CN109377515A - A kind of moving target detecting method and system based on improvement ViBe algorithm - Google Patents
A kind of moving target detecting method and system based on improvement ViBe algorithm Download PDFInfo
- Publication number
- CN109377515A CN109377515A CN201810879170.XA CN201810879170A CN109377515A CN 109377515 A CN109377515 A CN 109377515A CN 201810879170 A CN201810879170 A CN 201810879170A CN 109377515 A CN109377515 A CN 109377515A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- vibe algorithm
- foreground image
- pixel 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000001514 detection method Methods 0.000 claims abstract description 41
- 238000002156 mixing Methods 0.000 claims abstract description 26
- 230000004927 fusion Effects 0.000 claims abstract description 10
- 238000003708 edge detection Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 16
- 230000007797 corrosion Effects 0.000 claims description 13
- 238000005260 corrosion Methods 0.000 claims description 13
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000010339 dilation Effects 0.000 description 3
- 230000003628 erosive effect Effects 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/254—Analysis of motion involving subtraction of images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of based on the moving target detecting method and system that improve ViBe algorithm, wherein method from inputted video image the following steps are included: obtain the first sport foreground image using the ViBe algorithm of three frame difference methods of fusion；Edge extracting is carried out to inputted video image using the ViBe algorithm of combination of edge detection, and obtains the second sport foreground image；First sport foreground image, the second sport foreground image and preset blank image are subjected to fusion treatment, and obtain prospect blending image；After prospect blending image is carried out image restoration, the prospect blending image after restoring is exported, to complete moving object detection.The present invention is combined using three frame difference algorithms and ViBe algorithm, influence of the ViBe algorithm bring " ghost " to moving object detection precision is reduced, accuracy in detection is improved, meets the detection of moving target, monitoring cost is reduced, can be widely applied to Video Image process field.
Description
Technical field
The present invention relates to Video Image process field more particularly to a kind of movement mesh based on improvement ViBe algorithm
Mark detection method and system.
Background technique
Requirement of the society to safety now is higher and higher, video monitoring using more and more extensive.For security protection, need
It wants biggish human resources to check video for a long time, and is judged according to video, and the content that video is shown is largely safety
, situation of not causing danger, therefore cause the waste of human resources.So the detection of moving target be it is most important,
Because this can be sounded an alarm in the case where there are doubtful situations, people's real time inspection is allowed, reduce the waste of resource.
And ViBe is moving object detection algorithm more popular in recent years, which can meet instantly in real-time
Requirement to moving object detection, but there is also more obvious disadvantages.Algorithm detection moving target can have ghost, this meeting
The phenomenon that generating wrong report, this reduces the accuracy of moving object detection.
In order to solve the problems, such as ghost, current some innovatory algorithms such as: the foreground statistical method based on Pixel-level, to same
One pixel counts, if continuous occur being greater than the threshold number of setting, is judged as ghost, removes it, but this method
True foreground pixel point can be also removed.In addition the background model such as based on colouring information and the background mould based on texture information
The improved methods computation complexity such as type is higher, sacrifices real-time.These improved methods all miss to adapt to current demand.
Summary of the invention
In order to solve the above-mentioned technical problem, the object of the present invention is to provide a kind of inspections for improving moving object detection accuracy
Survey method.
It is a further object of the present invention to provide a kind of detection systems for improving moving object detection accuracy.
The technical scheme adopted by the invention is that:
A kind of moving target detecting method based on improvement ViBe algorithm, comprising the following steps:
S1, the first sport foreground image is obtained from inputted video image using the ViBe algorithm of three frame difference methods of fusion；
S2, edge extracting is carried out to inputted video image using the ViBe algorithm of combination of edge detection, and obtains the second fortune
Dynamic foreground image；
S3, the first sport foreground image, the second sport foreground image and preset blank image are subjected to fusion treatment, and
Acquisition prospect blending image；
S4, after prospect blending image is carried out image restoration, the prospect blending image after restoring is exported, to complete to move
Target detection.
Further, the step S1, specifically includes the following steps:
S11, after inputted video image is carried out gray processing processing, inputted video image is transported using three frame difference algorithms
Moving-target detection, and obtain the first foreground image；
S12, using ViBe algorithm to inputted video image carry out moving object detection after, to inputted video image carry out in
Value filtering, and obtain the second foreground image；
S13, the first foreground image and the second foreground image are subjected to point multiplication operation, to obtain the first motion foreground picture
Picture.
Further, the step S13, specifically:
Obtain the pixel value of the first foreground image and the second foreground image respectively, and by the corresponding position pixel of two images
Pixel value carry out pixel value as the first sport foreground image corresponding position after point multiplication operation.
Further, the edge detection be with Canny operator carry out edge extracting, the step S2, specifically include with
Lower step:
S21, edge extracting is carried out to inputted video image using Canny operator, to obtain edge contour image；
S22, the pixel value for obtaining edge contour image and the second foreground image respectively, and by the corresponding position picture of two images
Pixel value after the pixel value progress point multiplication operation of vegetarian refreshments as bianry image corresponding position；
S23, preset expansion structure element and corrosion structure element are obtained, and using expansion structure element to bianry image
After being expanded, then using corrosion structure element to the image progress dilation erosion processing after expansion；
S24, it after the image after corrosion treatment is carried out holes filling processing, is obtained on the image according to preset mode more
A connected region, and to connected region carry out assignment after, obtain the second sport foreground image.
Further, the preset mode be on the pixel of image by 8 neighborhoods detection in the way of.
Further, the connected region includes multiple pixels, and the pixel of the same connected region assigns identical picture
Element is worth, and the pixel between different connected regions assigns different pixel values.
Further, the step S3, specifically includes the following steps:
S31, the pixel value for obtaining the first sport foreground image and the second sport foreground image respectively, and by the phase of two images
Pixel value after answering the pixel value of position pixel to carry out point multiplication operation as third sport foreground image corresponding position；
S32, acquisition and the identical blank image of inputted video image size, and all pixels point in blank image is set
Pixel value is 0；
After pixel value in S33, acquisition third sport foreground image, connected region is obtained on third sport foreground image
Domain, and the connected region that will acquire is extracted to the respective pixel position of blank image, to obtain prospect blending image.
Further, in the S33 the step of obtaining connected region on third sport foreground image, specifically:
Connected region is obtained on third sport foreground image according to the pixel value in the second motion foreground picture.
Further, the step S4, specifically includes the following steps:
S41, after obtaining the non-zero pixel of pixel value in prospect blending image, the pixel value of the pixel of acquisition is set
It is set to 255；
S42, figure progress median filter process is merged to prospect, to realize image restoration, the prospect after output is restored is melted
Image is closed, moving object detection is completed.
It is of the present invention another solution is that
A kind of moving object detection system based on improvement ViBe algorithm, including
Memory, for storing at least one program；
Processor executes described above a kind of based on improving ViBe algorithm for loading at least one described program
Moving target method.
The beneficial effects of the present invention are: the present invention is combined using three frame difference algorithms and ViBe algorithm, ViBe algorithm is reduced
Influence of the bring " ghost " to moving object detection precision, improves accuracy in detection, meets the detection of moving target, reduces
Monitoring cost.
Detailed description of the invention
Fig. 1 is a kind of step flow chart based on the moving target detecting method for improving ViBe algorithm of the present invention；
Fig. 2 is the schematic diagram that connected region is obtained in bianry image；
Fig. 3 is the number schematic diagram in bianry image in same connected region；
Fig. 4 is the number schematic diagram in bianry image between different connected regions.
Specific embodiment
As shown in Figure 1, a kind of based on the moving target detecting method for improving ViBe algorithm, comprising the following steps:
A1, the first sport foreground image is obtained from inputted video image using the ViBe algorithm of three frame difference methods of fusion.
Wherein, step A1 includes A11~A13:
A11, after inputted video image is carried out gray processing processing, inputted video image is transported using three frame difference algorithms
Moving-target detection, and obtain the first foreground image.
A12, using ViBe algorithm to inputted video image carry out moving object detection after, to inputted video image carry out in
Value filtering, and obtain the second foreground image.
A13, the first foreground image and the second foreground image are subjected to point multiplication operation, to obtain the first motion foreground picture
Picture.The step specifically: obtain the pixel value of the first foreground image and the second foreground image respectively, and by the corresponding positions of two images
The pixel value for setting pixel carries out pixel value as the first sport foreground image corresponding position after point multiplication operation.
A2, edge extracting is carried out to inputted video image using the ViBe algorithm of combination of edge detection, and obtains the second fortune
Dynamic foreground image.
Wherein, step A2 includes A21~A24:
A21, edge extracting is carried out to inputted video image using Canny operator, to obtain edge contour image.
A22, the pixel value for obtaining edge contour image and the second foreground image respectively, and by the corresponding position picture of two images
Pixel value after the pixel value progress point multiplication operation of vegetarian refreshments as bianry image corresponding position.
A23, preset expansion structure element and corrosion structure element are obtained, and using expansion structure element to bianry image
After being expanded, then using corrosion structure element to the image progress dilation erosion processing after expansion.
A24, it after the image after corrosion treatment is carried out holes filling processing, is obtained on the image according to preset mode more
A connected region, and to connected region carry out assignment after, obtain the second sport foreground image.The connected region includes multiple pictures
Vegetarian refreshments, the pixel of the same connected region assign identical pixel value, and the pixel between different connected regions assigns different
Pixel value.The preset mode be on the pixel of image by 8 neighborhoods detection in the way of.
A3, the first sport foreground image, the second sport foreground image and preset blank image are subjected to fusion treatment, and
Acquisition prospect blending image.
Wherein, step A3 includes step A31~A33:
A31, the pixel value for obtaining the first sport foreground image and the second sport foreground image respectively, and by the phase of two images
Pixel value after answering the pixel value of position pixel to carry out point multiplication operation as third sport foreground image corresponding position.
A32, acquisition and the identical blank image of inputted video image size, and all pixels point in blank image is set
Pixel value is 0.
After pixel value in A33, acquisition third sport foreground image, connected region is obtained on third sport foreground image
Domain, and the connected region that will acquire is extracted to the respective pixel position of blank image, to obtain prospect blending image.Its
In, on third sport foreground image obtain connected region the step of, specifically: according to the pixel value in the second motion foreground picture
Connected region is obtained on third sport foreground image.
A4, after prospect blending image is carried out image restoration, the prospect blending image after restoring is exported, to complete to move
Target detection.
Wherein, A4 includes step A41~A42:
A41, after obtaining the non-zero pixel of pixel value in prospect blending image, the pixel value of the pixel of acquisition is set
It is set to 255.
A42, figure progress median filter process is merged to prospect, to realize image restoration, the prospect after output is restored is melted
Image is closed, moving object detection is completed.
It in the above method, is combined by using three frame difference algorithms and ViBe algorithm, reduces ViBe algorithm bring " ghost
Influence of the shadow " to moving object detection precision, the accuracy of the detection of raising.By merge Canny operator edge detection and
The lookup of connected region, can be to three frame difference algorithms and ViBe algorithm process bring profile is imperfect makes up, and improve fortune
Moving-target detection performance.
Embodiment two
Detailed explanation is carried out to implementation two below with reference to Fig. 2 to Fig. 4.
A kind of moving target detecting method based on improvement ViBe algorithm, comprising the following steps:
B1, the first sport foreground image is obtained from inputted video image using the ViBe algorithm of three frame difference methods of fusion.
The inputted video image can be the image of any video capture device acquisition, in the present embodiment, use
The PetsD1TeC1 video of IBM.Step B1 is specifically included: after inputted video image is carried out gray processing processing, being regarded according to input
Corresponding threshold value is arranged in frequency situation, carries out moving object detection using three frame difference algorithms, obtains the first foreground image.It utilizes
ViBe algorithm carries out moving object detection to inputted video image, and carries out median filtering to image, to obtain the second prospect
Image.First foreground image and the second foreground image are subjected to point multiplication operation, to obtain the first sport foreground image.
B2, edge extracting is carried out to inputted video image using the ViBe algorithm of combination of edge detection, to obtain second
Sport foreground image.
Step B2 has mainly used the ViBe algorithm of edge detection, which specifically includes: using Canny operator to current
Input picture carries out edge extracting, obtains edge contour image.The picture of edge contour image and the second foreground image is obtained respectively
Element value, and the pixel value of the corresponding position pixel of two images is subjected to picture as bianry image corresponding position after point multiplication operation
Element value.Preset expansion structure element and corrosion structure element are obtained, and swollen to bianry image progress using expansion structure element
After swollen, then using corrosion structure element to the image progress dilation erosion processing after expansion, then progress holes filling processing.It will
Obtained image obtains multiple connected regions according to preset mode, and carries out assignment to connected region, in same connected region
Pixel assign identical pixel value, the pixel of different connected regions assigns different pixel values, before obtaining the second movement
Scape image.It is described to search multiple connected regions specially according to 8 neighborhoods lookup connected region according to preset mode；Search connection
Region can also select 8 neighborhoods, can provide the efficiency of lookup according to 8 neighborhoods according to 16 neighborhoods, in the present embodiment.
As shown in Fig. 2, Fig. 2 is the bianry image after expansion and corrosion treatment, there are multiple pixels in image, obtains
Then first pixel searches connected region according to 8 neighborhood modes as label, the connected region of mark point is all tracked
It arrives, is then return to mark point, until the bottom right pixel point for inquiring image finally obtains three connected regions in the picture.
As shown in figure 3, in same connected region so pixel assigns same pixel value, in the present embodiment, to first tax
The pixel of the connected region of value is assigned a value of 1.As shown in figure 4, the pixel value that the pixel of different connected regions assigns is different,
In the present embodiment, 2 are assigned a value of to the pixel of the connected region of second assignment, to the picture of the connected region of third assignment
Vegetarian refreshments is assigned a value of 3.
B3, the first sport foreground image, the second sport foreground image and preset blank image are subjected to fusion treatment, and
Acquisition prospect blending image.
Step B3 is specifically included: the pixel value of the first sport foreground image and the second sport foreground image is obtained respectively, and
The pixel value of the corresponding position pixel of two images is carried out after point multiplication operation as third sport foreground image corresponding position
Pixel value.Blank image identical with inputted video image size is obtained, and the pixel of all pixels point of blank image is set
Value is 0.After obtaining the pixel value in third sport foreground image, connected region is obtained on third sport foreground image, and will
The connected region got is extracted to the respective pixel position of blank image, to obtain prospect blending image.Wherein,
The step of connected region is obtained on three sport foreground images specifically: transported according to the pixel value in the second motion foreground picture in third
Connected region is obtained on dynamic foreground image.
The pixel that connected region and pixel value in the second motion foreground picture including assignment are 0, before the first movement
After scape image and the second sport foreground image carry out point multiplication operation, third sport foreground image is obtained, in third motion foreground picture
There are multiple connected regions as in, the pixel value of these connected regions may be identical possible different, obtain pixel value and the second fortune
The identical connected region of pixel value on dynamic foreground picture.Such as in the second motion foreground picture by three connected regions, three connections
The pixel value in region is respectively 1,2 and 3, obtains the connected region that pixel value is 1,2 or 3 on third sport foreground image respectively
Domain, and the connected region of acquisition is extracted to the respective pixel position of blank image, obtain prospect blending image.
B4, after prospect blending image is carried out image restoration, the prospect blending image after restoring is exported, to complete to move
Target detection.
Step B4 is specifically included: after obtaining the non-zero pixel of pixel value in prospect blending image, by the pixel of acquisition
Pixel value be set as 255.Figure is merged to prospect and carries out median filter process, so that image restoration is realized, after output is restored
Prospect blending image completes moving object detection.It can detect that moving target by the prospect blending image after restoring.
By analysis of experiments, slowly occurs the view of moving object below for input first frame for no moving target
Frequently, the present invention is based on improving ViBe algorithm and general ViBe algorithm can timely detect slow moving object, and GMM
Algorithm fails to detect in time, needs just detect object by one section of move distance.And when detecting moving object, it is based on
Ghost can be inhibited than general ViBe algorithm faster by improving ViBe algorithm.So occurring slowly movement in processing video
When object, the present invention is based on improving, ViBe algorithm is more superior than general ViBe algorithm and GMM algorithm.
For input first frame, there are the videos of moving target, and the present invention is based on improvement ViBe algorithms and GMM algorithm can
It detects sport foreground image in time, ghost is generated there are moving target due to first frame image in general ViBe algorithm
Image, and the ghost images not can be removed also in video playing to 100 frame.So the present invention is based on improve ViBe algorithm and
GMM algorithm can detect that sport foreground image can inhibit ghost in real time in time.
Generally speaking, the present invention is based on improvement ViBe algorithms is better than GMM algorithm based on the test result of slow mobile target,
Based on initial frame, there are the test results of moving target can effectively inhibit ghost than general ViBe algorithm.So base of the present invention
It is more accurate in the test result for the moving target for improving ViBe algorithm.
Embodiment three
A kind of moving object detection system based on improvement ViBe algorithm, comprising:
Memory, for storing at least one program；
Processor executes above-mentioned a kind of based on the movement mesh for improving ViBe algorithm for loading at least one described program
Mark method.
It is to be illustrated to preferable implementation of the invention, but the invention is not limited to the implementation above
Example, those skilled in the art can also make various equivalent variations on the premise of without prejudice to spirit of the invention or replace
It changes, these equivalent deformations or replacement are all included in the scope defined by the claims of the present application.
Claims (10)
1. a kind of based on the moving target detecting method for improving ViBe algorithm, which comprises the following steps:
S1, the first sport foreground image is obtained from inputted video image using the ViBe algorithm of three frame difference methods of fusion；
S2, edge extracting is carried out to inputted video image using the ViBe algorithm of combination of edge detection, and before the second movement of acquisition
Scape image；
S3, the first sport foreground image, the second sport foreground image and preset blank image are subjected to fusion treatment, and obtained
Prospect blending image；
S4, after prospect blending image is carried out image restoration, the prospect blending image after restoring is exported, to complete moving target
Detection.
2. according to claim 1 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
Step S1 is stated, specifically includes the following steps:
S11, after inputted video image is carried out gray processing processing, movement mesh is carried out to inputted video image using three frame difference algorithms
Mark detection, and obtain the first foreground image；
S12, after carrying out moving object detection to inputted video image using ViBe algorithm, intermediate value filter is carried out to inputted video image
Wave, and obtain the second foreground image；
S13, the first foreground image and the second foreground image are subjected to point multiplication operation, to obtain the first sport foreground image.
3. according to claim 2 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
Step S13 is stated, specifically:
Obtain the pixel value of the first foreground image and the second foreground image respectively, and by the picture of the corresponding position pixel of two images
Pixel value after element value progress point multiplication operation as the first sport foreground image corresponding position.
4. according to claim 3 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
Stating edge detection is that edge extracting is carried out with Canny operator, the step S2, specifically includes the following steps:
S21, edge extracting is carried out to inputted video image using Canny operator, to obtain edge contour image；
S22, the pixel value for obtaining edge contour image and the second foreground image respectively, and by the corresponding position pixel of two images
Pixel value carry out with operation after pixel value as bianry image corresponding position；
S23, preset expansion structure element and corrosion structure element are obtained, and bianry image is carried out using expansion structure element
After expansion, then using corrosion structure element to the image progress corrosion treatment after expansion；
S24, after the image after corrosion treatment is carried out holes filling processing, multiple companies are obtained on the image according to preset mode
Logical region, and after carrying out assignment to connected region, obtain the second sport foreground image.
5. according to claim 4 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
State preset mode be on the pixel of image by 8 neighborhoods detection in the way of.
6. according to claim 4 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
Stating connected region includes multiple pixels, and the pixel of the same connected region assigns identical pixel value, different connected regions
Between pixel assign different pixel values.
7. according to claim 6 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
Step S3 is stated, specifically includes the following steps:
S31, the pixel value for obtaining the first sport foreground image and the second sport foreground image respectively, and by the corresponding positions of two images
The pixel value for setting pixel carries out pixel value as third sport foreground image corresponding position after point multiplication operation；
S32, acquisition and the identical blank image of inputted video image size, and the pixel of all pixels point in blank image is set
Value is 0；
After pixel value in S33, acquisition third sport foreground image, connected region is obtained on third sport foreground image, and
The connected region that will acquire is extracted to the respective pixel position of blank image, to obtain prospect blending image.
8. according to claim 7 a kind of based on the moving target detecting method for improving ViBe algorithm, the in the S33
The step of connected region is obtained on three sport foreground images, specifically:
Connected region is obtained on third sport foreground image according to the pixel value in the second motion foreground picture.
9. according to claim 8 a kind of based on the moving target detecting method for improving ViBe algorithm, which is characterized in that institute
Step S4 is stated, specifically includes the following steps:
S41, it after obtaining the non-zero pixel of pixel value in prospect blending image, sets the pixel value of the pixel of acquisition to
255；
S42, figure progress median filter process is merged to prospect, to realize image restoration, the prospect after output is restored merges figure
Picture completes moving object detection.
10. a kind of based on the moving object detection system for improving ViBe algorithm characterized by comprising
Memory, for storing at least one program；
Processor requires any one of 1-9 described a kind of based on improvement for loading at least one described program with perform claim
The moving target method of ViBe algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810879170.XA CN109377515A (en) | 2018-08-03 | 2018-08-03 | A kind of moving target detecting method and system based on improvement ViBe algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810879170.XA CN109377515A (en) | 2018-08-03 | 2018-08-03 | A kind of moving target detecting method and system based on improvement ViBe algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109377515A true CN109377515A (en) | 2019-02-22 |
Family
ID=65404410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810879170.XA Pending CN109377515A (en) | 2018-08-03 | 2018-08-03 | A kind of moving target detecting method and system based on improvement ViBe algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109377515A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765964A (en) * | 2019-10-30 | 2020-02-07 | 常熟理工学院 | Method for detecting abnormal behaviors in elevator car based on computer vision |
| CN110781721A (en) * | 2019-09-05 | 2020-02-11 | 温州视鑫科技有限公司 | Outdoor scene moving object detection method based on improved VIBE algorithm |
| CN111476747A (en) * | 2020-03-31 | 2020-07-31 | 深圳奥比中光科技有限公司 | System and method for eliminating ghost image in under-screen imaging |
| CN111950523A (en) * | 2020-08-28 | 2020-11-17 | 珠海大横琴科技发展有限公司 | Ship detection optimization method and device based on aerial photography, electronic equipment and medium |
| CN112184773A (en) * | 2020-09-30 | 2021-01-05 | 华中科技大学 | Helmet wearing detection method and system based on deep learning |
| CN115022535A (en) * | 2022-05-20 | 2022-09-06 | 深圳福鸽科技有限公司 | Image processing method and device and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103824070A (en) * | 2014-03-24 | 2014-05-28 | 重庆邮电大学 | Rapid pedestrian detection method based on computer vision |
| KR101527962B1 (en) * | 2014-03-27 | 2015-06-11 | 전남대학교산학협력단 | method of detecting foreground in video |
| CN106846359A (en) * | 2017-01-17 | 2017-06-13 | 湖南优象科技有限公司 | Moving target method for quick based on video sequence |
| 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 |
| CN107833242A (en) * | 2017-10-30 | 2018-03-23 | 南京理工大学 | One kind is based on marginal information and improves VIBE moving target detecting methods |
-
2018
- 2018-08-03 CN CN201810879170.XA patent/CN109377515A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103824070A (en) * | 2014-03-24 | 2014-05-28 | 重庆邮电大学 | Rapid pedestrian detection method based on computer vision |
| KR101527962B1 (en) * | 2014-03-27 | 2015-06-11 | 전남대학교산학협력단 | method of detecting foreground in video |
| CN106846359A (en) * | 2017-01-17 | 2017-06-13 | 湖南优象科技有限公司 | Moving target method for quick based on video sequence |
| 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 |
| CN107833242A (en) * | 2017-10-30 | 2018-03-23 | 南京理工大学 | One kind is based on marginal information and improves VIBE moving target detecting methods |
Non-Patent Citations (5)
| Title |
|---|
| 丁莹 等: "《复杂环境运动目标检测技术及应用》", 31 January 2014, 国防工业出版社 * |
| 张杜等: "基于ViBe算法及Canny边缘检测的运动目标提取", 《微型机与应用》 * |
| 殷闻强 等: "基于Canny与Frame-ViBE的运动目标检测方法", 《软件导刊》 * |
| 王丹等: "一种基于梯度和Vibe算法的运动检测算法", 《电视技术-视频应用与工程》 * |
| 赵健: "基于三帧差法的运动目标检测方法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110781721A (en) * | 2019-09-05 | 2020-02-11 | 温州视鑫科技有限公司 | Outdoor scene moving object detection method based on improved VIBE algorithm |
| CN110781721B (en) * | 2019-09-05 | 2024-04-30 | 温州视鑫科技有限公司 | Outdoor scene moving object detection method based on improved VIBE algorithm |
| CN110765964A (en) * | 2019-10-30 | 2020-02-07 | 常熟理工学院 | Method for detecting abnormal behaviors in elevator car based on computer vision |
| CN110765964B (en) * | 2019-10-30 | 2022-07-15 | 常熟理工学院 | Method for detecting abnormal behaviors in elevator car based on computer vision |
| CN111476747A (en) * | 2020-03-31 | 2020-07-31 | 深圳奥比中光科技有限公司 | System and method for eliminating ghost image in under-screen imaging |
| CN111950523A (en) * | 2020-08-28 | 2020-11-17 | 珠海大横琴科技发展有限公司 | Ship detection optimization method and device based on aerial photography, electronic equipment and medium |
| CN112184773A (en) * | 2020-09-30 | 2021-01-05 | 华中科技大学 | Helmet wearing detection method and system based on deep learning |
| CN115022535A (en) * | 2022-05-20 | 2022-09-06 | 深圳福鸽科技有限公司 | Image processing method and device and electronic equipment |
| CN115022535B (en) * | 2022-05-20 | 2024-03-08 | 深圳福鸽科技有限公司 | Image processing method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109377515A (en) | A kind of moving target detecting method and system based on improvement ViBe algorithm | |
| Rosin et al. | Image difference threshold strategies and shadow detection. | |
| KR101722803B1 (en) | Method, computer program, and device for hybrid tracking of real-time representations of objects in image sequence | |
| US11156564B2 (en) | Dirt detection on screen | |
| WO2012096768A2 (en) | Method of detecting logos, titles, or sub-titles in video frames | |
| EP2680226A1 (en) | Temporally consistent superpixels | |
| CN109035287B (en) | Foreground image extraction method and device and moving vehicle identification method and device | |
| US20070025592A1 (en) | Target-region detection apparatus, method and program | |
| CN109961016B (en) | Multi-gesture accurate segmentation method for smart home scene | |
| KR20130072073A (en) | Apparatus and method for extracting edge in image | |
| CN111260675B (en) | High-precision extraction method and system for image real boundary | |
| CN113688820A (en) | Stroboscopic stripe information identification method and device and electronic equipment | |
| CN115471476A (en) | Method, device, equipment and medium for detecting component defects | |
| US20220230332A1 (en) | Systems and methods for detecting motion during 3d data reconstruction | |
| CN114674826A (en) | Visual detection method and detection system based on cloth | |
| Widyawan et al. | Adaptive motion detection algorithm using frame differences and dynamic template matching method | |
| CN116721145B (en) | Liquid medium leakage detection method and device, electronic equipment and storage medium | |
| JP2005165387A (en) | Method and device for detecting stripe defective of picture and display device | |
| CN108769521B (en) | Photographing method, mobile terminal and computer readable storage medium | |
| CN114926508B (en) | Visual field boundary determining method, device, equipment and storage medium | |
| CN110349178A (en) | A kind of human body unusual checking and identifying system and method | |
| Turakhia et al. | Automatic crack detection in heritage site images for image inpainting | |
| Padalkar et al. | Svd based automatic detection of target regions for image inpainting | |
| CN108470326B (en) | Image completion method and device | |
| CN113298904A (en) | Monocular vision-based positioning and map construction method |
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: 20190222 |