CN105046285B - A kind of abnormal behaviour discrimination method based on kinematic constraint - Google Patents
A kind of abnormal behaviour discrimination method based on kinematic constraint Download PDFInfo
- Publication number
- CN105046285B CN105046285B CN201510551661.8A CN201510551661A CN105046285B CN 105046285 B CN105046285 B CN 105046285B CN 201510551661 A CN201510551661 A CN 201510551661A CN 105046285 B CN105046285 B CN 105046285B
- Authority
- CN
- China
- Prior art keywords
- optical
- flow feature
- kinematic constraint
- svm
- abnormal behaviour
- 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
- 206010000117 Abnormal behaviour Diseases 0.000 title claims abstract description 27
- 238000012850 discrimination method Methods 0.000 title claims abstract description 12
- 238000000034 method Methods 0.000 claims abstract description 52
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 230000003287 optical effect Effects 0.000 claims abstract description 13
- 238000004458 analytical method Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 abstract description 3
- 238000012706 support-vector machine Methods 0.000 description 21
- 230000006399 behavior Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005021 gait Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of abnormal behaviour discrimination method based on kinematic constraint, belongs to field of video monitoring, including training process, training process includes the following steps：1, first group of video frame images of input are demarcated using quick calibrating method；2, background modeling is carried out to calibrated input video frame image, while Optical-flow Feature is extracted from frame image with optical flow method；3, the Optical-flow Feature in foreground is analyzed and is chosen with the kinematic constraint of pedestrian；4, selected Optical-flow Feature is learnt with SVM, obtains the SVM classifier model parameter that can be used for abnormal behaviour identification；Further include detection process, detection process input is second group of video, and the SVM parameters used are obtained in training process, for the detection and judgement to model parameter.Present invention optimizes Optical-flow Feature calculating, eliminate noise Optical-flow Feature, and the Optical-flow Feature with pedestrian is more notable.And the abnormal behaviours such as run can be judged.
Description
Technical field
The present invention relates to a kind of abnormal behaviour discrimination method based on kinematic constraint.
Background technology
When monitor environment it is a bit crowded but not serious when, can the work based on forefathers to being based on support vector machines (SVM)
Abnormal behaviour discrimination method be improved, such as middle tracking using robust can mutually block not serious between pedestrian
When in the environment of track 20 people or so, in be proposed for the human body image of defect into the method for line trace.
But in especially crowded environment, when the mutual coverage extent between pedestrian is more than certain limit, it can not obtain
It is whole or clearly pedestrian contour feature when, above method will encounter difficulties when understanding crowd behaviour, need at this time new
Method solve the problems, such as this.It is proposed to this end that the concept of kinematic constraint, and it has been combined with the study of Optical-flow Feature
Come, realizes abnormal behaviour identification and the monitoring under crowded environment.
Invention content
The technical solution that the present invention solves above-mentioned technical problem is as follows：
A kind of abnormal behaviour discrimination method based on kinematic constraint, including training process, training process include the following steps：
(1), input first group of video, input video frame image is pre-processed, including using quick calibrating method into
Fixed, denoising of rower etc.；
(2), background modeling is carried out to calibrated input video frame image, is then extracted from frame image with optical flow method
Optical-flow Feature；
(3), the Optical-flow Feature in foreground is analyzed and is chosen with the kinematic constraint of pedestrian；
(4), selected Optical-flow Feature is learnt with SVM, obtains the SVM classifier that can be used for abnormal behaviour identification.
(5), the parameter of SVM classifier model is write down.
Based on the above technical solution, the present invention can also be improved as follows.
Further, extract that Optical-flow Feature takes in the image from frame is Lucas-Kanade algorithms.
Further, the kinematic constraint model is established by Bayesian formula, and formula is:
P (u, v | I (x, y, t))=α P (u, v) Π P (I (xi,yi,ti)|u,v)
Wherein, P (u, v) is prior probability, represents the two dimensional motion speed probability of object, and α is the direct proportion factor, P (u, v |
I (x, y, t)) it is its two dimensional motion speed probability under the conditions of gray value when space (x, y) is in moment t, P (I (x, y, t) | u,
V) be the grey scale change of object during exercise probability, i.e. posterior probability.
Further, the Optical-flow Feature in foreground is analyzed and is selected with the kinematic constraint of pedestrian in the step 3
It takes, specifically includes following steps：
(1), from the foreground image after background detection, Optical-flow Feature is selected from the image-region of pedestrian；
(2), go out to meet the Optical-flow Feature of threshold requirement by kinematic constraint model discrimination, the threshold value is to posteriority
Probability P (I (x, y, t) | u, v) carries out the threshold value of Select to use.
Further, the range of the α between zero and one,
Further, the threshold value is optimized during the iterative solution of SVM modeling process, obtains optimal value.
Further, the invention also includes detection process, the step of detection process, is：
(1), input second group of video, input video frame image is pre-processed, including using quick calibrating method into
Fixed, denoising of rower etc.；
(2), background modeling is carried out to calibrated input video frame image, is then extracted from frame image with optical flow method
Optical-flow Feature；
(3), with the kinematic constraint of pedestrian, the threshold value obtained according to the training stage divides the Optical-flow Feature in foreground
Analysis and selection；
(4), the SVM classifier obtained with training process judges selected Optical-flow Feature, if to abnormal behaviour (thing
First have learned that the input is to belong to abnormal behaviour) correct judgment, then the provable grader is reliable.
The beneficial effects of the invention are as follows：The present invention is combined by choosing Optical-flow Feature with kinematic constraint, in training process
The SVM classifier that can be used for abnormal behaviour judgement is obtained, by input is second known to behavior classification in detection process
Group video, the model obtained to training are tested, the correctness and reliability of testing model, while can also be played to model
The effect being further generalized.Pedestrian movement's constraint is used for optimizing Optical-flow Feature selection by the present invention for the first time, has been effectively eliminated and has been made an uproar
Sound Optical-flow Feature and interested Optical-flow Feature (flow features of interest) is enhanced, and can be to running
Abnormal behaviour makes accurate judgement.
Description of the drawings
Fig. 1 is the method for the present invention schematic diagram；
Fig. 2 is the schematic diagram using Bayes Modeling；
Fig. 3 is the calculating for using the Optical-flow Feature result of calculation of kinematic constraint and not using the Optical-flow Feature of kinematic constraint
Comparative result figure.
Specific implementation mode
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the present invention.
A kind of abnormal behaviour identification system based on kinematic constraint, including preprocessing module, background modeling module, light stream carry
Modulus block, flow field selecting module, kinematic constraint module and SVM classifier module.
The preprocessing module is used to pre-process input video frame image, including calibration, denoising etc.；
The background modeling module is used to carry out background modeling to pretreated input video frame image；
The light stream extraction module is used for from extracting Optical-flow Feature in frame image after background modeling；
The kinematic constraint module is used for the kinematic constraint with pedestrian, and it is special to filter out qualified light stream according to threshold value
Sign,
The SVM classifier module is for learning selected Optical-flow Feature.In the training stage, pass through SVM algorithm
Iteration and Optimization Solution, obtain optimal model parameter and threshold value；In test phase, the model obtained with training is to known sample
This is tested, the accuracy and reliability of detection model, if can make standard to the behavior (including abnormal behaviour) of pedestrian
True judgement.
Embodiment 1
As shown in Figure 1, a kind of abnormal behaviour discrimination method based on kinematic constraint, including training process and detection process；
The training process includes the following steps：
(1), input first group of video, input video frame image is pre-processed, including using quick calibrating method into
Fixed, denoising of rower etc.；
(2), background modeling is carried out to calibrated input video frame image, is then extracted from frame image with optical flow method
Optical-flow Feature；
(3), with the kinematic constraint of pedestrian, the Optical-flow Feature in foreground is analyzed and is chosen according to certain threshold value；
(4), selected Optical-flow Feature is learnt with SVM, obtains the SVM classifier that can be used for abnormal behaviour identification.
(5), the parameter of SVM classifier model is write down.
1, Optical-flow Feature defines
Light stream is widely used in Computer Vision Task, such as recognition of face, gait modeling, object tracking, at present institute
Two kinds of optical flow algorithms are Horn-Schunck algorithms and Lucas-Kanade algorithms.Horn-Schunck algorithms have more smooth
Light stream, global information and accurate time diffusion, but it is relatively slow and have rough boundary profile.Lucas-
The boundary error of Kanade algorithms is more, but algorithm more simply can be calculated quickly.In view of calculating speed, we using
Lucas-Kanade algorithms.
When not considering kinematic constraint, it includes optical flow computation and light stream extraction step to obtain Optical-flow Feature step.It obtains at this time
Optical-flow Feature vector, be the vector for the movable information that characteristic point is implied in frame image, without containing any pedestrian movement constrain
Priori.
The generation of Optical-flow Feature vector is divided into two steps：
The first step finds the feature vector point of human body image block, such as the angle vertex on the head of some pedestrian and left shoulder.
Second step, by the feature vector point of the previous frame image corresponding to the feature vector found point, for example (,) it is same
The angle vertex on the head of pedestrian and left shoulder.
2, Optical-flow Feature extracts
All light streams in selected foreground image can write P=[p (1), p (2) ..., p (i)], i=1..N,
The number of middle p (i)=[Lx, Ly, Mx, My, α], light stream are N (can be calculated from optical flow algorithm), and Lx and Ly indicate to scheme respectively
As the abscissa value and ordinate value of an optical flow position in block, Mx and My indicate that corresponding light stream amplitude (is equivalent to speed respectively
Value) abscissa value and ordinate value, α indicate deflection angle value.Due to being different in each image light stream number in the block, institute
To also need to be standardized each image block, treated, and Optical-flow Feature can write Q=[p (x1),p
(x2),...,p(xi)], wherein p (xi) elected from p (i).Input parameter may finally use formula (2.1) to describe：
(2.1) I=[p (x1),p(x2),...,p(xi)], i=1..N
Selected Optical-flow Feature as the input of grader after, problem can be attributed to two classes or the classification of multiclass is asked
Topic.Since SVM is used as the grader of pedestrian movement's behavior, selected input parameter is image Optical-flow Feature position in the block
It sets, direction and speed etc..
3, the introducing of kinematic constraint
When not considering kinematic constraint, the Optical-flow Feature of pedestrian behavior is trained the input of grader directly as SVM.Although
Optical-flow Feature can play a role behavior identification, but since robustness of the optical flow method under complex environment be not high, need to introduce
Kinematic constraint.
It is found after the Optical-flow Feature to pedestrian under congested conditions is analyzed, the motor pattern of pedestrian meets related with people
Sports rule, compared with the vehicle that other moving objects such as travel, it is apparent different (movement of vehicle be it is a kind of just
Body moves, and movement velocity and track are all different with pedestrian)；And the motor pattern of pedestrian, especially walk under normal circumstances
Velocity characteristic, be have very much it is regular.Using bayesian criterion come to pedestrian movement's constraint modeling, as shown in Figure 2.
There are two steps for modeling process：
The first step, it is ensured that only select Optical-flow Feature (in crowded environment, it is difficult to judge one from the image-region of pedestrian
Whether pixel is located among the image-region of people, a kind of solution of this problem is using a kind of robust and quick
The algorithm of people, or the selected element from the foreground image after background detection are detected, the latter is the method that this method uses),
The ratio that the Optical-flow Feature chosen accounts for Optical-flow Feature sum is P (A)；
Second step goes out to meet the Optical-flow Feature of threshold requirement by kinematic constraint model discrimination, so that Optical-flow Feature
Calculating is more accurate, to judge whether selected pixel meets human motion constraint, before the Optical-flow Feature filtered out accounts for screening
The ratio of Optical-flow Feature be P (B), in the training process, P (A) and P (B) can reach one and dynamically balance；
In second step, this method models kinematic constraint using bayes method (Bayesian).Bayes method is used
In human motion analysis purpose be calculate given image in posterior probability.Posterior probability be by calculate t moment be located at (x,
Y) the gray scale I (x, y, t) at place and the two dimensional motion speed (u, v) of object obtain.
(3.1) P (u, v | I (x, y, t))=α P (u, v) P (I (x, y, t) | u, v)
In formula (3.1), if it is assumed that image is to be continuous independent from the different positions and time, then have：
(3.2) P (u, v | I (x, y, t))=α P (u, v) Π P (I (xi,yi,ti)|u,v)
Wherein, P (u, v) is prior probability, represents the two dimensional motion speed probability of object, and α is the direct proportion factor, P (u, v |
I (x, y, t)) it is its two dimensional motion speed probability under the conditions of gray value when space (x, y) is in moment t, it can be in optical flow computation
In obtain, P (I (x, y, t) | u, v) is the probability of the grey scale change of object during exercise, i.e. posterior probability, then according to threshold value
P (I (x, y, t) | u, v) is carried out to filter out qualified Optical-flow Feature, threshold value is arbitrary value in the starting stage；
Gone out by kinematic constraint model discrimination after meeting the Optical-flow Feature of threshold requirement, selected Optical-flow Feature is as SVM
The input and study of algorithm, in the iterative solution of SVM algorithm, threshold value and SVM model parameters can be optimized constantly, finally be obtained
Obtain optimal threshold value and SVM model parameters.
The method of the present invention further includes detection process, and the detection process includes the following steps：
(1), input second group of video, input video frame image is pre-processed, including using quick calibrating method into
Fixed, denoising of rower etc.；
(2), background modeling is carried out to calibrated input video frame image, is then extracted from frame image with optical flow method
Optical-flow Feature；
(3), with the kinematic constraint of pedestrian, the threshold value obtained according to the training stage divides the Optical-flow Feature in foreground
Analysis and selection；
(4), the SVM classifier that training process obtains judges selected Optical-flow Feature, if to abnormal behaviour (thing
First have learned that the input is to belong to abnormal behaviour) correct judgment, then the provable grader is reliable.
4, experimental result
Kinematic constraint is calculated the knot with a kind of abnormal behaviour monitoring by this experiment shows for Optical-flow Feature under crowded environment
Fruit.In experiment, the resolution of monitor video is 320 × 240, and average length of time is 30 minutes.For every section of 30 minutes figure
Picture, wherein be used for model training for first 10 minutes, be used for model measurement in latter 20 minutes.
Consider the Optical-flow Feature result of calculation of kinematic constraint.
Fig. 3 illustrates that the Optical-flow Feature result of calculation for using kinematic constraint, left figure are not use kinematic constraint
Result of calculation, right figure are the result of calculation using kinematic constraint.It can be seen in figure 3 that kinematic constraint optimizes Optical-flow Feature
It calculates, eliminates noise Optical-flow Feature, the Optical-flow Feature with pedestrian is more notable.Simultaneously it can also be seen that vehicle
Kinematic constraint and the kinematic constraint of pedestrian have apparent difference.
The foregoing is merely the preferable case study on implementation of the present invention, are not intended to limit the invention, all spirit in the present invention
Within principle, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
Claims (5)
1. a kind of abnormal behaviour discrimination method based on kinematic constraint, which is characterized in that including training process and detection process, institute
Training process is stated to include the following steps：
(1), first group of video is inputted, frame image is demarcated using quick calibrating method；
(2), background modeling is carried out to calibrated input video frame image, then extracts light stream from frame image with optical flow method
Feature；
(3), the Optical-flow Feature in foreground is analyzed and is chosen with the kinematic constraint of pedestrian；
(4), selected Optical-flow Feature is learnt with SVM, obtains the SVM classifier that can be used for abnormal behaviour identification；
(5), the parameter of SVM classifier model is write down；
The kinematic constraint is modeled by Bayesian formula, and the Bayesian formula is:
P (u, v | I (x, y, t))=α P (u, v) Π P (I (xi,yi,ti)|u,v)
Wherein, P (u, v) is prior probability, represents the two dimensional motion speed probability of object, and α is the direct proportion factor, P (u, v | I (x,
Y, t)) it is its two dimensional motion speed probability under the conditions of gray value when space (x, y) is in moment t, P (I (x, y, t) | u, v) be
The probability of the grey scale change of object during exercise, i.e. posterior probability；
The Optical-flow Feature in foreground is analyzed and being chosen with the kinematic constraint of pedestrian in the step (3), specifically includes
Following steps：
(1), from the foreground image after background detection, Optical-flow Feature is selected from the image-region of pedestrian；
(2), go out to meet the Optical-flow Feature of threshold requirement by kinematic constraint model discrimination, the threshold value is to posterior probability
P (I (x, y, t) | u, v) carries out the threshold value of Select to use.
2. utilizing the abnormal behaviour discrimination method described in claim 1 based on kinematic constraint, which is characterized in that described from frame figure
Extract that Optical-flow Feature takes as in is Lucas-Kanade algorithms.
3. utilizing the abnormal behaviour discrimination method described in claim 1 based on kinematic constraint, which is characterized in that the model of the α
It encloses between zero and one.
4. utilizing the abnormal behaviour discrimination method described in claim 1 based on kinematic constraint, which is characterized in that the threshold value is
It is optimized during the iterative solution of SVM modeling process, obtains optimal value.
5. utilizing the abnormal behaviour discrimination method described in claim 1 based on kinematic constraint, which is characterized in that described to detect
The step of journey is：
(1), second group of video is inputted, input video frame image is demarcated using quick calibrating method；
(2), background modeling is carried out to calibrated input video frame image, while light stream is extracted from frame image with optical flow method
Feature；
(3), with the kinematic constraint of pedestrian, the threshold value obtained according to the training stage, to the Optical-flow Feature in foreground carry out analysis and
It chooses；
(4), the SVM model parameters for obtaining training process input SVM models, are carried out to selected Optical-flow Feature with SVM models
Judge.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510551661.8A CN105046285B (en) | 2015-08-31 | 2015-08-31 | A kind of abnormal behaviour discrimination method based on kinematic constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510551661.8A CN105046285B (en) | 2015-08-31 | 2015-08-31 | A kind of abnormal behaviour discrimination method based on kinematic constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105046285A CN105046285A (en) | 2015-11-11 |
| CN105046285B true CN105046285B (en) | 2018-08-17 |
Family
ID=54452814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510551661.8A Expired - Fee Related CN105046285B (en) | 2015-08-31 | 2015-08-31 | A kind of abnormal behaviour discrimination method based on kinematic constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105046285B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105469054B (en) * | 2015-11-25 | 2019-05-07 | 天津光电高斯通信工程技术股份有限公司 | The model building method of normal behaviour and the detection method of abnormal behaviour |
| CN110222616B (en) * | 2019-05-28 | 2021-08-31 | 浙江大华技术股份有限公司 | Pedestrian abnormal behavior detection method, image processing device and storage device |
| CN111046797A (en) * | 2019-12-12 | 2020-04-21 | 天地伟业技术有限公司 | Oil pipeline warning method based on personnel and vehicle behavior analysis |
| CN113762027B (en) * | 2021-03-15 | 2023-09-08 | 北京京东振世信息技术有限公司 | Abnormal behavior identification method, device, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271527A (en) * | 2008-02-25 | 2008-09-24 | 北京理工大学 | Exception action detecting method based on athletic ground partial statistics characteristic analysis |
| CN102663429A (en) * | 2012-04-11 | 2012-09-12 | 上海交通大学 | Method for motion pattern classification and action recognition of moving target |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101733131B1 (en) * | 2010-12-14 | 2017-05-10 | 한국전자통신연구원 | 3D motion recognition method and apparatus |
-
2015
- 2015-08-31 CN CN201510551661.8A patent/CN105046285B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101271527A (en) * | 2008-02-25 | 2008-09-24 | 北京理工大学 | Exception action detecting method based on athletic ground partial statistics characteristic analysis |
| CN102663429A (en) * | 2012-04-11 | 2012-09-12 | 上海交通大学 | Method for motion pattern classification and action recognition of moving target |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105046285A (en) | 2015-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mancini et al. | Fast robust monocular depth estimation for obstacle detection with fully convolutional networks | |
| CN106875424B (en) | A kind of urban environment driving vehicle Activity recognition method based on machine vision | |
| CN106462976B (en) | Method for tracking shape in scene observed by asynchronous sensor | |
| Ryan et al. | Textures of optical flow for real-time anomaly detection in crowds | |
| US10482609B2 (en) | Optical flow determination system | |
| US8340357B2 (en) | Moving object detection method and moving object detection apparatus | |
| CN101339655B (en) | Visual sense tracking method based on target characteristic and bayesian filtering | |
| US20220004818A1 (en) | Systems and Methods for Evaluating Perception System Quality | |
| CN105046285B (en) | A kind of abnormal behaviour discrimination method based on kinematic constraint | |
| CN106803265A (en) | Multi-object tracking method based on optical flow method and Kalman filtering | |
| CN102609720B (en) | Pedestrian detection method based on position correction model | |
| CN104281853A (en) | Behavior identification method based on 3D convolution neural network | |
| CN108985204A (en) | Pedestrian detection tracking and device | |
| CN104992453A (en) | Target tracking method under complicated background based on extreme learning machine | |
| CN103177454A (en) | Dynamic image moving object detection method | |
| CN118096815B (en) | Road abnormal event detection system based on machine vision | |
| CN108629792A (en) | Laser eyepiece detection method and device based on background modeling Yu background difference | |
| CN105809954A (en) | Traffic event detection method and system | |
| CN109166137A (en) | For shake Moving Object in Video Sequences detection algorithm | |
| CN107085703A (en) | Merge face detection and the automobile passenger method of counting of tracking | |
| CN109740609A (en) | A kind of gauge detection method and device | |
| CN109636834A (en) | Video frequency vehicle target tracking algorism based on TLD innovatory algorithm | |
| CN112418149A (en) | Abnormal behavior detection method based on deep convolutional neural network | |
| Głowacz et al. | Video detection algorithm using an optical flow calculation method | |
| Min et al. | Coeb-slam: A robust vslam in dynamic environments combined object detection, epipolar geometry constraint, and blur filtering |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200709 Address after: 350015 No. 99 Denglong Road, Fuzhou Economic and Technological Development Zone, Fujian Province Patentee after: YANGO University Address before: 430000, No. 31, Wuhuan Avenue, Dongxihu District, Hubei, Wuhan Patentee before: WUHAN YINGSHI INTELLIGENT SCIENCE & TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180817 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |