating the relative position of human body head and shoulder, gram
When having taken the prior art and directly acquiring the specific features such as eyes of user or mouth, because individual difference is different to camera requirement for height
Sitting posture detects error situation caused by cause.The method of the present invention can be applied to the application such as picture pick-up device, desk lamp, desk, computer
Scene.
Example IV
Corresponding to the above method embodiment, the embodiment of the invention also provides a kind of electronic equipment.Fig. 5 is the present invention
The structural schematic diagram for the electronic equipment that embodiment provides, the electronic equipment include: processor 510, communication interface 520, storage
Device 530 and communication bus 540, in which:
Processor 510, communication interface 520, memory 530 complete mutual communication, memory by communication bus 540
530, for storing computer program;
Processor 510 when for executing the program stored on memory 530, realizes that the present invention implements the sitting posture provided
Detection method.Specifically, the sitting posture detecting method, comprising:
S01 generates user's human body contour outline figure by preset rules according to the user video image of the predeterminable area of acquisition, described
Predeterminable area includes user's head, neck, shoulder.
S02 judges head and shoulder position, calculates the difference on head and shoulder position.
S03 judges near-sighted sitting posture occurred if difference meets preset threshold.
Specifically, including: to utilize background subtraction or frame-to-frame differences by the method that preset rules generate user's human body contour outline figure
Point-score obtains human body contour outline.
Judge head and shoulder position, the method for calculating the difference of head and shoulder position includes: to user
Human body contour outline figure carries out gradient calculating, and using second largest position of change of gradient in human body contour outline figure as head position, gradient becomes
Change maximum position as shoulder position, calculates the difference of head position and shoulder position;Or human body contour outline is calculated from bottom to top
The width of adjacent position in figure, if current width judges a upper position as shoulder than upper position reduced width presupposition multiple
Wing position;If current width increases presupposition multiple than upper position width, judge that current location for head position, calculates head
The difference of portion position and shoulder position.
The implementation of above-mentioned sitting posture detecting method is identical as the sitting posture detecting method that preceding method embodiment part provides,
Which is not described herein again.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.Each reality in this specification
It applies example and is all made of relevant mode and describe, the same or similar parts between the embodiments can be referred to each other, each embodiment
What is stressed is the difference from other embodiments.For device, electronic equipment embodiment, due to it
It is substantially similar to embodiment of the method, so being described relatively simple, the relevent part can refer to the partial explaination of embodiments of method.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should all include
Within protection scope of the present invention.
Claims (10)
1. a kind of intelligent desk lamp, including table lamp body, which is characterized in that further include: camera, control module, reminding module,
In:
Camera is fixed on table lamp body, for acquiring the user video image of predeterminable area, and by collected video image
It is sent to control module, the predeterminable area includes user's head, neck, shoulder;
Control module generates user's human body contour outline figure by preset rules, judges head for receiving the video image of camera acquisition
Portion and shoulder position calculate the difference on head and shoulder position, if difference meets preset threshold, judge occur
Near-sighted sitting posture simultaneously controls reminding module output prompting message.
2. intelligent desk lamp as described in claim 1, which is characterized in that control module generates user's human body contour outline by preset rules
Figure, comprising: obtain human body contour outline using background subtraction or frame differential method.
3. intelligent desk lamp as described in claim 1, which is characterized in that the difference on control module calculating head and shoulder position
Value, comprising: gradient calculating is carried out to user's human body contour outline figure, using second largest position of change of gradient in human body contour outline figure as head
Portion position, the maximum position of change of gradient calculate the difference of head position and shoulder position as shoulder position;
Or, the width of adjacent position in human body contour outline figure is calculated from bottom to top, if current width is than upper position reduced width
Presupposition multiple then judges a upper position as shoulder position;If current width increases presupposition multiple than upper position width, sentence
Disconnected current location is head position, calculates the difference of head position and shoulder position.
4. intelligent desk lamp as described in claim 1, which is characterized in that the control module is also used to video figure based on the received
As identification human eye, and the number of winks in preset time period inner video image is calculated, if number of winks is more than preset threshold, sentenced
It is disconnected fatigue state occur, and control reminding module output prompting message.
5. intelligent desk lamp as described in claim 1, which is characterized in that the control module is also used to video figure based on the received
As identification mouth, and the number of yawning in preset time period inner video image is calculated, if number of yawning is more than preset threshold,
Then judge fatigue state occurred, and controls reminding module output prompting message.
6. intelligent desk lamp as described in claim 4 or 5, which is characterized in that control module identifies human eye or mouth, comprising: control
Image recognition is carried out using the library function of preset OpenCV after the video image of molding block reception camera acquisition.
7. intelligent desk lamp as described in claim 1, which is characterized in that the desk lamp further includes communication module, and control module is logical
It crosses communication module and the information that user has near-sighted sitting posture or fatigue state is sent to exterior terminal equipment.
8. a kind of sitting posture detecting method, which comprises the following steps:
User's human body contour outline figure, the predeterminable area are generated by preset rules according to the user video image of the predeterminable area of acquisition
Including user's head, neck, shoulder;
Judge head and shoulder position, calculates the difference on head and shoulder position, if difference meets preset threshold,
There is near-sighted sitting posture in judgement.
9. sitting posture detecting method as claimed in claim 8, which is characterized in that the judgement head and shoulder position calculate
The method of the difference on head and shoulder position includes: to carry out gradient calculating to user's human body contour outline figure, by human body contour outline figure
Second largest position of middle change of gradient calculates head position as shoulder position as head position, the maximum position of change of gradient
Set the difference with shoulder position;
Or, the width of adjacent position in human body contour outline figure is calculated from bottom to top, if current width is than upper position reduced width
Presupposition multiple then judges a upper position as shoulder position;If current width increases presupposition multiple than upper position width, sentence
Disconnected current location is head position, calculates the difference of head position and shoulder position.
10. a kind of electronic equipment, including processor, communication interface, memory and communication bus, in which:
Processor, communication interface, memory complete mutual communication by communication bus;
Memory, for storing computer program;
Processor when for executing the program stored on memory, realizes any method and step of claim 8-9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910601482.9A CN110458010A (en) | 2019-07-05 | 2019-07-05 | A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910601482.9A CN110458010A (en) | 2019-07-05 | 2019-07-05 | A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110458010A true CN110458010A (en) | 2019-11-15 |
Family
ID=68482265
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910601482.9A Pending CN110458010A (en) | 2019-07-05 | 2019-07-05 | A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110458010A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110969099A (en) * | 2019-11-20 | 2020-04-07 | 湖南检信智能科技有限公司 | Threshold value calculation method for myopia prevention and early warning linear distance and intelligent desk lamp |
CN111265220A (en) * | 2020-01-21 | 2020-06-12 | 王力安防科技股份有限公司 | Myopia early warning method, device and equipment |
CN111401382A (en) * | 2019-12-04 | 2020-07-10 | 浙江凯耀照明有限责任公司 | Method for automatically detecting camera range and desk lamp using same |
CN112483924A (en) * | 2020-11-20 | 2021-03-12 | 杭州慧芯达科技有限公司 | AI eye-protecting desk lamp with sitting posture correction reminding function and implementation method thereof |
CN112633411A (en) * | 2021-01-04 | 2021-04-09 | 欧普照明股份有限公司 | Sitting posture detection device and sitting posture detection method |
CN112907569A (en) * | 2021-03-24 | 2021-06-04 | 北京房江湖科技有限公司 | Head image area segmentation method and device, electronic equipment and storage medium |
US11568561B2 (en) * | 2019-06-17 | 2023-01-31 | Qisda Corporation | Lamp and method for detecting a sitting posture of a user |
CN116092129A (en) * | 2023-04-06 | 2023-05-09 | 合肥工业大学 | Intelligent bookshelf and control method thereof |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103400110A (en) * | 2013-07-10 | 2013-11-20 | 上海交通大学 | Abnormal face detection method in front of ATM (automatic teller machine) |
CN104217554A (en) * | 2014-09-19 | 2014-12-17 | 武汉理工大学 | Reminding system and method for health study posture for student |
CN104424772A (en) * | 2013-08-27 | 2015-03-18 | 宋俍宏 | Sitting posture monitoring and adjusting device |
CN106022304A (en) * | 2016-06-03 | 2016-10-12 | 浙江大学 | Binocular camera-based real time human sitting posture condition detection method |
CN109117705A (en) * | 2018-06-14 | 2019-01-01 | 浙江凯耀照明股份有限公司 | A kind of system and method carrying out sitting posture identification by desk lamp |
CN109340607A (en) * | 2018-10-23 | 2019-02-15 | 电子科技大学 | A kind of anti-visual fatigue multi-function desk lamp |
CN109584522A (en) * | 2018-12-11 | 2019-04-05 | 深圳市优点智联科技有限公司 | Desk lamp and its control method, computer storage medium |
-
2019
- 2019-07-05 CN CN201910601482.9A patent/CN110458010A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103400110A (en) * | 2013-07-10 | 2013-11-20 | 上海交通大学 | Abnormal face detection method in front of ATM (automatic teller machine) |
CN104424772A (en) * | 2013-08-27 | 2015-03-18 | 宋俍宏 | Sitting posture monitoring and adjusting device |
CN104217554A (en) * | 2014-09-19 | 2014-12-17 | 武汉理工大学 | Reminding system and method for health study posture for student |
CN106022304A (en) * | 2016-06-03 | 2016-10-12 | 浙江大学 | Binocular camera-based real time human sitting posture condition detection method |
CN109117705A (en) * | 2018-06-14 | 2019-01-01 | 浙江凯耀照明股份有限公司 | A kind of system and method carrying out sitting posture identification by desk lamp |
CN109340607A (en) * | 2018-10-23 | 2019-02-15 | 电子科技大学 | A kind of anti-visual fatigue multi-function desk lamp |
CN109584522A (en) * | 2018-12-11 | 2019-04-05 | 深圳市优点智联科技有限公司 | Desk lamp and its control method, computer storage medium |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11568561B2 (en) * | 2019-06-17 | 2023-01-31 | Qisda Corporation | Lamp and method for detecting a sitting posture of a user |
CN110969099A (en) * | 2019-11-20 | 2020-04-07 | 湖南检信智能科技有限公司 | Threshold value calculation method for myopia prevention and early warning linear distance and intelligent desk lamp |
CN111401382A (en) * | 2019-12-04 | 2020-07-10 | 浙江凯耀照明有限责任公司 | Method for automatically detecting camera range and desk lamp using same |
CN111401382B (en) * | 2019-12-04 | 2023-05-26 | 浙江凯耀照明有限责任公司 | Method for automatically detecting camera range and table lamp using same |
CN111265220A (en) * | 2020-01-21 | 2020-06-12 | 王力安防科技股份有限公司 | Myopia early warning method, device and equipment |
CN112483924A (en) * | 2020-11-20 | 2021-03-12 | 杭州慧芯达科技有限公司 | AI eye-protecting desk lamp with sitting posture correction reminding function and implementation method thereof |
CN112633411A (en) * | 2021-01-04 | 2021-04-09 | 欧普照明股份有限公司 | Sitting posture detection device and sitting posture detection method |
CN112907569A (en) * | 2021-03-24 | 2021-06-04 | 北京房江湖科技有限公司 | Head image area segmentation method and device, electronic equipment and storage medium |
CN112907569B (en) * | 2021-03-24 | 2024-03-15 | 贝壳找房(北京)科技有限公司 | Head image region segmentation method, device, electronic equipment and storage medium |
CN116092129A (en) * | 2023-04-06 | 2023-05-09 | 合肥工业大学 | Intelligent bookshelf and control method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110458010A (en) | A kind of intelligent desk lamp, a kind of sitting posture detecting method and a kind of electronic equipment | |
CN206863380U (en) | A kind of intelligent glasses | |
CN104239860B (en) | Sitting posture detection and reminding method and device during use of intelligent terminal | |
BR102016029263A2 (en) | ELECTRONIC OPTICAL LENS WITH OSCILLATOR FREQUENCY ADJUSTMENT | |
CN105095885B (en) | Human eye state detection method and detection device | |
CN109035699A (en) | The elderly house overall view monitoring nursing system and method based on artificial intelligence | |
CN103902043B (en) | Intelligent terminal and its based reminding method | |
CN105183170B (en) | Wear-type wearable device and its information processing method, device | |
CN106919250A (en) | A kind of based reminding method and device | |
US11084424B2 (en) | Video image output apparatus, video image output method, and medium | |
CN107340849A (en) | Mobile device and its eyeshield control method | |
CN107958572B (en) | Baby monitoring system | |
CN112069949A (en) | Artificial intelligence-based infant sleep monitoring system and monitoring method | |
CN109948435A (en) | Sitting posture prompting method and device | |
CN102956090A (en) | Double-lens image sensing myopia-preventing reminder | |
CN110870771A (en) | Image detection method and image detection device for determining posture of user | |
CN106648081A (en) | Method for turning off screen and electronic equipment | |
KR100766592B1 (en) | drowsiness detection using reflected image on eye iris | |
CN106851409A (en) | Video-assisted equipment and its control method with posture error correction | |
CN208172977U (en) | A kind of Intelligent sitting posture monitoring device based on structured light technique | |
CN111734974B (en) | Intelligent desk lamp with sitting posture reminding function | |
CN106445135B (en) | Data interaction method and system for posture correction through intelligent robot | |
CN107072524A (en) | Eye movement detection means | |
CN109426797A (en) | Doze detection device and doze detection method thereof | |
CN206381163U (en) | Instant eye monitoring glasses |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191115 |