CN101699469A - Method for automatically identifying action of writing on blackboard of teacher in class video recording - Google Patents
Method for automatically identifying action of writing on blackboard of teacher in class video recording Download PDFInfo
- Publication number
- CN101699469A CN101699469A CN200910185427A CN200910185427A CN101699469A CN 101699469 A CN101699469 A CN 101699469A CN 200910185427 A CN200910185427 A CN 200910185427A CN 200910185427 A CN200910185427 A CN 200910185427A CN 101699469 A CN101699469 A CN 101699469A
- Authority
- CN
- China
- Prior art keywords
- blackboard
- teacher
- writing
- difference image
- frame
- 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
Images
Abstract
The invention discloses a method for automatically identifying the action of writing on blackboard of a teacher in class video recording, which belongs to image recognition technology in a long-distance educational system, which comprises the following processing steps: a, capturing a frame of image by using a frame grabber, subtracting a previous frame of image from the current image to obtain a binary frame difference image and analyzing the binary frame difference image; b, subjecting the binary frame difference image to closed operation in mathematical morphology, marking communicated regions and removing obvious noise regions in the binary frame difference image by setting a threshold; c, acquiring a target rectangular frame and calculating the area of the rectangle, determining the movement of a human body when the area of the rectangle meets a certain condition and detecting and identifying the behavior of the writing of the teacher on the blackboard; d, when the more than 5 successive frames of images of behavior of writing on the blackboard are detected, stretching a camera; and e, if no 50 successive frames of images of the behavior of writing on the blackboard are detected, restoring the stretching magnification to an initial magnification.
Description
Technical field
The present invention is the automatic identification algorithm of teacher's blackboard action of writing in the video recording of a kind of classroom, belongs to the image recognition technology in the long-distance educational system.
Background technology
Long-distance education is more and more admitted as a kind of effective and economic mode of learning.And one of key factor of restriction long-distance education development is the shortage of course video resource, and its reason is the artificial production process of comparatively loaded down with trivial details teaching live video.At present, taking instructional video needs special cameraman to go omnidistance the shooting in the classroom, and perhaps the teacher gives lessons in the special classroom of recording, and cost is than higher.Therefore, make production equipment automatization, intellectuality, do not need special photographer's participation, can be conveniently laying in common classroom, be with a wide range of applications.
Realize the automatic recorded broadcast of teaching process, intelligent video analysis is its core technology, comprises target detection, target following and behavior identification etc.Target detection is utilized in the video information of motion or feature, detects objects in video, belongs to rudimentary Video processing.Target following is followed the tracks of specific target according to the result of target detection, belongs to intermediate Video processing.Behavior identification is discerned the behavior of tracking target, belongs to advanced video and handles.
Be used for the video analysis of remote teaching, in main target to taking---after the teacher positions, follows the tracks of, also need the behavior of the teacher in the video is discerned, make video camera make different reactions to different behaviors automatically, obtain to change good photographic effect.Wherein, the action that the teacher writes blackboard (writing words or drawing on blackboard) is one of modal important behavior in the classroom instruction, writing blackboard if detect the teacher, then to focus on the hand with writing on the blackboard be on the blackboard at center to camera, makes on the blackboard content written high-visible.When detect write blackboard and finish after, then video camera recovers original state.
Summary of the invention
Technical matters: the purpose of this invention is to provide the automatic identifying method of teacher's blackboard action of writing in the video recording of a kind of classroom, behind the position of finding the teacher exactly, judge whether the teacher has the action of writing blackboard, stretch if having then carry out video camera.The name of the game is in the target area of teacher's writing on the blackboard, writes the bigger characteristics of the common upper right quarter action of writing on the blackboard according to the teacher and comes the amount of exercise size of comparing motion target each several part, and judge that thus whether the teacher is at writing on the blackboard and stretching video camera.
Technical scheme: the automatic identifying method of teacher's blackboard action of writing may further comprise the steps in the video recording of classroom of the present invention:
A. use capture card to gather a two field picture, do the image subtraction computing, obtain the frame difference image of binaryzation and analyze with present frame and former frame;
B. after the closed operation in the two-value frame difference image doing mathematics morphology being handled, carry out connected component labeling again,, remove in the two-value frame difference image comparatively obvious noise zone by threshold value is set;
C. obtain the target rectangle frame and calculate the area of rectangle, when rectangular area satisfies institute's fixed condition, be judged as human motion, then teacher's writing on the blackboard behavior is detected and discerns;
D. if be consecutively detected when writing the blackboard behavior, then video camera is carried out stretch processing above 5 two field pictures;
E. if continuous detecting when writing the blackboard behavior and surpass 50 two field pictures, then is reduced to initial multiple to the draw ratio of video camera.
In described two-value frame difference image, the momental method that the judgement teacher writes the blackboard action is:
1) frame difference image analysis: the object of frame difference image analysis is the boundary rectangle frame of target body, be to the motion target detection result, at first the closed operation in the frame difference image doing mathematics morphology is handled, then the result after handling is carried out connected component labeling, remove the noise in the two-value frame difference image; When the pixel number of certain connected region less than 1000 the time, think that this zone is a noise region, at this moment, the black pixel point of in the two-value frame difference image, forming target body, remainder is the white pixel point, travel through the coordinate of all black pixel points in the whole two-value frame difference image, and in the middle of these coordinates, find out the minimum and maximum coordinate figure of level and vertical direction respectively, determine the top left corner apex A of target body boundary rectangle frame, summit, upper right corner B, summit, lower right corner C, the coordinate of summit, lower left corner D, when rectangular area 3000 between 10000 the time, just illustrate that the teacher might write blackboard;
2) divide the moving region: the attitude that the teacher is writing blackboard generally is to lift the right hand, hand has bigger motion and the action of other positions is very little, therefore at first target body boundary rectangle frame to be divided into several zones, since the action of writing blackboard be the teacher dorsad camera reach and write writing on the blackboard, and the teacher satisfies certain proportion towards the upper right side usually to hand that stretches out writing on the blackboard and health, therefore can determine near underarm region that cuts apart a unique point G, by cutting apart the straight line that unique point G does level and vertical direction respectively, and crossing, thereby the boundary rectangle frame can be divided into 4 rectangular areas with the boundary rectangle frame;
Image when selecting 100 width of cloth teacher writings on the blackboard calculates hand width pixel value and hand height pixel value in every width of cloth image as sample, and all samples is averaged just can roughly calculate H
hAnd H
wAgain by the coordinate on four summits of boundary rectangle frame that obtained, just can calculate the coordinate of feature cut-point G, after the coordinate of having determined feature cut-point G, the level and the vertical direction straight line of undue cutpoint is with the boundary rectangle frame is left and right, the coordinate of intersection point E, the F of upper and lower four edges, H, I is also definite thereupon, thereby obtain four moving regions by target body boundary rectangle frame, wherein the upper right portion zone of boundary rectangle frame has comprised the teacher and has stretched out the hand of writing blackboard, and whole left part has comprised teacher's health;
3) judge amount of exercise: according to ready-portioned 4 moving regions, calculate each regional amount of exercise respectively, certain regional amount of exercise can be that the number of 1 pixel is described by this regional value in the two-value frame difference image, and the teacher writes the right regions that the attitude of blackboard is reflected as in the target body boundary rectangle frame bigger motion is arranged, and lower right area not motion basically, lower-left and top left region only have small motion;
Therefore when each regional amount of exercise satisfies following three conditions, judge that this moment, the objects in video human body was write blackboard:
A. the amount of exercise of upper right hand region HBFG is bigger than the amount of exercise of bottom right white space GFCI;
B. the amount of exercise of the body region EGID of lower-left is bigger than the amount of exercise of bottom right white space GFCI;
C. upper right hand region HBFG has certain amount of exercise, and promptly the ratio of hand region amount of exercise and this region area reaches certain ratio.
Beneficial effect: the present invention utilize the method for comparing motion amount to judge teacher's writing on the blackboard behavior not only calculated amount is little, and in the writing on the blackboard behavior of judging the teacher, the writing on the blackboard content can be amplified to a suitable ratio, make recording and broadcasting system can for the learner provide a kind of clear and intuitively the sensation.
1. owing to adopt the video camera in the automatic recording and broadcasting system of long-distance education of the present invention to move, do not need to consider specific background, have better effect so the scene variation is carried out motion detection with frame difference method.
2. relatively the each several part amount of exercise is that colour-difference promptly compares the number of motion pixel in the frame difference image each several part zone apart from bigger pixel number in comparison two width of cloth image each several part zones.This method calculated amount is less and the result is accurate, is suitable for real-time detection.
3. when having detected writing on the blackboard, can stretch according to scale-up factor adaptively in different scenes reaches effect clearly, is applicable to most of long-distance educational systems.
Description of drawings
Frame difference figure when Fig. 1 writes blackboard,
Fig. 2 frame difference image is analyzed,
Fig. 3 writes the process flow diagram of blackboard behavior recognizer.
Embodiment
To write this behavior of blackboard to the teacher and discern, must analyze, describe this characteristics, design a sorter then, judge whether the teacher in the video is writing blackboard with mathematical linguistics to the characteristics of writing this behavior of blackboard.The attitude that the teacher is writing blackboard generally is to lift the right hand in the blackboard writing on the blackboard, and the hand that it is characterized in that the teacher has bigger motion and other positions actions of health are very little, therefore is necessary earlier target body boundary rectangle frame to be divided into 4 zones.There is bigger motion in the zone, the upper right corner that wherein comprises hand, and the not motion basically of the white space in the lower right corner, the body part in the lower left corner also has slight motion.Each regional amount of exercise also compares amount of exercise between each zone in the calculating target body boundary rectangle frame, judges that thus whether the teacher is at writing on the blackboard and stretching video camera.If be consecutively detected when writing the blackboard behavior, then video camera carried out stretch processing above 5 two field pictures; If continuous detecting when writing the blackboard behavior and surpass 50 two field pictures, then is reduced to initial multiple to the draw ratio of video camera.
Human body behavior in the video discerned exactly the motor behavior in the video is analyzed, and the inter-frame difference image movable information in the reflecting video well is commonly used to describe the movable information in the video.The inter-frame difference image is that adjacent two frame video images carry out image subtraction computing (absolute value of difference) and the image that obtains:
D
k(x，y)＝|f
k-1(x，y)-f
k(x，y)| (1)
Because there is noise more or less in natural image, in directly carrying out the inter-frame difference image that image subtraction obtains, there are significant difference in noise and moving region.Be necessary to adopt a fixing threshold value that the inter-frame difference image is carried out binaryzation, obtain the frame difference image (as shown in Figure 1) of binaryzation:
In formula (2), T
hSize issue to determine according to the degree of motion.If T
hToo big, then detect less than the smaller part of human motion, can cause detected target body discontinuous; Otherwise detected target body can comprise the partial noise of background.Measure by experiment, determine T
hValue be 35 o'clock comparatively suitable.After passing threshold is cut apart, D
k(x, some y)=1 are exactly the black pixel point of expression target body among Fig. 1, D
k(x, some y)=0 has been formed white background area.
Analyze the action that the teacher writes blackboard, at first will carry out moving object detection the two-value frame difference image.Only in the zone, have in the motor behavior, just write the blackboard behavior and detect.Detection process of moving target is to realize by carrying out connected component labeling again after the closed operation during two-value frame difference image doing mathematics morphology is handled.When handling the binaryzation frame difference image, the zone of forming to the black pixel point of each connection by the zone marker algorithm adds identical mark, also promptly: when handling a black pixel point, enclosing identical mark with the black pixel point that its is communicated with on every side, this has just formed the zone of a mark.Simultaneously, owing to more or less exist the noise pixel point in the frame difference image after the binaryzation, the zone marker algorithm can also carry out denoising to the binaryzation frame difference image.When the black pixel point in certain noise region constitutes a little connected region, the number of supposing all black pixel points in the zone then all becomes the white pixel zone to whole pixels in this connected region less than a special value (this threshold value is 1000 in this algorithm).If the action of writing blackboard arranged, then the connected region that constituted of target body pixel is bigger usually, and its testing process realizes by the boundary rectangle frame that obtains target.
In the two-value frame difference image, except forming the black pixel point of target body, remainder is white pixel point (as shown in Figure 1).Travel through whole two-value frame difference image, relatively all black pixel point horizontal ordinate and ordinate values are found out the minimum value P of level respectively
lWith maximal value P
r, and the minimum value P of vertical direction
tWith maximal value P
b, and then four summit A, B, the coordinate of C, D of target body boundary rectangle frame among definite Fig. 2.Therefore, the coordinate on four summits is respectively: the A point coordinate is (P
l, P
t), the B point coordinate is (P
r, P
t), the C point coordinate is (P
r, P
b), the coordinate of D is (P
l, P
b).When rectangular area 3000 between 10000 the time, just illustrate that the teacher might write blackboard and write the identification of blackboard action.
Since the action of writing blackboard be the teacher dorsad camera reach and write writing on the blackboard (as shown in Figure 1), and the teacher satisfies certain proportion towards the upper right side usually to the hand that stretches out writing on the blackboard and health, so can determine near underarm region that cuts apart a unique point G.By cutting apart the straight line that unique point G does level and vertical direction respectively, and intersect at an E, F, H, I (as shown in Figure 2) with left and right, the upper and lower four edges of crossing boundary rectangle frame respectively, thereby the boundary rectangle frame can be divided into 4 rectangular areas.After target body boundary rectangle frame ABCD is divided into 4 zones, just can differentiate the behavior that the teacher writes blackboard according to the contact between each regional movement amount.Can find that by analyzing the reflection of attitude on the two-value frame difference image that the teacher writes blackboard mainly shows as simultaneously: zone, the upper right corner HBGF of human body boundary rectangle ABCD has bigger motion; And zone, lower right corner GFCI not motion basically; Lower left corner EGID and upper left corner AHGE also have a spot of motion.Like this, varying in size of the motion amplitude in four zones of human body target and frequency just formed the basic characteristics that the teacher writes the blackboard action.Therefore,, can design and judge whether target body is writing the sorter of blackboard in the video the analysis that teacher on the classroom writes the blackboard behavior according to these.
At first determine the position that G is ordered, the target body boundary rectangle is divided into 4 rectangular areas according to the ratio of human body.Image when selecting 100 width of cloth teacher writings on the blackboard calculates hand width pixel value and hand height pixel value in every width of cloth image as sample, and all samples is averaged just can roughly calculate H
hAnd H
wIn Fig. 2 rectangle, by H
wCan obtain the coordinate that H orders with the B point coordinate is (P
r-H
w, P
t), by H
hCan obtain the coordinate that F orders with the B point coordinate is (P
r, P
t+ H
h), and then the coordinate that obtains cutting apart unique point G is (P
r-H
w, P
t+ H
h).Similarly, can obtain the coordinate that E orders is (P
l, P
t+ H
h), the coordinate that I is ordered is (P
r-H
w, P
b).
After the coordinate that obtains these points, just can determine 4 moving regions (as shown in Figure 2).Each regional amount of exercise can be that the number of 1 pixel is described by this regional value in the two-value frame difference image, and the amount of exercise in note zone is Motion
Area, be Motion as the amount of exercise of regional ABCD
ABCD
When target meets following three conditions simultaneously, judge that then target body is being write blackboard in video this moment:
1) amount of exercise of upper right hand region HBFG is than the amount of exercise big (a is a scale-up factor) of bottom right white space GFCI
Motion
HBFG-a×Motion
GFCI＞0 (3)
2) amount of exercise of lower-left body region EGID is than the amount of exercise big (b is a scale-up factor) of bottom right white space GFCI
Motion
EGID-b×Motion
GFCI＞0 (4)
3) upper right hand region HBFG has certain amount of exercise, amount of exercise and its area A rea that promptly should the zone
HBFGRatio reach certain ratio c.
According to regional HBFG and the regional GFCI ratio in vertical direction, the value that can determine a is H
h/ (100-H
h).Ratio by area E GID and regional GFCI can obtain, and the value of b is defined as (100-H
w)/H
WAccording to repeatedly test, can draw among the upper right moving region HBFG, the moving region accounts for more than 30% of whole region area greatly.So the value of c is defined as 0.3.
In conjunction with above-mentioned three formulas, obtain judging whether target body is as follows at the two class sorters of writing blackboard in the video:
Wherein, Motion is each regional amount of exercise, and Area is the area in zone, and a, b, c are parameter.When Is_writing is 1 as a result, represent to write that it is not to write blackboard that blackboard, result are 0.
The teacher is write process flow diagram that the blackboard behavior carries out recognition methods as shown in Figure 3, and implementation step is:
1) gathers a two field picture by video frequency collection card, do the image subtraction computing, thereby obtain the frame difference image of binaryzation with present frame and former frame;
2) by after the closed operation in the two-value frame difference image doing mathematics morphology is handled, carry out connected component labeling again the two-value frame difference image is carried out denoising;
3) detected all black pixel points of traversal obtain comprising the boundary rectangle frame of target, if boundary rectangle frame area then carries out teacher's writing on the blackboard behavior and detects 3000 between 10000 the time;
4) external rectangle frame is carried out area dividing, it is upper right to calculate target body boundary rectangle frame, the amount of exercise of bottom right and bottom left section moving region;
5) satisfy formula (3) simultaneously when this trizonal amount of exercise, the relation of (4) and (5) then shows to detect the teacher in writing on the blackboard;
6) if when being consecutively detected teacher's writing on the blackboard behavior and surpassing 5 two field pictures, video camera then stretches.If when continuous detecting surpassed 50 two field pictures less than teacher's writing on the blackboard behavior, then draw ratio was reduced to initial multiple.
Claims (2)
1. the automatic identifying method of teacher's blackboard action of writing during record a video in a classroom is characterized in that this method may further comprise the steps:
A. use capture card to gather a two field picture, do the image subtraction computing, obtain the frame difference image of binaryzation and analyze with present frame and former frame;
B. after the closed operation in the two-value frame difference image doing mathematics morphology being handled, carry out connected component labeling again,, remove in the two-value frame difference image comparatively obvious noise zone by threshold value is set;
C. obtain the target rectangle frame and calculate the area of rectangle, when rectangular area satisfies institute's fixed condition, be judged as human motion, then teacher's writing on the blackboard behavior is detected and discerns;
D. if be consecutively detected when writing the blackboard behavior, then video camera is carried out stretch processing above 5 two field pictures;
E. if continuous detecting when writing the blackboard behavior and surpass 50 two field pictures, then is reduced to initial multiple to the draw ratio of video camera.
2. the automatic identifying method of teacher's blackboard action of writing is characterized by in described two-value frame difference image in the video recording of classroom according to claim 1, and the momental method that the judgement teacher writes the blackboard action is:
1) frame difference image analysis: the object of frame difference image analysis is the boundary rectangle frame of target body, be to the motion target detection result, at first the closed operation in the frame difference image doing mathematics morphology is handled, then the result after handling is carried out connected component labeling, remove the noise in the two-value frame difference image; When the pixel number of certain connected region less than 1000 the time, think that this zone is a noise region, at this moment, the black pixel point of in the two-value frame difference image, forming target body, remainder is the white pixel point, travel through the coordinate of all black pixel points in the whole two-value frame difference image, and in the middle of these coordinates, find out the minimum and maximum coordinate figure of level and vertical direction respectively, determine the top left corner apex A of target body boundary rectangle frame, summit, upper right corner B, summit, lower right corner C, the coordinate of summit, lower left corner D, when rectangular area 3000 between 10000 the time, just illustrate that the teacher might write blackboard;
2) divide the moving region: the attitude that the teacher is writing blackboard generally is to lift the right hand, hand has bigger motion and the action of other positions is very little, therefore at first target body boundary rectangle frame to be divided into several zones, since the action of writing blackboard be the teacher dorsad camera reach and write writing on the blackboard, and the teacher satisfies certain proportion towards the upper right side usually to hand that stretches out writing on the blackboard and health, therefore can determine near underarm region that cuts apart a unique point G, by cutting apart the straight line that unique point G does level and vertical direction respectively, and crossing, thereby the boundary rectangle frame can be divided into 4 rectangular areas with the boundary rectangle frame;
Image when selecting 100 width of cloth teacher writings on the blackboard calculates hand width pixel value and hand height pixel value in every width of cloth image as sample, and all samples is averaged just can roughly calculate H
hAnd H
wAgain by the coordinate on four summits of boundary rectangle frame that obtained, just can calculate the coordinate of feature cut-point G, after the coordinate of having determined feature cut-point G, the level and the vertical direction straight line of undue cutpoint is with the boundary rectangle frame is left and right, the coordinate of intersection point E, the F of upper and lower four edges, H, I is also definite thereupon, thereby obtain four moving regions by target body boundary rectangle frame, wherein the upper right portion zone of boundary rectangle frame has comprised the teacher and has stretched out the hand of writing blackboard, and whole left part has comprised teacher's health;
3) judge amount of exercise: according to ready-portioned 4 moving regions, calculate each regional amount of exercise respectively, certain regional amount of exercise can be that the number of 1 pixel is described by this regional value in the two-value frame difference image, and the teacher writes the right regions that the attitude of blackboard is reflected as in the target body boundary rectangle frame bigger motion is arranged, and lower right area not motion basically, lower-left and top left region only have small motion;
Therefore when each regional amount of exercise satisfies following three conditions, judge that this moment, the objects in video human body was write blackboard:
A. the amount of exercise of upper right hand region HBFG is bigger than the amount of exercise of bottom right white space GFCI;
B. the amount of exercise of the body region EGID of lower-left is bigger than the amount of exercise of bottom right white space GFCI;
C. upper right hand region HBFG has certain amount of exercise, and promptly the ratio of hand region amount of exercise and this region area reaches certain ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910185427A CN101699469A (en) | 2009-11-09 | 2009-11-09 | Method for automatically identifying action of writing on blackboard of teacher in class video recording |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910185427A CN101699469A (en) | 2009-11-09 | 2009-11-09 | Method for automatically identifying action of writing on blackboard of teacher in class video recording |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101699469A true CN101699469A (en) | 2010-04-28 |
Family
ID=42147927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910185427A Pending CN101699469A (en) | 2009-11-09 | 2009-11-09 | Method for automatically identifying action of writing on blackboard of teacher in class video recording |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101699469A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679177A (en) * | 2013-12-27 | 2014-03-26 | 天津森宇科技发展有限公司 | Location extracting processor for image point group |
| CN104333735A (en) * | 2014-10-27 | 2015-02-04 | 燕山大学 | Anywhere defecation behavior monitoring method based on machine vision |
| CN104537672A (en) * | 2015-01-05 | 2015-04-22 | 长安大学 | Road environment monotony detection method based on videos |
| CN104539888A (en) * | 2014-12-16 | 2015-04-22 | 广西科技大学 | Video monitoring method for closed-chest cardiac massage in cardio-pulmonary resuscitation first aid training |
| CN104540026A (en) * | 2014-12-22 | 2015-04-22 | 杭州施强网络科技有限公司 | Lesson content video recoding and replaying method and system |
| CN104574359A (en) * | 2014-11-03 | 2015-04-29 | 南京邮电大学 | Student tracking and positioning method based on primary and secondary cameras |
| CN107454395A (en) * | 2017-08-23 | 2017-12-08 | 上海安威士科技股份有限公司 | A kind of high-definition network camera and intelligent code stream control method |
| CN107452018A (en) * | 2017-08-02 | 2017-12-08 | 北京翰博尔信息技术股份有限公司 | Speaker's tracking and system |
| CN107527418A (en) * | 2017-07-11 | 2017-12-29 | 深圳怡化电脑股份有限公司 | A kind of Heisui River lettering position method, apparatus, terminal device and readable storage medium storing program for executing |
| CN108171282A (en) * | 2017-12-29 | 2018-06-15 | 安徽慧视金瞳科技有限公司 | A kind of blackboard person's handwriting automatic synthesis method |
| CN108881765A (en) * | 2018-05-25 | 2018-11-23 | 讯飞幻境(北京)科技有限公司 | Light weight recorded broadcast method, apparatus and system |
| WO2019000653A1 (en) * | 2017-06-30 | 2019-01-03 | 清华大学深圳研究生院 | Image target identification method and apparatus |
| CN109543530A (en) * | 2018-10-19 | 2019-03-29 | 福州瑞芯微电子股份有限公司 | A kind of writing on the blackboard method for detecting position, storage medium and system |
| CN109583352A (en) * | 2018-11-22 | 2019-04-05 | 广州市保伦电子有限公司 | Classroom teacher's behaviors acquisition methods, device and medium based on video analysis |
| CN109800662A (en) * | 2018-12-28 | 2019-05-24 | 广州海昇计算机科技有限公司 | A kind of teacher teaches Activity recognition method, system, device and storage medium |
| CN110532863A (en) * | 2019-07-19 | 2019-12-03 | 平安科技（深圳）有限公司 | Gesture operation method, device and computer equipment |
| CN110765814A (en) * | 2018-07-26 | 2020-02-07 | 杭州海康威视数字技术股份有限公司 | Blackboard writing behavior recognition method and device and camera |
| CN110866473A (en) * | 2019-11-04 | 2020-03-06 | 浙江大华技术股份有限公司 | Target object tracking detection method and device, storage medium and electronic device |
| CN110933331A (en) * | 2019-12-06 | 2020-03-27 | 浙江蓝鸽科技有限公司 | Teaching video synthesis method and system |
| CN111008969A (en) * | 2019-12-06 | 2020-04-14 | 浙江蓝鸽科技有限公司 | Blackboard writing extraction and intelligent recognition method and system |
| CN111325073A (en) * | 2018-12-17 | 2020-06-23 | 上海交通大学 | Monitoring video abnormal behavior detection method based on motion information clustering |
| WO2020168818A1 (en) * | 2019-02-20 | 2020-08-27 | 广州视源电子科技股份有限公司 | Image processing method, system and device, storage medium, and blackboard device |
| CN111814587A (en) * | 2020-06-18 | 2020-10-23 | 浙江大华技术股份有限公司 | Human behavior detection method, teacher behavior detection method, and related system and device |
| CN113572965A (en) * | 2021-08-06 | 2021-10-29 | 广州市奥威亚电子科技有限公司 | Double-blackboard-writing tracking shooting method and device and shooting system |
| CN114898409A (en) * | 2022-07-14 | 2022-08-12 | 深圳市海清视讯科技有限公司 | Data processing method and device |
-
2009
- 2009-11-09 CN CN200910185427A patent/CN101699469A/en active Pending
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679177A (en) * | 2013-12-27 | 2014-03-26 | 天津森宇科技发展有限公司 | Location extracting processor for image point group |
| CN104333735A (en) * | 2014-10-27 | 2015-02-04 | 燕山大学 | Anywhere defecation behavior monitoring method based on machine vision |
| CN104574359A (en) * | 2014-11-03 | 2015-04-29 | 南京邮电大学 | Student tracking and positioning method based on primary and secondary cameras |
| CN104574359B (en) * | 2014-11-03 | 2017-09-29 | 南京邮电大学 | A kind of student trace localization method based on principal and subordinate's video camera |
| CN104539888A (en) * | 2014-12-16 | 2015-04-22 | 广西科技大学 | Video monitoring method for closed-chest cardiac massage in cardio-pulmonary resuscitation first aid training |
| CN104539888B (en) * | 2014-12-16 | 2018-06-05 | 广西科技大学 | The video frequency monitoring method of closed cardiac massage art in CPR first aid training |
| CN104540026A (en) * | 2014-12-22 | 2015-04-22 | 杭州施强网络科技有限公司 | Lesson content video recoding and replaying method and system |
| CN104540026B (en) * | 2014-12-22 | 2018-02-09 | 杭州施强教育科技有限公司 | A kind of class offerings video recording and back method and system |
| CN104537672B (en) * | 2015-01-05 | 2018-11-30 | 长安大学 | A kind of detection method of the road environment monotonicity based on video |
| CN104537672A (en) * | 2015-01-05 | 2015-04-22 | 长安大学 | Road environment monotony detection method based on videos |
| WO2019000653A1 (en) * | 2017-06-30 | 2019-01-03 | 清华大学深圳研究生院 | Image target identification method and apparatus |
| CN107527418B (en) * | 2017-07-11 | 2019-08-23 | 深圳怡化电脑股份有限公司 | A kind of Heisui River lettering position method, apparatus, terminal device and readable storage medium storing program for executing |
| CN107527418A (en) * | 2017-07-11 | 2017-12-29 | 深圳怡化电脑股份有限公司 | A kind of Heisui River lettering position method, apparatus, terminal device and readable storage medium storing program for executing |
| CN107452018A (en) * | 2017-08-02 | 2017-12-08 | 北京翰博尔信息技术股份有限公司 | Speaker's tracking and system |
| CN107454395A (en) * | 2017-08-23 | 2017-12-08 | 上海安威士科技股份有限公司 | A kind of high-definition network camera and intelligent code stream control method |
| CN108171282A (en) * | 2017-12-29 | 2018-06-15 | 安徽慧视金瞳科技有限公司 | A kind of blackboard person's handwriting automatic synthesis method |
| CN108171282B (en) * | 2017-12-29 | 2021-08-31 | 安徽慧视金瞳科技有限公司 | Automatic blackboard handwriting synthesis method |
| CN108881765A (en) * | 2018-05-25 | 2018-11-23 | 讯飞幻境(北京)科技有限公司 | Light weight recorded broadcast method, apparatus and system |
| CN110765814A (en) * | 2018-07-26 | 2020-02-07 | 杭州海康威视数字技术股份有限公司 | Blackboard writing behavior recognition method and device and camera |
| CN109543530A (en) * | 2018-10-19 | 2019-03-29 | 福州瑞芯微电子股份有限公司 | A kind of writing on the blackboard method for detecting position, storage medium and system |
| CN109543530B (en) * | 2018-10-19 | 2020-11-06 | 瑞芯微电子股份有限公司 | Blackboard writing position detection method, storage medium and system |
| CN109583352A (en) * | 2018-11-22 | 2019-04-05 | 广州市保伦电子有限公司 | Classroom teacher's behaviors acquisition methods, device and medium based on video analysis |
| CN111325073B (en) * | 2018-12-17 | 2024-02-20 | 上海交通大学 | Monitoring video abnormal behavior detection method based on motion information clustering |
| CN111325073A (en) * | 2018-12-17 | 2020-06-23 | 上海交通大学 | Monitoring video abnormal behavior detection method based on motion information clustering |
| CN109800662A (en) * | 2018-12-28 | 2019-05-24 | 广州海昇计算机科技有限公司 | A kind of teacher teaches Activity recognition method, system, device and storage medium |
| WO2020168818A1 (en) * | 2019-02-20 | 2020-08-27 | 广州视源电子科技股份有限公司 | Image processing method, system and device, storage medium, and blackboard device |
| WO2021012513A1 (en) * | 2019-07-19 | 2021-01-28 | 平安科技（深圳）有限公司 | Gesture operation method and apparatus, and computer device |
| CN110532863A (en) * | 2019-07-19 | 2019-12-03 | 平安科技（深圳）有限公司 | Gesture operation method, device and computer equipment |
| CN110866473B (en) * | 2019-11-04 | 2022-11-18 | 浙江大华技术股份有限公司 | Target object tracking detection method and device, storage medium and electronic device |
| CN110866473A (en) * | 2019-11-04 | 2020-03-06 | 浙江大华技术股份有限公司 | Target object tracking detection method and device, storage medium and electronic device |
| CN111008969A (en) * | 2019-12-06 | 2020-04-14 | 浙江蓝鸽科技有限公司 | Blackboard writing extraction and intelligent recognition method and system |
| CN110933331A (en) * | 2019-12-06 | 2020-03-27 | 浙江蓝鸽科技有限公司 | Teaching video synthesis method and system |
| CN111814587A (en) * | 2020-06-18 | 2020-10-23 | 浙江大华技术股份有限公司 | Human behavior detection method, teacher behavior detection method, and related system and device |
| CN113572965A (en) * | 2021-08-06 | 2021-10-29 | 广州市奥威亚电子科技有限公司 | Double-blackboard-writing tracking shooting method and device and shooting system |
| CN114898409A (en) * | 2022-07-14 | 2022-08-12 | 深圳市海清视讯科技有限公司 | Data processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101699469A (en) | Method for automatically identifying action of writing on blackboard of teacher in class video recording | |
| CN108090857B (en) | Multi-mode student classroom behavior analysis system and method | |
| CN101447082B (en) | Detection method of moving target on a real-time basis | |
| CN101406390B (en) | Method and apparatus for detecting part of human body and human, and method and apparatus for detecting objects | |
| CN108805116B (en) | Image text detection method and system | |
| CN112633231B (en) | Fire disaster identification method and device | |
| CN107403175A (en) | Visual tracking method and Visual Tracking System under a kind of movement background | |
| CN105160310A (en) | 3D (three-dimensional) convolutional neural network based human body behavior recognition method | |
| CN102087790B (en) | Method and system for low-altitude ground vehicle detection and motion analysis | |
| CN106204633B (en) | Student tracking method and device based on computer vision | |
| CN112488071B (en) | Method, device, electronic equipment and storage medium for extracting pedestrian features | |
| CN105930794A (en) | Indoor scene identification method based on cloud computing | |
| CN104794479A (en) | Method for detecting text in natural scene picture based on local width change of strokes | |
| CN102096812A (en) | Teacher blackboard writing action detection method for intelligent teaching recording and playing system | |
| CN112001219A (en) | Multi-angle multi-face recognition attendance checking method and system | |
| CN107948586B (en) | Trans-regional moving target detecting method and device based on video-splicing | |
| CN103096117B (en) | Video noise detection method and device | |
| CN113283334B (en) | Classroom concentration analysis method, device and storage medium | |
| CN112686122B (en) | Human body and shadow detection method and device, electronic equipment and storage medium | |
| CN115471773B (en) | Intelligent classroom-oriented student tracking method and system | |
| CN115272340B (en) | Industrial product defect detection method and device | |
| CN110334703B (en) | Ship detection and identification method in day and night image | |
| CN110084146B (en) | Pedestrian detection method and device based on shielding perception self-supervision learning | |
| CN106502246A (en) | A kind of intelligent vehicle automated induction systems based on grader | |
| CN112949621A (en) | Method and device for marking test paper answering area, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100428 |