CN104751487A - Method for detecting movement target based on colored RGB three-pane color-change frame difference - Google Patents
Method for detecting movement target based on colored RGB three-pane color-change frame difference Download PDFInfo
- Publication number
- CN104751487A CN104751487A CN201510134601.6A CN201510134601A CN104751487A CN 104751487 A CN104751487 A CN 104751487A CN 201510134601 A CN201510134601 A CN 201510134601A CN 104751487 A CN104751487 A CN 104751487A
- Authority
- CN
- China
- Prior art keywords
- plane
- frame
- pixel
- color
- 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
Links
Landscapes
- Processing Of Color Television Signals (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method for detecting a movement target based on colored RGB three-pane color-change frame difference. The method comprises the steps of acquiring colored images of two adjacent frames in the scene to be measured; extracting red, green and blue planes from the color images and storing; performing differential operation for the red, green and blue planes by the frame difference method according to the extracted six planes so as to generate three differential images; sequentially calculating for pixel points of the three differential images; performing binarization for each pixel on the obtained new differential image so as to obtain the contour of the movement target. According to the method, the frame differential calculation is respectively performed for three RGB planes; whether the color difference changes of the pixel points at the same coordinate point are the same; the color difference changes are overlapped and calculated; therefore, a large difference occurs between the movement target contour pixel value and the pixel value of other areas, and as a result, the movement target can be accurately extracted, and the complete information of the movement target contour can be maintained to the maximum.
Description
Technical field
The present invention relates to a kind of detection method of moving target, especially a kind of moving target detecting method becoming frame difference based on colored RGB tri-film color.
Background technology
Video technique has vast potential for future development in scientific research and engineer applied field.In Computer Vision process, the detection of moving target and extraction are gordian techniquies.The object of moving object detection and extraction is in order to moving target is separated from background image, is the segmentation problem of moving target and background.
The top priority of Computer Vision is the target detecting motion from sequence of video images, and how being separated with target background is fast and effectively the focus and emphasis of current research.Detection algorithm conventional at present has optical flow method, Background difference and frame-to-frame differences method.Wherein, optical flow method operand is comparatively large, is not suitable for real-time process; Background difference needs modeling, and comparatively responsive to dynamic scene change, and incorrect setting of parameter (such as learning rate) directly will affect whole structure; Frame-to-frame differences method is a kind of method of carrying out difference in time series, and owing to not needing modeling, therefore real-time is best compared with additive method.The paper " Moving target classification and tracking from real time video " that such as A.Lipton delivers has carried out detailed elaboration to gray scale plane frame difference; Patent documentation " a kind of recognition methods of moving target, device " (CN103826102A), based on frame difference method, combines with Background difference, by present frame modeling and and next frame image carry out difference operation and determine motion target area.But, the frame difference method used in above-mentioned document is only carried out frame difference to gray scale plane and is calculated, due to single plane each pixel span little and be subject to the brightness of external environment condition light change interference, movement destination image is made to be difficult to accurately distinguish with background image, easily cause the loss of effective information, add the difficulty of complete extraction moving target profile, have impact on the accuracy of later stage Computer Vision.
Moving Object in Video Sequences detection is the difficult point in field of video image processing, and in intelligent video monitoring application, computing machine needs to carry out fast and accurate process gathered image, and this process proposes requirements at the higher level to software algorithm.Especially little to guarded region, dark or brightness change video monitoring scene frequently, and the accuracy of existing algorithm is just difficult to meet actual requirement.
Summary of the invention
For above-mentioned prior art Problems existing, the invention provides a kind of moving target detecting method becoming frame difference based on colored RGB tri-film color.
The technical scheme realizing above-mentioned moving target detecting method is as follows.
Become a moving target detecting method for frame difference based on colored RGB tri-film color, it is realized by following steps:
(1) the is obtained according to the video content under scene to be measured
k-1frame and
kframe scene coloured image,
kfor being greater than the integer of 1;
(2) step (1) is obtained
k-1color image frame, extracts the red plane of this coloured image
f k-1 (
x, y, r), green color plane
f k-1 (
x, y, g) and blue color planes
f k-1 (
x, y, b), be stored in memory body,
x,yrepresent pixel coordinate
, r, g, brepresent the red plane of coloured image, green color plane and blue color planes respectively, subscript
k-1represent video frame number;
(3) step (1) is obtained
kcolor image frame, extracts the red plane of this coloured image
f k (
x, y, r), green color plane
f k (
x, y, g) and blue color planes
f k (
x, y, b), be stored in memory body, subscript
krepresent video frame number;
(4) according to six planes extracted in step (2) and step (3), use frame difference method by the
kframe and
k-1the red plane of two field picture, green color plane and blue color planes carry out difference operation respectively, generate the three width error images corresponding to RGB tri-plane, are respectively
d k (
x, y, r),
d k (
x, y, g),
d k (
x, y, b); Operation expression is:
(5) to meeting
d k (
x, y, r),
d k (
x, y, g),
d k (
x, y, b) be greater than simultaneously 0 or be less than simultaneously 0 pixel calculate
d k (
x,y);
c k (
x,y) operation expression be:
In formula, | | represent and take absolute value, Min{ represent the minimum value in each entry value in { } bracket;
(6) according to whether meeting
d k (
x, y, r),
d k (
x, y, g),
d k (
x,y,
b) be greater than simultaneously 0 or be less than simultaneously 0 condition, each pixel is calculated corresponding
d k (
x,y); For qualified pixel,
d k (
x,y) operation expression be:
In formula, δ be (0,3] between real number; For the pixel not meeting Rule of judgment,
d k (
x,y) operation expression be:
(7) in step (6)
d k (
x,y) value calculates new error image
e k (
x,y);
e k (
x,y) operation expression be:
In formula, floor [] function representation carries out downward rounding operation to the numerical value in bracket [];
(8) to the new error image that step (7) obtains
e k (
x,y) in each pixel do binary conversion treatment, thus obtain the profile of moving target, binary conversion treatment operation expression is:
In formula,
a k (
x,y) be image after binaryzation, T is threshold value.
Realize a kind of technical scheme becoming the moving target detecting method of frame difference based on colored RGB tri-film color that the invention described above provides, compared with prior art, the present invention calculates by carrying out frame difference respectively to RGB tri-planes, determines whether as foreground moving image with the size of same pixel chromatic aberration; Meanwhile, by whether in the same way judging same coordinate pixel chromatic aberration, eliminate the impact that external environment condition causes because of brightness change; Finally, superposition calculation is carried out to chromatic aberration, makes moving target outliner pixel values and other area pixel values form larger contrast, thus to moving target carry out complete, accurately extract, remain the complete information of moving target profile to greatest extent.
Embodiment
The specific embodiment of the present invention will be elaborated below.
Implement a kind of technical scheme becoming the moving target detecting method of frame difference based on colored RGB tri-film color that the present invention is above-mentioned provided, it is realized by following steps.
Step one, obtain the according to the video content under scene to be measured
k-1frame and
kframe scene coloured image,
kfor being greater than the integer of 1;
Step 2, step one is obtained the
k-1color image frame, extracts the red plane of this coloured image
f k-1 (
x,
y,r), green color plane
f k-1 (
x, y, g) and blue color planes
f k-1 (
x, y, b), be stored in memory body,
x,yrepresent pixel coordinate,
r, g, brepresent the red plane of coloured image, green color plane and blue color planes respectively, subscript
k-1represent video frame number;
Step 3, step one is obtained the
kcolor image frame, extracts the red plane of this coloured image
f k-1 (
x, y, r), green color plane
f k-1 (
x, y, g) and blue color planes
f k-1 (
x, y, b), be stored in memory body, subscript
krepresent video frame number;
Step 4, according to six planes extracted in step 2 and step 3, use frame difference method that the red plane of kth frame and kth-1 color image frame, green color plane and blue color planes are carried out difference operation respectively, generate the three width error images corresponding to RGB tri-plane, be respectively
d k (
x, y, r),
d k (
x, y, g),
d k (
x, y, b); Operation expression is:
f k image is coloured image,
f k (
x, y, r)
, f k (
x, y, g) and
f k (
x, y, b) be
f k rGB three planes of image, wherein the value of each pixel is the integer being greater than zero;
d k (
x, y, r)
, D k (
x, y, g),
d k (
x, y, b) representing the error image of colored RGB tri-plane respectively, the value of each pixel of error image just may be, also may be negative value.Below with the 2nd frame and the 1st two field picture (namely
kwhen=2) in get the concrete numerical evaluation of 3 coordinates (0,0) (1,1) (2,2) corresponding pixel points and be described.The R plane f that false coordinate (0,0) is corresponding
1(0,0, r)=1, f
2(0,0, r)=200, G plane f
1(0,0, g)=10, f
2(0,0, g)=160, B plane f
1(0,0, b)=8, f
2(0,0, b)=100; The R plane D that coordinate (0,0) is corresponding
2(0,0, r)=f
2(0,0, r)-f
1(0,0, r)=200-1=199, G plane D
2(0,0, g)=f
2(0,0, g)-f
1(0,0, g)=160-10=150, B plane D
2(0,0, b)=f
2(0,0, b)-f
1(0,0, b)=100-8=92.The R plane f that coordinate (1,1) is corresponding
1(1,1, r)=5, f
2(1,1, r)=0, G plane f
1(1,1, g)=20, f
2(1,1, g)=1, B plane f
1(1,1, b)=40, f
2(1,1, b)=3; The R plane D that coordinate (1,1) is corresponding
2(1,1, r)=f
2(1,1, r)-f
1(1,1, r)=0-5=-5, G plane D
2(1,1, g)=f
2(1,1, g)-f
1(1,1, g)=1-20=-19, B plane D
2(1,1, b)=f
2(1,1, b)-f
1(1,1, b)=3-40=-37.The R plane f that coordinate (2,2) is corresponding
1(2,2, r)=50, f
2(2,2, r)=0, G plane f
1(2,2, g)=5, f
2(2,2, g)=20, B plane f
1(2,2, b)=3, f
2(2,2, b)=30; The R plane D that coordinate (2,2) is corresponding
2(2,2, r)=f
2(2,2, r)-f
1(2,2, r)=0-50=-50, G plane D
2(2,2, g)=f
2(2,2, g)-f
1(2,2, g)=20-5=15, B plane D
2(2,2, b)=f
2(2,2, b)-f
1(2,2, b)=30-3=27.
Step 5, to meeting
d k (
x, y, r),
d k (
x, y, g),
d k (
x, y, b) be greater than simultaneously 0 or be less than simultaneously 0 pixel calculate
d k (
x,y);
c k (
x,y) operation expression be:
In formula, | | represent and take absolute value, Min{ represent the minimum value in each entry value in { } bracket;
For coordinate (0,0) point, D
2(0,0, r)=199, D
2(0,0, g)=150, D
2(0,0, b)=92 are all greater than 0, meet Rule of judgment, 0 < 92 < 150 < 199, therefore C
k(0,0) value is 92.For coordinate (1,1) point, D
2(1,1, r)=-5, D
2(1,1, g)=-19, D
2(1,1, b)=-37 are all less than 0, meet Rule of judgment ,-37 <-19 <-5 < 0, therefore C
k(1,1) value is 5.For coordinate (2,2) point, D
2(2,2, r)=-50, D
2(2,2, g)=15, D
2(2,2, b)=27, do not meet Rule of judgment.
Whether step 6, according to meeting
d k (
x, y, r),
d k (
x, y, g),
d k (
x,y,
b) be greater than simultaneously 0 or be less than simultaneously 0 condition, each pixel is calculated corresponding
d k (
x,y); For qualified pixel,
d k (
x,y) operation expression be:
In formula, δ be (0,3] between real number; For the pixel not meeting Rule of judgment,
d k (
x,y) operation expression be:
For the pixel meeting Rule of judgment in step 5, suppose that δ value is 2, coordinate (0,0) point, D
2(0,0)=199+150+92-2 × 92=461-184=257; Coordinate (1,1) point, D
2(1,1)=5+19+37-2 × 5=51.For the pixel not meeting Rule of judgment, D
2(2,2)=50+15+27=92.
Step 7, in step 6
d k (
x,y) value calculates new error image
e k (
x,y);
e k (
x,y) operation expression be:
In formula, floor [] function representation carries out downward rounding operation to the numerical value in bracket [];
For coordinate (0,0) point, D
2(0,0)=257 > 255, therefore E
2(0,0)=255; Coordinate (1,1) point, D
2(1,1)=51 < 255, therefore E
2(1,1)=51; Coordinate (2,2) point, D
2(2,2)=92 < 255, therefore E
2(2,2)=92.
Step 8, to the new error image that step 7 obtains
e k (
x,y) in each pixel do binary conversion treatment, thus obtain the profile of moving target, binary conversion treatment operation expression is:
In formula,
a k (
x,y) be image after binaryzation, T is threshold value.
Suppose that T is 100, for coordinate (0,0) point, E
2(0,0)=255 > 100, therefore A
2(0,0)=1; Coordinate (1,1) point, E
2(1,1)=51 < 100, therefore A
2(1,1)=0; Coordinate (2,2) point, E
2(2,2)=92 < 100, therefore A
2(2,2)=0.Through the process of abovementioned steps, moving target outliner pixel values and other area pixel values form larger contrast, threshold value T will be easy to determine, thus ensure that to moving target carry out complete, accurately extract, remain the complete information of moving target profile to greatest extent.
Claims (1)
1. become a moving target detecting method for frame difference based on colored RGB tri-film color, it is realized by following steps:
(1) the is obtained according to the video content under scene to be measured
k-1frame and
kframe scene coloured image,
kfor being greater than the integer of 1;
(2) to kth-1 color image frame that step (1) obtains, the red plane of this coloured image is extracted
f k-1 (
x, y, r), green color plane
f k-1 (
x, y, g) and blue color planes
f k-1 (
x, y, b), be stored in memory body,
x,yrepresent pixel coordinate
, r, g, brepresent the red plane of coloured image, green color plane and blue color planes respectively, subscript
k-1represent video frame number;
(3) step (1) is obtained
kcolor image frame, extracts the red plane of this coloured image
f k (
x, y, r), green color plane
f k (
x, y, g) and blue color planes
f k (
x, y, b), be stored in memory body, subscript
krepresent video frame number;
(4) according to six planes extracted in step (2) and step (3), use frame difference method by the
kframe and
k-1the red plane of two field picture, green color plane and blue color planes carry out difference operation respectively, generate the three width error images corresponding to RGB tri-plane, are respectively
d k (
x, y, r),
d k (
x, y, g),
d k (
x, y, b); Operation expression is:
(5) to meeting
d k (
x, y, r),
d k (
x, y, g),
d k (
x, y, b) be greater than simultaneously 0 or be less than simultaneously 0 pixel calculate
d k (
x,y);
c k (
x,y) operation expression be:
In formula, | | represent and take absolute value, Min{ represent the minimum value in each entry value in { } bracket;
(6) according to whether meeting
d k (
x, y, r),
d k (
x, y, g),
d k (
x,y,
b) be greater than simultaneously 0 or be less than simultaneously 0 condition, each pixel is calculated corresponding
d k (
x,y); For qualified pixel,
d k (
x,y) operation expression be:
In formula, δ be (0,3] between real number; For the pixel not meeting Rule of judgment,
d k (
x,y) operation expression be:
(7) in step (6)
d k (
x,y) value calculates new error image
e k (
x,y);
e k (
x,y) operation expression be:
In formula, floor [] function representation carries out downward rounding operation to the numerical value in bracket [];
(8) to the new error image that step (7) obtains
e k (
x,y) in each pixel do binary conversion treatment, thus obtain the profile of moving target, binary conversion treatment operation expression is:
In formula,
a k (
x,y) be image after binaryzation, T is threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510134601.6A CN104751487B (en) | 2015-03-26 | 2015-03-26 | A kind of moving target detecting method based on the plane discoloration frames of colored RGB tri- difference |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510134601.6A CN104751487B (en) | 2015-03-26 | 2015-03-26 | A kind of moving target detecting method based on the plane discoloration frames of colored RGB tri- difference |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104751487A true CN104751487A (en) | 2015-07-01 |
| CN104751487B CN104751487B (en) | 2017-06-06 |
Family
ID=53591106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510134601.6A Active CN104751487B (en) | 2015-03-26 | 2015-03-26 | A kind of moving target detecting method based on the plane discoloration frames of colored RGB tri- difference |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104751487B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205835A (en) * | 2015-09-02 | 2015-12-30 | 湖南工业大学 | Colored moving target recognizing and tracking method based on Fourier descriptor |
| CN109766831A (en) * | 2019-01-09 | 2019-05-17 | 深圳市三宝创新智能有限公司 | A kind of road colour band recognition methods, device, computer equipment and storage medium |
| CN109887045A (en) * | 2019-03-29 | 2019-06-14 | 南京维沃软件技术有限公司 | A kind of image display method and terminal device |
| CN111951302A (en) * | 2020-08-12 | 2020-11-17 | 南京师范大学 | Robot vision positioning method based on feature matching |
| US12026862B2 (en) | 2019-10-11 | 2024-07-02 | 3M Innovative Properties Company | Apparatus and methods for preprocessing images having elements of interest |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004282368A (en) * | 2003-03-14 | 2004-10-07 | Omron Corp | Mobile object monitoring device |
| CN101098462A (en) * | 2007-07-12 | 2008-01-02 | 上海交通大学 | Chroma deviation and brightness deviation combined video moving object detection method |
| CN102915543A (en) * | 2012-09-12 | 2013-02-06 | 西安电子科技大学 | Figure motion change detecting method based on extracting function and three-channel separation |
| US20130094694A1 (en) * | 2011-10-12 | 2013-04-18 | Raytheon Company | Three-frame difference moving target acquisition system and method for target track identification |
-
2015
- 2015-03-26 CN CN201510134601.6A patent/CN104751487B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004282368A (en) * | 2003-03-14 | 2004-10-07 | Omron Corp | Mobile object monitoring device |
| CN101098462A (en) * | 2007-07-12 | 2008-01-02 | 上海交通大学 | Chroma deviation and brightness deviation combined video moving object detection method |
| US20130094694A1 (en) * | 2011-10-12 | 2013-04-18 | Raytheon Company | Three-frame difference moving target acquisition system and method for target track identification |
| CN102915543A (en) * | 2012-09-12 | 2013-02-06 | 西安电子科技大学 | Figure motion change detecting method based on extracting function and three-channel separation |
Non-Patent Citations (4)
| Title |
|---|
| CINA MOTAMED: "Motion detection and tracking using belief indicators for an automatic visual-surveillance system", 《PERFORMANCE EVALUATION OF TRACKING AND SURVEILLANCE》 * |
| SJ MCKENNA ET AL.: "Tracking Groups of People", 《COMPUTER VISION & IMAGE UNDERSTANDING》 * |
| 沈盼盼 等: "基于RGB三通道分离的运动目标检测方法", 《视频应用与工程》 * |
| 苏礼坤 等: "基于帧间差分的彩色视频背景提取", 《成都信息工程学院学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205835A (en) * | 2015-09-02 | 2015-12-30 | 湖南工业大学 | Colored moving target recognizing and tracking method based on Fourier descriptor |
| CN109766831A (en) * | 2019-01-09 | 2019-05-17 | 深圳市三宝创新智能有限公司 | A kind of road colour band recognition methods, device, computer equipment and storage medium |
| CN109887045A (en) * | 2019-03-29 | 2019-06-14 | 南京维沃软件技术有限公司 | A kind of image display method and terminal device |
| CN109887045B (en) * | 2019-03-29 | 2022-02-01 | 南京维沃软件技术有限公司 | Image display method and terminal equipment |
| US12026862B2 (en) | 2019-10-11 | 2024-07-02 | 3M Innovative Properties Company | Apparatus and methods for preprocessing images having elements of interest |
| CN111951302A (en) * | 2020-08-12 | 2020-11-17 | 南京师范大学 | Robot vision positioning method based on feature matching |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104751487B (en) | 2017-06-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104392468B (en) | Based on the moving target detecting method for improving visual background extraction | |
| CN102307274B (en) | Motion detection method based on edge detection and frame difference | |
| CN102799883B (en) | Method and device for extracting movement target from video image | |
| CN107833236A (en) | Semantic vision positioning system and method are combined under a kind of dynamic environment | |
| CN110276264B (en) | Crowd density estimation method based on foreground segmentation graph | |
| CN107452015B (en) | Target tracking system with re-detection mechanism | |
| CN105488811B (en) | A kind of method for tracking target and system based on concentration gradient | |
| CN111311666A (en) | Monocular vision odometer method integrating edge features and deep learning | |
| CN104598915A (en) | Gesture recognition method and gesture recognition device | |
| CN106952294B (en) | A kind of video tracing method based on RGB-D data | |
| CN107014294A (en) | A kind of contact net geometric parameter detection method and system based on infrared image | |
| CN104751487A (en) | Method for detecting movement target based on colored RGB three-pane color-change frame difference | |
| CN105809716B (en) | Foreground extraction method integrating superpixel and three-dimensional self-organizing background subtraction method | |
| CN104063885A (en) | Improved movement target detecting and tracking method | |
| CN107146239A (en) | Satellite video moving target detecting method and system | |
| CN110298297A (en) | Flame identification method and device | |
| CN110334703B (en) | Ship detection and identification method in day and night image | |
| CN110006444B (en) | Anti-interference visual odometer construction method based on optimized Gaussian mixture model | |
| CN106204594A (en) | A kind of direction detection method of dispersivity moving object based on video image | |
| Schorghuber et al. | SLAMANTIC-leveraging semantics to improve VSLAM in dynamic environments | |
| CN104318559A (en) | Quick feature point detecting method for video image matching | |
| CN109166137A (en) | For shake Moving Object in Video Sequences detection algorithm | |
| Yu et al. | Drso-slam: A dynamic rgb-d slam algorithm for indoor dynamic scenes | |
| CN111191535A (en) | Pedestrian detection model construction method based on deep learning and pedestrian detection method | |
| CN115049954A (en) | Target identification method, device, electronic equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |