CN103258332A - Moving object detection method resisting illumination variation - Google Patents
Moving object detection method resisting illumination variation Download PDFInfo
- Publication number
- CN103258332A CN103258332A CN2013102003724A CN201310200372A CN103258332A CN 103258332 A CN103258332 A CN 103258332A CN 2013102003724 A CN2013102003724 A CN 2013102003724A CN 201310200372 A CN201310200372 A CN 201310200372A CN 103258332 A CN103258332 A CN 103258332A
- Authority
- CN
- China
- Prior art keywords
- pixel
- moving object
- model
- gaussian profile
- background
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a moving object detection method resisting illumination variation. An existing moving object detection method is sensitive to the illumination variation frequently, and therefore when illumination variation occurs in a detection environment, accurate detection for a moving object is difficult. The moving object detection method resisting the illumination variation integrates an asymmetrical inter-frame difference method and a self-adaptation mixed gauss background difference method in inter-frame difference. The moving object detection method resisting the illumination variation includes a first step of preprocessing an input moving image, a second step of respectively conducting asymmetrical inter-frame difference moving object detection and self-adaptation mixed gauss background difference moving object detection, and a third step of comparing moving object results independently obtained through the two detection methods, calculating the moving object results, extracting a comprehensive moving object, filling cavities inside the moving object and filtering out some small interference through mathematics morphological processing. Through the moving object detection method resisting the illumination variation, the moving object can be accurately detected under the normal condition, and meanwhile the accuracy rate can reach over 90% under the condition that the illumination variation of the detecting environment occurs. Therefore, the moving object detection method resisting the illumination variation can meet needs of a monitoring system.
Description
Technical field
The invention belongs to image processing field in video monitoring system, it is related to a kind of moving target detecting method of illumination variation in anti-detection environment.
Background technology
With developing rapidly for Digital Network technology, video image turns into the important carrier of information transmission.A large amount of, abundant movable informations that sequence of video images is included cause the great interest of people.Although the eyes of people can directly differentiate the object of motion from sequence of video images, extract the information of motion, the demand of social development can not be met to obtain with working motion information by relying solely on the natural intelligence of the mankind.Human vision is substituted with computer vision, extract, analyze and understand that movable information turns into a popular direction in modern scientific research field from image sequence, the detection of Moving Object in Video Sequences all has very great value as the basic link of computer vision motion analysis in theoretical research and practical application.
The moving object detection of sequence of video images is the analysis to bottom video information, with Digital Signal Processing, remove the background of image sequence, extract people's foreground moving object of interest and its movable information of carrying, it is to realize motion target tracking, target classification and the basis of analysis, therefore the detection of moving target is a vital step, decides the quality of subsequent result.Existing moving object detection algorithm can be summarized as four kinds：Optical flow method, frame differential method, the method for feature based and Background difference.Optical flow method is the size and Orientation that each pixel motion is calculated according to consecutive image frame sequence, so that it is determined that background and Moving Objects；Inter-frame difference is the change of the image pixel intensities between inspection consecutive frame in a shorter time range, and non-zero pixels are considered as what Moving Objects were caused；Characteristic matching method, the method for feature based includes two main steps：One is the extraction feature from adjacent two width or several images not in the same time, and sets up correspondence；Two be come the structure (shape, position etc.) for calculating object and motion according to the corresponding relation between these features.Background difference, it is to be compared the intensity of each pixel of present frame one by one with the permission value in its background model, and the pixel for not meeting model is considered as Moving Objects pixel.
Video motion analysis is developed so far, and produces the algorithm of many moving object detection and trackings, but most of algorithm is proposed for specific scene, is had some limitations in terms of versatility.Actual scene is complicated, weather, the change of illumination condition, the shade of associated movement target, and other objects are movable into and out, and mutually blocking between target and the requirement to algorithm real-time etc. all cause the difficulty of moving object detection and tracking.So, the sane accurate and high performance moving object detection of research and track algorithm are still a challenging problem.
The content of the invention
There is provided a kind of detection method of the moving target of resisting illumination variation in view of the shortcomings of the prior art by the present invention.The technical solution adopted for solving the technical problem of the present invention mainly includes two large divisions：Asymmetric inter-frame difference and ADAPTIVE MIXED Gaussian Background difference.Mainly include the following steps that：
Step (1) is pre-processed to video image.
Step (2) detects moving target to the video image after pretreatment using asymmetric frame differential method.
Step (3) detects moving target to the video image after pretreatment using ADAPTIVE MIXED Gaussian Background calculus of finite differences.
The result that step (4) is individually detected to step (2) and step (3) is carried out and computing, and then extracts comprehensive moving target.
The integrated motion target that step (5) is obtained to step (4) carries out morphology processing, the result of further optimizing detection.
Beneficial effects of the present invention：
（1）The problem of for detecting illumination variation in environment, the present invention proposes a kind of effective solution, utilize the complementarity of two kinds of algorithms of asymmetric inter-frame difference and adaptive background difference, moving target can be accurately detected in the case of illumination variation, accuracy rate is up to more than 90%, disclosure satisfy that demand.
（2）This method is not only pre-processed to image, and is converted into gray level image, has so been saved the memory headroom of consumption, has finally also been post-processed so that detection becomes apparent from, accurately.
（3）Used in this method in the background difference algorithm of self adaptation, the algorithm and propose the viewpoint of oneself, so that background model is timely updated, in favor of the accurate detection of target.
Brief description of the drawings
Fig. 1 is main flow chart of the invention；
Fig. 2 is the flow chart of step (2) in the present invention；
Fig. 3 is the flow chart of step (3) in the present invention.
Embodiment
Below in conjunction with accompanying drawing, the invention will be further described.
As shown in figure 1, the inventive method comprises the following steps：
Step (1) is pre-processed to video image, main to include carrying out denoising, several processing procedures such as gradation conversion and image sharpening to image.
Step (2) detects moving target on the basis that step (1) is handled with asymmetric inter-frame difference.
Step (3) detects moving target on the basis that step (1) is handled with the method for ADAPTIVE MIXED Gaussian Background difference.
Step (4) is after step (2) and step (3) detection, the testing result bianry image of step (2) and the testing result bianry image of step (3) are carried out, with operating, finally advanceing to integrated motion target.
Operations of step (5) Jing Guo step (4), is post-processed to its result, and the process is mainly carry out morphology processing, and the cavity filled up inside moving target filters some noise spots of surrounding, obtains last testing result.
The specific method that asymmetric target is detected in step (2) is as described below, and flow chart is as shown in 2：
If
Represent current
Frame
The pixel value at place, corresponding former frame is
, a later frame is
, the two-value difference image of present frame and former frame is
, the two-value difference image of former frame and a later frame is
,
It is the threshold value of setting, then to the
The asymmetric calculus of differences of two field picture is expressed as follows：
(1)
Formula (1) shows only have
=1 HeDuring=1 establishment simultaneously,
=1 just sets up.It can so eliminate the displaying background in bianry image, obtain theMotion target area in two field picture.
The specific method of ADAPTIVE MIXED Gauss target detection is as described below in step (3), and particular flow sheet is as shown in Figure 3：
Mixed Gauss model is a kind of multi-modal background model, and its basic thought is：To each pixel, while withIndividual complementary related Gauss model describes its state jointly,
Span be 3 ~ 7, each Gauss model has respective average, variance and weight.As long as pixel meets any one in the Gaussian Profile of K expression background during the detection, then being considered as the pixel has the pixel of background characteristics, that is, background pixel point；Conversely, the pixel is then judged as target.
Moment pixel
Probability be expressed as：
In formula
What is represented is at (x, y) point theThe of frame
Individual Gaussian Profile, it is defined as shown：
N represents vector in formula (3)
Dimension,
、
、
It is at point (x, y) place respectively
FrameAverage, covariance matrix and the weight of individual Gaussian Profile, and
, because gray level image is single pass, port number n=1 when background modeling is carried out with mixed Gauss model to gray level image, and assign larger variances and less weights to them as average with the pixel value of piece image every.
According to the background model having built up, to each pixel in the frame, its pixel value is matched respectively with K Gaussian Profile in mixed Gauss model, if one of them meets formula (4), then it is assumed that
Matched with the background model.Wherein D is the value for being empirically for determination similarity and setting, span
.
RepresentPlace the
In Gauss model
The standard deviation of individual Gaussian Profile；
If the pixel value is matched with its background model, its average, variance and weight are updated according to formula (5), to update background model.But if all matched with any one Gaussian Profile, it is believed that distribution of the pixel value on single model does not have any influence, not more in new model each Gaussian Profile parameter.
If any one Gaussian Profile in the pixel value and model in new video frame is all unsatisfactory for formula (4), it is believed that occur in that new distribution form, it is necessary to add new distribution in original model.If added after new distribution, the number of Gaussian Profile is more than K, then to K Gaussian Profile in the model according to
Size be ranked up, replaced with the distribution of current pixel
It is worth minimum distribution, using the pixel value of the point as average, while assigning a larger variance and less weight。
After model modification, 1 is not necessarily after the weight of each pixel Gaussian Profile, so doing normalized according to formula (6) to the weight of each Gaussian Profile of new model:
After the Gaussian Profile in each pixel point model is ranked up, if precedingThe cumulative probability of individual state be more than T and
When minimum, then it is assumed that they are background states, remaining state is determined as prospect, so drives off moving target.If the ratio between the number of object pixel and sum of all pixels mesh are more than 0.85, then think that ambient lighting intensity changes, parameter to the maximum Gaussian Profile of priority in all pixels point Gauss model is updated, average is replaced with the pixel value of current frame pixel point, variance assigns a higher value, the maximum value of weight in weight modulus type, so that new distribution turns into uncertain background, the accurate detection to next two field picture.
(7)
B is the result that each pixel is judged as prospect or background.For the Gauss model of each pixel, when according to threshold value T, there is a corresponding b, then B=1, represents that the pixel is identified as background, otherwise B=0, represents that the pixel is identified as prospect.
The specific method of step (4) is：
For
Two field picture set the result of step (2) as
, the result of step (3) is
, then by
The integrated motion target of frame detection
For：
Formula (8) shows only have
With
When being 0 simultaneously
It is just 0.Thus obtain the motion target area of synthesis with reference to the result of step (2) and the result of step (3), and the shortcoming that complementary step (2) and step (3) are individually detected.
Claims (4)
1. a kind of method of the anti-moving object detection for looking after change, it is characterised in that this method comprises the following steps：
Step (1) is pre-processed to video image；
Step (2) detects moving target to the video image after pretreatment using asymmetric frame differential method；
Step (3) detects moving target to the video image after pretreatment using ADAPTIVE MIXED Gaussian Background calculus of finite differences；
The result that step (4) is individually detected to step (2) and step (3) is carried out and computing, and then extracts comprehensive moving target；
The integrated motion target that step (5) is obtained to step (4) carries out morphology processing, the result of further optimizing detection.
2. moving target detecting method according to claim 1, it is characterised in that：Motion target area is individually detected with asymmetric frame differential method in step (2), its specific method is：
If
Represent current
Frame
The pixel value at place, corresponding former frame is
, a later frame is
, the two-value difference image of present frame and former frame is, the two-value difference image of former frame and a later frame is
,
It is the threshold value of setting, then to theThe asymmetric calculus of differences of two field picture is expressed as follows：
Formula (1) shows only have
=1 He
During=1 establishment simultaneously,
=1 just sets up, and can so eliminate the displaying background in bianry image, obtains the
Motion target area in two field picture.
3. the detection method of moving target according to claim 1, it is characterised in that：Moving target is individually detected with ADAPTIVE MIXED Gaussian Background calculus of finite differences in step (3), its specific method is as follows：
1) sets up background model：To each pixel, while with
Individual complementary related Gauss model describes its state jointly,
Span be 3 ~ 7, each Gauss model has respective average, variance and weight；As long as pixel meets during detectionIt is individual represent background Gaussian Profile in any one, then being considered as the pixel has the pixel of background characteristics, that is, background pixel point；Conversely, the pixel is then judged as target；
Moment pixel
Probability be expressed as：
In formula
Represent be at (x, y) point,
The of moment
Individual Gaussian Profile, it is defined as shown：
N represents vector in formula
Dimension,
、
、
It is at point (x, y) place respectively
Frame
Average, covariance matrix and the weight of individual Gaussian Profile, and；
2) model modifications：According to the background model having built up, to each pixel in the frame, its pixel value is matched respectively with K Gaussian Profile in mixed Gauss model, if one of them meets formula (4), then it is assumed thatMatched with the background model, wherein D is the value for being empirically for determination similarity and setting；
If 1. the pixel values and any one Gaussian Profile all meet formula (4), then it is assumed that distribution of the pixel value on single model does not have any influence, not each Gaussian Distribution Parameters more in new model；
If 2. the pixel values meet formula (4) with a Gaussian Profile in model, the average, variance and weight of the Gaussian Profile in model are updated according to formula；
If 3. any one Gaussian Profile in the pixel value and model in new video frames is all unsatisfactory for formula (4), it is believed that occur in that new distribution form, it is necessary to add new distribution in original model；If added after new distribution, the number of Gaussian Profile is more than K, then to K Gaussian Profile in the model according toSize be ranked up, replaced with the distribution of current pixel
It is worth minimum distribution, using the pixel value of the point as average, while assigning a larger variance and less weight
；
3) weights are normalized：After model modification, the weight of each pixel Gaussian Profile is not necessarily 1, so doing normalized according to formula (6) to the weight of each Gaussian Profile of new model:
4) segmentation of foreground targets：After the Gaussian Profile in each pixel point model is ranked up, if precedingThe cumulative probability of individual state be more than T andWhen minimum, then it is assumed that they are background states, remaining state is determined as prospect, so drives off moving target；If the ratio between the number of object pixel and sum of all pixels mesh are more than 0.85, then think that ambient lighting intensity changes, parameter to the maximum Gaussian Profile of priority in all pixels point Gauss model is updated, average is replaced with the pixel value of current frame pixel point, variance assigns a higher value, the maximum value of weight in weight modulus type, so that new distribution turns into uncertain background, the accurate detection to next two field picture；
(7)
B represents that each pixel is judged as the result of prospect or background.
4. the detection method of moving target according to claim 1, it is characterised in that：Step (4) is specifically：
For
Two field picture set the result of step (2) as
, the result of step (3) is
, then by
The integrated motion target of frame detection
For：
Formula (8) shows only have
With
When being 0 simultaneously
It is just 0；Thus obtain the motion target area of synthesis with reference to the result of step (2) and the result of step (3), and the shortcoming that complementary step (2) and step (3) are individually detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310200372.4A CN103258332B (en) | 2013-05-24 | 2013-05-24 | A kind of detection method of the moving target of resisting illumination variation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310200372.4A CN103258332B (en) | 2013-05-24 | 2013-05-24 | A kind of detection method of the moving target of resisting illumination variation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103258332A true CN103258332A (en) | 2013-08-21 |
| CN103258332B CN103258332B (en) | 2016-06-29 |
Family
ID=48962225
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310200372.4A Active CN103258332B (en) | 2013-05-24 | 2013-05-24 | A kind of detection method of the moving target of resisting illumination variation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103258332B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103674104A (en) * | 2013-10-28 | 2014-03-26 | 中国科学院长春光学精密机械与物理研究所 | Method for increasing tracking distance of photoelectric device |
| CN103793914A (en) * | 2014-02-17 | 2014-05-14 | 沈阳工业大学 | Method for extracting and updating static background in video with self-adaptation parameters |
| CN104268899A (en) * | 2014-09-22 | 2015-01-07 | 河海大学 | Moving object detection method based on frame difference and background difference |
| CN104301669A (en) * | 2014-09-12 | 2015-01-21 | 重庆大学 | Suspicious target detection tracking and recognition method based on dual-camera cooperation |
| CN105447890A (en) * | 2015-12-08 | 2016-03-30 | 南京航空航天大学 | Motion vehicle detection method resisting light effect |
| CN106205217A (en) * | 2016-06-24 | 2016-12-07 | 华中科技大学 | Unmanned plane automatic testing method based on machine vision and unmanned plane method of control |
| CN107766814A (en) * | 2017-10-18 | 2018-03-06 | 山东科技大学 | The recognition methods of crowd behaviour in a kind of video based on Adaboost algorithm |
| CN107798689A (en) * | 2016-08-28 | 2018-03-13 | 淮安信息职业技术学院 | Traffic video image background extracting method |
| CN108765461A (en) * | 2018-05-29 | 2018-11-06 | 北大青鸟环宇消防设备股份有限公司 | A kind of extraction of fire image block and recognition methods and its device |
| CN110049206A (en) * | 2015-04-28 | 2019-07-23 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN113362375A (en) * | 2021-06-16 | 2021-09-07 | 江铃汽车股份有限公司 | Moving object detection method for vehicle |
| CN115830515A (en) * | 2023-01-31 | 2023-03-21 | 中国电子科技集团公司第二十八研究所 | Video target behavior identification method based on spatial grid |
| CN117132948A (en) * | 2023-10-27 | 2023-11-28 | 南昌理工学院 | Scenic spot tourist flow monitoring method, system, readable storage medium and computer |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101159859A (en) * | 2007-11-29 | 2008-04-09 | 北京中星微电子有限公司 | Motion detection method, device and an intelligent monitoring system |
| CN101635852A (en) * | 2009-08-26 | 2010-01-27 | 北京航空航天大学 | Method for detecting real-time moving object based on adaptive background modeling |
| CN102184552A (en) * | 2011-05-11 | 2011-09-14 | 上海理工大学 | Moving target detecting method based on differential fusion and image edge information |
-
2013
- 2013-05-24 CN CN201310200372.4A patent/CN103258332B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101159859A (en) * | 2007-11-29 | 2008-04-09 | 北京中星微电子有限公司 | Motion detection method, device and an intelligent monitoring system |
| CN101635852A (en) * | 2009-08-26 | 2010-01-27 | 北京航空航天大学 | Method for detecting real-time moving object based on adaptive background modeling |
| CN102184552A (en) * | 2011-05-11 | 2011-09-14 | 上海理工大学 | Moving target detecting method based on differential fusion and image edge information |
Non-Patent Citations (2)
| Title |
|---|
| JIANLIN WANG ET AL: "A Moving Object Detection Algorithm based on Background Subtraction Model", 《2011 3RD IEEE INTERNATIONAL CONFERENCE ON INFORMATION MANAGEMENT AND ENGINEERING(ICIME 2011)》 * |
| 敖邦乾 等: "对光照变化鲁棒的一种运动目标检测方法", 《光电工程》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103674104A (en) * | 2013-10-28 | 2014-03-26 | 中国科学院长春光学精密机械与物理研究所 | Method for increasing tracking distance of photoelectric device |
| CN103674104B (en) * | 2013-10-28 | 2017-01-25 | 中国科学院长春光学精密机械与物理研究所 | Method for increasing tracking distance of photoelectric device |
| CN103793914A (en) * | 2014-02-17 | 2014-05-14 | 沈阳工业大学 | Method for extracting and updating static background in video with self-adaptation parameters |
| CN103793914B (en) * | 2014-02-17 | 2017-05-24 | 沈阳工业大学 | Method for extracting and updating static background in video with self-adaptation parameters |
| CN104301669A (en) * | 2014-09-12 | 2015-01-21 | 重庆大学 | Suspicious target detection tracking and recognition method based on dual-camera cooperation |
| CN104268899A (en) * | 2014-09-22 | 2015-01-07 | 河海大学 | Moving object detection method based on frame difference and background difference |
| CN110049206A (en) * | 2015-04-28 | 2019-07-23 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN110049206B (en) * | 2015-04-28 | 2021-08-10 | 腾讯科技（深圳）有限公司 | Image processing method, image processing apparatus, and computer-readable storage medium |
| CN105447890A (en) * | 2015-12-08 | 2016-03-30 | 南京航空航天大学 | Motion vehicle detection method resisting light effect |
| CN106205217B (en) * | 2016-06-24 | 2018-07-13 | 华中科技大学 | Unmanned plane automatic testing method based on machine vision and unmanned plane method of control |
| CN106205217A (en) * | 2016-06-24 | 2016-12-07 | 华中科技大学 | Unmanned plane automatic testing method based on machine vision and unmanned plane method of control |
| CN107798689A (en) * | 2016-08-28 | 2018-03-13 | 淮安信息职业技术学院 | Traffic video image background extracting method |
| CN107766814A (en) * | 2017-10-18 | 2018-03-06 | 山东科技大学 | The recognition methods of crowd behaviour in a kind of video based on Adaboost algorithm |
| CN108765461A (en) * | 2018-05-29 | 2018-11-06 | 北大青鸟环宇消防设备股份有限公司 | A kind of extraction of fire image block and recognition methods and its device |
| CN108765461B (en) * | 2018-05-29 | 2022-07-12 | 青鸟消防股份有限公司 | Fire-fighting fire image block extraction and identification method and device |
| CN113362375A (en) * | 2021-06-16 | 2021-09-07 | 江铃汽车股份有限公司 | Moving object detection method for vehicle |
| CN115830515A (en) * | 2023-01-31 | 2023-03-21 | 中国电子科技集团公司第二十八研究所 | Video target behavior identification method based on spatial grid |
| CN117132948A (en) * | 2023-10-27 | 2023-11-28 | 南昌理工学院 | Scenic spot tourist flow monitoring method, system, readable storage medium and computer |
| CN117132948B (en) * | 2023-10-27 | 2024-01-30 | 南昌理工学院 | Scenic spot tourist flow monitoring method, system, readable storage medium and computer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103258332B (en) | 2016-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103258332A (en) | Moving object detection method resisting illumination variation | |
| CN106157267B (en) | Image defogging transmissivity optimization method based on dark channel prior | |
| CN102184552B (en) | Moving target detecting method based on differential fusion and image edge information | |
| CN112036254B (en) | Moving vehicle foreground detection method based on video image | |
| CN103700116B (en) | Background modeling method for movement target detection | |
| CN105631831B (en) | Video image enhancing method under the conditions of a kind of haze | |
| CN105138987B (en) | A kind of vehicle checking method based on converging channels feature and estimation | |
| CN103729858B (en) | A kind of video monitoring system is left over the detection method of article | |
| CN104616290A (en) | Target detection algorithm in combination of statistical matrix model and adaptive threshold | |
| CN111062293B (en) | Unmanned aerial vehicle forest flame identification method based on deep learning | |
| CN110782477A (en) | Moving target rapid detection method based on sequence image and computer vision system | |
| CN103679677B (en) | A kind of bimodulus image decision level fusion tracking updating mutually based on model | |
| CN110428450B (en) | Scale-adaptive target tracking method applied to mine tunnel mobile inspection image | |
| CN110298297A (en) | Flame identification method and device | |
| CN107944354B (en) | Vehicle detection method based on deep learning | |
| CN109919053A (en) | A kind of deep learning vehicle parking detection method based on monitor video | |
| CN111127360B (en) | Gray image transfer learning method based on automatic encoder | |
| CN105741319B (en) | Improvement visual background extracting method based on blindly more new strategy and foreground model | |
| CN104268899A (en) | Moving object detection method based on frame difference and background difference | |
| CN105513053A (en) | Background modeling method for video analysis | |
| CN104598929A (en) | HOG (Histograms of Oriented Gradients) type quick feature extracting method | |
| CN103106796A (en) | Vehicle detection method and device of intelligent traffic surveillance and control system | |
| CN104717400A (en) | Real-time defogging method of monitoring video | |
| CN104933728A (en) | Mixed motion target detection method | |
| CN107122732B (en) | High-robustness rapid license plate positioning method in monitoring scene |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201217 Address after: No.477 Yongle Road, Wangdian Town, Xiuzhou District, Jiaxing City, Zhejiang Province Patentee after: ZHEJIANG OUTUO ELECTRICAL Co.,Ltd. Address before: 310018, No. 18 Jiao Tong Street, Xiasha Higher Education Park, Hangzhou, Zhejiang Patentee before: ZHEJIANG GONGSHANG University |
|
| TR01 | Transfer of patent right |