CN110288630A - A kind of moving target ghost suppressing method of background modeling - Google Patents

A kind of moving target ghost suppressing method of background modeling Download PDF

Info

Publication number
CN110288630A
CN110288630A CN201910569625.2A CN201910569625A CN110288630A CN 110288630 A CN110288630 A CN 110288630A CN 201910569625 A CN201910569625 A CN 201910569625A CN 110288630 A CN110288630 A CN 110288630A
Authority
CN
China
Prior art keywords
frame
subregion
gray
mask
image
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
Application number
CN201910569625.2A
Other languages
Chinese (zh)
Other versions
CN110288630B (en
Inventor
高飞
李云阳
葛一粟
王金超
卢书芳
张元鸣
程振波
肖刚
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.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
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 Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN201910569625.2A priority Critical patent/CN110288630B/en
Publication of CN110288630A publication Critical patent/CN110288630A/en
Application granted granted Critical
Publication of CN110288630B publication Critical patent/CN110288630B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/215—Motion-based segmentation
    • 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

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a kind of moving target ghost suppressing methods of background modeling, include the following steps: step 1: extracting the gray level image gray of ViBe algorithm segmentation bianry image mask and corresponding video frame frame and frame generated, wherein frame is RGB image, gray is grayscale image, and the size of mask, frame, gray are width × height；Step 2: segmentation frame, step 3: inhibiting the foreground point mask, the invention has the advantages that the present invention can preferably overcome the problems, such as the ghost of ViBe algorithm under the premise of background is uncomplicated in the video sequence.

Description

A kind of moving target ghost suppressing method of background modeling
Technical field
The present invention relates to video sequence automatic measurement technique fields, and in particular to a kind of moving target ghost of background modeling Suppressing method.
Background technique
With the development in the world, the available number of cameras in the whole world is sharply increased.It is this to increase a large amount of increasings for leading to data Add, it means that video data can not be stored or be handled manually.
Moving object detection is an important research direction of computer vision field, is various subsequent advanced processes, such as The basis of target classification, behavior understanding etc. suffers from fields such as security monitoring, intelligent transportation and is widely applied.And it is calculating It is a key technology that machine vision, which fields, the background modeling such as monitors with intelligent screen, is to realize moving object detection and tracking Basis.Therefore, it has great significance for the research of screen background modeling.
Background modeling is the basic work of sequence image analysis, is the hot issue of the outer scholar's research of Now Domestic.It builds The result of mould generates the subsequent processings such as motion detection, moving object classification, tracking and behavior understanding to video image important It influences.But due to the difference of actual application environment and the diversity of background, it is difficult to establish good background sample.So It is needed in practical application through different algorithms just obtain preferable achievement come optimizing design scheme.
Since the paper publishing of ViBe algorithm in 2011, ViBe algorithm is applied to the mobile target inspection of screen sequence It surveys, but ghost problem existing for the technology (mobile object being detected does not correspond to any real motion object).Ghost Problem will lead to downstream technique false judgment.
Summary of the invention
In order to overcome the problems, such as ghost existing for ViBe algorithm, the present invention provides a kind of ghost detecting method.
Technical scheme is as follows:
A kind of moving target ghost suppressing method of background modeling, includes the following steps:
Step 1: extract ViBe algorithm segmentation bianry image mask and corresponding video frame frame generated and The gray level image gray of frame, wherein frame is RGB image, and gray is grayscale image, and the size of mask, frame, gray are width×height；
Step 2: segmentation frame；
Step 3: inhibiting the foreground point mask.
A kind of moving target ghost suppressing method of background modeling, which is characterized in that step 2 specific steps It is as follows:
Step 2.1: using edge detection algorithm, detect the edge gray, obtain edge image grayEdge；
Step 2.2: frame being divided into using watershed algorithm by N number of subregion according to grayEdge, each subregion has MiIt is a Pixel, the coordinate record in each subregion is in Ai={ (rij,cij)|j∈[0,Mi- 1] }, [0, N-1] i ∈, wherein 0≤rij < height, 0≤cij< width；
A kind of moving target ghost suppressing method of background modeling, which is characterized in that step 3 specific steps It is as follows:
Step 3.1: the accounting d of the foreground pixel of subregion is calculated according to formula (1) and (2)i:
Wherein, I () indicates indicative function, if its parameter is very, to return to 1, otherwise return to 0；mask(rij,cij) table Show bianry image mask in coordinate (rij,cij) at gray value；
Step 3.2: to all subregion Ai, i=0,1 ..., N-1 be handled as follows: if diCoordinate in mask is then by < D (rij,cij) gray value be set to 0, (rij,cij)∈Ai；Wherein, D is threshold value.
The beneficial effects of the present invention are: the present invention can be under the premise of background be uncomplicated in the video sequence, preferable gram The ghost and smear problem of ViBe algorithm are taken, and the target area of significant change in video sequence or movement can be extracted Come, in addition, the present invention can split mobile object pixel grade, to provide other algorithms accurate information.
Specific embodiment
A specific embodiment of the invention is elaborated below with reference to embodiment.
Step 1: extract ViBe algorithm segmentation bianry image mask and corresponding video frame frame generated and The gray level image gray of frame, wherein frame is RGB image, and gray is grayscale image, and the size of mask, frame, gray are width×height。
Step 2: segmentation frame, the specific steps are as follows:
Step 2.1: using edge detection algorithm (as included Canny, sobel), detecting the edge gray, obtain edge image grayEdge；
Step 2.2: frame being divided into using watershed algorithm by N number of subregion according to grayEdge, each subregion has MiIt is a Pixel, the coordinate record in each subregion is in Ai={ (rij,cij)|j∈[0,Mi- 1] }, [0, N-1] i ∈, wherein 0≤rij < height, 0≤cij< width；Wherein AiIndicate i-th of subregion, AiIt is a set, includes all the points in the subregion； (rij,ciJ) it indicates in i-th of subregion, j-th of pixel.
Step 3: inhibiting the foreground point mask, the specific steps are as follows:
Step 3.1: the accounting d of the foreground pixel of subregion is calculated according to formula (1) and (2)i:
Wherein, I () indicates indicative function, if its parameter is very, to return to 1, otherwise return to 0；mask(rij,cij) table Show bianry image mask in coordinate (rij,cij) at gray value；
Step 3.2: to all subregion Ai, i=0,1 ..., N-1 be handled as follows: if diCoordinate in mask is then by < D (rij,cij) gray value be set to 0, (rij,cij)∈Ai；Wherein, D is threshold value, in this example, D 0.2.
Content described in this specification embodiment is only enumerating to the way of realization of inventive concept, protection of the invention Range should not be construed as being limited to the specific forms stated in the embodiments, and protection scope of the present invention is also and in this field skill Art personnel conceive according to the present invention it is conceivable that equivalent technologies mean.

Claims (3)

1. a kind of moving target ghost suppressing method of background modeling, which comprises the steps of:
Step 1: extracting ViBe algorithm segmentation bianry image mask's and corresponding video frame frame and frame generated Gray level image gray, wherein frame is RGB image, and gray is grayscale image, the size of mask, frame, gray be width × height；
Step 2: segmentation frame；
Step 3: inhibiting the foreground point mask.
2. a kind of moving target ghost suppressing method of background modeling according to claim 1, which is characterized in that the step Rapid 2 specific step is as follows:
Step 2.1: using edge detection algorithm, detect the edge gray, obtain edge image grayEdge；
Step 2.2: frame being divided into using watershed algorithm by N number of subregion according to grayEdge, each subregion has MiA pixel Point, the coordinate record in each subregion is in Ai={ (rij,cij)|j∈[0,Mi- 1] }, [0, N-1] i ∈, wherein 0≤rij< Height, 0≤cij< width；Wherein AiIndicate i-th of subregion, AiIt is a set, includes all the points in the subregion；(rij, cij) indicate in i-th of subregion, j-th of pixel.
3. a kind of moving target ghost suppressing method of background modeling according to claim 1, which is characterized in that the step Rapid 3 specific step is as follows:
Step 3.1: the accounting d of the foreground pixel of subregion is calculated according to formula (1) and (2)i:
Wherein, I () indicates indicative function, if its parameter is very, to return to 1, otherwise return to 0；
mask(rij,cij) indicate bianry image mask in coordinate (rij,cij) at gray value；
Step 3.2: to all subregion Ai, i=0,1 ..., N-1 be handled as follows: if diCoordinate in mask is then (r by < Dij, cij) gray value be set to 0, (rij,cij)∈Ai；Wherein, D is threshold value.
CN201910569625.2A 2019-06-27 2019-06-27 Moving target ghost suppression method for background modeling Active CN110288630B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910569625.2A CN110288630B (en) 2019-06-27 2019-06-27 Moving target ghost suppression method for background modeling

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910569625.2A CN110288630B (en) 2019-06-27 2019-06-27 Moving target ghost suppression method for background modeling

Publications (2)

Publication Number Publication Date
CN110288630A true CN110288630A (en) 2019-09-27
CN110288630B CN110288630B (en) 2021-11-09

Family

ID=68019304

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910569625.2A Active CN110288630B (en) 2019-06-27 2019-06-27 Moving target ghost suppression method for background modeling

Country Status (1)

Country Link
CN (1) CN110288630B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104463165A (en) * 2014-10-24 2015-03-25 南京邮电大学 Target detection method integrating Canny operator with Vibe algorithm
CN106022230A (en) * 2016-05-11 2016-10-12 太原理工大学 Video-based detection method for drowning event in swimming pool
CN106683062A (en) * 2017-01-10 2017-05-17 厦门大学 Method of checking the moving target on the basis of ViBe under a stationary camera
CN107085836A (en) * 2017-05-16 2017-08-22 合肥工业大学 A kind of general ghost removing method in moving object segmentation
US20180075586A1 (en) * 2016-09-09 2018-03-15 Omnivision Technologies, Inc. Ghost artifact removal system and method
CN108805897A (en) * 2018-05-22 2018-11-13 安徽大学 Improved moving target detection VIBE algorithm

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104463165A (en) * 2014-10-24 2015-03-25 南京邮电大学 Target detection method integrating Canny operator with Vibe algorithm
CN106022230A (en) * 2016-05-11 2016-10-12 太原理工大学 Video-based detection method for drowning event in swimming pool
US20180075586A1 (en) * 2016-09-09 2018-03-15 Omnivision Technologies, Inc. Ghost artifact removal system and method
CN106683062A (en) * 2017-01-10 2017-05-17 厦门大学 Method of checking the moving target on the basis of ViBe under a stationary camera
CN107085836A (en) * 2017-05-16 2017-08-22 合肥工业大学 A kind of general ghost removing method in moving object segmentation
CN108805897A (en) * 2018-05-22 2018-11-13 安徽大学 Improved moving target detection VIBE algorithm

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
E. MAHOOR: ""An improved motion detection algorithm using ViBe"", 《ICCCNT》 *
贺超宇: ""基于改进的Vibe和Canny边缘检测算法的运动目标检测"", 《数据通信》 *

Also Published As

Publication number Publication date
CN110288630B (en) 2021-11-09

Similar Documents

Publication Publication Date Title
CN103150549B (en) A kind of road tunnel fire detection method based on the early stage motion feature of smog
TWI409718B (en) Method of locating license plate of moving vehicle
CN100589561C (en) Dubious static object detecting method based on video content analysis
CN102915433B (en) Character combination-based license plate positioning and identifying method
CN104978567B (en) Vehicle checking method based on scene classification
CN105279772B (en) A kind of trackability method of discrimination of infrared sequence image
CN109919002B (en) Yellow stop line identification method and device, computer equipment and storage medium
CN111310645A (en) Overflow bin early warning method, device, equipment and storage medium for cargo accumulation amount
Torabi et al. Local self-similarity-based registration of human ROIs in pairs of stereo thermal-visible videos
CN106709530A (en) License plate recognition method based on video
CN104050684B (en) A kind of video frequency motion target sorting technique based on on-line training and system
CN111611907A (en) Image-enhanced infrared target detection method
CN105740751A (en) Object detection and identification method and system
CN110751619A (en) Insulator defect detection method
CN102610104B (en) Onboard front vehicle detection method
CN110345918A (en) Space junk detection method and system based on fixed star retrieval
CN103699876B (en) Method and device for identifying vehicle number based on linear array CCD (Charge Coupled Device) images
CN105678737A (en) Digital image corner point detection method based on Radon transform
CN103607558A (en) Video monitoring system, target matching method and apparatus thereof
CN109978916B (en) Vibe moving target detection method based on gray level image feature matching
CN117456371B (en) Group string hot spot detection method, device, equipment and medium
CN103337080A (en) Registration technology of infrared image and visible image based on Hausdorff distance in gradient direction
US9727780B2 (en) Pedestrian detecting system
CN106530292A (en) Strip steel surface defect image rapid identification method based on line scanning camera
CN110345919A (en) Space junk detection method based on three-dimensional space vector and two-dimensional plane coordinate

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant