CN109887007A - The detection method and device of space base moving target over the ground - Google Patents
The detection method and device of space base moving target over the ground Download PDFInfo
- Publication number
- CN109887007A CN109887007A CN201711276006.1A CN201711276006A CN109887007A CN 109887007 A CN109887007 A CN 109887007A CN 201711276006 A CN201711276006 A CN 201711276006A CN 109887007 A CN109887007 A CN 109887007A
- Authority
- CN
- China
- Prior art keywords
- target
- frame
- image
- detection
- difference
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 93
- 230000033001 locomotion Effects 0.000 claims abstract description 25
- 238000000034 method Methods 0.000 claims description 36
- 230000008569 process Effects 0.000 claims description 13
- 238000010586 diagram Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000007689 inspection Methods 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000010429 evolutionary process Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of detection method and device of space base moving target over the ground.Wherein, the detection method of a kind of space base moving target over the ground, comprising: obtain target image frame and simultaneously the target image frame is pre-processed, to obtain the surely successive image frame as after;The successive image frame obtained using after pretreatment carries out the detection of moving target as input, wherein, it is detected using the target that three-frame differencing is greater than given threshold to movement velocity, is detected using the target that background difference algorithm is less than given threshold to movement velocity;It is further detected using the algorithm of target detection based on tracking.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to the detection method and dress of a kind of space base moving target over the ground
It sets.
Background technique
Using high and medium flying platform and high-resolution imaging system, vast region can be monitored in real time and be continued
Track and monitor all kinds of interested targets and target complex, such as naval vessel, aircraft, vehicle, tank, personnel.For common city
City or operation battlefield surroundings, the wide view field of this kind of camera can usually observe thousands of ground moving objects, including
Personnel, vehicle, naval vessel etc..In order to obtain biggish observation visual field, the flying height of the flying platform where camera is usually higher,
The imaging pixel dimension of the target observed in the picture is usually smaller;Simultaneously because the movement of platform itself, the figure observed
For earth background as in also with image motion, this causes certain difficulty to the detection of moving target and target complex.
Target detection is the image partition method of geometrical property based on target area, kinetic characteristic and statistical nature.Target detection side
Method is widely used in Face datection, humanoid detection, vehicle detection, with the development of the times, existing object detection method
Effect needs to be further increased.
Summary of the invention
In view of the above technical problems, the present invention proposes a kind of detection method and device of space base moving target over the ground.
Technical solution is as follows:
In one aspect, the detection method of a kind of space base of proposition moving target over the ground, comprising:
It obtains target image frame and the target image frame is pre-processed, to obtain the surely successive image frame as after;
The successive image frame obtained using after pretreatment carries out the detection of moving target as input, wherein poor using three frames
The target for dividing algorithm to be greater than given threshold to movement velocity detects, and is less than setting to movement velocity using background difference algorithm
The target of threshold value is detected;
It is further detected using the algorithm of target detection based on tracking, wherein the target inspection based on tracking
The process of method of determining and calculating includes:
For each target, two image districts of the target are recorded, one is the appearance images comprising original gradation information
δa, the other is the shape image δ after the soft segmentation of targets;
By to target predicted position x0Field search, find maximum probability target detection, calculation formula is as follows:
Wherein, Ω (x0) it is target predicted position x0Field search, Ps(x|Ds,Db,δs) it is shape matching probability, DsIt is
Three-frame difference figure, DbIt is background difference diagram, Pa(x|It,δa) it is appearance matching probability, ItIt is present frame, Pv(x|It) be based on pair
Claim the vehicle detection probability of attribute.
On the other hand, the detection device of a kind of space base of proposition moving target over the ground comprising:
Pretreatment unit, for obtaining target image frame and being pre-processed to the target image frame, to obtain surely as after
Successive image frame;
First object detection unit, for carrying out moving target as input using the successive image frame obtained after pre-processing
Detection, wherein detected using the target that three-frame differencing is greater than given threshold to movement velocity, calculated using background difference
The target that method is less than given threshold to movement velocity detects;
Second object detection unit, for further being detected using the algorithm of target detection based on tracking, wherein
The process of the algorithm of target detection based on tracking includes:
For each target, two image districts of the target are recorded, one is the appearance images δ comprising original gradation informationa, the other is the shape image δ after the soft segmentation of targets;
By to target predicted position x0Field search, find maximum probability target detection, calculation formula is as follows:
Wherein, Ω (x0) it is target predicted position x0Field search, Ps(x|Ds,Db,δs) it is shape matching probability, DsIt is
Three-frame difference figure, DbIt is background difference diagram, Pa(x|It,δa) it is appearance matching probability, ItIt is present frame, Pv(x|It) be based on pair
Claim the vehicle detection probability of attribute.
Effect caused by technical solution provided by the embodiment of the present invention includes:
For the low frame rate motion image sequence of space base wide area monitoring, the invention proposes utilize image difference technology and mesh
The quick detection that feature pickup technology realizes moving target is marked, algorithm calculation amount is small, time-consuming short.In terms of image difference, it will tie
Inter-frame difference and background difference are closed, realizes to the detection of the small objects of low frame rate video image, significantly reduces the mistake of target
Inspection rate provides reliable input for subsequent multiple target tracking.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
This specification embodiment can be limited.
In addition, any embodiment in this specification embodiment does not need to reach above-mentioned whole effects.
Detailed description of the invention
In order to illustrate more clearly of this specification embodiment or technical solution in the prior art, below will to embodiment or
Attached drawing needed to be used in the description of the prior art is briefly described, it should be apparent that, the accompanying drawings in the following description is only
The some embodiments recorded in this specification embodiment for those of ordinary skill in the art can also be attached according to these
Figure obtains other attached drawings.
Fig. 1 is a kind of flow diagram of space base detection method of moving target over the ground of the embodiment of the present invention;
Fig. 2 is the principle that three-frame difference technology carries out target detection;
Fig. 3 is illustrative Background evolutionary process;
Fig. 4 is a kind of module diagram of space base detection device of moving target over the ground of the embodiment of the present invention.
Fig. 5 shows the detection performance comparing result of two frame differences and three-frame difference;
Fig. 6 shows application of the balanced-filter on wide area image.
Specific embodiment
Hereinafter, with reference to the accompanying drawings to detailed description of the present invention embodiment.However, it is possible to come in many different forms real
The present invention is applied, and the present invention should not be construed as limited to the specific embodiment illustrated here.On the contrary, providing these implementations
Example is in order to explain the principle of the present invention and its practical application, to make others skilled in the art it will be appreciated that the present invention
Various embodiments and be suitable for the various modifications of specific intended application.
For the low frame rate motion image sequence of space base wide area monitoring, the present invention proposes to utilize image difference technology and target
The quick detection of feature pickup technology realization moving target.It, will be real in conjunction with inter-frame difference and background difference in terms of image difference
Now to the detection of the small objects of low frame rate video image;Meanwhile our also further research and analysis moving vehicles are in high-altitude
The imaging characteristics of image propose to realize using this feature of symmetric properties to vehicle/non-vehicle target classification, to reduce target
False detection rate provides reliable input for subsequent multiple target tracking.
Referring to Fig.1, in an embodiment of the present invention, the detection method of a kind of space base moving target over the ground, may include as
Lower step: step 101: obtaining target image frame and the target image frame is pre-processed, to obtain the surely sequential chart as after
As frame.
Pre-processing image data process includes at least steady picture and is registrated the two more crucial steps, its precision will certainly
The quality and subsequent tracking of maneuvering target precision of fixed entire target detection.It is straight since the grid cell size of target is usually smaller
Diameter may be about 5-10 pixel, a large amount of steady as making an uproar if required target size precision is not achieved as precision for the steady of image
Sound and motion detection mistake will be introduced in subsequent tracking computing module, influence the final precision and track matter of tracking
Amount.
In general, the steady successive video frames as after, since sample frequency is lower, target relative movement is very fast.In order to realize with
The reliable registration on ground and high-precision steady picture, processing system introduce reference data figure and three dimensional topographic data, and image is allowed to match
Irregular three-D geometry deformation is carried out during quasi-, ensure that steady precision and flexibility as registration.Meanwhile in the process of registration
In, processing system has also been constructed a feedback loop and is modified to original reference map, enables the process being entirely registrated not
It is disconnected to tend to stable convergence.If without reference to reference map and terrain data, it may be considered that carry out substitution using first frame and with plane
As initial landform.
Step 102: the successive image frame obtained using after pretreatment carries out the detection of moving target as input, wherein adopts
It is detected with the target that three-frame differencing is greater than given threshold to movement velocity, using background difference algorithm to movement velocity
Target less than given threshold is detected.
After completing pretreatment image data early period, the stable continuous videos image of background is obtained, and can be in this, as
Input carries out moving target (such as vehicle) detection of next step.Since the size of wide area video image is big, sample frequency phase
To lower, usually can control in 2Hz, quick ground moving object this kind of for vehicle in this way, the moving displacement of interframe can
To reach dozens of pixel.If using two conventional frame differences, it will form double objects testing in difference image.For this purpose, I
Propose using three-frame difference technology carry out target detection, its principle is as shown in Figure 2, i.e., it is described use three-frame differencing pair
The target that movement velocity is greater than given threshold is detected, comprising:
T-1 frame, t frame and t+1 frame image are extracted respectively;
Calculate the difference result y1 of t-1 frame and t frame image;
Calculate the difference result y2 of t frame and t+1 frame image;
Y1 and y2 are asked and operation, to obtain testing result.
The Computing Principle of three-frame difference, the double the problem of detecting of target caused by two frame differences can be effectively avoided in it
(one of them is decoy).
However, for microinching target, (threshold speed that can be set herein with one is come to distinguish moving target be at a slow speed
Or effective target detection can not necessarily quickly), be obtained using three-frame difference.For this purpose, background differential technique is introduced, it can be with
Further realize the lasting detection to the target stopped after movement.In background differential technique, it is necessary to have reference background image
As the foundation of difference, this just needs to construct corresponding background image first.In the actual operation process, background image is not
In the presence of it can be obtained from the video sequence observed by the accumulative of information.In one embodiment, easier
Mode can be through method that pixel value part updates and realize, specific formula is as follows:
IB(x)=α IB(x)+(1-α)It(x),wherex∈ΩB
Wherein, IBIt (x) is background image, It(x) be t moment input picture, ΩBIt is background image region.With the time
Extension, Background is by the Background being formed on t to t-N Δ t time window that gradually develops from the first frame of initialization.
Slow down when vehicle or stop completely, Three image difference cannot provide effective target detection, so that subsequent
Pursuit path terminates.In order to solve this problem, it is proposed that the algorithm of target detection (following step 103) based on tracking is into one
Step strengthens target detection process.It is understood that background modeling and background difference have certain advantage detection mobile slow or stop
Object, but due to the noise and three-dimensional parallax of image, background difference is also easy to produce more empty inspection target.Therefore, I
Limit background difference be served only for by track based on target detection during, any new mesh is initialized without the use of it
Track is marked, to guarantee that target can be still accurately tracked under halted state.
Step 103: further being detected using the algorithm of target detection based on tracking.
This method is by resolving appearance, the joint probability of shape and background difference, moreover it is possible to which advanced optimizing target position is
Target detection is optimal.Wherein, the process of the algorithm of target detection based on tracking includes:
For each target, two image districts of the target are recorded, one is the appearance images comprising original gradation information
δa, the other is the shape image δ after the soft segmentation of targets;
By to target predicted position x0Field search, find maximum probability target detection, calculation formula is as follows:
Wherein, Ω (x0) it is target predicted position x0Field search, Ps(x|Ds,Db,δs) it is shape matching probability, DsIt is
Three-frame difference figure, DbIt is background difference diagram, Pa(x|It,δa) it is appearance matching probability, ItIt is present frame, Pv(x|It) be based on pair
Claim the vehicle detection probability of attribute.In the algorithm, in order to avoid target missing inspection, highest three regions of probability value can be chosen and made
For the result of target detection.
Fig. 3 shows the evolutionary process of Background, wherein the target area (a) gray scale and shape image;(b) current input figure
Picture;(c) appearance matching probability Pa;(d) shape matching probability Ps;(e) the vehicle detection probability P based on symmetric propertiesv;(f) it merges
Target detection probability afterwards.
In an alternate embodiment of the invention, after above-mentioned steps 103, the method can also include:
Step 104: using the filter based on symmetric properties to by three-frame differencing or background difference algorithm or
The target that algorithm of target detection based on tracking generates is screened.
In order to further decrease the empty inspection rate of target, the present invention is also by research and analysis image feature extraction to target detection
Influence.Since target area very little pixel is few (usually in 8 × 8 pixels), using conventional Feature Extraction Technology (such as HOG or
Color histogram), the accurate description of the target area cannot be obtained.For this purpose, devising the vehicle shape based on symmetric properties
Filter screens target caused by several sections in front.
Because space base wide area video monitoring image resolution ratio it is lower, with reference to defend figure it is matched during, we
Image is subjected to scaling first, the ground resolution of pixel is controlled in 0.5 meter/pixel, can guarantee that image has in this way
There is unified physical size.Under this scale, common vehicle (such as car, buggy) usually has about 8 × 8 pixel sizes,
If applying balanced-filter on this image, vehicle class target would generally generate stronger detection signal, can be used as additional
Information carry out vehicle and Fei Che classification.In our experiment, the radius of balanced-filter is set 4 pixel (vehicles by we
The half of diameter) and obtain preferable result.
The present invention is by combining multi-frame difference and background difference that can significantly improve the detection efficiency of moving target, but simultaneously
Also it will increase target detection false alarm rate.In order to reduce false alarm rate, the present invention designs the algorithm of the object filtering based on priori knowledge,
It can quickly and effectively be classified to vehicle/non-vehicle target using the geometrically symmetric attribute of target in the present invention, be reduced
Moving vehicle detection wrong report.For the low frame rate motion image sequence of space base wide area monitoring, the invention proposes utilize image difference
The technology of dividing and target signature pickup technology realize the quick detection of moving target, and algorithm calculation amount is small, time-consuming short.In image difference
Aspect will be realized to the detection of the small objects of low frame rate video image, be significantly reduced in conjunction with inter-frame difference and background difference
The false detection rate of target provides reliable input for subsequent multiple target tracking.
Fig. 4 is a kind of module diagram of space base detection device of moving target over the ground of the embodiment of the present invention, such as Fig. 4 institute
Show, in inventing a kind of embodiment, device may include:
Pretreatment unit 201, for obtaining target image frame and being pre-processed to the target image frame, to obtain steady picture
Successive image frame afterwards;
First object detection unit 202, for carrying out movement mesh as input using the successive image frame obtained after pre-processing
Target detection, wherein detected using the target that three-frame differencing is greater than given threshold to movement velocity, using background subtraction
The target for dividing algorithm to be less than given threshold to movement velocity detects;
Second object detection unit 203, for further being detected using the algorithm of target detection based on tracking,
In, the process of the algorithm of target detection based on tracking includes:
For each target, two image districts of the target are recorded, one is the appearance images comprising original gradation information
δa, the other is the shape image δ after the soft segmentation of targets;
By to target predicted position x0Field search, find maximum probability target detection, calculation formula is as follows:
Wherein, Ω (x0) it is target predicted position x0Field search, Ps(x|Ds,Db,δs) it is shape matching probability, DsIt is
Three-frame difference figure, DbIt is background difference diagram, Pa(x|It,δa) it is appearance matching probability, ItIt is present frame, Pv(x|It) be based on pair
Claim the vehicle detection probability of attribute.
Further, described device can also include:
Screening unit, for being calculated using the filter based on symmetric properties by three-frame differencing or background difference
Method or the target of the algorithm of target detection generation based on tracking are screened.
The function of modules and the realization process of effect are specifically detailed in the above method and correspond to step in above-mentioned apparatus
Realization process, details are not described herein.
Fig. 5 shows the detection performance comparing result of two frame differences and three-frame difference.Wherein, the image of the leftmost side in figure
It is: input picture and result;The image of centre in figure is: three-frame difference result;The image of the rightmost side in figure is: two frames are poor
Divide result (wherein double detected artifacts are obvious), it is seen then that the performance of three-frame difference is more preferably.
Fig. 6 shows application of the balanced-filter on wide area image.Wherein, the left-side images in Fig. 6 are as follows: input figure
Picture;Image right are as follows: using the convolution signal of balanced-filter.As it can be seen that the signal of filter is darker, for white for illegal vehicle
The signal of vehicle, filter is brighter, and convolution signal is taken absolute value, and can obtain shown in (f) in Fig. 3 based on symmetric properties
Vehicle detection probability Pv。
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device reality
For applying example, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to embodiment of the method
Part explanation.The apparatus embodiments described above are merely exemplary, wherein described be used as separate part description
Module may or may not be physically separated, can be each module when implementing this specification example scheme
Function realize in the same or multiple software and or hardware.Can also select according to the actual needs part therein or
Person's whole module achieves the purpose of the solution of this embodiment.Those of ordinary skill in the art are not the case where making the creative labor
Under, it can it understands and implements.The above is only the specific embodiment of this specification embodiment, it is noted that for this
For the those of ordinary skill of technical field, under the premise of not departing from this specification embodiment principle, it can also make several
Improvements and modifications, these improvements and modifications also should be regarded as the protection scope of this specification embodiment.
Claims (10)
1. a kind of detection method of space base moving target over the ground characterized by comprising
It obtains target image frame and the target image frame is pre-processed, to obtain the surely successive image frame as after;
The successive image frame obtained using after pretreatment carries out the detection of moving target as input, wherein is calculated using three-frame difference
The target that method is greater than given threshold to movement velocity detects, and is less than given threshold to movement velocity using background difference algorithm
Target detected;
It is further detected using the algorithm of target detection based on tracking, wherein the target detection based on tracking is calculated
The process of method includes:
For each target, two image districts of the target are recorded, one is the appearance images δ comprising original gradation informationa, separately
One is shape image δ after the soft segmentation of targets;
By to target predicted position x0Field search, find maximum probability target detection, calculation formula is as follows:
Wherein, Ω (x0) it is target predicted position x0Field search, Ps(x|Ds, Db, δs) it is shape matching probability, DsIt is three frames
Difference diagram, DbIt is background difference diagram, Pa(x|It,δa) it is appearance matching probability, ItIt is present frame, Pv(x|It) it is to be belonged to based on symmetrical
The vehicle detection probability of property.
2. the method according to claim 1, wherein the method also includes:
Using the filter based on symmetric properties to by three-frame differencing or background difference algorithm or based on the mesh of tracking
The target that mark detection algorithm generates is screened.
3. being set the method according to claim 1, wherein the use three-frame differencing is greater than movement velocity
The target for determining threshold value is detected, comprising:
T-1 frame, t frame and t+1 frame image are extracted respectively;
Calculate the difference result y1 of t-1 frame and t frame image;
Calculate the difference result y2 of t frame and t+1 frame image;
Y1 and y2 are asked and operation, to obtain testing result.
4. the method according to claim 1, wherein the method that the background image is updated by pixel value part
It obtains, formula is as follows:
IB(x)=α IB(x)+(1-α)It(x),wherex∈ΩB
Wherein, IBIt (x) is background image, It(x) be t moment input picture, ΩBIt is background image region.
5. according to the method described in claim 2, it is characterized in that, the radius of the filter is set to 4 pixels.
6. a kind of detection device of space base moving target over the ground characterized by comprising
Pretreatment unit, for obtaining target image frame and being pre-processed to the target image frame, to obtain the surely company as after
Continuous picture frame;
First object detection unit, the successive image frame for being obtained using after pretreatment carry out the inspection of moving target as input
It surveys, wherein detected using the target that three-frame differencing is greater than given threshold to movement velocity, using background difference algorithm
The target for being less than given threshold to movement velocity detects;
Second object detection unit, for further being detected using the algorithm of target detection based on tracking, wherein described
The process of algorithm of target detection based on tracking includes:
For each target, two image districts of the target are recorded, one is the appearance images δ comprising original gradation informationa, separately
One is shape image δ after the soft segmentation of targets;
By to target predicted position x0Field search, find maximum probability target detection, calculation formula is as follows:
Wherein, Ω (x0) it is target predicted position x0Field search, Ps(x|Ds,Db,δs) it is shape matching probability, DsIt is three frames
Difference diagram, DbIt is background difference diagram, Pa(x|It,δa) it is appearance matching probability, ItIt is present frame, Pv(x|It) it is to be belonged to based on symmetrical
The vehicle detection probability of property.
7. device according to claim 6, which is characterized in that described device further include:
Screening unit, for using the filter based on symmetric properties to by three-frame differencing or background difference algorithm or
The target that algorithm of target detection based on tracking generates is screened.
8. device according to claim 6, which is characterized in that the first object detection unit is at least used for:
T-1 frame, t frame and t+1 frame image are extracted respectively;
Calculate the difference result y1 of t-1 frame and t frame image;
Calculate the difference result y2 of t frame and t+1 frame image;
Y1 and y2 are asked and operation, to obtain testing result.
9. device according to claim 6, which is characterized in that the method that the background image is updated by pixel value part
It obtains, formula is as follows:
IB(x)=α IB(x)+(1-α)It(x),wherex∈ΩB
Wherein, IBIt (x) is background image, It(x) be t moment input picture, ΩBIt is background image region.
10. device according to claim 7, which is characterized in that the radius of the filter is set to 4 pixels.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711276006.1A CN109887007A (en) | 2017-12-06 | 2017-12-06 | The detection method and device of space base moving target over the ground |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711276006.1A CN109887007A (en) | 2017-12-06 | 2017-12-06 | The detection method and device of space base moving target over the ground |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109887007A true CN109887007A (en) | 2019-06-14 |
Family
ID=66923673
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711276006.1A Pending CN109887007A (en) | 2017-12-06 | 2017-12-06 | The detection method and device of space base moving target over the ground |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109887007A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112734796A (en) * | 2021-01-18 | 2021-04-30 | 烟台南山学院 | Moving target detection method and detection system for realizing target detection method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9008366B1 (en) * | 2012-01-23 | 2015-04-14 | Hrl Laboratories, Llc | Bio-inspired method of ground object cueing in airborne motion imagery |
CN106296725A (en) * | 2015-06-12 | 2017-01-04 | 富泰华工业(深圳)有限公司 | Moving target detects and tracking and object detecting device in real time |
CN107067416A (en) * | 2017-05-11 | 2017-08-18 | 南宁市正祥科技有限公司 | A kind of detection method of moving target |
-
2017
- 2017-12-06 CN CN201711276006.1A patent/CN109887007A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9008366B1 (en) * | 2012-01-23 | 2015-04-14 | Hrl Laboratories, Llc | Bio-inspired method of ground object cueing in airborne motion imagery |
CN106296725A (en) * | 2015-06-12 | 2017-01-04 | 富泰华工业(深圳)有限公司 | Moving target detects and tracking and object detecting device in real time |
CN107067416A (en) * | 2017-05-11 | 2017-08-18 | 南宁市正祥科技有限公司 | A kind of detection method of moving target |
Non-Patent Citations (5)
Title |
---|
CHIN-TENG LIN等: "Boosted Vehicle Detection Using Local and Global Features", 《JOURNAL OF SIGNAL AND INFORMATION PROCESSING》 * |
JIANGJIAN XIAO等: "Vehicle Detection and Tracking in Wide Field-of-View Aerial Video", 《2010 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
周晓: "基于三帧差分和滑动平均背景的运动目标检测", 《计算机测量与控制.》 * |
孙学彬: "基于多对象特征获取的多目标跟踪算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
孙水发等: "《视频前景检测及其在水电工程监测中的应用》", 北京:国防工业出版社 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112734796A (en) * | 2021-01-18 | 2021-04-30 | 烟台南山学院 | Moving target detection method and detection system for realizing target detection method |
CN112734796B (en) * | 2021-01-18 | 2023-04-18 | 烟台南山学院 | Moving target detection method and detection system for realizing moving target detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109272530B (en) | Target tracking method and device for space-based monitoring scene | |
US9866815B2 (en) | 3D object segmentation | |
CN108154520B (en) | A kind of moving target detecting method based on light stream and frame matching | |
US8718328B1 (en) | Digital processing method and system for determination of object occlusion in an image sequence | |
US8073196B2 (en) | Detection and tracking of moving objects from a moving platform in presence of strong parallax | |
US8712096B2 (en) | Method and apparatus for detecting and tracking vehicles | |
CN107146239B (en) | Satellite video moving target detection method and system | |
CN107146200B (en) | Unmanned aerial vehicle remote sensing image splicing method based on image splicing quality evaluation | |
CN110688905B (en) | Three-dimensional object detection and tracking method based on key frame | |
KR20150027291A (en) | Optical flow tracking method and apparatus | |
CN105374049B (en) | Multi-corner point tracking method and device based on sparse optical flow method | |
CN108022249A (en) | A kind of remote sensing video satellite moving vehicle target region of interest extraction method | |
CN108010065A (en) | Low target quick determination method and device, storage medium and electric terminal | |
CN108629792A (en) | Laser eyepiece detection method and device based on background modeling Yu background difference | |
US11410459B2 (en) | Face detection and recognition method using light field camera system | |
Pan et al. | Depth map completion by jointly exploiting blurry color images and sparse depth maps | |
CN104217442A (en) | Aerial video moving object detection method based on multiple model estimation | |
CN116978009A (en) | Dynamic object filtering method based on 4D millimeter wave radar | |
Colombari et al. | Patch-based background initialization in heavily cluttered video | |
Pollard et al. | GM-PHD filters for multi-object tracking in uncalibrated aerial videos | |
Zhang et al. | An optical flow based moving objects detection algorithm for the UAV | |
May et al. | Moving target detection for sense and avoid using regional phase correlation | |
Patel et al. | Comparison of optical flow algorithms for speed determination of moving objects | |
CN109887007A (en) | The detection method and device of space base moving target over the ground | |
Sincan et al. | Moving object detection by a mounted moving camera |
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 |
Application publication date: 20190614 |
|
RJ01 | Rejection of invention patent application after publication |