CN110334631A - A kind of sitting posture detecting method based on Face datection and Binary Operation - Google Patents
A kind of sitting posture detecting method based on Face datection and Binary Operation Download PDFInfo
- Publication number
- CN110334631A CN110334631A CN201910568748.4A CN201910568748A CN110334631A CN 110334631 A CN110334631 A CN 110334631A CN 201910568748 A CN201910568748 A CN 201910568748A CN 110334631 A CN110334631 A CN 110334631A
- Authority
- CN
- China
- Prior art keywords
- sitting posture
- user
- image
- standard
- head position
- 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
Classifications
-
- 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
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
A kind of sitting posture detecting method based on Face datection and Binary Operation disclosed by the invention, first acquisition standard sitting posture picture;Differentiate the brightness of detection environment again whether in detectable range, for range outside picture pre-process;Using Adaboost Face datection algorithm localization criteria sitting posture head position, and with the location information, detection margin is set;Whether detection user's sitting posture exceeds detection margin, it is first detected whether in the case of exceeding to lean forward or the wrong sitting posture of hypsokinesis, if front-rear direction inerrancy sitting posture, subtracts each other simultaneously piecemeal using the bianry image of standard picture and realtime graphic, detect whether the wrong sitting posture for left-leaning or Right deviation.The present invention compensates for the problem of being affected using Adaboost Face datection algorithm by light merely, solve the problems, such as that head position is limited when the acquisition of user's specification sitting posture, front-rear direction uses different detection methods from left and right directions mistake sitting posture, simplifies judgment condition.
Description
Technical field
The invention belongs to image processing method technical fields, and in particular to a kind of seat based on Face datection and Binary Operation
Posture detection method.
Background technique
As office worker or teenager, often needs to conscientiously stare at computer screen work at a desk or bow one's head in book
Learn before table.When maintaining incorrect sitting posture for a long time in work or learning process, it is easy for causing the discomfort on body, increases
Add the probability for suffering from myopia with cervical spondylosis.More or less there is myopia, and the illness age of cervical spondylosis in most of teenager
Also gradually rejuvenation, therefore form good sitting posture and be accustomed to having very important meaning to office worker and teenager.
Currently, existing sitting posture detecting method includes, using traditional skin color segmentation algorithm extract characteristic point carry out detection or
It is trained using neural network and distinguishes different wrong sitting postures.But presence is influenced by light, feature is difficult to extract, trains sample
The problems such as this quantity is more.
Summary of the invention
The purpose of the present invention is to provide a kind of sitting posture detecting method based on Face datection and Binary Operation solves existing
Some sitting posture detecting methods are influenced and the problem of judgment condition complexity by light environment.
The technical scheme adopted by the invention is that: a kind of sitting posture detecting method based on Face datection and Binary Operation, packet
Include following steps:
Step 1: acquisition user's specification sitting posture image;
Step 2: calculating collected standard sitting posture brightness of image in step 1, the lower image of brightness enhance bright
The pretreatment of degree is not processed the higher image of brightness;
Step 3: in the standard sitting posture image obtained using Adaboost Face datection algorithm to step 2 the head of user into
Row positioning;
Step 4: using head position location obtained in step 3, detection margin is set;
Step 5: starting acquisition user's sitting posture in real time, according to the detection margin being arranged in step 4 to the real-time sitting posture of user
Head position is detected, and is prompted according to testing result user.
The features of the present invention also characterized in that
Step 2 is specifically implemented according to the following steps: standard sitting posture image is transformed into YCbCr model by the RGB model space
Space, conversion formula are as follows:
In formula (1), Y-component, i.e. luminance component are extracted, calculates the average gray of the image under Y-component, it is assumed that image ruler
Very little size is m × n, and average gray isCalculation formula are as follows:
If average gray is compared with standard value height, directly progress step 3;If average gray is low compared with standard value, need
Following pretreatment is carried out to image, wherein standard value is 85:
Pretreatment is enhancing brightness of image, it is assumed that original image is I (x, y), and the image after highlighting is G (x, y), then has
G (x, y)=a × I (x, y) (3)
A value is 3 in formula (3).
Step 3 specifically: detected, positioned using standard sitting posture image of the Adaboost Face datection algorithm to input
The head position of user is D (x, y, w, h), and is marked with wire frame, and (x, y) indicates to detect that the upper left corner of head position wire frame is sat
Mark, w indicate that the width of wire frame, h indicate the length of wire frame.
Step 4 specifically: enabling d=0.3 × w, d is the detection margin of user's sitting posture, i.e. the head movement range of user exists
Between (x-d, y-d) and (x+w+d, y+h+d), whether screen is exceeded according to the standard sitting posture of the size detection user of d, if excessively
Close to screen edge, it is unable to satisfy detection margin d, then user is prompted to resurvey standard sitting posture.
Using the head position of the Adaboost Face datection algorithm detection real-time sitting posture of user in step 5, if user's head
Prompt is not made to user without departing from detection margin in position;If user's head position exceeds detection margin, user is first determined whether
Whether sitting posture there is sitting posture lack of standardization in the longitudinal direction, when user does not detect sitting posture lack of standardization in the longitudinal direction, then
Whether detection user there is sitting posture lack of standardization in left and right directions, specifically:
Step 5.1: judging whether user's sitting posture sitting posture lack of standardization, the head of standard sitting posture image occurs in the longitudinal direction
Position is D (x, y, w, h), it is assumed that the head position of real-time sitting posture image is D ' (x ', y ', w ', h '), following adjudicates item when meeting
Part relationship:
I.e. real-time sitting posture head area ratio standard sitting posture head area is big, and user's head position is higher than the head of standard sitting posture
Portion position then determines that user's sitting posture leans forward;
When meeting following judgment condition relationship:
I.e. real-time sitting posture head area ratio standard sitting posture head area is small, and user's head position is lower, holds beyond detection
Limit, then determine that user's sitting posture tilts backwards;
Step 5.2: when step 5.1 does not detect that user's sitting posture has mistake in the longitudinal direction, then to standard sitting posture figure
Picture and real-time sitting posture image carry out binaryzation, when carrying out binary conversion treatment, are using the average gray calculated in step 2Make
For threshold value, it is consistent that the picture binaryzation standard under different brightness is arranged by the threshold value;
Two width bianry images are done into additive operation later, then divide error image according to user's specification sitting posture head position
For 3 × 3 block, be expressed as S (1,1), S (1,2), S (1,3), S (2,1), S (2,2), S (2,3), S (3,1), S (3,2),
The position of S (3,3), central block are consistent with standard sitting posture head position;
Seek the pixel number that the pixel value in S (2,1) block and S (2,3) block is 1, respectively h (2,1) and h (2,3), if meeting
H (2,1) < h (2,3), then it is assumed that user's sitting posture is tilted to the left;If meeting h (2,1) > h (2,3), then it is assumed that user's sitting posture is to Right deviation
Tiltedly.
The beneficial effects of the present invention are: a kind of sitting posture detecting method based on Face datection and Binary Operation of the present invention, is adopted
With improved Adaboost Face datection algorithm, judge whether to pre-process picture by brightness, improves script method
The situation of inaccuracy is detected when to dark;And different methods are taken from left and right directions in the front-rear direction of detection sitting posture,
Judgment condition is simplified, the wrong sitting posture of four direction can be clearly distinguished.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the sitting posture detecting method based on Face datection and Binary Operation of the present invention;
Fig. 2 is detection margin setting signal in a kind of sitting posture detecting method based on Face datection and Binary Operation of the present invention
Figure;
Fig. 3 is Binary Image against Block signal in a kind of sitting posture detecting method based on Face datection and Binary Operation of the present invention
Figure.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention is described in detail.
The present invention provides a kind of sitting posture detecting method based on Face datection and Binary Operation, as shown in Figure 1, specifically pressing
Implement according to following steps:
Step 1: acquiring the standard sitting posture image of user, detection margin coefficient is set, and range is 0.2-0.5;
Step 2: standard sitting posture image is transformed into the YCbCr model space, conversion formula by the RGB model space are as follows:
Y-component is extracted, i.e. luminance component calculates the average gray of the image under Y-component, it is assumed that picture size size is
M × n, average gray areCalculation formula are as follows:
By many experiments, judge that the higher or relatively low threshold value of average gray takes 85, as standard value, if gray scale
Average value then directly carries out step 3 compared with standard value height;If average gray is low compared with the standard value, need to carry out image
Pretreatment:
Pretreatment is enhancing brightness of image, it is assumed that original image is I (x, y), and the image after highlighting is G (x, y), then has
G (x, y)=a × I (x, y) (3)
Wherein a is the value greater than 1, and for enhancing brightness of image, by many experiments, this method is 3 to a value;
Step 3: being detected using standard sitting posture image of the Adaboost Face datection algorithm to input, position user's
Head position is D (x, y, w, h), and is marked with wire frame, and (x, y) indicates to detect the top left co-ordinate of head position wire frame, w table
The width of timberline frame, h indicate the length of wire frame;
Step 4: as shown in Fig. 2, enabling d=0.3 × w, defining the detection margin that d is user's sitting posture, i.e. the head of user is living
Whether dynamic range exceeds screen according to the standard sitting posture of the size detection user of d between (x-d, y-d) and (x+w+d, y+h+d)
Curtain, if being unable to satisfy detection margin d too close to screen edge, then prompts user to resurvey standard sitting posture;
Step 5: utilizing the head position of the Adaboost algorithm detection real-time sitting posture of user.If user's head position without departing from
Detection margin does not make prompt to user;If user's head position exceeds detection margin, first determine whether user's sitting posture in front and back
Whether lack of standardization sitting posture is occurred on direction, when user does not detect sitting posture lack of standardization in the longitudinal direction, then in left and right directions
Whether detection user there is sitting posture lack of standardization.
Step 5.1: front-rear direction: the head position of tentative standard sitting posture image is D (x, y, w, h), real-time sitting posture image
Head position be D (x', y', w', h'), when meeting following judgment condition relationship:
I.e. real-time sitting posture head area ratio standard sitting posture head area is big, and user's head position is higher than the head of standard sitting posture
Portion position then determines that user's sitting posture leans forward.
When meeting following judgment condition relationship:
I.e. real-time sitting posture head area ratio standard sitting posture head area is small, and user's head position is lower, holds beyond detection
Limit, then determine that user's sitting posture tilts backwards;
Step 5.2: binaryzation left and right directions: being carried out to standard sitting posture picture and real-time sitting posture picture first.Carry out two-value
When changing processing, due to pixel value different problems after using adaptive threshold binaryzation that can have processing under different light, i.e., pair
When certain figure processes, the foreground pixel value of the figure is 1, and when being processed to another width figure, the foreground pixel value of the figure is 0, therefore
The average gray calculated in step 2 is utilized at thisFor threshold value, (by many experiments, 50) which is taken, and is arranged by the threshold value
Picture binaryzation standard under different brightness is consistent.
Two width bianry images are done into additive operation later.Error image is divided according to user's specification sitting posture head position again
For 3 × 3 block, be expressed as S (1,1), S (1,2), S (1,3), S (2,1), S (2,2), S (2,3), S (3,1), S (3,2),
The position of S (3,3), central block are consistent with standard sitting posture head position, and schematic diagram is as shown in Figure 3.
Seek the pixel number that the pixel value in S (2,1) block and S (2,3) block is 1, respectively h (2,1) and h (2,3): if meeting
H (2,1) < h (2,3), then it is assumed that user's sitting posture is tilted to the left (there are mirror images for the photo of shooting);If meeting h (2,1) > h (2,3),
Then think that user's sitting posture is tilted to the right.
Interpretation of result
(1) pretreated Adaboost algorithm, experiment acquisition 100 is added in more classical Adaboost algorithm and the present invention
The more black picture of backlighting environment human face is detected with two kinds of algorithms respectively, is calculated two kinds of algorithms respectively and can accurately be detected and is more
Face (i.e. exclusion missing inspection and erroneous detection) in few picture, as a result such as table 1:
Table 1
As it can be seen from table 1 reusing Adaboost algorithm after being pre-processed to the lower image of brightness and carrying out face
Detection, accuracy rate are promoted.
(2) it is compared with sitting posture detecting method and sitting posture detecting method of the invention that skin color segmentation extracts characteristic point, point
Not under the environment of backlight, the normal environment of background complexity light and the normal environment of the simple light of background, four mistakes are detected
The Detection accuracy of direction sitting posture (fix, and can verifying respective algorithms correctly detect by the picture sum in each mistake sitting posture direction
Make mistake), as a result such as table 2:
Table 2
Since skin color segmentation algorithm is very big with background influence by light, and colour of skin Clustering Model cannot accurately divide the colour of skin
With the non-colour of skin, therefore use skin color segmentation thought sitting posture detecting method to inventive algorithm detect environment under picture process,
Effect is undesirable.
Claims (5)
1. a kind of sitting posture detecting method based on Face datection and Binary Operation, which comprises the following steps:
Step 1: acquisition user's specification sitting posture image;
Step 2: calculating collected standard sitting posture brightness of image in step 1, highlight to the lower image of brightness
Pretreatment, is not processed the higher image of brightness;
Step 3: the head of user is determined in the standard sitting posture image obtained using Adaboost Face datection algorithm to step 2
Position;
Step 4: using head position location obtained in step 3, detection margin is set;
Step 5: starting acquisition user's sitting posture in real time, according to the detection margin being arranged in step 4 to the head of the real-time sitting posture of user
Position is detected, and is prompted according to testing result user.
2. a kind of sitting posture detecting method based on Face datection and Binary Operation as described in claim 1, which is characterized in that institute
It states step 2 to be specifically implemented according to the following steps: standard sitting posture image is transformed into the YCbCr model space by the RGB model space, turn
Change formula are as follows:
In formula (1), Y-component, i.e. luminance component are extracted, calculates the average gray of the image under Y-component, it is assumed that picture size is big
Small is m × n, and average gray isCalculation formula are as follows:
If average gray is compared with standard value height, directly progress step 3;If average gray is low compared with standard value, need to figure
As carrying out following pretreatment, wherein standard value is 85:
Pretreatment is enhancing brightness of image, it is assumed that original image is I (x, y), and the image after highlighting is G (x, y), then has
G (x, y)=a × I (x, y) (3)
A value is 3 in formula (3).
3. a kind of sitting posture detecting method based on Face datection and Binary Operation as claimed in claim 2, which is characterized in that institute
State step 3 specifically: detect using standard sitting posture image of the Adaboost Face datection algorithm to input, position user's
Head position is D (x, y, w, h), and is marked with wire frame, and (x, y) indicates to detect the top left co-ordinate of head position wire frame, w table
The width of timberline frame, h indicate the length of wire frame.
4. a kind of sitting posture detecting method based on Face datection and Binary Operation as claimed in claim 3, which is characterized in that institute
State step 4 specifically: enabling d=0.3 × w, d is the detection margin of user's sitting posture, i.e., the head movement range of user is in (x-d, y-
D) between (x+w+d, y+h+d), whether screen is exceeded according to the standard sitting posture of the size detection user of d, if too close to screen
Certain edge of curtain, is unable to satisfy detection margin d, then user is prompted to resurvey standard sitting posture.
5. a kind of sitting posture detecting method based on Face datection and Binary Operation as claimed in claim 4, which is characterized in that institute
It states using the head position of the Adaboost Face datection algorithm detection real-time sitting posture of user in step 5, if user's head position is not
Beyond detection margin, prompt is not made to user;If user's head position exceeds detection margin, first determine whether that user's sitting posture exists
Whether lack of standardization sitting posture is occurred on front-rear direction, when user does not detect sitting posture lack of standardization in the longitudinal direction, then in left and right
Whether angle detecting user there is sitting posture lack of standardization, specifically:
Step 5.1: judging whether user's sitting posture sitting posture lack of standardization, the head position of standard sitting posture image occurs in the longitudinal direction
For D (x, y, w, h), it is assumed that the head position of real-time sitting posture image is D ' (x ', y ', w ', h '), is closed when meeting following judgment condition
System:
I.e. real-time sitting posture head area ratio standard sitting posture head area is big, and user's head position is higher than the head position of standard sitting posture
It sets, then determines that user's sitting posture leans forward;
When meeting following judgment condition relationship:
I.e. real-time sitting posture head area ratio standard sitting posture head area is small, and user's head position is lower, exceeds detection margin, then
Determine that user's sitting posture tilts backwards;
Step 5.2: when step 5.1 does not detect that user's sitting posture has mistake in the longitudinal direction, then to standard sitting posture image and
Real-time sitting posture image carries out binaryzation, when carrying out binary conversion treatment, is using the average gray calculated in step 2As threshold
Value, it is consistent to be arranged the picture binaryzation standard under different brightness by the threshold value;
Two width bianry images are done into additive operation later, then error image is divided into 3 according to user's specification sitting posture head position
× 3 block is expressed as S (1,1), S (1,2), S (1,3), S (2,1), S (2,2), S (2,3), S (3,1), S (3,2), S
The position of (3,3), central block is consistent with standard sitting posture head position;
Seek the pixel number that the pixel value in S (2,1) block and S (2,3) block is 1, respectively h (2,1) and h (2,3), if meet h (2,
1) < h (2,3), then it is assumed that user's sitting posture is tilted to the left;If meeting h (2,1) > h (2,3), then it is assumed that user's sitting posture is tilted to the right.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910568748.4A CN110334631B (en) | 2019-06-27 | 2019-06-27 | Sitting posture detection method based on face detection and binary operation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910568748.4A CN110334631B (en) | 2019-06-27 | 2019-06-27 | Sitting posture detection method based on face detection and binary operation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110334631A true CN110334631A (en) | 2019-10-15 |
CN110334631B CN110334631B (en) | 2021-06-15 |
Family
ID=68144621
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910568748.4A Active CN110334631B (en) | 2019-06-27 | 2019-06-27 | Sitting posture detection method based on face detection and binary operation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110334631B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111734974A (en) * | 2020-01-22 | 2020-10-02 | 中山明易智能家居科技有限公司 | Intelligent desk lamp with sitting posture reminding function |
CN113313917A (en) * | 2020-02-26 | 2021-08-27 | 北京君正集成电路股份有限公司 | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection |
CN113837044A (en) * | 2021-09-13 | 2021-12-24 | 深圳市罗湖医院集团 | Organ positioning method based on ambient brightness and related equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706953A (en) * | 2009-11-13 | 2010-05-12 | 北京中星微电子有限公司 | Histogram equalization based image enhancement method and device |
CN103488980A (en) * | 2013-10-10 | 2014-01-01 | 广东小天才科技有限公司 | Sitting posture judging method and device based on camera |
KR101722131B1 (en) * | 2015-11-25 | 2017-03-31 | 국민대학교 산학협력단 | Posture and Space Recognition System of a Human Body Using Multimodal Sensors |
CN107392860A (en) * | 2017-06-23 | 2017-11-24 | 歌尔科技有限公司 | Image enchancing method and equipment, AR equipment |
CN108095333A (en) * | 2017-12-18 | 2018-06-01 | 珠海瞳印科技有限公司 | Sitting posture correcting device and its bearing calibration |
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 |
-
2019
- 2019-06-27 CN CN201910568748.4A patent/CN110334631B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706953A (en) * | 2009-11-13 | 2010-05-12 | 北京中星微电子有限公司 | Histogram equalization based image enhancement method and device |
CN103488980A (en) * | 2013-10-10 | 2014-01-01 | 广东小天才科技有限公司 | Sitting posture judging method and device based on camera |
KR101722131B1 (en) * | 2015-11-25 | 2017-03-31 | 국민대학교 산학협력단 | Posture and Space Recognition System of a Human Body Using Multimodal Sensors |
CN107392860A (en) * | 2017-06-23 | 2017-11-24 | 歌尔科技有限公司 | Image enchancing method and equipment, AR equipment |
CN108095333A (en) * | 2017-12-18 | 2018-06-01 | 珠海瞳印科技有限公司 | Sitting posture correcting device and its bearing calibration |
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 |
Non-Patent Citations (4)
Title |
---|
张宇: "基于人脸检测与肤色统计的坐姿行为监测", 《计算机与网络》 * |
蔡利梅等: "《数字图像处理》", 31 August 2014, 中国矿业大学出版社 * |
谢剑斌等: "《视觉感知与智能视频监控》", 31 March 2012, 国防科技大学出版社 * |
闫敬文: "《数字图像处理 MATLAB版》", 31 August 2011, 国防工业出版社 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111734974A (en) * | 2020-01-22 | 2020-10-02 | 中山明易智能家居科技有限公司 | Intelligent desk lamp with sitting posture reminding function |
CN111734974B (en) * | 2020-01-22 | 2022-06-03 | 中山明易智能家居科技有限公司 | Intelligent desk lamp with sitting posture reminding function |
CN113313917A (en) * | 2020-02-26 | 2021-08-27 | 北京君正集成电路股份有限公司 | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection |
CN113313917B (en) * | 2020-02-26 | 2022-12-16 | 北京君正集成电路股份有限公司 | Method for solving false alarm generated when no target exists in front of detector in sitting posture detection |
CN113837044A (en) * | 2021-09-13 | 2021-12-24 | 深圳市罗湖医院集团 | Organ positioning method based on ambient brightness and related equipment |
CN113837044B (en) * | 2021-09-13 | 2024-01-23 | 深圳市罗湖医院集团 | Organ positioning method based on ambient brightness and related equipment |
Also Published As
Publication number | Publication date |
---|---|
CN110334631B (en) | 2021-06-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10747988B2 (en) | Method and device for face tracking and smart terminal | |
CN110334631A (en) | A kind of sitting posture detecting method based on Face datection and Binary Operation | |
CN103810478B (en) | A kind of sitting posture detecting method and device | |
CN109785396B (en) | Writing posture monitoring method, system and device based on binocular camera | |
CN114241548A (en) | Small target detection algorithm based on improved YOLOv5 | |
CN103914833B (en) | The method and system that a kind of automatic detection bill is the most incomplete | |
CN105205437B (en) | Side face detection method and device based on contouring head verifying | |
CN109685025A (en) | Shoulder feature and sitting posture Activity recognition method | |
CN110148092B (en) | Method for analyzing sitting posture and emotional state of teenager based on machine vision | |
CN109978907A (en) | A kind of sitting posture of student detection method towards household scene | |
CN109725721A (en) | Human-eye positioning method and system for naked eye 3D display system | |
JP2007048172A (en) | Information classification device | |
CN105046230B (en) | The method and system of people canthus detection based on image | |
Skutkova et al. | Preprocessing and classification of electrophoresis gel images using dynamic time warping | |
Bonteanu et al. | A robust pupil detection algorithm based on a new adaptive thresholding procedure | |
CN114283448A (en) | Child sitting posture reminding method and system based on head posture estimation | |
CN105844235A (en) | Visual saliency-based complex environment face detection method | |
Manaf et al. | Color recognition system with augmented reality concept and finger interaction: Case study for color blind aid system | |
CN107944393A (en) | Face nose localization method | |
CN104077562A (en) | Judgment method for scanning direction of answer sheet | |
CN206363347U (en) | Based on Corner Detection and the medicine identifying system that matches | |
CN107798316B (en) | Method for judging eye state based on pupil characteristics | |
Kim et al. | Gaze tracking based on pupil estimation using multilayer perception | |
Hassan et al. | Enhance iris segmentation method for person recognition based on image processing techniques | |
CN109145748A (en) | The intravenous line extraction algorithm of visible light vein visualising image |
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 |