CN101789127A - Method for extracting target from video image - Google Patents
Method for extracting target from video image Download PDFInfo
- Publication number
- CN101789127A CN101789127A CN 201010114922 CN201010114922A CN101789127A CN 101789127 A CN101789127 A CN 101789127A CN 201010114922 CN201010114922 CN 201010114922 CN 201010114922 A CN201010114922 A CN 201010114922A CN 101789127 A CN101789127 A CN 101789127A
- Authority
- CN
- China
- Prior art keywords
- frame image
- target
- image
- foreground area
- angle point
- 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
- 238000000034 method Methods 0.000 title claims abstract description 19
- 238000001514 detection method Methods 0.000 claims abstract description 22
- 238000005286 illumination Methods 0.000 claims abstract description 20
- 238000011410 subtraction method Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 3
- 238000011960 computer-aided design Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005314 correlation function Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Abstract
The invention relates to a method for extracting a target from a video image, and discloses a method for extracting the target from the video image capable of improving the accuracy of the target detection aiming at the problems that the target detection method in the prior art is easily influenced by illumination and the detection accuracy is low. The technical scheme of the invention is that: the method for extracting the target from the video image comprises the following steps: a, acquiring a previous frame image and a current frame image; b, detecting the angular points of the previous frame image and the current frame image; c, acquiring a foreground area according to the difference between the previous frame image and the current frame image; d, counting the number of the angular points of the previous frame image and the current frame image in the foreground area and counting the number of position-matched angular points of the previous frame image and the current frame image in the foreground area; and e, judging whether a target is present in the foreground area according to a relationship between the number of the angular points counted by the step d and set parameters. The method for extracting the target from the video image is particularly suitable for detecting a target in a fixed area under the condition of illumination influence.
Description
Technical field
The present invention relates to video image processing technology, particularly a kind of method of in video image, extracting target.
Background technology
The extraction of target in the video image (or being called foreground extraction, foreground detection) is the common technology in fields such as computer vision, video monitoring, target following.The background subtraction method is the most frequently used method, and this method obtains foreground image F with the difference of current frame image imgC and background image (or former frame image) imgB: F=imgC-imgB, thereby determines foreground area.
But the variation of illumination can cause the foreground extraction mistake, has very important significance for foreground extraction so eliminate the influence of illumination.
For eliminating the influence that illumination is extracted target, a lot of algorithms have been proposed now.The Video Segmentation of gauss hybrid models and gradient information " under the illumination sudden change environment based on " (Chinese image graphics journal, 2007.12 (11): 2068-2072) propose the structure gradient cross correlation function is applied to the foreground detection that illumination causes, zone in illumination variation, brightness and color component all do not change, but gradient and background correlativity are big, and the foreground area that target causes, bigger change takes place in its gradient, and is little with the background correlativity.Find that through experiment gradient has reflected texture information, the variation of texture has been represented in the variation of gradient information to a certain extent, and during integral image generation illumination variation, the texture information of image is constant, detects feasible.But two kinds of situations are arranged, and this method is inapplicable: (1) produces hot spot when image local generation illumination variation, hot spot has the edge, and its graded is very big, can be surveyed by mistake to be target.When (2) color of object is evenly distributed, target internal can be considered to equal proportion generation change, the cavity can appear in target internal, influences target detection.
Summary of the invention
Technical matters to be solved by this invention, the object detection method at prior art is subjected to illumination effect easily exactly, and the problem that detection accuracy is low provides target extraction method in a kind of video image, improves the accuracy of target detection.
The technical scheme that the present invention solve the technical problem employing is that the method for extraction target in the video image may further comprise the steps:
A, collection former frame image and current frame image;
The angle point of b, detection former frame image and current frame image;
C, poor according to current frame image and former frame image obtain foreground area;
D, add up former frame image and the current frame image angle point number in described foreground area and the angle point number of former frame image and current frame image location matches in this zone respectively;
E, according to the angle point number of steps d statistics and the relation of setup parameter, judge whether described foreground area has target;
Concrete, in the steps d, adopt the background subtraction method to obtain the poor of current frame image and former frame image;
Further, the angle point number of note current frame image in foreground area is N
c, former frame image angle point number in the same area is N
b, former frame image and the current frame image angle point number of location matches in foreground area is N
m
The concrete determination methods of step e is as follows:
E1, if
N
m＜T；N
b＞T
1；N
c＞T
1；
Wherein T, T
1Be setting threshold, then judging has target to enter;
E2, if
N
b≥T；N
c＜T；N
m＜T；
Then judging has target to be removed;
E3, if
N
c＜T；N
b＜T
2；
T wherein
2Be setting threshold, judge that then this moment and driftlessness occur, foreground area changes and is caused by illumination variation;
E4, if
N
c＞T
1；N
m≥N
b×T
3；N
m≥T；
T wherein
3Be setting threshold, a number percent judges that then this moment and driftlessness occur, and foreground area changes and caused by illumination;
Concrete, described former frame image is the background image of fixed area.
The invention has the beneficial effects as follows, can eliminate illumination effect fast and effectively, target is made accurately judged, and can judge also that by corners Matching target removes or brings into, the subsequent treatment and the video early warning of video had significance.
Embodiment
Below in conjunction with embodiment, describe technical scheme of the present invention in detail.
The present invention as characteristics of image, according to the variation of angle point in the video image, with the comparison of setting value (parameter), judges whether the variation of foreground area is caused by target with angle point, or owing to the variation of illumination causes, thereby target is judged accurately.
The present invention adopts Harris Corner Detection technology, introduces the angular-point detection method of this technology below.
1) computed image I (x, partial structurtes matrix M y)
Wherein:
2) calculate each pixel angle point metric C (x, y)
C(x，y)＝det(M)-k×(trace(M))
2
Wherein:
Det (M)=AB-C
2Trace (M)=A+B; K is a constant.
3) angle point is judged
(x, y)＜P (P is an empirical value, is generally zero), (x is zero y), and whether the angle point metric that detects each pixel then is place regional area maximum, if maximum, then keeps, otherwise zero setting then to put C if C.At last, (x, y) non-zero think that then it is an angle point as if C.
Adopt said method that former frame image and current frame image are handled, just can get their angle point image.
Can certainly adopt other Corner Detection technology, as: based on the Corner Detection of gradient (referring to DericheR, Giraudon G.A Computational Approach for Corner and Vertex Detection[J] .Computer Vision, 1993,10 (2): 101-120), based on the Corner Detection of contour curve (referring to Xiao Qian, the Shandong magnificence. the self-adaptive angular-point level and smooth based on Gauss detects [J]. computer-aided design (CAD) and graphics journal, 2003,15 (11): 1358-1361).
The present invention adopts background subtraction method (or being called the background subtraction method) to determine foreground area.
Embodiment
Background image and present image with the monitoring camera collection illustrates object detection method of the present invention as former frame image and current frame image respectively below.The background image here is exactly the background image of fixed area.
The first step, collection background image and present image.
The angle point of second step, employing Harris angular-point detection method difference detection background image and present image.
The 3rd step, adopt the background subtraction method to make the calculus of differences of background image and present image, poor according to background image and present image obtains foreground area.
The 4th step, statistics angle point number: the angle point number of former frame image and current frame image location matches in foreground area is designated as N
mBe in the angle point number in the foreground area in the present image, be designated as N
cThe angle point number of background image in the same area is designated as N
b
The 5th the step, the angle point in described foreground area is counted N according to present image and background image
c, N
bAnd the angle point that mates in this zone is counted N
m, with setup parameter T, T
1, T
2, T
3Relation, judge whether described foreground area has target.Wherein: T, T
1, T
2, T
3Be setting threshold, their value is relevant with actual image content, as angle point number in the image etc.
Concrete determination methods is as follows:
If N
m＜T; N
b＞T
1N
c＞T
1Here T
1Approximate the angle point number of background image in foreground area, T can get less value;
This situation represents that present image and background image angle point in foreground area is all a lot, and should mate angle point situation seldom in the zone, and then judging has new object (target) to occur.
If N
b〉=T; N
c＜T; N
m＜T; Here T approximates the angle point number of background image in foreground area, T
1Can get less value;
This situation represents that the angle point of present image in foreground area is few, and the angle point of background image in foreground area is many, the few situation of coupling angle point in this zone, and then judging has object (target) to be removed.
If N
c＜T; N
b＜T
2Here T, T
2Can get less value;
This situation is represented present image and background image angle point seldom situation all in foreground area, judges that there is no object (target) this moment occurs, and prospect is caused by illumination.
If N
c＞T
1N
m〉=N
b* T
3N
m〉=T; Here T
1Approximate the angle point number of background image in foreground area, T≤T
1, T
3Be a number percent;
This situation represents that present image angle point in foreground area is many, background image angle point in foreground area is many, coupling angle point number reaches the situation of the certain percentage of background image angle point number in this zone, judges that there is no object (target) this moment occurs, and prospect is caused by illumination.
The present invention compares with existing relevant art, this target extraction method based on Corner Detection even can produce in the situation that hot spot and color of object be evenly distributed, is eliminated illumination effect in the high light environment, extract more accurately target, improved the accuracy of target detection.
Claims (5)
1. extract the method for target in the video image, may further comprise the steps:
A, collection former frame image and current frame image;
The angle point of b, detection former frame image and current frame image;
C, poor according to current frame image and former frame image obtain foreground area;
D, add up former frame image and the current frame image angle point number in described foreground area and the angle point number of former frame image and current frame image location matches in this zone respectively;
E, according to the angle point number of steps d statistics and the relation of setup parameter, judge whether described foreground area has target.
2. extract the method for target in the video image according to claim 1, it is characterized in that, in the steps d, adopt the background subtraction method to obtain the poor of current frame image and former frame image.
3. extract the method for target in the video image according to claim 1 and 2, it is characterized in that, the angle point number of note current frame image in foreground area is N
c, former frame image angle point number in the same area is N
b, former frame image and the current frame image angle point number of location matches in foreground area is N
m
The concrete determination methods of step e is as follows:
E1, if
N
m＜T；N
b＞T
1；N
c＞T
1；
Wherein T, T
1Be setting threshold, then judging has target to enter;
E2, if
N
b≥T；N
c＜T；N
m＜T；
Then judging has target to be removed;
E3, if
N
c＜T；N
b＜T
2；
T wherein
2Be setting threshold, judge that then this moment and driftlessness occur, foreground area is caused by illumination;
E4, if
N
c＞T
1；N
m≥N
b×T
3；N
m≥T；
T wherein
3Be setting threshold, a number percent judges that then this moment and driftlessness occur, and foreground area is caused by illumination.
4. extract the method for target in the video image according to claim 3, it is characterized in that described former frame image is the background image of fixed area.
5. extract the method for target in the video image according to claim 1 and 2, it is characterized in that described former frame image is the background image of fixed area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101149227A CN101789127B (en) | 2010-02-26 | 2010-02-26 | Method for extracting target from video image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101149227A CN101789127B (en) | 2010-02-26 | 2010-02-26 | Method for extracting target from video image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101789127A true CN101789127A (en) | 2010-07-28 |
| CN101789127B CN101789127B (en) | 2012-01-18 |
Family
ID=42532330
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101149227A Expired - Fee Related CN101789127B (en) | 2010-02-26 | 2010-02-26 | Method for extracting target from video image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101789127B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102044080A (en) * | 2010-12-16 | 2011-05-04 | 北京航空航天大学 | Mobile object detection method and device |
| CN102685398A (en) * | 2011-09-06 | 2012-09-19 | 天脉聚源(北京)传媒科技有限公司 | News video scene generating method |
| CN104182993A (en) * | 2014-09-10 | 2014-12-03 | 四川九洲电器集团有限责任公司 | Target tracking method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101097601A (en) * | 2006-06-26 | 2008-01-02 | 北京航空航天大学 | Image rapid edge matching method based on angle point guiding |
| CN101216941A (en) * | 2008-01-17 | 2008-07-09 | 上海交通大学 | Motion estimation method under violent illumination variation based on corner matching and optic flow method |
| CN101655982A (en) * | 2009-09-04 | 2010-02-24 | 上海交通大学 | Image registration method based on improved Harris angular point |
-
2010
- 2010-02-26 CN CN2010101149227A patent/CN101789127B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101097601A (en) * | 2006-06-26 | 2008-01-02 | 北京航空航天大学 | Image rapid edge matching method based on angle point guiding |
| CN101216941A (en) * | 2008-01-17 | 2008-07-09 | 上海交通大学 | Motion estimation method under violent illumination variation based on corner matching and optic flow method |
| CN101655982A (en) * | 2009-09-04 | 2010-02-24 | 上海交通大学 | Image registration method based on improved Harris angular point |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102044080A (en) * | 2010-12-16 | 2011-05-04 | 北京航空航天大学 | Mobile object detection method and device |
| CN102044080B (en) * | 2010-12-16 | 2014-04-23 | 北京航空航天大学 | Mobile object detection method and device |
| CN102685398A (en) * | 2011-09-06 | 2012-09-19 | 天脉聚源(北京)传媒科技有限公司 | News video scene generating method |
| CN102685398B (en) * | 2011-09-06 | 2014-08-13 | 天脉聚源(北京)传媒科技有限公司 | News video scene generating method |
| CN104182993A (en) * | 2014-09-10 | 2014-12-03 | 四川九洲电器集团有限责任公司 | Target tracking method |
| CN104182993B (en) * | 2014-09-10 | 2017-02-15 | 四川九洲电器集团有限责任公司 | Target tracking method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101789127B (en) | 2012-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yan et al. | One extended OTSU flame image recognition method using RGBL and stripe segmentation | |
| CN107230188B (en) | Method for eliminating video motion shadow | |
| CN105740945A (en) | People counting method based on video analysis | |
| CN101923637B (en) | A kind of mobile terminal and method for detecting human face thereof and device | |
| CN113902641B (en) | Data center hot zone judging method and system based on infrared image | |
| CN115249246A (en) | Optical glass surface defect detection method | |
| CN111611907A (en) | Image-enhanced infrared target detection method | |
| CN114494318B (en) | Cornea contour extraction method based on cornea dynamic deformation video of Ojin algorithm | |
| Yang et al. | An efficient method for vehicle model identification via logo recognition | |
| CN111754538A (en) | Threshold segmentation method for USB surface defect detection | |
| CN112241693A (en) | Illegal welding fire image identification method based on YOLOv3 | |
| CN101789127B (en) | Method for extracting target from video image | |
| Jing et al. | Automatic recognition of weave pattern and repeat for yarn-dyed fabric based on KFCM and IDMF | |
| CN108520539B (en) | Image target detection method based on sparse learning variable model | |
| CN107871315B (en) | Video image motion detection method and device | |
| Gong et al. | An improved Canny algorithm based on adaptive 2D-Otsu and Newton Iterative | |
| CN108010076B (en) | End face appearance modeling method for intensive industrial bar image detection | |
| Liang et al. | An effective automatic tracking algorithm based on Camshift and Kalman filter | |
| CN106339709A (en) | Real-time image extraction method | |
| CN107067411B (en) | Mean-shift tracking method combined with dense features | |
| CN115147613A (en) | Infrared small target detection method based on multidirectional fusion | |
| Bao et al. | Solar panel segmentation under low contrast condition | |
| Wu et al. | Moving-object Detection Based on Shadow Removal and Prospect Reconstruction | |
| Zhen-Jie et al. | Research on Detection and Tracking of Moving Vehicles in Complex Environment Based on Real-Time Surveillance Video | |
| Yang et al. | A method of steel strip image segmentation based on local gray information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB03 | Change of inventor or designer information |
Inventor after: Supplementary construction Inventor after: Luo An Inventor after: Zhou Congjun Inventor after: Bai Yun Inventor after: Hu Ruhuan Inventor after: Yang Yun Inventor before: Lu Lu Inventor before: Zou Jianhua Inventor before: Bai Yun |
|
| COR | Change of bibliographic data |
Free format text: CORRECT: INVENTOR; FROM: LU LU ZOU JIANHUA BAI YUN TO: BU JIAN LUO AN ZHOU CONGJUN BAI YUN HU RUHUAN YANG YUN |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Method for extracting target from video image Effective date of registration: 20150119 Granted publication date: 20120118 Pledgee: The Agricultural Bank of Chengdu branch of Limited by Share Ltd. China Chengdu Pledgor: CHENGDU SANTAI ELECTRONIC INDUSTRY Co.,Ltd. Registration number: 2015510000005 |
|
| PLDC | Enforcement, change and cancellation of contracts on pledge of patent right or utility model | ||
| C56 | Change in the name or address of the patentee |
Owner name: CHENGDU SANTAI HOLDINGS GROUP CO., LTD. Free format text: FORMER NAME: CHENGDU SANTAI ELECTRONIC INDUSTRY CO., LTD. |
|
| CP01 | Change in the name or title of a patent holder |
Address after: 610091 Jinniu District high tech Industrial Park, Sichuan, Sichuan West Road, No. 42, No. Patentee after: CHENGDU SANTAI HOLDING GROUP CO.,LTD. Address before: 610091 Jinniu District high tech Industrial Park, Sichuan, Sichuan West Road, No. 42, No. Patentee before: CHENGDU SANTAI ELECTRONIC INDUSTRY Co.,Ltd. |
|
| PLDC | Enforcement, change and cancellation of contracts on pledge of patent right or utility model | ||
| PM01 | Change of the registration of the contract for pledge of patent right |
Change date: 20160205 Registration number: 2015510000005 Pledgor after: CHENGDU SANTAI HOLDING GROUP CO.,LTD. Pledgor before: CHENGDU SANTAI ELECTRONIC INDUSTRY Co.,Ltd. |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20160229 Granted publication date: 20120118 Pledgee: The Agricultural Bank of Chengdu branch of Limited by Share Ltd. China Chengdu Pledgor: CHENGDU SANTAI HOLDING GROUP CO.,LTD. Registration number: 2015510000005 |
|
| PLDC | Enforcement, change and cancellation of contracts on pledge of patent right or utility model | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Method for extracting target from video image Effective date of registration: 20160317 Granted publication date: 20120118 Pledgee: The Agricultural Bank of Chengdu branch of Limited by Share Ltd. China Chengdu Pledgor: CHENGDU SANTAI HOLDING GROUP CO.,LTD. Registration number: 2016510000008 |
|
| PLDC | Enforcement, change and cancellation of contracts on pledge of patent right or utility model | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20210129 Granted publication date: 20120118 Pledgee: The Agricultural Bank of Chengdu branch of Limited by Share Ltd. China Chengdu Pledgor: CHENGDU SANTAI HOLDING GROUP Co.,Ltd. Registration number: 2016510000008 |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210305 Address after: No. 1305, unit 1, building 1, No. 1700, North Tianfu Avenue, Chengdu hi tech Zone, China (Sichuan) pilot Free Trade Zone, Chengdu, Sichuan 610093 Patentee after: CHENGDU SANTAI INTELLIGENT TECHNOLOGY Co.,Ltd. Address before: 610091 No. 42 Shu West Road, Jinniu District hi tech Industrial Park, Sichuan, Chengdu Patentee before: CHENGDU SANTAI HOLDING GROUP Co.,Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120118 |