CN104504724B - A kind of moving body track and extraction algorithm not influenceed by barrier - Google Patents
A kind of moving body track and extraction algorithm not influenceed by barrier Download PDFInfo
- Publication number
- CN104504724B CN104504724B CN201510020446.5A CN201510020446A CN104504724B CN 104504724 B CN104504724 B CN 104504724B CN 201510020446 A CN201510020446 A CN 201510020446A CN 104504724 B CN104504724 B CN 104504724B
- Authority
- CN
- China
- Prior art keywords
- mrow
- characteristic point
- mtr
- mtd
- 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.)
- Expired - Fee Related
Links
- 230000004888 barrier function Effects 0.000 title claims abstract description 15
- 238000000605 extraction Methods 0.000 title claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims abstract description 21
- 230000009286 beneficial effect Effects 0.000 claims description 15
- 230000000903 blocking effect Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 5
- 230000008901 benefit Effects 0.000 claims description 4
- 239000000284 extract Substances 0.000 abstract description 2
- 238000000034 method Methods 0.000 description 11
- 230000000694 effects Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000007796 conventional method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/292—Multi-camera tracking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/254—Analysis of motion involving subtraction of images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The present invention designs a kind of Moving object extraction and track algorithm not influenceed by barrier, while the algorithm can also provide the Detailed motion information of target object various pieces while target object is followed the trail of.The algorithm extracts characteristic point to the continuous videos picture frame obtained by camera, and the characteristic point of acquisition is established into matching matrix, and the extraction of moving object and tracking are converted into an Assignment Problems, the matching relationship of characteristic point is obtained by solving an Assignment Problems, and realizes the extraction and tracking to mobile object.Because the algorithm handles multiple image frame simultaneously in matching, therefore when moving object is blocked by barrier, tracking is unaffected.
Description
Technical field
The present invention relates to the technical field of computer vision, more particularly to it is a kind of not by the moving object that barrier is influenceed with
Track and extraction algorithm.
Background technology
The tracking of mobile object in the video that camera obtains is a focus in computer vision research field,
Had broad application prospects in the systems such as robot vision, video monitoring.Conventional method such as background subtraction, continuous frame-to-frame differences
Point-score can realize more good tracking effect in the case where camera is fixed, but in situation about being influenceed by barrier
Under, the tracking of moving object would generally be interrupted.
The method for following the trail of mobile object that many researchers propose at present, such as optical flow method, based on SIFT feature
Matching method, Mean-Shift methods and particle filter method can not solve the problems, such as to run into obstructing objects in tracking.Do not hindered
It in application fields such as such as security protection, robot vision is essential to hinder moving body track that thing influences and extraction algorithm.
The content of the invention
It is an object of the present invention to provide a kind of moving body track and extraction algorithm not influenceed by barrier, the party
Method can also provide the Detailed motion information of target object various pieces while target object is followed the trail of.Therefore, the algorithm is not
Traditional method is only compared in the tracking of mobile object and outside more preferable effect, the Activity recognition of personage can also be applied to
In field.
In order to realize foregoing invention purpose, the invention provides a kind of new movement matched simultaneously based on multiframe characteristic point
Object tracking methods.When this method is first by the picture frame at the current time of the video image obtained by dollying head and before
The picture frame at quarter extracts, and by the feature point extraction in image:
Assuming that the picture frame of current t is framet,
T frames before t are extracted simultaneously, and the T frames are framet-h, h=1 ... T },
To frametAnd framet-h, h=1 ... and T } extraction characteristic point, and protected in the characteristic point that each frame is extracted
N number of characteristic point that stability is best is stayed, then the characteristic point sum being extracted is individual for N × (T+1).
Because the trajectory track problem of mobile object can be converted to of the various pieces in successive frame in mobile object
With problem, therefore, by the matching result for trying to achieve each characteristic point of each characteristic point in present frame therewith in previous frame
Obtain track of the mobile object in successive frame.
The matching result of the N × T of N number of characteristic point in present frame therewith in previous frame characteristic point is sought, N × N × T can be established
Original beneficial matrix, matching result is motion track of each position of mobile object in successive frame.
The characteristic point of present frame therewith previous frame characteristic point matching when should meet following condition:
The cumulative total value of similarity between the characteristic point to match is maximum, i.e., the matching problem of 1 couple 1 can be considered as one
The individual maximized optimization problem of global similarity, i.e. an Assignment Problems,.
When a new mobile object appears in picture frame picture first, the characteristic point of the mobile object may be therewith
Any characteristic point of previous frame can not all be matched, and one kind as above-mentioned technical proposal is improved, former benefit of this programme to N × N × T
Matrix is extended.
N × N × T original beneficial matrix is extended to N × (N × T+1).The element value of the expansion of beneficial matrix is default
Threshold value.
Then the optimization matching of the matching problem of the characteristic point of previous frame is mathematically represented as the characteristic point of present frame therewith:
Minimize
s.t
pij={ 0,1 },
Wherein T be present frame before frame number, similarity { Curr (t)i, Prev (t-h)jIt is present frame characteristic point i
Feature space space length between the characteristic point j of previous frame therewith, thresholdforcreating are default threshold value.
In actually calculating, similarity is calculated by the distance between feature histogram of two characteristic points.Therefore, it is global
It is to minimize C that similarity, which maximizes actual,.The optimization matching of this 1 couple 1 can be considered as 1 Assignment Problems, and solve.
The characteristic point of present frame and threshold portion when matching, then this characteristic point is considered as belonging to new mobile object.
Previous frame in when belonging in the characteristic point of mobile object in the current frame without match point, previous frame in moving body quilt
It is considered as and is blocked.The characteristic point being blocked within a certain period of time not again by the Feature Points Matching in current time picture frame,
This feature point is considered as disappearing from picture.Again by current time picture frame after the mobile object being tracked is being blocked
Feature Points Matching, this feature point is considered as the appearance again after blocking.
When the mobile object being tracked is being blocked, its position can do linear estimation by blocking front and rear position.
A kind of the advantage of the invention is that tracing algorithm that do not influenceed by barrier can be realized and interrupted.Hindered in object
Hinder thing to realize lasting tracking when blocking, and position when being blocked can be extrapolated.This feature makes this method exist
Relatively conventional method has more preferable applicability and practicality in terms of tracking object.In addition, technical scheme is due to can be with
The various pieces of mobile object are tracked, therefore track characteristic can be provided for the Activity recognition of personage.
Brief description of the drawings
Fig. 1 is basic beneficial matrix schematic diagram.
Fig. 2 is the beneficial matrix schematic diagram after extension.
Fig. 3 is the final form for the matrix that heterochromia most has smallization.
Fig. 4 is the schematic diagram of shape and structure change constraint.
Embodiment
Characteristic point is extracted to each picture frame using SIFT algorithms in practical operation.And pass through the pixel around characteristic point
Histograms of oriented gradients characteristic point is described.
In practical operation, after obtaining the feature histogram of each characteristic point, pass through Pasteur's distance 2 characteristic points of calculating
The distance between feature histogram, i.e., the similarity of two characteristic points.The calculation formula of wherein Pasteur's distance is as follows:
Wherein p and q represents 2 normalized histograms of process respectively.
Assuming that retain the best N number of characteristic point of stability, the then feature being extracted in the characteristic point that each frame is extracted
Point sum is individual for N × (T+1).
By be calculated all characteristic points of present frame and all characteristic points of previous frame between similarity after, can establish
One basic beneficial matrix.The size of the beneficial matrix is N × N × T, as shown in Figure 1.The matching result of beneficial matrix is feature
Motion track of the point in time space.
When a new mobile object appears in picture frame picture first, the characteristic point of the mobile object may be therewith
Any characteristic point of previous frame can not all be matched, therefore N × N × T former beneficial matrix is extended.Former beneficial matrix is extended to
N×(N×T+1).The element value of the expansion of beneficial matrix is default threshold value, as shown in Figure 2.The characteristic point of present frame with
Threshold portion when matching, then this characteristic point is considered as belonging to new mobile object.
In actual match, previous frame in when belonging in the characteristic point of mobile object in the current frame without match point, before
Moving body in frame is considered as being blocked.The characteristic point being blocked is not within a certain period of time again by current time picture frame
Feature Points Matching, this feature point is considered as disappearing from picture.Worked as again after the mobile object being tracked is being blocked
Feature Points Matching in preceding moment picture frame, this feature point are considered as the appearance again after blocking.
The matching example of the object that is actually blocked is provided in Fig. 3, the figure at characteristic point a and the t-3 moment in present frame t
As the Feature Points Matching success in frame, then show that characteristic point a is blocked in t-2 the and t-3 moment by barrier, and in t
Occur once again.
Fig. 4 provides the dead reckoning example of mobile object when being blocked by barrier, and the object of figure intermediate cam shape is in t+2
Moment is blocked, position 3 before being blocked according to it and the position 5 after being blocked can do linear reckoning, and it is when being blocked
Position is 4.
It should be noted last that the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted.This area
It will be appreciated by the skilled person that modified to technical scheme or equivalent substitution, such as the solution of optimization problem,
Calculating of feature difference etc. all should be covered in claim of the invention without departure from the spirit and scope of technical solution of the present invention
Among scope.
Claims (5)
- A kind of 1. moving body track and extraction algorithm not influenceed by barrier, it is characterised in that:S01:The picture frame of current t is designated as framet, and the T frames before t are designated as framet-h, h=1 ... T }, and Characteristic point is extracted respectively, and the feature point number of each frame is designated as N;S02:Then frame is calculatedtEach characteristic point and framet-h, the feature before h={ 1 ... T } each characteristic point is empty Between space length, establish a size be N × N × T original beneficial matrix;S03:Establish the expansion benefits matrix that a size is N × (T × N+1), including the original benefit square that T size is N × N The threshold matrix that battle array and a size are N × N;S04:Using the matching problem of beneficial matrix as a Solving Assignment Problem;S05:Whether it is new mobile object according to different matching result judging characteristic points, if the characteristic point and threshold of present frame The threshold value of value matrix matches, then this characteristic point is considered as belonging to new mobile object;S06:Whether disappeared according to continuous matching result judging characteristic point, when previous frame in belong in the characteristic point of mobile object In the current frame without can match point when, previous frame in mobile object be considered as being blocked or disappear;S07:Whether it is to occur again after blocking according to continuous matching result judging characteristic point;S08:Position of the characteristic point occurred again when blocking after being blocked according to the estimation of continuous matching result.
- A kind of 2. moving body track and extraction algorithm not influenceed by barrier according to claim 1, it is characterized in that N × (T × N+1) beneficial matrix includes the original beneficial matrix that T size is N × N and the threshold value square that a size is N × N Battle array.
- 3. a kind of moving body track and extraction algorithm not influenceed by barrier according to claim 1, it is characterized in that Its optimization matching is mathematically represented as:Minimize<mrow> <mi>C</mi> <mo>=</mo> <munderover> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>N</mi> <mo>&times;</mo> <mrow> <mo>(</mo> <mi>T</mi> <mo>&times;</mo> <mi>N</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow> </munderover> <msub> <mi>p</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <msub> <mi>c</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> </mrow>s.t<mrow> <munderover> <mi>&Sigma;</mi> <mrow> <mi>j</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>N</mi> <mo>&times;</mo> <mrow> <mo>(</mo> <mi>T</mi> <mo>&times;</mo> <mi>N</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow> </munderover> <msub> <mi>p</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mi>i</mi> <mo>=</mo> <mo>{</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>N</mi> <mo>}</mo> <mo>,</mo> </mrow><mrow> <munderover> <mi>&Sigma;</mi> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msub> <mi>p</mi> <mrow> <mi>i</mi> <mi>j</mi> </mrow> </msub> <mo>&le;</mo> <mn>1</mn> <mo>,</mo> <mi>j</mi> <mo>=</mo> <mo>{</mo> <mn>1</mn> <mo>,</mo> <mn>2</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>N</mi> <mo>&times;</mo> <mrow> <mo>(</mo> <mi>T</mi> <mo>&times;</mo> <mi>N</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>}</mo> <mo>,</mo> </mrow>pij={ 0,1 },<mrow> <mi>c</mi> <mo>=</mo> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <mi>s</mi> <mi>i</mi> <mi>m</mi> <mi>i</mi> <mi>l</mi> <mi>a</mi> <mi>r</mi> <mi>i</mi> <mi>t</mi> <mi>y</mi> <mo>{</mo> <mi>C</mi> <mi>u</mi> <mi>r</mi> <mi>r</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mi>i</mi> <mo>,</mo> <mi>Pr</mi> <mi>e</mi> <mi>v</mi> <msub> <mrow> <mo>(</mo> <mi>t</mi> <mo>-</mo> <mi>h</mi> <mo>)</mo> </mrow> <mi>j</mi> </msub> <mo>}</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>h</mi> <mo>=</mo> <mo>{</mo> <mn>1</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>T</mi> <mo>}</mo> <mo>,</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>i</mi> <mo>=</mo> <mo>{</mo> <mn>1</mn> <mo>,</mo> <mn>...</mn> <mi>N</mi> <mo>}</mo> <mo>,</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>j</mi> <mo>=</mo> <mo>{</mo> <mrow> <mo>(</mo> <mi>h</mi> <mo>-</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&times;</mo> <mi>N</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mi>h</mi> <mo>&times;</mo> <mi>N</mi> <mo>}</mo> <mo>,</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>T</mi> <mi>h</mi> <mi>r</mi> <mi>e</mi> <mi>s</mi> <mi>h</mi> <mi>o</mi> <mi>l</mi> <mi>d</mi> <mi> </mi> <mi>f</mi> <mi>o</mi> <mi>r</mi> <mi> </mi> <mi>c</mi> <mi>r</mi> <mi>e</mi> <mi>a</mi> <mi>t</mi> <mi>i</mi> <mi>n</mi> <mi>g</mi> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>i</mi> <mo>=</mo> <mo>{</mo> <mn>1</mn> <mo>,</mo> <mn>...</mn> <mi>N</mi> <mo>}</mo> <mo>,</mo> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <mi>j</mi> <mo>=</mo> <mo>{</mo> <mrow> <mo>(</mo> <mi>T</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&times;</mo> <mi>N</mi> <mo>+</mo> <mn>1</mn> <mo>,</mo> <mn>...</mn> <mo>,</mo> <mrow> <mo>(</mo> <mi>T</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&times;</mo> <mi>N</mi> <mo>}</mo> <mo>.</mo> </mrow> </mtd> </mtr> </mtable> </mfenced> </mrow>Wherein T be present frame before frame number, similarity { Curr (t)i, Prev (t-h)jFor present frame characteristic point i therewith Feature space space length between the characteristic point j of previous frame, threshold for creating are default threshold value.
- 4. a kind of moving body track and extraction algorithm not influenceed by barrier according to claim 3, it is characterized in that When the mobile object being tracked is being blocked, its position can do linear estimation by blocking front and rear position.
- 5. a kind of moving body track and extraction algorithm not influenceed by barrier according to claim 1, it is characterized in that When the mobile object being tracked is being blocked, the tracking to object will not be interrupted.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510020446.5A CN104504724B (en) | 2015-01-15 | 2015-01-15 | A kind of moving body track and extraction algorithm not influenceed by barrier |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510020446.5A CN104504724B (en) | 2015-01-15 | 2015-01-15 | A kind of moving body track and extraction algorithm not influenceed by barrier |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104504724A CN104504724A (en) | 2015-04-08 |
CN104504724B true CN104504724B (en) | 2018-04-06 |
Family
ID=52946118
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510020446.5A Expired - Fee Related CN104504724B (en) | 2015-01-15 | 2015-01-15 | A kind of moving body track and extraction algorithm not influenceed by barrier |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104504724B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106803264A (en) * | 2015-11-25 | 2017-06-06 | 天津工业大学 | A kind of image matching method for multiple target objects volume tracing |
WO2017147747A1 (en) * | 2016-02-29 | 2017-09-08 | SZ DJI Technology Co., Ltd. | Obstacle avoidance during target tracking |
CN106096508B (en) * | 2016-05-30 | 2019-09-13 | 无锡天脉聚源传媒科技有限公司 | The method and device that target is covered is determined in a kind of image |
CN106056630B (en) * | 2016-06-06 | 2018-09-21 | 南昌航空大学 | Occlusion area detection method based on image sequence light stream and triangle gridding |
CN108304766B (en) * | 2017-12-12 | 2019-01-04 | 交通运输部规划研究院 | A method of dangerous material stockyard is screened based on high-definition remote sensing |
CN110293551B (en) * | 2018-03-21 | 2020-10-16 | 北京猎户星空科技有限公司 | Visual sensor installation position determining method and device and electronic equipment |
CN111241979B (en) * | 2020-01-07 | 2023-06-23 | 浙江科技学院 | Real-time obstacle detection method based on image feature calibration |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101246547A (en) * | 2008-03-03 | 2008-08-20 | 北京航空航天大学 | Method for detecting moving objects in video according to scene variation characteristic |
CN102456225A (en) * | 2010-10-22 | 2012-05-16 | 深圳中兴力维技术有限公司 | Video monitoring system and moving target detecting and tracking method thereof |
CN103037140A (en) * | 2012-12-12 | 2013-04-10 | 杭州国策商图科技有限公司 | Target tracing algorithm with fortissimo robustness and based on block matching |
-
2015
- 2015-01-15 CN CN201510020446.5A patent/CN104504724B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101246547A (en) * | 2008-03-03 | 2008-08-20 | 北京航空航天大学 | Method for detecting moving objects in video according to scene variation characteristic |
CN102456225A (en) * | 2010-10-22 | 2012-05-16 | 深圳中兴力维技术有限公司 | Video monitoring system and moving target detecting and tracking method thereof |
CN103037140A (en) * | 2012-12-12 | 2013-04-10 | 杭州国策商图科技有限公司 | Target tracing algorithm with fortissimo robustness and based on block matching |
Also Published As
Publication number | Publication date |
---|---|
CN104504724A (en) | 2015-04-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104504724B (en) | A kind of moving body track and extraction algorithm not influenceed by barrier | |
Bergmann et al. | Tracking without bells and whistles | |
Beery et al. | Context r-cnn: Long term temporal context for per-camera object detection | |
CN106650620B (en) | A kind of target person identification method for tracing using unmanned plane monitoring | |
Ojha et al. | Image processing techniques for object tracking in video surveillance-A survey | |
US10140521B2 (en) | Method, system and apparatus for processing an image | |
Risse et al. | Visual tracking of small animals in cluttered natural environments using a freely moving camera | |
CN110569785B (en) | Face recognition method integrating tracking technology | |
Sun et al. | Moving foreground object detection via robust SIFT trajectories | |
Deilamani et al. | Moving object tracking based on mean shift algorithm and features fusion | |
Fazli et al. | Particle filter based object tracking with sift and color feature | |
Kothiya et al. | A review on real time object tracking in video sequences | |
Martin et al. | Multiple-object tracking in cluttered and crowded public spaces | |
Zhang et al. | A fast and robust people counting method in video surveillance | |
Ramos et al. | Fast-forward video based on semantic extraction | |
CN102194270B (en) | Statistical method for pedestrian flow based on heuristic information | |
Delibasoglu | UAV images dataset for moving object detection from moving cameras | |
Wang et al. | Video stabilization based on high degree b-spline smoothing | |
CN106023252A (en) | Multi-camera human body tracking method based on OAB algorithm | |
Delibaşoğlu | PESMOD: Small moving object detection benchmark dataset for moving cameras | |
Ladjailia et al. | Automated detection of similar human actions using motion descriptors | |
Ben Shitrit et al. | Tracking multiple players using a single camera | |
Figueiredo et al. | A video tensor self-descriptor based on block matching | |
Wang et al. | Multiple object tracking by multi-feature combination based on min-cost network flow | |
Fawzy et al. | 2DHOOF-2DPCA contour based optical flow algorithm for human activity recognition |
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 | ||
DD01 | Delivery of document by public notice | ||
DD01 | Delivery of document by public notice |
Addressee: HANGZHOU GUOCE MAP TECHNOLOGY Co.,Ltd. Person in charge of patents Document name: payment instructions |
|
DD01 | Delivery of document by public notice | ||
DD01 | Delivery of document by public notice |
Addressee: HANGZHOU GUOCE MAP TECHNOLOGY Co.,Ltd. Person in charge of patents Document name: Notice of Termination of Patent Rights |
|
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180406 |