CN103826071A - Three-dimensional camera shooting method for three-dimensional identification and continuous tracking - Google Patents
Three-dimensional camera shooting method for three-dimensional identification and continuous tracking Download PDFInfo
- Publication number
- CN103826071A CN103826071A CN201410087180.1A CN201410087180A CN103826071A CN 103826071 A CN103826071 A CN 103826071A CN 201410087180 A CN201410087180 A CN 201410087180A CN 103826071 A CN103826071 A CN 103826071A
- Authority
- CN
- China
- Prior art keywords
- video
- moving object
- camera shooting
- shooting method
- dimensional
- 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 239000007787 solid Substances 0.000 claims description 9
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 abstract 2
- 238000013459 approach Methods 0.000 description 4
- 239000000284 extract Substances 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 108010076504 Protein Sorting Signals Proteins 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000005021 gait Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Studio Devices (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a three-dimensional camera shooting method for three-dimensional identification and continuous tracking. The method comprises the steps of A. shooting video for the monitored moving object from different angles respectively by a lens No. 1 and a lens No. 2 of a camera; B. respectively carrying out feature extraction on the moving object and the static object according to the shot video, wherein the extracted features comprise the size, the speed, the moving direction and the color of the moving object as well as the geographic position, the size and the color of the static object; and C. outputting a set of video streaming data according to the video, and outputting a set of feature data according to the extracted features. After three-dimensional camera shooting method is adopted, three-dimensional identification and continuous tracking can be realized for the monitored moving object.
Description
Technical field
The present invention relates to electronic communication technology, in particular a kind of three-dimensional camera shooting method for solid identification and Continuous Tracking.
Background technology
For the analysis of dynamic image and finally identify moving target, mainly realize by two approach at present.
A kind of approach is the mechanism of imitating eyes imaging and recognition object, allow computer extract moving target from the two-dimentional consecutive image sequence obtaining, and the viewable portion that therefrom reconstructs three-dimensional body is to reach the object of recognition object, although have made some progress in this research on the one hand, but due to the research hysteresis of other side and the inherent shortcoming of computer (for example, computer system adopts two-dimentional logic at large, and eyes imaging and recognition object are not only two dimension), the progress that makes to be reduced completely and identified by this approach moving target is slow.
Another approach is image method identification moving target, the basic principle of image method identification kinematic parameter is: after imaging head object for (infrared or visible ray etc.) is taken in, the image signal sequence of formation is sent into computer, after the preliminary treatment to image, feature extraction, target identification, in consecutive image sequence, carry out Feature Points Matching, and then solve the kinematic parameter of target object, realize search, identification and tracking to target object.Once complete decision-making and the processing that just can be completed by servomechanism next step to the identification of target and kinematic parameter thereof.Image method identification is to start with from input picture and two aspects of target object, by the processing to input image sequence, it is mated with the image in target sample storehouse, to reach the object of identification target.
The method also exists defect, such as, Sample Storehouse may be very large, is unfavorable for real-time tracking and processing.
Therefore, there is defect in prior art, needs to improve.
Summary of the invention
Technical problem to be solved by this invention is: a kind of three-dimensional camera shooting method for solid identification and Continuous Tracking that can three-dimensionally identify, connect real time tracking motion object is provided.
Technical scheme of the present invention is as follows: a kind of three-dimensional camera shooting method for solid identification and Continuous Tracking, comprises the steps: A: adopt a camera lens and No. two camera lens moving object capture video video recordings to monitoring from different angles respectively of a video camera; B: according to the video record of taking, moving object and stationary body are carried out respectively to feature extraction, wherein, comprise size, speed, the direction of motion and the color of moving object, and the geographical position of stationary body, size and color; C: export a road video stream data according to video record, and, export a road characteristic according to the feature of extracting.
Be applied to technique scheme, in described three-dimensional camera shooting method, after step C, also perform step D: adopt N video camera repeating step A-C, wherein N is greater than 1 natural number.
Be applied to each technique scheme, in described three-dimensional camera shooting method, after step D, also perform step E: each video camera output video stream data He Yi road, Yi road characteristic is exported to respectively in a PC, PC is analyzed each data, what judge each video camera shooting is same moving object, and draws the mobile alignment of this moving object.
Be applied to each technique scheme, in described three-dimensional camera shooting method, before steps A, also perform step A0: the diverse geographic location that each video camera is fixed on to a setting regions.
Be applied to each technique scheme, in described three-dimensional camera shooting method, in step B, specifically adopt GPS/ Big Dipper chip and direction sensor to carry out feature extraction to moving object.
Be applied to each technique scheme, in described three-dimensional camera shooting method, in step B, specifically adopt GPS/ Big Dipper chip and compass to carry out feature extraction to moving object
Adopt such scheme, the present invention carries out capture video video recording from different angles to moving object respectively by a camera lens and No. two camera lenses of video camera, then, according to video record, moving object and stationary body are carried out respectively to feature extraction again, so, can obtain the characteristic of moving object and stationary body according to the feature of extracting, and can obtain the video stream data of moving object according to the video record of taking, so, by characteristic and video stream data, can realize the solid identification to moving object; Can form a guarded region covering continuously and people or the vehicle to this region realized continuously from motion tracking by multiple video cameras, draw shiftable haulage line, can realize the Continuous Tracking of moving object.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of invention.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.
The present embodiment provides a kind of three-dimensional camera shooting method for solid identification and Continuous Tracking, this three-dimensional camera shooting method, main two camera lenses that arrange by video camera are distinguished capture video video recording, then according to video record, realize the feature extraction of moving object and stationary body, by video record and extraction feature are analyzed, realize the solid identification of moving object; And, take continuously by the video camera that arranges many, then analyze by PC, can realize the Continuous Tracking of moving object.
Wherein, as shown in Figure 1, three-dimensional camera shooting method comprises the steps:
First, execution step A: adopt a camera lens and No. two camera lens moving object capture video video recordings to monitoring respectively of a video camera, for example, to be fixed on the position, control point of a certain guarded region with video camera, then absorb respectively the video record of moving object from different angles by the camera lens and No. two camera lenses that arrange on video camera.
And a camera lens and No. two camera lenses send to the video record of shooting in video camera DSP are set respectively, wherein, DSP(digital signal processor) be a kind of microprocessor of uniqueness, be the device of processing bulk information with digital signal; So, DSP processes respectively the video record of taking.
Then, execution step B: according to the video record of taking, moving object and stationary body are carried out respectively to feature extraction, wherein, by video camera, DSP to be set the moving object in video record and stationary body are carried out respectively to feature extraction, wherein, comprise size, speed, the direction of motion and the color of moving object, and comprise geographical position, size and the color of stationary body.
DSP is in the time analyzing, it is mainly the moving object imaging in the video record by analyzing a camera lens and No. two camera lens different angles shootings, can determine size and the color of moving object, and the size of stationary body and color, and, GPS/ Big Dipper chip and direction sensor can also be set in video camera, wherein, direction sensor can be set to compass, GPS/ Big Dipper chip is connected with DSP respectively with direction sensor, so, by GPS/ Big Dipper chip and DSP, can calculate according to video record the speed of moving object, and the geographical position of stationary body, and, also be connected with DSP by direction sensor, can obtain the direction of moving object.
Again then, execution step C: DSP exports a road video stream data according to video record, and, export a road characteristic according to the feature of extracting, wherein, be connected with outside PC by DSP, by DSP, video stream data and special data uploaded respectively in PC, PC, through processes and displays, can be realized the solid identification of moving object again.
For example,, when Moving Objects is behaved, by human figure, height, the girth of a garment, gait and face feature extract, and from multi-angle, Moving Objects are implemented to feature and conclude, thereby provide condition for follow-up work, and for example, when Moving Objects is vehicle, its shape, number plate of vehicle are extracted, obtained the shape facilities such as the brand, color, feature, the number-plate number of car, so, by the shape facility extracting, it is carried out to solid identification.
And for example, after step C, also perform step D: adopt N video camera repeating step A-C, at a video camera, for example, each video camera can be arranged on to different geographical position, then camera lens by its setting and No. two camera lenses remove the video record of taking moving object to the video camera by diverse geographic location from different angles respectively, then the DSP arranging by corresponding video camera, GPS/ Big Dipper chip, and direction sensor extracts the feature of moving object and stationary body according to video record, wherein, N is greater than 1 natural number, in reality, the quantity of video camera can be set according to actual conditions, for example, can arrange three, six etc.
Or, after step D, also perform step E, record a video and extract after feature in N platform video camera capture video, each video camera is all exported video stream data He Yi road, Yi road characteristic and is exported to respectively in a PC, there are N road video stream data and N road characteristic to export in same PC, so, PC is analyzed each circuit-switched data, when the characteristic of moving object of extracting is while being consistent, what can judge the shooting of N platform video camera is same moving object, so, according to the speed of each moving objects in video and direction, can draw the mobile alignment of this moving object.
For example, with artificial example, after First video camera is found mobile people, can draw its track route according to its direction of motion and speed, when on second video camera when finder, the people's that two video cameras are found position will overlap, and, the features such as size, color are identical, then can be defined as same person, so First video camera will be transferred on second video camera this people's supervision.
If First video camera, second camera review are discontinuous, there is neutral gear centre, according to this people's movement velocity and direction, calculate its time that enters second video camera and orientation according to inertia, when this time really the correspondence position in second video camera there is people, and the shape facilities such as size, color are identical, can judge that this people and a video camera are same people, so transfer second camera surveillance to.
Or, before steps A, also perform step A0: the diverse geographic location that each video camera is fixed on to a setting regions, can pass through the moving region of predicted motion object in advance, then, in this region, multiple cameras is installed, be connected with PC respectively by multiple cameras, for example, can form a guarded region covering continuously and people or the vehicle to this region realized continuously from motion tracking by multiple video cameras, draw shiftable haulage line.
These are only preferred embodiment of the present invention, be not limited to the present invention, all any modifications of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (6)
1. for a three-dimensional camera shooting method for solid identification and Continuous Tracking, it is characterized in that, comprise the steps:
A: adopt a camera lens and No. two camera lens moving object capture video video recordings to monitoring from different angles respectively of a video camera;
B: according to the video record of taking, moving object and stationary body are carried out respectively to feature extraction, wherein, comprise size, speed, the direction of motion and the color of moving object, and the geographical position of stationary body, size and color;
C: export a road video stream data according to video record, and, export a road characteristic according to the feature of extracting.
2. three-dimensional camera shooting method according to claim 1, is characterized in that, after step C, also performs step D: adopt N video camera repeating step A-C, wherein N is greater than 1 natural number.
3. three-dimensional camera shooting method according to claim 2, it is characterized in that, after step D, also perform step E: each video camera output video stream data He Yi road, Yi road characteristic is exported to respectively in same PC, PC is analyzed each data, what judge each video camera shooting is same moving object, and draws the mobile alignment of this moving object.
4. three-dimensional camera shooting method according to claim 3, is characterized in that, before steps A, also performs step A0: the diverse geographic location that each video camera is fixed on to a setting regions.
5. three-dimensional camera shooting method according to claim 1, is characterized in that, in step B, specifically adopts GPS/ Big Dipper chip and direction sensor to carry out feature extraction to moving object.
6. three-dimensional camera shooting method according to claim 5, is characterized in that, in step B, specifically adopts GPS/ Big Dipper chip and compass to carry out feature extraction to moving object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410087180.1A CN103826071A (en) | 2014-03-11 | 2014-03-11 | Three-dimensional camera shooting method for three-dimensional identification and continuous tracking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410087180.1A CN103826071A (en) | 2014-03-11 | 2014-03-11 | Three-dimensional camera shooting method for three-dimensional identification and continuous tracking |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103826071A true CN103826071A (en) | 2014-05-28 |
Family
ID=50760872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410087180.1A Pending CN103826071A (en) | 2014-03-11 | 2014-03-11 | Three-dimensional camera shooting method for three-dimensional identification and continuous tracking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103826071A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105072312A (en) * | 2015-07-23 | 2015-11-18 | 柳州正高科技有限公司 | Method for predicting image moving direction in dynamic video |
| CN105072384A (en) * | 2015-07-23 | 2015-11-18 | 柳州正高科技有限公司 | Method for obtaining football moving images |
| CN105136064A (en) * | 2015-09-13 | 2015-12-09 | 维希艾信息科技（无锡）有限公司 | Moving object three-dimensional size detection system and method |
| CN105376523A (en) * | 2014-08-21 | 2016-03-02 | 思创影像科技股份有限公司 | Stereoscopic vision detection method and system |
| CN108205664A (en) * | 2018-01-09 | 2018-06-26 | 美的集团股份有限公司 | A kind of food recognition methods and device, storage medium, computer equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040125207A1 (en) * | 2002-08-01 | 2004-07-01 | Anurag Mittal | Robust stereo-driven video-based surveillance |
| US20070031037A1 (en) * | 2005-08-02 | 2007-02-08 | Microsoft Corporation | Stereo image segmentation |
| CN101344965A (en) * | 2008-09-04 | 2009-01-14 | 上海交通大学 | Tracking system based on binocular camera |
| CN101572804A (en) * | 2009-03-30 | 2009-11-04 | 浙江大学 | Multi-camera intelligent control method and device |
| CN102034247A (en) * | 2010-12-23 | 2011-04-27 | 中国科学院自动化研究所 | Motion capture method for binocular vision image based on background modeling |
| CN102436662A (en) * | 2011-11-29 | 2012-05-02 | 南京信息工程大学 | Human body target tracking method in nonoverlapping vision field multi-camera network |
| CN103024350A (en) * | 2012-11-13 | 2013-04-03 | 清华大学 | Master-slave tracking method for binocular PTZ (Pan-Tilt-Zoom) visual system and system applying same |
| US20130123801A1 (en) * | 2011-11-15 | 2013-05-16 | Macdonald Dettwiler & Associates | Method of real-time tracking of moving/flexible surfaces |
-
2014
- 2014-03-11 CN CN201410087180.1A patent/CN103826071A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040125207A1 (en) * | 2002-08-01 | 2004-07-01 | Anurag Mittal | Robust stereo-driven video-based surveillance |
| US20070031037A1 (en) * | 2005-08-02 | 2007-02-08 | Microsoft Corporation | Stereo image segmentation |
| CN101344965A (en) * | 2008-09-04 | 2009-01-14 | 上海交通大学 | Tracking system based on binocular camera |
| CN101572804A (en) * | 2009-03-30 | 2009-11-04 | 浙江大学 | Multi-camera intelligent control method and device |
| CN102034247A (en) * | 2010-12-23 | 2011-04-27 | 中国科学院自动化研究所 | Motion capture method for binocular vision image based on background modeling |
| US20130123801A1 (en) * | 2011-11-15 | 2013-05-16 | Macdonald Dettwiler & Associates | Method of real-time tracking of moving/flexible surfaces |
| CN102436662A (en) * | 2011-11-29 | 2012-05-02 | 南京信息工程大学 | Human body target tracking method in nonoverlapping vision field multi-camera network |
| CN103024350A (en) * | 2012-11-13 | 2013-04-03 | 清华大学 | Master-slave tracking method for binocular PTZ (Pan-Tilt-Zoom) visual system and system applying same |
Non-Patent Citations (2)
| Title |
|---|
| EMANUEL E.ZELNIKER .ET AL: "Global Abnormal Behaviour Detection Using a Network of CCTV Cameras", 《THE EIGHTH INTERNATIONAL WORKSHOP ON VISUAL SURVEILLANCE-VS2008，MARSEILLE：FRANCE》 * |
| TAO ZHAO，MANOJ AGGARWAL .ET AL: "Real-time Wide Area Multi-camera Stereo Tracking", 《IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105376523A (en) * | 2014-08-21 | 2016-03-02 | 思创影像科技股份有限公司 | Stereoscopic vision detection method and system |
| CN105072312A (en) * | 2015-07-23 | 2015-11-18 | 柳州正高科技有限公司 | Method for predicting image moving direction in dynamic video |
| CN105072384A (en) * | 2015-07-23 | 2015-11-18 | 柳州正高科技有限公司 | Method for obtaining football moving images |
| CN105136064A (en) * | 2015-09-13 | 2015-12-09 | 维希艾信息科技（无锡）有限公司 | Moving object three-dimensional size detection system and method |
| CN108205664A (en) * | 2018-01-09 | 2018-06-26 | 美的集团股份有限公司 | A kind of food recognition methods and device, storage medium, computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102189262B1 (en) | Apparatus and method for collecting traffic information using edge computing | |
| CN102243687A (en) | Physical education teaching auxiliary system based on motion identification technology and implementation method of physical education teaching auxiliary system | |
| CN103680291B (en) | The method synchronizing location and mapping based on ceiling vision | |
| Abughalieh et al. | Predicting pedestrian intention to cross the road | |
| CN107851318A (en) | System and method for Object tracking | |
| CA3037805A1 (en) | A method and system for creating a virtual 3d model | |
| CN103826071A (en) | Three-dimensional camera shooting method for three-dimensional identification and continuous tracking | |
| CN105830093A (en) | Systems, methods, and apparatus for generating metadata relating to spatial regions of non-uniform size | |
| CN109758756B (en) | Gymnastics video analysis method and system based on 3D camera | |
| CN112861808B (en) | Dynamic gesture recognition method, device, computer equipment and readable storage medium | |
| CN101909206A (en) | Video-based intelligent aircraft tracking system | |
| CN109117838B (en) | Target detection method and device applied to unmanned ship sensing system | |
| CN105844659A (en) | Moving part tracking method and device | |
| CN104571511A (en) | System and method for reproducing objects in 3D scene | |
| TWI732374B (en) | Method and apparatus for object recognition | |
| CN106504274A (en) | A kind of visual tracking method and system based under infrared camera | |
| CN103903279B (en) | Parallel tracking system and method based on bionic binocular vision airborne platform | |
| CN114966696A (en) | Transformer-based cross-modal fusion target detection method | |
| Urgo et al. | AI-based pose estimation of human operators in manufacturing environments | |
| Nielsen et al. | Taking the temperature of pedestrian movement in public spaces | |
| Jain et al. | Gestarlite: An on-device pointing finger based gestural interface for smartphones and video see-through head-mounts | |
| CN108058170A (en) | A kind of vision robot's data acquisition processing system | |
| Ma et al. | Localization and mapping method based on multimodal information fusion and deep learning for dynamic object removal | |
| CN116259001A (en) | Multi-view fusion three-dimensional pedestrian posture estimation and tracking method | |
| Spevakov et al. | Detecting objects moving in space from a mobile vision system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140528 |
|
| RJ01 | Rejection of invention patent application after publication |