CN1885346A - Detection method for moving target in infrared image sequence under complex background - Google Patents
Detection method for moving target in infrared image sequence under complex background Download PDFInfo
- Publication number
- CN1885346A CN1885346A CN 200610021067 CN200610021067A CN1885346A CN 1885346 A CN1885346 A CN 1885346A CN 200610021067 CN200610021067 CN 200610021067 CN 200610021067 A CN200610021067 A CN 200610021067A CN 1885346 A CN1885346 A CN 1885346A
- Authority
- CN
- China
- Prior art keywords
- target
- difference image
- image
- template
- frame
- 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
Images
Abstract
The invention relates to a method for testing the motion target in infrared image sequence at complex background, belonging to the digit image processing technique, wherein the inventive method comprises: first, calculating the differential images between last frame and present frame, present frame and next frame, in infrared image sequence; then extracting change testing template from two differential images; combining two templates (selecting the crossed region), to obtain the motion target template; at last, marking the connecting area on the motion target template, to obtain the motion target; the step that extracting change testing template comprises: evaluating the differential image noise parameter (average value and variance) to extract initial change testing template, that iterating from two ends of histogram to the middle; using small unit to corrode and expand to process closed calculation with large unit. The invention can divide target accurately, have better real-time property, and strong robustness.
Description
Technical field
Motion target detection method in a kind of infrared image sequence under complex background belongs to the digital image processing techniques field, particularly the moving Object Segmentation technology in the infrared image sequence.
Background technology
Owing to have passive work, advantages such as the strong and accurate all weather operations of antijamming capability, infrared imagery technique has been widely used in military fields such as scouting, supervision and guidance at present, civil areas such as this external aircraft navigation, highway intelligent transportation system also have wide application prospect (referring to document: Cai Yi, Tang Jinya. to some view of infrared thermal imaging technique development.Infrared technique, 2000,22 (2): 2-6).
Detection for Moving Target is the important subject of Flame Image Process and computer vision always, has a wide range of applications in a lot of fields.Moving object detection is judged exactly and is had or not moving target in the image sequence, and it is split.Moving target detecting method at present commonly used have powerful connections subtraction, time differencing method and optical flow method etc.
1. background subtraction is (referring to document: Surendra G, Osama M, Robert F K, et al.Detection andClassification of Vehicles.IEEE Trans.On intelligent transportation systems, 2002,3 (1): 37-47)
Its basic thought is that current frame image and the background image of storing in advance or obtain are in real time subtracted each other, if the pixel absolute value judges then that greater than a certain threshold value this picture element belongs to motion target area in the difference image, otherwise, just declare this picture element and belong to the background area.Background subtraction is simple to operate, and the detection position is accurate and speed is fast.But common background subtraction is very responsive to the variation of illumination conditions such as light, weather, poor anti jamming capability, and in addition, the background elimination is not suitable for camera motion or the very big situation of background grey scale change yet.
2. time differencing method is (referring to document: Nerr A, Colonnese S, Russb G, et al.Automatic Moving Objectand Background Sep aration.Signal Processing, 1998,66 (2): 219-232)
Time differencing method is the frame-to-frame differences point-score again, is to adopt between the consecutive frame in image sequence based on the time difference of pixel and thresholding to extract moving region in the image.Time differencing method has excellent adaptability for dynamic environment, and shortcoming is to be difficult for moving target is split fully.The principle that time difference and background subtract all is a Changing Area Detection, promptly by detecting region of variation and the invariant region in the sequence chart picture frame, moving object and static background is cut apart.
3. optical flow method is (referring to document: Barron J, Fleet D.Performance of optical flow techniques.Internatronal Journal of Computer Vision.199412 (1): 42-77)
Optical flow method is by the research optical flow field, obtains the sports ground that is similar to from sequence image, carries out target according to the motion feature of sports ground then and cuts apart.The major advantage of optical flow method is to detect moving target in motion cameras, to the less-restrictive of target in the motion of interframe, can handle big interframe displacement; Major defect is that most of optical flow computation methods are quite complicated, and noiseproof feature is poor, if there is not specific hardware supported, generally is difficult to be applied to the real-time operation of moving target in the sequence image.
Also have many other moving target detecting methods in addition, as based on the method for curve evolution, based on morphology methods with based on method of random field etc.Though moving target detecting method is varied, be mostly to propose to come at concrete task, do not have a kind of method in common.
Because characteristics such as infrared image have that the target-to-background contrast is low, object edge is fuzzy and noise is big, the relative visible light moving object detection of infrared moving target detection is more difficult usually.(Zhao Yingnan such as Zhao Yingnan, Yang Jingyu. the infrared vehicle based on Gabor wave filter and svm classifier device detects. computer engineering, 2005,31 (10): 191-192) extract area-of-interest by histogram analysis earlier, Threshold Segmentation obtains target then, and this target is the high gray scale part of target; (Wang Cheng, Wang Runsheng such as Wang Cheng.The target detection of infrared image under the open-air complex background.Infrared and laser engineering, 2000,29 (1): 5-8) proposed the algorithm of target detection of infrared image under a kind of open-air complex background, this algorithm comprises two treatment steps: at first, behind the scene registration, utilize the frame-to-frame differences image to extract the movable information of target, and carry out the rough detection of target in view of the above; Secondly, the correlativity of combining target motion on time and space carried out essence and detected, and the major defect of this algorithm is that the target segmentation precision is not high.
Summary of the invention
The invention provides a kind of based on the moving target detecting method in the infrared image sequence with complex background of time difference, this method have target cut apart accurately complete, real-time is good and characteristics such as strong robustness.
Content of the present invention for convenience of description, make some term definitions at this:
1. difference image: two two field picture respective pixel are subtracted each other the image that obtains, and difference image has been given prominence to the movable information in the scene.
2. motion change zone: because the zone with bigger gray scale absolute value that target travel causes comprises the background that appears, the background and the overlay region of covering, wherein the present frame target is made up of the background and the overlay region that cover, as shown in Figure 1 in the difference image.
3. change detection mask (CDM, change detection mask): the template that obtains after the difference image binary conversion treatment, pixel value is 1 region representation motion change zone on the template, is 0 region representation invariant region.
4. three-frame difference method: a kind of moving target detecting method, utilize continuous three two field pictures to calculate two width of cloth change detection masks, again they are sought common ground to obtain the exact position of intermediate frame moving target, its principle is as shown in Figure 2.
5. morphologic filtering: use the method for mathematical morphology to carry out filtering.Usually use based on morphology dilation operation and erosion operation, be combined as opening operation and closed operation and remove the noise of different binary images respectively (referring to document: work such as Paul Gonzales, Ruan Qiuqi, Ruan Yuzhi translates. Digital Image Processing (second edition). and Beijing: electronics worker publishing house, 2003).
6. erosion operation: one of the most basic morphological operator, X corrodes with B and is designated as X Θ B, be defined as E=X Θ B={ (x, y) | B
X, y X}, wherein X is an image collection, B is a structural unit, B
X, yThe expression B move to (x, y).
7. dilation operation: one of the most basic morphological operator, X expands with B and is designated as X B, definition be D=X B={ (x, y) | B
X, y∩ X ≠ }, wherein X is an image collection, B is a structural unit, B
X, yThe expression B move to (x, y).
8. closed operation: first expansion post-etching.
9. connected component labeling: the image to binaryzation is communicated with demarcation, for the zone of each connection be provided with a sign (referring to document: Zhang Yujin. Image Engineering (first volume)-Flame Image Process and analysis.Beijing: Tsing-Hua University publishes, and 1999,3).
Technical solution of the present invention is as follows:
The motion target detection method is characterized in that in a kind of infrared image sequence under complex background, comprises the following step (as shown in Figure 3):
The difference image of previous frame and present frame in step 1, the calculating infrared image sequence.
Step 2, the difference image that utilizes step 1 to obtain extract change detection mask.
As shown in Figure 4, concrete steps comprise following 3 step by step:
The 1st step: extract initial change and detect template, concrete grammar is as follows:
(1) calculate difference image histogram h (n), wherein, n is a gray level, and h (n) is the number of pixels of n for gray level in the image.
(2) calculate difference image average μ by h (n)
0And standard deviation sigma
0As initial value.
(3) by current estimated value μ
iAnd σ
i, calculate h (μ
i-3 σ
i, μ
i+ 3 σ
i) the average μ of interval interior pixel
I+1And standard deviation sigma
I+1
(4) as σ
I+1/ σ
i＞=T, then iteration stops, and wherein, T is a threshold value, and 0＜T≤1, chooses μ
I+1And σ
I+1As best estimate; Otherwise return (3).
(5) extract initial change and detect template:
CDM (x, y)=λ | if (d (x, y)＜μ-3 σ or d (x, y)＞μ+3 σ), λ=1, else λ=0}, wherein, (x y) is difference image to d, and x, y are respectively the horizontal ordinate and the ordinate of picture element.
The 2nd step: the initial change detection template that the 1st step obtained is done following morphology processing:
(1) corrosion (structural unit is 3 * 3 crosses).
(2) expand (structural unit is that radius is 3 circle) back and initial change detection template with.
(3) closed operation (structural unit is that radius is 5 circle).
The 3rd step: the change detection mask procession scanning after the processing of the 2nd step is repaired: at first do connected component labeling, to distinguish different targets, procession scanning respectively then, the background dot of 2 centres of same target (connected region) is changed to impact point in the same delegation (row).
The difference image of step 3, calculating present frame and next frame.
Step 4, the difference image that utilizes step 3 to obtain extract change detection mask, the same step 2 of method.
Step 5, two width of cloth change detection masks that step 2 and step 4 are obtained are done union (getting common factor), obtain templates of moving objects.
Step 6, the templates of moving objects that step 5 is obtained are done connected component labeling, are considered as moving target greater than the connected region of certain area.
Just can be partitioned into moving target by above-mentioned steps.
Need to prove:
1: step 1, three to dynamic background, need be carried out overall motion estimation and compensation at static background before calculating difference image.
2: extracting the principle that initial change detects template in the step 2, four is: ground unrest can be regarded Gauss as (referring to document: Kim M in the difference image, Choi J G, Lee M H, A VOP generation tool:Automatic segmentation of movingobjects in image sequences based on spatio-temporal information.IEEE Trans.on Circuits andSystems for Video Technology, 1999,9 (8): 1216-1226), and the pixel distribution in motion change zone is non-Gauss.Therefore, in difference image, distinguish the problem that motion change zone and invariant region can be regarded the non-Gaussian data of identification in Gaussian data as.In the difference image histogram, target is distributed in two ends mostly, noise concentrates near middle 0 value, the present invention from the histogram two ends to middle iteration estimated difference partial image noise parameter, comprise average and variance, as shown in Figure 5 (because of average near 0 value, change little, do not mark average), threshold value T generally gets about 0.75.Under the smaller situation of the area of target and entire image, iteration can restrain.Experimental result shows, generally needs iteration 1～3 time.
3: the principle that morphology is handled in the step 2, four is: the most of isolated noise in (1) corrosion back is removed, and target also suffers damage; (2) expand after, the target area increases, but has brought distortion, with the result after expanding and former CDM and, the target area that can recover to be corroded, and isolated noise can not be resumed owing to being corroded in (1).The main cause that adopts 3 kinds of size structure unit is that often the target disappearance is seriously for infrared image CDM complex background under, connective poor, so corrosion will cause big infringement with the exempt from customs examination target with little unit, and closed operation has only with the effective target of filling of big unit ability.
4: the necessity of rank scanning reparation is that target still may exist than macroscopic-void and disappearance among the CDM after morphology is handled owing to target internal in the infrared image lacks texture in the step 2, four.
Characteristic of the present invention and innovation part are:
1, based on the three-frame difference method, real-time is good, and it is accurate to cut apart target, as long as target all exists in three frames, and target velocity is greater than certain value, and testing result is not blocked by target speed and multiple goal to be influenced.
2, in the Changing Area Detection, propose a kind ofly, calculate simply, solved adaptive threshold preferably and chosen problem based on histogrammic difference image noise parameter method of estimation.
3, a kind of morphology disposal route is proposed, can be in CDM target disappearance and ground unrest all very under the serious situation, comparatively intactly extract the target area.
Characteristics such as the present invention is directed under the complex background that the target-to-background contrast of infrared image is low, object edge fuzzy and noise is big, a kind of moving target detecting method based on the three-frame difference method is provided, this method adopts technology such as the detection of adaptive change zone, the processing of antinoise morphology and rank scanning reparation, have target cut apart accurately complete, real-time is good and characteristics such as strong robustness, in military fields such as battle reconnaissance, firepower control and guidances, and civil areas such as aircraft navigation and highway intelligent transportation system have wide application prospect.
Description of drawings
Fig. 1: difference image synoptic diagram.
Fig. 2: three-frame difference method principle schematic.
Fig. 3: schematic flow sheet of the present invention.
Fig. 4: change detection mask extracts process flow diagram.
Fig. 5: the difference image noise parameter is estimated synoptic diagram.
Fig. 6: infrared moving vehicle detecting system experimental result under the field environment.Wherein (a) is the former frame image of present frame in the infrared image sequence, (b) be current frame image in the infrared image sequence, (c) be the next frame image of present frame in the infrared image sequence, (d) be front cross frame region of variation image, (e) be back two frame region of variation images, (f) be the present frame target image.
Embodiment
Use the present invention, realized infrared moving vehicle detecting system under the field environment.
(1) data source collection
The IR235 type infrared video camera that uses Wuhan Gao De company to produce, 8～14 microns of service bands, focal plane pixel 320*240.During shooting thermal camera is erected at and follows the tracks of vehicle on the tripod and take, the place is certain open-air skid pad, and vehicle is four kinds in offroad vehicle, minibus, jeep and a car, shooting time from afternoon to evening.
(2) The simulation experiment result
Under different automobile types, target numbers, shooting distance and angle case, adopt the present invention to carry out moving object detection and all achieve satisfactory results.Fig. 6 is offroad vehicle test experience result, and the integrality of target and accuracy are good as can be seen.
The present invention has execution speed faster.With the simulated program of VC6.0 exploitation, at match poplar 2.4G CPU, on the PC platform of 256 MB of memory, to 320*240 resolution infrared image, the time of carrying out once complete three-frame difference algorithm is about 100ms.Owing to need the three-frame difference of complete except detecting target the 1st time, all the other are each to detect only needs to calculate 1 difference and get final product (in addition 1 time obtained by detection last time), and program can reach the processing speed about detection/second 20 times, can be used in real time environment.
Claims (1)
1, motion target detection method in a kind of infrared image sequence under complex background is characterized in that, comprises the following step:
The difference image of previous frame and present frame in step 1, the calculating infrared image sequence;
Step 2, utilize the difference image that step 1 obtains to extract change detection mask, concrete steps comprise following 3 step by step:
The 1st step: extract initial change and detect template, concrete grammar is as follows:
(1), calculate difference image histogram h (n), wherein, n is a gray level, h (n) is the number of pixels of n for gray level in the image;
(2), calculate difference image average μ by h (n)
0And standard deviation sigma
0As initial value;
(3), by current estimated value μ
iAnd σ
i, calculate h (μ
i-3 σ
i, μ
i+ 3 σ
i) the average μ of interval interior pixel
I+1And standard deviation sigma
I+1
(4), as σ
I+1/ σ
i＞=T, then iteration stops, and wherein, T is a threshold value, and 0＜T≤1, chooses μ
I+1And σ
I+1As best estimate; Otherwise return (3);
(5) extract initial change and detect template:
CDM (x, y)=λ | if (d (x, y)＜μ-3 σ or d (x, y)＞μ+3 σ), λ=1, else λ=0}, wherein, (x y) is difference image to d, and x, y are respectively the horizontal ordinate and the ordinate of picture element;
The 2nd step: the initial change that the 1st step was obtained detects the morphology processing that template is done following order:
(1), the structural unit with 3 * 3 crosses carries out erosion operation;
(2), with radius be the structural unit of 3 circle carry out behind the dilation operation with initial change detect template with;
(3), be that the structural unit of 5 circle carries out closed operation with radius;
The 3rd step: the change detection mask procession scanning after the processing of the 2nd step is repaired: at first do connected component labeling, to distinguish different targets, procession scanning respectively then, the background dot of 2 centres of same target (connected region) is changed to impact point in the same delegation (row);
The difference image of step 3, calculating present frame and next frame;
Step 4, the difference image that utilizes step 3 to obtain extract change detection mask, the same step 2 of method;
Step 5, two width of cloth change detection masks that step 2 and step 4 are obtained are done union, obtain templates of moving objects;
Step 6, the templates of moving objects that step 5 is obtained are done connected component labeling, are considered as moving target greater than the connected region of certain area;
Just can be partitioned into moving target by above-mentioned steps.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100210679A CN100385461C (en) | 2006-06-01 | 2006-06-01 | Detection method for moving target in infrared image sequence under complex background |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006100210679A CN100385461C (en) | 2006-06-01 | 2006-06-01 | Detection method for moving target in infrared image sequence under complex background |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1885346A true CN1885346A (en) | 2006-12-27 |
| CN100385461C CN100385461C (en) | 2008-04-30 |
Family
ID=37583485
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100210679A Expired - Fee Related CN100385461C (en) | 2006-06-01 | 2006-06-01 | Detection method for moving target in infrared image sequence under complex background |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100385461C (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101408978B (en) * | 2008-11-27 | 2010-12-01 | 东软集团股份有限公司 | Method and apparatus for detecting barrier based on monocular vision |
| CN101976437A (en) * | 2010-09-29 | 2011-02-16 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
| CN102129692A (en) * | 2011-03-31 | 2011-07-20 | 中国民用航空总局第二研究所 | Method and system for detecting motion target in double threshold scene |
| CN102156984A (en) * | 2011-04-06 | 2011-08-17 | 南京大学 | Method for determining optimal mark image by adaptive threshold segmentation |
| CN102279191A (en) * | 2010-06-13 | 2011-12-14 | 中钞特种防伪科技有限公司 | Detection method and apparatus for defects in periodic texture images |
| CN101739553B (en) * | 2009-12-10 | 2012-01-11 | 青岛海信网络科技股份有限公司 | Method for identifying target in parallax image |
| CN102494676A (en) * | 2011-12-12 | 2012-06-13 | 中国科学院长春光学精密机械与物理研究所 | Satellite automatic recognition device under complicated backgrounds |
| CN103093471A (en) * | 2013-01-24 | 2013-05-08 | 青岛智目科技有限公司 | Foreground extraction method under complicated background |
| CN103402074A (en) * | 2013-07-05 | 2013-11-20 | 中山大学 | Recognition device for television image moving target |
| CN104966306A (en) * | 2015-07-07 | 2015-10-07 | 长春理工大学 | Measurement based method for detecting quantum video moving-object in complex background |
| CN105245809A (en) * | 2015-09-30 | 2016-01-13 | 小米科技有限责任公司 | Video recording method and video recording device |
| CN106091984A (en) * | 2016-06-06 | 2016-11-09 | 中国人民解放军信息工程大学 | A kind of three dimensional point cloud acquisition methods based on line laser |
| CN106874949A (en) * | 2017-02-10 | 2017-06-20 | 华中科技大学 | A kind of moving platform moving target detecting method and system based on infrared image |
| CN106878668A (en) * | 2015-12-10 | 2017-06-20 | 微软技术许可有限责任公司 | Mobile detection to object |
| CN106951863A (en) * | 2017-03-20 | 2017-07-14 | 贵州电网有限责任公司电力科学研究院 | A kind of substation equipment infrared image change detecting method based on random forest |
| CN107993254A (en) * | 2017-12-14 | 2018-05-04 | 广西科技大学 | Moving target detecting method based on disassociation frame calculus of finite differences |
| CN108548846A (en) * | 2018-06-21 | 2018-09-18 | 电子科技大学 | Bearing device subsurface defect extracting method based on thermal image AP clusters |
| CN109512402A (en) * | 2018-11-09 | 2019-03-26 | 吉林大学 | The method of the accurate medical human infrared thermal imaging of short distance various dimensions |
| CN109731258A (en) * | 2018-12-19 | 2019-05-10 | 沈阳天眼智云信息科技有限公司 | The monitoring of transformer ignition point and extinguishing method and system based on infrared technique |
| CN111539975A (en) * | 2020-04-09 | 2020-08-14 | 普联技术有限公司 | Method, device and equipment for detecting moving target and storage medium |
| CN112418105A (en) * | 2020-11-25 | 2021-02-26 | 湖北工业大学 | High maneuvering satellite time sequence remote sensing image moving ship target detection method based on difference method |
| CN112907626A (en) * | 2021-02-08 | 2021-06-04 | 中国空间技术研究院 | Moving object extraction method based on satellite time-exceeding phase data multi-source information |
| CN114037633A (en) * | 2021-11-18 | 2022-02-11 | 南京智谱科技有限公司 | Infrared image processing method and device |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101320475B (en) * | 2008-06-10 | 2010-12-29 | 北京航空航天大学 | Operating range estimation method of infrared imaging system under complex background condition |
| TWI494899B (en) | 2012-12-19 | 2015-08-01 | Ind Tech Res Inst | Method for in-image periodic noise reparation |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3507857B2 (en) * | 1999-10-27 | 2004-03-15 | 国土交通省国土技術政策総合研究所長 | Traffic flow detector |
| JP3871900B2 (en) * | 2001-05-25 | 2007-01-24 | エンブリッジ レイク プロプリエタリー リミテッド | Motion detection and warning system for mobile devices |
| CN1251144C (en) * | 2002-09-12 | 2006-04-12 | 上海交通大学 | Infra red detecting and tracing method for weak target under complex background condition |
| CN1207687C (en) * | 2003-07-03 | 2005-06-22 | 上海交通大学 | Multi-frame fuzzy detection method for point object in low S/N ratio infrared image sequence |
-
2006
- 2006-06-01 CN CNB2006100210679A patent/CN100385461C/en not_active Expired - Fee Related
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101408978B (en) * | 2008-11-27 | 2010-12-01 | 东软集团股份有限公司 | Method and apparatus for detecting barrier based on monocular vision |
| CN101739553B (en) * | 2009-12-10 | 2012-01-11 | 青岛海信网络科技股份有限公司 | Method for identifying target in parallax image |
| CN102279191B (en) * | 2010-06-13 | 2014-04-02 | 中钞特种防伪科技有限公司 | Detection method and apparatus for defects in periodic texture images |
| CN102279191A (en) * | 2010-06-13 | 2011-12-14 | 中钞特种防伪科技有限公司 | Detection method and apparatus for defects in periodic texture images |
| CN101976437A (en) * | 2010-09-29 | 2011-02-16 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
| CN101976437B (en) * | 2010-09-29 | 2012-10-03 | 中国资源卫星应用中心 | High-resolution remote sensing image variation detection method based on self-adaptive threshold division |
| CN102129692A (en) * | 2011-03-31 | 2011-07-20 | 中国民用航空总局第二研究所 | Method and system for detecting motion target in double threshold scene |
| CN102156984A (en) * | 2011-04-06 | 2011-08-17 | 南京大学 | Method for determining optimal mark image by adaptive threshold segmentation |
| CN102494676A (en) * | 2011-12-12 | 2012-06-13 | 中国科学院长春光学精密机械与物理研究所 | Satellite automatic recognition device under complicated backgrounds |
| CN102494676B (en) * | 2011-12-12 | 2013-07-03 | 中国科学院长春光学精密机械与物理研究所 | Satellite automatic recognition device under complicated backgrounds |
| CN103093471A (en) * | 2013-01-24 | 2013-05-08 | 青岛智目科技有限公司 | Foreground extraction method under complicated background |
| CN103402074A (en) * | 2013-07-05 | 2013-11-20 | 中山大学 | Recognition device for television image moving target |
| CN104966306A (en) * | 2015-07-07 | 2015-10-07 | 长春理工大学 | Measurement based method for detecting quantum video moving-object in complex background |
| CN105245809A (en) * | 2015-09-30 | 2016-01-13 | 小米科技有限责任公司 | Video recording method and video recording device |
| CN106878668A (en) * | 2015-12-10 | 2017-06-20 | 微软技术许可有限责任公司 | Mobile detection to object |
| US10460456B2 (en) | 2015-12-10 | 2019-10-29 | Microsoft Technology Licensing, Llc | Motion detection of object |
| CN106091984A (en) * | 2016-06-06 | 2016-11-09 | 中国人民解放军信息工程大学 | A kind of three dimensional point cloud acquisition methods based on line laser |
| CN106091984B (en) * | 2016-06-06 | 2019-01-25 | 中国人民解放军信息工程大学 | A kind of three dimensional point cloud acquisition methods based on line laser |
| CN106874949A (en) * | 2017-02-10 | 2017-06-20 | 华中科技大学 | A kind of moving platform moving target detecting method and system based on infrared image |
| CN106951863A (en) * | 2017-03-20 | 2017-07-14 | 贵州电网有限责任公司电力科学研究院 | A kind of substation equipment infrared image change detecting method based on random forest |
| CN106951863B (en) * | 2017-03-20 | 2023-09-26 | 贵州电网有限责任公司电力科学研究院 | Method for detecting change of infrared image of substation equipment based on random forest |
| CN107993254A (en) * | 2017-12-14 | 2018-05-04 | 广西科技大学 | Moving target detecting method based on disassociation frame calculus of finite differences |
| CN108548846A (en) * | 2018-06-21 | 2018-09-18 | 电子科技大学 | Bearing device subsurface defect extracting method based on thermal image AP clusters |
| CN109512402A (en) * | 2018-11-09 | 2019-03-26 | 吉林大学 | The method of the accurate medical human infrared thermal imaging of short distance various dimensions |
| CN109731258A (en) * | 2018-12-19 | 2019-05-10 | 沈阳天眼智云信息科技有限公司 | The monitoring of transformer ignition point and extinguishing method and system based on infrared technique |
| CN111539975A (en) * | 2020-04-09 | 2020-08-14 | 普联技术有限公司 | Method, device and equipment for detecting moving target and storage medium |
| CN111539975B (en) * | 2020-04-09 | 2023-12-22 | 成都市联洲国际技术有限公司 | Method, device, equipment and storage medium for detecting moving object |
| CN112418105A (en) * | 2020-11-25 | 2021-02-26 | 湖北工业大学 | High maneuvering satellite time sequence remote sensing image moving ship target detection method based on difference method |
| CN112418105B (en) * | 2020-11-25 | 2022-09-27 | 湖北工业大学 | High maneuvering satellite time sequence remote sensing image moving ship target detection method based on difference method |
| CN112907626A (en) * | 2021-02-08 | 2021-06-04 | 中国空间技术研究院 | Moving object extraction method based on satellite time-exceeding phase data multi-source information |
| CN114037633A (en) * | 2021-11-18 | 2022-02-11 | 南京智谱科技有限公司 | Infrared image processing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100385461C (en) | 2008-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1885346A (en) | Detection method for moving target in infrared image sequence under complex background | |
| EP2381416B1 (en) | Method for image restoration in a computer vision system | |
| Wu et al. | Lane-mark extraction for automobiles under complex conditions | |
| CN103325112B (en) | Moving target method for quick in dynamic scene | |
| US8867790B2 (en) | Object detection device, object detection method, and program | |
| CN108960190B (en) | SAR video target detection method based on FCN image sequence model | |
| KR101483742B1 (en) | Lane Detection method for Advanced Vehicle | |
| CN109816606B (en) | Method for tracking target by using optical remote sensing satellite | |
| Ravanfar et al. | Low contrast sperm detection and tracking by watershed algorithm and particle filter | |
| CN111444778A (en) | Lane line detection method | |
| Kumar et al. | Vehicle speed detection using corner detection | |
| Li et al. | Foreground detection based on optical flow and background subtract | |
| Soundrapandiyan et al. | Adaptive pedestrian detection in infrared images using fuzzy enhancement and top-hat transform | |
| FAN et al. | Robust lane detection and tracking based on machine vision | |
| Rahaman et al. | Lane detection for autonomous vehicle management: PHT approach | |
| Wu et al. | Registration-based moving vehicle detection for low-altitude urban traffic surveillance | |
| Cao et al. | Visual attention accelerated vehicle detection in low-altitude airborne video of urban environment | |
| Sindoori et al. | Adaboost technique for vehicle detection in aerial surveillance | |
| Liao et al. | An approach for detecting moving target in VideoSAR imagery sequence | |
| Lin et al. | Robust vehicle detection in vision systems based on fast wavelet transform and texture analysis | |
| Greeshma et al. | Object tracking using filtering and optimized sum of absolute difference | |
| Sarhangi Fard | Improving Moving Vehicles Detection and Vehicles Trajectory Plotting Using Optimized Background Subtraction Method | |
| Wojcikowski et al. | An intelligent image processing sensor-the algorithm and the hardware implementation | |
| Wang et al. | A Tracking Algorithm for a Moving Object in the Image Sequence Based on the Adaptive Threshold | |
| CN112270657A (en) | Sky background-based target detection and tracking algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080430 Termination date: 20100601 |