CN107798689A - Traffic video image background extracting method - Google Patents

Traffic video image background extracting method Download PDF

Info

Publication number
CN107798689A
CN107798689A CN201610812063.6A CN201610812063A CN107798689A CN 107798689 A CN107798689 A CN 107798689A CN 201610812063 A CN201610812063 A CN 201610812063A CN 107798689 A CN107798689 A CN 107798689A
Authority
CN
China
Prior art keywords
image
background
sum
present
threshold value
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
Application number
CN201610812063.6A
Other languages
Chinese (zh)
Inventor
许鹏飞
贾银洁
孙秀英
王成林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huaian Vocational College of Information Technology
Original Assignee
Huaian Vocational College of Information Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huaian Vocational College of Information Technology filed Critical Huaian Vocational College of Information Technology
Priority to CN201610812063.6A priority Critical patent/CN107798689A/en
Publication of CN107798689A publication Critical patent/CN107798689A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence

Landscapes

  • Image Analysis (AREA)

Abstract

Traffic video image background extracting method of the present invention, comprises the following steps：1) difference of present image and background image is sought | Ik(x, y) Bk‑1(x, y) |；2) the bianry image BW of present image and background image difference is soughti BW i = 1 , | I k - B k - 1 | &GreaterEqual; T 1 0 , | I k - B k - 1 | < T 1 Wherein, Bk‑1For current background image threshold value, IkFor next frame image threshold, threshold value when T1 is bianry image；The change of actual motion scene illumination condition can be preferably adapted to, improves speed again, it is easy to accomplish, processing is quick, can adapt to illumination brightness, the change of gray scale, timely and effectively updates background.

Description

Traffic video image background extracting method
Technical field
The present invention relates to a kind of traffic video image background extracting method, belong to traffic video image field.
Background technology
At present, moving object detection algorithm the most frequently used in traffic video image background extracting is background subtraction, the back of the body The problem of basic task of scape calculus of finite differences is background constructing, and traditional background constructing algorithm is present is computing complexity, poor real, obtained The background effect arrived is bad.The present invention relates to traffic video image background extracting method, algorithm has multiple image method of average mould The characteristics of type is simple, arithmetic speed is fast, while there is the advantages of continuous frame difference method again, it can be determined that go out in new image whether With moving target, so as to effectively eliminate the influence that moving target is brought to background.In addition, by being put down with to image The method that equal gray value is compared, can preferably adapt to the change of actual motion scene illumination condition, the multiframe after improvement Image averaging method need not store former image, it is only necessary to which present image can extract background image, the computer of occupancy Resource is few, and the sampling to video and the extraction to background image can be carried out synchronously, so as to save the time, improve speed, This is advantage possessed by the multiple image method of average after improving, and new algorithm is easily achieved, and processing is quick, and it is strong to can adapt to illumination The change of degree, timely and effectively updates background.Test result indicates that the effect that new algorithm obtains is preferable.
The content of the invention
Deficiency of the prior art, the technical problem to be solved in the present invention are more than：There is provided one kind can be preferably The change of actual motion scene illumination condition is adapted to, improves speed again, it is easy to accomplish, processing is quick, and it is bright to can adapt to illumination Degree, the change of gray scale, timely and effectively update background.
A kind of traffic video image background extracting method of the present invention, it is characterised in that comprise the following steps：
1) difference of present image and background image is sought | IkCx, y)-Bk-1(x, y) |；
2) the bianry image BW of present image and background image difference is soughti
Wherein, Bk-1For current background image threshold value, IkFor next frame image threshold, threshold value when T1 is bianry image；
If 3) sum (sum (BWi)) ＜ T2；Then background is updated
If sum (sum (BWi))≥T2；Then background does not update
Wherein, T2 is that the gray value of bianry image is cumulative and threshold value；
4) the average value mean2 (I of next two field picture are soughtk)；
5) the average value mean2 (B of current background image are soughtk-1)；
6) α=mean2 (Ik)/mean2(Bk-1)；
If 7) | α -1 | ＞ T3, Bk=aBk-1
Wherein, α is desired level, and T3 is that intensity of illumination updates threshold value；
The step 1), 2) it is used for threshold value when present image and background image carry out calculus of differences and obtain bianry image T1。
Described step 3) model employs the multiple image method of average and background difference combines and extracts and update background image Method, this method is as follows：
Judge whether there is motion mesh in present image by the calculus of differences of present image and current background image first Mark is present；
Determine whether present image is updated in background image according to judged result；
By the gray value of the average value to present image and background image and bianry image is cumulative and T2 compared with sum (sum(BWi)) ＜ T2 or sum (sum (BWi)) >=T2 methods come judge whether to background image carry out grayness renewal.
α in described step 6) is by mean2 (Ik) and mean2 (Bk-1) determine, specially cause in illumination variation bright While degree changes, by the reduced value of the average value and the average value of next two field picture of current background image, with illumination Intensity renewal threshold value T3 makes contrast and is confirmed whether to make intensity of illumination renewal.
Brief description of the drawings
Fig. 1 is schematic flow sheet of the present invention.
Embodiment
Embodiments of the invention are described further below in conjunction with the accompanying drawings：
A kind of as shown in figure 1, traffic video image background extracting method of the present invention, it is characterised in that including with Lower step：
8) difference of present image and background image is sought | Ik(x, y)-Bk-1(x, y) |；
9) the bianry image BW of present image and background image difference is soughti
Wherein, Bk-1For current background image threshold value, IkFor next frame image threshold, threshold value when T1 is bianry image；
If 10) sum (sum (BWi)) ＜ T2；Then background is updated
If sum (sum (BWi))≥T2；Then background does not update
Wherein, T2 is that the gray value of bianry image is cumulative and threshold value；
11) the average value mean2 (I of next two field picture are soughtk)；
12) the average value mean2 (B of current background image are soughtk-1)；
13) α=mean2 (Ik)/mean2(Bk-1)；
If 14) | α -1 | ＞ T3, Bk=α Bk-1
Wherein, α is desired level, and T3 is that intensity of illumination updates threshold value；
The step 1), 2) it is used for threshold value when present image and background image carry out calculus of differences and obtain bianry image T1。
Described step 3) model employs the multiple image method of average and background difference combines and extracts and update background image Method, this method is as follows：
Judge whether there is motion mesh in present image by the calculus of differences of present image and current background image first Mark is present；
Determine whether present image is updated in background image according to judged result；
By the gray value of the average value to present image and background image and bianry image is cumulative and T2 compared with sum (sum(BWi)) ＜ T2 or sum (sum (BWi)) >=T2 methods come judge whether to background image carry out grayness renewal.
α in described step 6) is by mean2 (Ik) and mean2 (Bk-1) determine, specially cause in illumination variation bright While degree changes, by the reduced value of the average value and the average value of next two field picture of current background image, with illumination Intensity renewal threshold value T3 makes contrast and is confirmed whether to make intensity of illumination renewal.

Claims (4)

1. a kind of traffic video image background extracting method, it is characterised in that comprise the following steps：
1) difference of present image and background image is sought | Ik(x, y)-Bk-1(x, y) |；
2) the bianry image BW of present image and background image difference is soughti
Wherein, Bk-1For current background image threshold value, IkFor next frame image threshold, threshold value when T1 is bianry image；
If 3) sum (sum (BWi)) ＜ T2；Then background is updated
If sum (sum (BWi))≥T2；Then background does not update
Wherein, T2 is that the gray value of bianry image is cumulative and threshold value；
4) the average value mean2 (I of next two field picture are soughtk)；
5) the average value mean2 (B of current background image are soughtk-1)；
6) α=mean2 (Ik)/mean2(Bk-1)；
If 7) | α -1 | ＞ T3, Bk=α Bk-1
Wherein, α is desired level, and T3 is that intensity of illumination updates threshold value.
A kind of 2. traffic video image background extracting method according to claim 1, it is characterised in that：The step 1), 2) it is used for threshold value T1 when present image carries out calculus of differences with background image and obtains bianry image.
A kind of 3. traffic video image background extracting method according to claims 1, it is characterised in that：Described step Rapid 3) model employs the multiple image method of average and background difference combines the method extracted and update background image, and this method is such as Under：
Judge whether there is moving target to deposit in present image by the calculus of differences of present image and current background image first ；
Determine whether present image is updated in background image according to judged result；
By the gray value of the average value to present image and background image and bianry image is cumulative and T2 compared with sum (sum (BWi)) ＜ T2 or sum (sum (BWi)) >=T2 methods come judge whether to background image carry out grayness renewal.
A kind of 4. traffic video image background extracting method according to claims 1, it is characterised in that：Described step It is rapid 6) in α by mean2 (Ik) and mean2 (Bk-1) determine, specially cause brightness to change in illumination variation same When, by the reduced value of the average value and the average value of next two field picture of current background image, with intensity of illumination renewal threshold value T3 Contrast is made to be confirmed whether to make intensity of illumination renewal.
CN201610812063.6A 2016-08-28 2016-08-28 Traffic video image background extracting method Pending CN107798689A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610812063.6A CN107798689A (en) 2016-08-28 2016-08-28 Traffic video image background extracting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610812063.6A CN107798689A (en) 2016-08-28 2016-08-28 Traffic video image background extracting method

Publications (1)

Publication Number Publication Date
CN107798689A true CN107798689A (en) 2018-03-13

Family

ID=61530975

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610812063.6A Pending CN107798689A (en) 2016-08-28 2016-08-28 Traffic video image background extracting method

Country Status (1)

Country Link
CN (1) CN107798689A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110580706A (en) * 2018-06-11 2019-12-17 北京中科晶上超媒体信息技术有限公司 Method and device for extracting video background model

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1738426A (en) * 2005-09-09 2006-02-22 南京大学 Video motion goal division and track method
CN101017573A (en) * 2007-02-09 2007-08-15 南京大学 Method for detecting and identifying moving target based on video monitoring
CN101221663A (en) * 2008-01-18 2008-07-16 电子科技大学中山学院 Intelligent monitoring and alarming method based on movement object detection
CN101621615A (en) * 2009-07-24 2010-01-06 南京邮电大学 Self-adaptive background modeling and moving target detecting method
CN102930541A (en) * 2012-10-29 2013-02-13 深圳市开天源自动化工程有限公司 Background extracting and updating method of video images
CN103179325A (en) * 2013-03-26 2013-06-26 北京理工大学 Self-adaptive 3D (Three-Dimensional) noise reduction method for low signal-to-noise ratio video under fixed scene
CN103258332A (en) * 2013-05-24 2013-08-21 浙江工商大学 Moving object detection method resisting illumination variation
CN104616290A (en) * 2015-01-14 2015-05-13 合肥工业大学 Target detection algorithm in combination of statistical matrix model and adaptive threshold
CN105844671A (en) * 2016-04-12 2016-08-10 河北大学 Rapid background subtraction method under changing illumination conditions

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1738426A (en) * 2005-09-09 2006-02-22 南京大学 Video motion goal division and track method
CN101017573A (en) * 2007-02-09 2007-08-15 南京大学 Method for detecting and identifying moving target based on video monitoring
CN101221663A (en) * 2008-01-18 2008-07-16 电子科技大学中山学院 Intelligent monitoring and alarming method based on movement object detection
CN101621615A (en) * 2009-07-24 2010-01-06 南京邮电大学 Self-adaptive background modeling and moving target detecting method
CN102930541A (en) * 2012-10-29 2013-02-13 深圳市开天源自动化工程有限公司 Background extracting and updating method of video images
CN103179325A (en) * 2013-03-26 2013-06-26 北京理工大学 Self-adaptive 3D (Three-Dimensional) noise reduction method for low signal-to-noise ratio video under fixed scene
CN103258332A (en) * 2013-05-24 2013-08-21 浙江工商大学 Moving object detection method resisting illumination variation
CN104616290A (en) * 2015-01-14 2015-05-13 合肥工业大学 Target detection algorithm in combination of statistical matrix model and adaptive threshold
CN105844671A (en) * 2016-04-12 2016-08-10 河北大学 Rapid background subtraction method under changing illumination conditions

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110580706A (en) * 2018-06-11 2019-12-17 北京中科晶上超媒体信息技术有限公司 Method and device for extracting video background model

Similar Documents

Publication Publication Date Title
CN105608456A (en) Multi-directional text detection method based on full convolution network
EP2458335A3 (en) Road estimation device and method for estimating road
CN105303581B (en) A kind of moving target detecting method of auto-adaptive parameter
CN110956681B (en) Portrait background automatic replacement method combining convolution network and neighborhood similarity
CN109087309A (en) A kind of image partition method of amalgamation of global and local information level collection
EP2458332A3 (en) Road estimation device and method for estimating road
CN104680521B (en) A kind of improved background modeling and foreground detection method
CN104123734A (en) Visible light and infrared detection result integration based moving target detection method
EP2458331A3 (en) Road estimation device and method for estimating road
CN110363769B (en) Image segmentation method for cantilever system of high-speed rail contact net supporting device
CN106204617A (en) Adapting to image binarization method based on residual image rectangular histogram cyclic shift
CN106447673A (en) Chip pin extraction method under non-uniform illumination condition
CN106650638A (en) Abandoned object detection method
CN104504722A (en) Method for correcting image colors through gray points
CN115482165A (en) Image defogging method based on dark channel prior
CN104156959A (en) Video matting method and device
TW201025198A (en) Method for tracking and processing image
CN104091307A (en) Frog day image rapid restoration method based on feedback mean value filtering
CN103295205B (en) A kind of low-light-level image quick enhancement method based on Retinex and device
CN107944437A (en) A kind of Face detection method based on neutral net and integral image
CN104361566B (en) A kind of image processing method in optimization dark portion region
CN110751606A (en) Foam image processing method and system based on neural network algorithm
TW201032180A (en) Method and device for keeping image background by multiple gauss models
CN107798689A (en) Traffic video image background extracting method
CN104978734B (en) The extracting method and device of foreground image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20180313

WD01 Invention patent application deemed withdrawn after publication