CN107767403A - A kind of method and apparatus based on moving object detection under dynamic background - Google Patents
A kind of method and apparatus based on moving object detection under dynamic background Download PDFInfo
- Publication number
- CN107767403A CN107767403A CN201711077301.4A CN201711077301A CN107767403A CN 107767403 A CN107767403 A CN 107767403A CN 201711077301 A CN201711077301 A CN 201711077301A CN 107767403 A CN107767403 A CN 107767403A
- Authority
- CN
- China
- Prior art keywords
- image
- characteristic point
- motion
- moving
- target
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The present invention provides a kind of method and apparatus based on moving object detection under dynamic background, can not detect complete motion target area relative to prior art presence, motion feature point testing result is vulnerable to the influence of error hiding characteristic point, causes error detection；It is not ideal enough to the interior zone Detection results of moving target, easily cause detection target interior void problem the problem of.A kind of method and apparatus based on moving object detection under dynamic background provided by the invention, for the moving object detection current situation under dynamic background and problem be present, moving object detection based on monocular vision under dynamic background is furtherd investigate, discrete point can only be detected in order to overcome characteristic point, make up the defects of motion compensation process can only detect moving target edge contour, detect moving target in real time, improve the integrality and accuracy rate of moving target, it is proposed merges characteristic point with motion compensation information method, detects moving target.
Description
Technical field
The present invention relates to dynamic detection field, more particularly to a kind of method based on moving object detection under dynamic background
And device.
Background technology
Moving target detecting method can substantially be divided into based on spy according to the difference using feature under existing dynamic background
Sign point and the moving target detecting method based on motion compensation.
The former is by extracting the characteristic point in image, and according to motion feature point and background characteristics point, in camera motion
During the movement differential information that shows, motion feature point is detected, so as to obtain the position of moving target.Such method can be examined
Characteristic point discrete on moving target is measured, but complete motion target area can not be detected, motion feature point testing result
It is vulnerable to the influence of error hiding characteristic point, causes error detection.The latter is by estimating the motion compensation model of image, by image pixel
Point substitutes into motion compensation model, calculates motion compensated image, according to luminance difference value of the pixel in motion compensation figure, detection
Motion target area.Such method is detected preferably to the marginal portion of moving target, but the interior zone of moving target is examined
It is not ideal enough to survey effect, when the movement velocity of target is slower, easily causes the interior void problem of detection target.
The content of the invention
It is an object of the present invention to provide a kind of method and apparatus based on moving object detection under dynamic background,
Solving prior art distinguished point based and the moving target detecting method based on motion compensation, the former presence can not detect completely
Motion target area, motion feature point testing result is vulnerable to the influence of error hiding characteristic point, the problem of causing error detection；Afterwards
Person is not ideal enough in the presence of the interior zone Detection results to moving target, when the movement velocity of target is slower, easily causes inspection
The problem of surveying the interior void problem of target.
In a first aspect, the embodiment of the present application provides a kind of method based on moving object detection under dynamic background, the side
Method includes：
Step S101：Gather image；
Step S102：Motion feature point is detected, obtains the characteristic point of two continuous frames image, movement function is built with characteristic point
Model, obtain characteristic point movable information image；
Step S103：Compensation campaign information is extracted, region is divided the image into and compensates, the motion for improving whole two field picture is mended
Repay effect；
Step S104：Output result, propose moving image target.
With reference to the application's in a first aspect, in second of embodiment of the application first aspect, step S101：Adopt
Collect image, including collection image is preserved.
With reference to the application's in a first aspect, in the third embodiment of the application first aspect, step S102：Inspection
Motion feature point is surveyed, obtains the characteristic point of two continuous frames image, movement function model is built with characteristic point, obtains characteristic point motion
Frame, the model are 6 parameter affine transform model construction motion models.
With reference to the application's in a first aspect, in the 4th kind of embodiment of the application first aspect, step S103：Carry
Compensation campaign information is taken, region is divided the image into and compensates, the motion compensation effect of whole two field picture is improved, also using true
Real two field picture does difference with motion compensated image, obtains difference image.
Second aspect, the embodiment of the present application provide a kind of device based on moving object detection under dynamic background, the side
Method includes：
Gather elementary area：For gathering image；
Detect moving cell：For detecting motion feature point, the characteristic point of two continuous frames image is obtained, is built with characteristic point
Movement function model, obtain characteristic point movable information image；
Extract compensation campaign unit：For extracting compensation campaign information, divide the image into region and compensate, improve whole frame
The motion compensation effect of image；
Obtain object element：Output result, obtain moving image target.
With reference to the second aspect of the application, in second of embodiment of the application first aspect, image list is gathered
Member：For gathering image, including for being preserved to collection image.
With reference to the second aspect of the application, in the third embodiment of the application first aspect, detection motion is single
Member：For detecting motion feature point, the characteristic point of two continuous frames image is obtained, movement function model is built with characteristic point, obtains
Characteristic point movable information image, the model are 6 parameter affine transform model construction motion models.
With reference to the second aspect of the application, in the 4th kind of embodiment of the application first aspect, extraction compensation fortune
Moving cell：For extracting compensation campaign information, divide the image into region and compensate, improve the motion compensation effect of whole two field picture
Fruit, in addition to for doing difference using real two field picture and motion compensated image, obtain difference image.
There is above-mentioned technical proposal to understand：It is a kind of based on moving target under dynamic background it is an object of the present invention to provide
The method and apparatus of detection, it is preceding relative to prior art distinguished point based and the moving target detecting method based on motion compensation
Person's presence can not detect complete motion target area, and motion feature point testing result is vulnerable to the shadow of error hiding characteristic point
Ring, the problem of causing error detection；The latter is not ideal enough in the presence of the interior zone Detection results to moving target, when the motion of target
When speed is slower, easily cause detection target interior void problem the problem of.One kind provided by the invention is based on dynamic background
The method and apparatus of lower moving object detection, for the moving object detection current situation under dynamic background and problem be present, it is right
The moving object detection based on monocular vision is furtherd investigate under dynamic background, in order to overcome characteristic point can only detect from
Scatterplot, the defects of motion compensation process can only detect moving target edge contour is made up, detects moving target in real time, is carried
The integrality and accuracy rate of high moving target, propose to merge characteristic point with motion compensation information method, detect moving target.
Brief description of the drawings
Some specific embodiments of the present invention are described in detail by way of example, and not by way of limitation with reference to the accompanying drawings hereinafter.
Identical reference denotes same or similar part or part in accompanying drawing.It should be appreciated by those skilled in the art that these
What accompanying drawing was not necessarily drawn to scale.In accompanying drawing：
Fig. 1 is a kind of method flow diagram based on moving object detection under dynamic background.
Embodiment
The present invention is carried out clear below in conjunction with the accompanying drawing in the embodiment of the present invention to the technical scheme in the embodiment of the present invention
Chu, complete description, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.
Based on the embodiment in the present invention, those of ordinary skill in the art are obtained all under the premise of creative work is not made
Other embodiment, belong to the scope of protection of the invention.
In a first aspect, the embodiment of the present application provides a kind of method based on moving object detection under dynamic background, the side
Method includes：
Step S101：Gather image；
Step S102：Motion feature point is detected, obtains the characteristic point of two continuous frames image, movement function is built with characteristic point
Model, obtain characteristic point movable information image；
Step S103：Compensation campaign information is extracted, region is divided the image into and compensates, the motion for improving whole two field picture is mended
Repay effect；
Step S104：Output result, propose moving image target.
It is existing to solve it is an object of the present invention to provide a kind of method based on moving object detection under dynamic background
There are technology distinguished point based and the moving target detecting method based on motion compensation, the former presence can not detect completely to move
Target area, motion feature point testing result are vulnerable to the influence of error hiding characteristic point, the problem of causing error detection；The latter is present
It is not ideal enough to the interior zone Detection results of moving target, when the movement velocity of target is slower, easily cause detection target
Interior void problem the problem of.
Further, step S101：Image is gathered, including collection image is preserved.
Further, step S102：Motion feature point is detected, the characteristic point of two continuous frames image is obtained, with characteristic point structure
Movement function model is built, obtains characteristic point movable information image, the model is that 6 parameter affine transform model constructions move mould
Type.
Further, step S103：Compensation campaign information is extracted, region is divided the image into and compensates, improve whole frame figure
The motion compensation effect of picture, difference also is done using real two field picture and motion compensated image, obtain difference image.
Second aspect, the embodiment of the present application provide a kind of device based on moving object detection under dynamic background, the side
Method includes：
Gather elementary area：For gathering image；
Detect moving cell：For detecting motion feature point, the characteristic point of two continuous frames image is obtained, is built with characteristic point
Movement function model, obtain characteristic point movable information image；
Extract compensation campaign unit：For extracting compensation campaign information, divide the image into region and compensate, improve whole frame
The motion compensation effect of image；
Obtain object element：Output result, obtain moving image target.
It is existing to solve it is an object of the present invention to provide a kind of device based on moving object detection under dynamic background
There are technology distinguished point based and the moving target detecting method based on motion compensation, the former presence can not detect completely to move
Target area, motion feature point testing result are vulnerable to the influence of error hiding characteristic point, the problem of causing error detection；The latter is present
It is not ideal enough to the interior zone Detection results of moving target, when the movement velocity of target is slower, easily cause detection target
Interior void problem the problem of.
Further, elementary area is gathered：For gathering image, including for being preserved to collection image.
Further, moving cell is detected：For detecting motion feature point, the characteristic point of two continuous frames image is obtained, is used
Characteristic point builds movement function model, obtains characteristic point movable information image, the model is 6 parameter affine transform model constructions
Motion model.
Further, compensation campaign unit is extracted：For extracting compensation campaign information, divide the image into region and mended
Repay, improve the motion compensation effect of whole two field picture, in addition to for doing difference using real two field picture and motion compensated image,
Obtain difference image.
There is above-mentioned technical proposal to understand：It is a kind of based on moving target under dynamic background it is an object of the present invention to provide
The method and apparatus of detection, it is preceding relative to prior art distinguished point based and the moving target detecting method based on motion compensation
Person's presence can not detect complete motion target area, and motion feature point testing result is vulnerable to the shadow of error hiding characteristic point
Ring, the problem of causing error detection；The latter is not ideal enough in the presence of the interior zone Detection results to moving target, when the motion of target
When speed is slower, easily cause detection target interior void problem the problem of.One kind provided by the invention is based on dynamic background
The method and apparatus of lower moving object detection, for the moving object detection current situation under dynamic background and problem be present, it is right
The moving object detection based on monocular vision is furtherd investigate under dynamic background, in order to overcome characteristic point can only detect from
Scatterplot, the defects of motion compensation process can only detect moving target edge contour is made up, detects moving target in real time, is carried
The integrality and accuracy rate of high moving target, propose to merge characteristic point with motion compensation information method, detect moving target.
So far, although those skilled in the art will appreciate that detailed herein have shown and described multiple showing for the present invention
Example property embodiment, still, still can be direct according to present disclosure without departing from the spirit and scope of the present invention
It is determined that or derive many other variations or modifications for meeting the principle of the invention.Therefore, the scope of the present invention is understood that and recognized
It is set to and covers other all these variations or modifications.
Claims (8)
- A kind of 1. method based on moving object detection under dynamic background, it is characterised in that methods described includes：Step S101：Gather image；Step S102：Motion feature point is detected, obtains the characteristic point of two continuous frames image, movement function mould is built with characteristic point Type, obtain characteristic point movable information image；Step S103：Compensation campaign information is extracted, region is divided the image into and compensates, improves the motion compensation effect of whole two field picture Fruit；Step S104：Output result, propose moving image target.
- 2. according to the method for claim 1, it is characterised in that step S101：Image is gathered, including collection image is carried out Preserve.
- 3. according to the method for claim 1, it is characterised in that step S102：Motion feature point is detected, obtains two continuous frames The characteristic point of image, movement function model is built with characteristic point, obtains characteristic point movable information image, the model is 6 parameters Affine Transform Model builds motion model.
- 4. according to the method for claim 1, it is characterised in that step S103：Compensation campaign information is extracted, is divided the image into Region compensates, and the motion compensation effect of whole two field picture is improved, also using real two field picture and motion compensated image Difference is done, obtains difference image.
- 5. a kind of device based on moving object detection under dynamic background, it is characterised in that described device includes：Gather elementary area：For gathering image；Detect moving cell：For detecting motion feature point, the characteristic point of two continuous frames image is obtained, is built and moved with characteristic point Function model, obtain characteristic point movable information image；Extract compensation campaign unit：For extracting compensation campaign information, divide the image into region and compensate, improve whole two field picture Motion compensation effect；Obtain object element：Output result, obtain moving image target.
- 6. according to the method for claim 1, it is characterised in that collection elementary area：For gathering image, including for pair Collection image is preserved.
- 7. according to the method for claim 1, it is characterised in that detection moving cell：For detecting motion feature point, obtain The characteristic point of two continuous frames image, movement function model is built with characteristic point, obtains characteristic point movable information image, the model For 6 parameter affine transform model construction motion models.
- 8. according to the method for claim 1, it is characterised in that extraction compensation campaign unit：For extracting compensation campaign letter Breath, divides the image into region and compensates, and improves the motion compensation effect of whole two field picture, in addition to for utilizing real frame figure As doing difference with motion compensated image, difference image is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711077301.4A CN107767403A (en) | 2017-11-06 | 2017-11-06 | A kind of method and apparatus based on moving object detection under dynamic background |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711077301.4A CN107767403A (en) | 2017-11-06 | 2017-11-06 | A kind of method and apparatus based on moving object detection under dynamic background |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107767403A true CN107767403A (en) | 2018-03-06 |
Family
ID=61273620
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711077301.4A Pending CN107767403A (en) | 2017-11-06 | 2017-11-06 | A kind of method and apparatus based on moving object detection under dynamic background |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107767403A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163334A (en) * | 2011-03-04 | 2011-08-24 | 北京航空航天大学 | Method for extracting video object under dynamic background based on fisher linear discriminant analysis |
| CN106651903A (en) * | 2016-12-30 | 2017-05-10 | 明见（厦门）技术有限公司 | Moving object detection method |
-
2017
- 2017-11-06 CN CN201711077301.4A patent/CN107767403A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102163334A (en) * | 2011-03-04 | 2011-08-24 | 北京航空航天大学 | Method for extracting video object under dynamic background based on fisher linear discriminant analysis |
| CN106651903A (en) * | 2016-12-30 | 2017-05-10 | 明见（厦门）技术有限公司 | Moving object detection method |
Non-Patent Citations (2)
| Title |
|---|
| 沈晋原: "《多媒体通信技术基础》", 31 May 2015 * |
| 王亮芬: "基于 SIFT 特征匹配和动态更新背景模型的运动目标检测算法", 《计算机应用与软件》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103077521B (en) | A kind of area-of-interest exacting method for video monitoring | |
| CN103955948B (en) | A kind of space movement target detection method under dynamic environment | |
| CN103679749B (en) | A kind of image processing method and device based on motion target tracking | |
| EP2858008A2 (en) | Target detecting method and system | |
| CN107292297A (en) | A kind of video car flow quantity measuring method tracked based on deep learning and Duplication | |
| CN104077577A (en) | Trademark detection method based on convolutional neural network | |
| CN105930795A (en) | Walking state identification method based on space vector between human body skeleton joints | |
| WO2008020598A1 (en) | Subject number detecting device and subject number detecting method | |
| CN104268853A (en) | Infrared image and visible image registering method | |
| CN109636771A (en) | Airbound target detection method and system based on image procossing | |
| CN104200492B (en) | Video object automatic detection tracking of taking photo by plane based on profile constraints | |
| CN106375706A (en) | Method and apparatus for velocity measurement of moving object by use of double pick-up heads, and terminal | |
| CN104899839A (en) | Ghost quick-inhibition method based on ViBe algorithm | |
| CN109166125A (en) | A kind of three dimensional depth image partitioning algorithm based on multiple edge syncretizing mechanism | |
| CN106815583A (en) | A kind of vehicle at night license plate locating method being combined based on MSER and SWT | |
| CN108009556A (en) | A kind of floater in river detection method based on fixed point graphical analysis | |
| CN104537688A (en) | Moving object detecting method based on background subtraction and HOG features | |
| CN101930606A (en) | Field depth extending method for image edge detection | |
| CN105869148A (en) | Target detection method and device | |
| CN103914829B (en) | Method for detecting edge of noisy image | |
| CN107545557A (en) | Egg detecting method and device in excrement image | |
| CN105787870A (en) | Graphic image splicing fusion system | |
| CN103714314A (en) | Television video station caption identification method combining edge and color information | |
| CN108021857B (en) | Building detection method based on unmanned aerial vehicle aerial image sequence depth recovery | |
| CN114463619B (en) | Infrared dim target detection method based on integrated fusion features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180306 |