CN105550670A - Target object dynamic tracking and measurement positioning method - Google Patents
Target object dynamic tracking and measurement positioning method Download PDFInfo
- Publication number
- CN105550670A CN105550670A CN201610054348.8A CN201610054348A CN105550670A CN 105550670 A CN105550670 A CN 105550670A CN 201610054348 A CN201610054348 A CN 201610054348A CN 105550670 A CN105550670 A CN 105550670A
- Authority
- CN
- China
- Prior art keywords
- image
- target object
- background
- camera
- dynamic
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- 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
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/247—Aligning, centring, orientation detection or correction of the image by affine transforms, e.g. correction due to perspective effects; Quadrilaterals, e.g. trapezoids
Abstract
The invention discloses a target object dynamic tracking and measurement positioning method. According to the method, two cameras are used for collecting images of a monitoring region; the backgrounds of the monitoring region are dynamically updated; a target object is extracted; a three-dimensional point cloud of a viewing angle region is generated by using a binocular identification positioning principle; the target object is dynamically tracked and positioned through combining the target object extraction and the binocular identification positioning principle. According to the method of the invention, vision distance measurement and target tracking are combined in the video monitoring field; the pixel coordinate of the image in which the target is located is determined; the target object is locked through combining the three-dimensional point cloud generated by the vision distance measurement; the three-dimensional coordinate of the target object is determined. When the target object enters a warning region, the system warns; therefore real time early warning is realized; the captured position information of the target object provides scientific basis for actual operation of a backstage worker.
Description
Technical field
The present invention relates to video security protection and human-computer interaction technique field, particularly relate to the dynamic Tracking and Measurment localization method of a kind of field of video monitoring target object.
Background technology
At present, video surveillance applications is more prevalent, brings immeasurable effect at safety-security area to the work of people, but existing monitoring technique intelligence degree is low, still depending on a large amount of human resources comes video content recognition, to tackle danger and accident.Most of traditional video monitoring system only can the video information in acquisition monitoring region, this monitor mode depends on artificial continuous firing with test and monitoring area burst and unsafe condition, lack and intelligent early-warning is carried out to the dangerous information in guarded region, actual motion need drop into a large amount of manpower to carry out in real time or ex-post analysis, and the image that this video monitoring mode is passed back can not provide target object accurate positional information, manipulation personnel only rule of thumb can infer the approximate location of target object, make the tracking of target object and location inaccurate and lack of wisdom.
Summary of the invention
The invention provides the dynamic Tracking and Measurment localization method of a kind of target object, make up the deficiency that picture that traditional video surveillance passes back can not provide target object exact position, improve the present situation that it relies on a large amount of human resources, improve the intelligent level of video monitoring system.
For this reason, the technical scheme adopted is:
The dynamic Tracking and Measurment localization method of a kind of target object, the method utilizes two camera collection guarded region images, is upgraded and target object extraction, utilize binocular identification positioning principle, generate field of view three-dimensional point cloud by guarded region background dynamics; Combining target Object Extraction and binocular identification positioning principle, dynamic track and localization target object.
Its concrete steps are as follows:
Step 1, target object extracts: Dynamic Establishing background picture library real-time update, different threshold values is given to the background of Different Dynamic degree, according to the calculus of differences result of image in present image and background picture library, distinguish the prospect in present image and background parts, and background parts is updated in background picture library;
Step 2, binocular is found range:
(1) removal of images distortion corrects with camera: utilize Taylor series expansion and combine and add correction factor, correction gather pattern distortion; Adopt 16
12 chessboards are demarcated camera as demarcation thing, guarantee that the unique point in checkerboard image is uniformly distributed by distance minimization, projection maximization principle, coordinate points is to equation to utilize the geometric relationship of chessboard unique point and image characteristic point to draw, thus solve camera inside and outside parameter, by the correcting distorted image of intrinsic parameter, draw the image of true nature more; By angle and the position of the relative chessboard of outer parameter adjustment two sub-picture, export the correcting image that row is aimed at;
(2) images match: simultaneously at the multiple image of different visual fields photographic subjects object, search the same characteristic features of left and right camera image captured by the different visual field of synchronization, analyze difference wherein, export the pixel coordinate difference of same unique point on left images;
(3) re-projection: left images same characteristic features point pixel coordinate difference result is changed into distance by triangulation, exports the three-dimensional point cloud of multi-view image;
Step 3, target following is located: any width current frame image in image captured by the camera of left and right and respective background image are made difference, target in dynamic lock image, and extract its pixel coordinate at present frame, in conjunction with the three-dimensional point cloud information that binocular range finding generates, determine the three-dimensional point cloud of this target, try to achieve the coordinate of target object in world coordinate system.
Use mixed Gauss model in described step 1, weaken in image the disturbing factor being similar to leaf and rocking, to reduce the mutual interference of prospect and background; Effectively be separated present frame prospect and background image according to dynamic threshold, and the background parts of present image is updated in background picture library; According to the foreground image extracted, determine the pixel coordinate of image residing for prospect, for the three-dimensional world coordinate calculating foreground image provides scientific basis.
Visual token combines in field of video monitoring with target following by the present invention, is followed the tracks of by target dynamic, determines the pixel coordinate of target place image, in conjunction with the three-dimensional point cloud that visual token generates, and lock onto target object, and determine its three-dimensional coordinate.When target object enters warning region, system just can give the alarm, and reaches the object of real-time early warning; The positional information of the target object captured provides scientific basis to the actual manipulation of background work personnel.
To sum up, the present invention, compared with existing video monitoring, has the following advantages: (1) dynamic background figure library model by setting up, and through image procossing, dynamic locking enters the target object of guarded region, for the real-time early warning of safety-security area provides support.(2) binocular range measurement principle is adopted, the foreground extraction of combining target object, can the positional information of Obtaining Accurate target object, make up the deficiency that traditional video surveillance can not provide target object precise position information, improve the intelligent level of video monitoring.
Accompanying drawing explanation
Fig. 1 is general principles schematic diagram of the present invention;
Fig. 2 is target object measurement and positioning process flow diagram of the present invention
Fig. 3 is camera imaging illustraton of model;
Fig. 4 is triangulation schematic diagram.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail.
General principles signal of the present invention as shown in Figure 1, by the left and right camera collection binocular image information of USB interface, through ARM11 development board, image is processed, captured target object space and geometric size information, for automatic early-warning and background work personnel take corresponding measure to provide foundation.
In the present invention, target object measurement and positioning flow process as shown in Figure 2, by left and right camera collection binocular image information, with 16
12 chessboards are demarcate thing, adopt distance minimization, projection maximization principle stereo calibration camera, try to achieve camera parameter and correcting distorted image, coupling left images unique point, and pass through the three-dimensional point cloud of triangulation synthetic image; Dynamically arrange similarity threshold accurately to extract target object, obtain target object pixel coordinate, in conjunction with the three-dimensional point cloud generated, captured target object space and geometric size information, for man-machine interaction and intelligent early-warning provide foundation.Concrete grammar is as follows:
Step 1, target object extracts: Dynamic Establishing background picture library real-time update, different threshold values is given to the background of different degree of dynamism, image in present image and background picture library is made difference, when difference result exceedes the threshold value of setting, can determine that present image and background image difference result exceed threshold portion is background, and remainder is then prospect.The background parts of image needs to be updated in background picture library.
Step 2, binocular is found range:
(1) removal of images distortion corrects with camera: desirable camera imaging model is pin-hole model, as shown in Figure 3, camera in order to increase light transmission capacity, employs lens when actual production, but lens can produce error in the manufacture and installation, the image of camera collection is caused to distort.In order to reduce pattern distortion to the impact of graphical analysis as far as possible, select employing 16
12 chessboards are demarcated camera as demarcation thing, solve camera inside and outside parameter.By the correcting distorted image of intrinsic parameter, make image true nature more; By angle and the position of the relative chessboard of outer parameter adjustment two sub-picture, export row alignment image.
(2) images match: simultaneously at the multiple image of different visual fields photographic subjects object, searches the same characteristic features of left and right camera image captured by the different visual field of synchronization, exports the pixel coordinate difference of same unique point on left images.
(3) re-projection: left images same characteristic features point pixel coordinate difference result is changed into distance by triangulation, exports the three-dimensional point cloud of multi-view image.
Step 3, target following is located: any width current frame image in image captured by the camera of left and right and respective background image are made difference, target in dynamic lock image, and extract its pixel coordinate at present frame, in conjunction with the three-dimensional point cloud information that binocular range finding generates, determine the three-dimensional point cloud of this target object, try to achieve the coordinate figure of target object in world coordinate system.
(1) elaborate about target object extraction in step 1
In video monitoring, dynamic target object is the focus paid close attention to of people often, and it is the core procedure of intelligent monitoring that target object extracts.Based on background model, the difference of image in Water demand current frame image and background picture library, to extract the prospect part of current frame image; But in actual extracting, background image is often subject to the impact of illumination or complex scene, threshold value for distinguishing current frame image prospect and background parts can not be fixed, therefore, need real-time update background model, constantly the threshold value of adjustment differentiate between images prospect and background parts.The present invention introduces mixed Gauss model and is similar to leaf the disturbing factor such as rocks, to reduce the mutual interference of prospect and background to weaken in image.Utilize the matching result s of background model and current frame image, dynamic conditioning matching similarity threshold k.The relation of matching result s and threshold k is as follows:
Wherein a, b, m are preset parameters; When background changes, threshold k can suitably adjust to adapt to background perturbation.
(2) elaborate about binocular range finding in step 2
Binocular range finding relates to the important content of two large divisions: camera calibration and binocular range finding.
First the ultimate principle of binocular range finding was introduced before introducing camera calibration.Desirable binocular range finding model triangulation as shown in Figure 4.In Fig. 4, the two sub-picture optical axis perfect parallelism (optical axis is the ray that projection centre is drawn towards principal point c direction) that pixel column is aimed at
with
be respectively left and right projection centre,
with
be respectively the focal length of two cameras and equal, principal point
with
left images has identical pixel coordinate, and the imaging point of unique point X on left images is respectively
with
,
with
in respective pixel coordinate system, horizontal shift is respectively
with
, parallax is:
if f is the focal length of camera, utilizes similar triangle theory, object can be released as follows from the equation of the distance Z of cam lens:
In order to set up desirable binocular range finding platform, need to carry out stereo calibration to camera, the imaging model of camera as shown in Figure 3.With 16
the chessboard that 12 black and white lattice intersect, as camera calibration thing, with black and white lattice point of crossing for unique point, sets up chessboard unique point by the conversion such as Matrix Translation, rotation
and image characteristic point
between contact, set up equation, utilize least square scheduling algorithm to solve the parameter such as focal length, distortion factor of camera.
Traditional camera calibration adopts 9
the chessboard that 6 black and white lattice intersect is as demarcation thing, and this chessboard one has 54 correction feature points, and correction feature point is less, causes there is blind area during fractional distortion regional correction, affects visual token precision.The present invention adopts intensive chessboard (16
12) as camera calibration thing, adopt distance minimization, projection maximization principle, tool has the following advantages: there is more correction feature point in unit area image simultaneously, can in the hope of the higher distortion factor of accuracy; Distance minimization, projection maximization principle make checkerboard image maximize in the screen accounting of field of view, can ensure that field of view unique point is uniformly distributed, improve stated accuracy.This low coverage multiple spot camera calibration mode can improve camera calibration precision, improves the flake phenomenon of image, promotes binocular ranging efficiency, improves visual token precision.
Effect of the present invention is further illustrated below by concrete application scenarios:
Scene 1: video monitoring safety-security area real-time early warning.Traditional monitor mode needs the long-time checking monitoring video of staff to reach the object of monitoring in real time, and depend on a large amount of human resources, efficiency and the intelligent level of monitoring are lower.Dynamic Establishing background model of the present invention real-time update, extract foreground image by image difference computing, in conjunction with binocular identification positioning principle, dynamically follows the tracks of and localizing objects object, realize real-time early warning.A large amount of human resources in traditional video surveillance free by the present invention from real work, improve the intelligent level of supervisory system.
Scene 2: visual token.Common distance-finding method has laser ranging, infrared distance measurement, ultrasonic ranging, radar range finding etc., visual token of the present invention is compared with this several distance-finding method, do not need during measurement to send any signal to testee, principle is simple, cost is low, can record target object location under complex environment.Meanwhile, if selected the unique point in space by mouse, utilize Pythagorean theorem, sine and cosine theorem etc. just can calculate unique point spacing and relative position relation, calculate the geometric size information of target object further.
Scene 3: object edge detects.Common Edge-Detection Algorithm, obtain the profile information of object often through the single order or second derivative of analyzing variation of image grayscale, the type edge detection algorithm, effectively can not extract objects' contour information in complex scene.The present invention generates the depth information of three-dimensional point cloud according to vision measurement, by drawing function, can draw the profile of different depth object, realizes accurately extracting intended target contour of object in multiple foreground object.The method can be used for autonomous intelligence operation and the field such as vision guided navigation of robot.
Scene 4: man-machine interaction.The video information in most of traditional video monitoring system only acquisition monitoring region, the interactive information quantity not sufficient that this monitor mode provides for staff, staff need in conjunction with the micro-judgment of oneself and the approximate location inferring target object, and workload is large, and precision is low.The present invention, according to binocular identification positioning principle, draws the three-dimensional point cloud of field of view, combining target Object Extraction, determines the positional information of target object, for the decision-making of staff provides foundation.
To sum up, the present invention simulates the mode of human eye process scenery, and part replaces human brain to carry out understanding and cognition to natural things, generates the three-dimensional point cloud of field of view based on binocular range measurement principle; Based on the Background library model dynamically updated, by image difference computing, obtain target object image pixel coordinate; In conjunction with three-dimensional point cloud information and the target object image pixel coordinate of field of view, the dynamic tracking of realize target object and location.
Claims (3)
1. the dynamic Tracking and Measurment localization method of target object, it is characterized in that: the method utilizes two camera collection guarded region images, upgraded by guarded region background dynamics and target object extraction, utilize binocular identification positioning principle, generate field of view three-dimensional point cloud; Combining target Object Extraction and binocular identification positioning principle, dynamic track and localization target object.
2. the dynamic Tracking and Measurment localization method of a kind of target object according to claim 1, the method is specific as follows:
Step 1, target object extracts: Dynamic Establishing background picture library real-time update, different threshold values is given to the background of Different Dynamic degree, according to the calculus of differences result of image in present image and background picture library, distinguish the prospect in present image and background parts, and background parts is updated in background picture library;
Step 2, binocular is found range:
(1) removal of images distortion corrects with camera: utilize Taylor series expansion and combine and add correction factor, correction gather pattern distortion; 16*12 chessboard is adopted to demarcate camera as demarcation thing, guarantee that the unique point in checkerboard image is uniformly distributed by distance minimization, projection maximization principle, coordinate points is to equation to utilize the geometric relationship of chessboard unique point and image characteristic point to draw, thus solve camera inside and outside parameter, by the correcting distorted image of intrinsic parameter, draw the image of true nature more; By angle and the position of the relative chessboard of outer parameter adjustment two sub-picture, export the correcting image that row is aimed at;
(2) images match: simultaneously at the multiple image of different visual fields photographic subjects object, search the same characteristic features of left and right camera image captured by the different visual field of synchronization, analyze difference wherein, export the pixel coordinate difference of same unique point on left images;
(3) re-projection: left images same characteristic features point pixel coordinate difference result is changed into distance by triangulation, exports the three-dimensional point cloud of multi-view image;
Step 3, target following is located: any width current frame image in image captured by the camera of left and right and respective background image are made difference, target in dynamic lock image, and extract its pixel coordinate at present frame, in conjunction with the three-dimensional point cloud information that binocular range finding generates, determine the three-dimensional point cloud of this target, try to achieve the coordinate of target object in world coordinate system.
3. the dynamic Tracking and Measurment localization method of a kind of target object according to claim 2, is characterized in that: use mixed Gauss model in described step 1, weakens in image the disturbing factor being similar to leaf and rocking, to reduce the mutual interference of prospect and background; Effectively be separated present frame prospect and background image according to dynamic threshold, and the background parts of present image is updated in background picture library; According to the foreground image extracted, determine the pixel coordinate of image residing for prospect, for the three-dimensional world coordinate calculating foreground image provides scientific basis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610054348.8A CN105550670B (en) | 2016-01-27 | 2016-01-27 | A kind of target object dynamically track and measurement and positioning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610054348.8A CN105550670B (en) | 2016-01-27 | 2016-01-27 | A kind of target object dynamically track and measurement and positioning method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105550670A true CN105550670A (en) | 2016-05-04 |
| CN105550670B CN105550670B (en) | 2019-07-12 |
Family
ID=55829853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610054348.8A Expired - Fee Related CN105550670B (en) | 2016-01-27 | 2016-01-27 | A kind of target object dynamically track and measurement and positioning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105550670B (en) |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106447680A (en) * | 2016-11-23 | 2017-02-22 | 湖南华诺星空电子技术有限公司 | Method for radar and vision fused target detecting and tracking in dynamic background environment |
| CN107122770A (en) * | 2017-06-13 | 2017-09-01 | 驭势（上海）汽车科技有限公司 | Many mesh camera systems, intelligent driving system, automobile, method and storage medium |
| CN107246696A (en) * | 2017-06-27 | 2017-10-13 | 上海卓思智能科技股份有限公司 | A kind of vent cabinet windowing area measuring method and system and a kind of controller |
| CN107367767A (en) * | 2017-06-27 | 2017-11-21 | 上海卓思智能科技股份有限公司 | A kind of vent cabinet window foreign matter detecting method and system and a kind of controller |
| WO2018014420A1 (en) * | 2016-07-21 | 2018-01-25 | 深圳曼塔智能科技有限公司 | Light-emitting target recognition-based unmanned aerial vehicle tracking control system and method |
| CN107773248A (en) * | 2017-09-30 | 2018-03-09 | 优视眼动科技（北京）有限公司 | Eye tracker and image processing method |
| CN107884767A (en) * | 2017-10-31 | 2018-04-06 | 暨南大学 | A kind of method of binocular vision system measurement ship distance and height |
| CN107992820A (en) * | 2017-11-29 | 2018-05-04 | 北京伟景智能科技有限公司 | Counter automatic selling method based on binocular vision |
| CN108090922A (en) * | 2016-11-21 | 2018-05-29 | 中国科学院沈阳计算技术研究所有限公司 | Intelligent Target pursuit path recording method |
| CN108087208A (en) * | 2016-11-21 | 2018-05-29 | 北京金风科创风电设备有限公司 | Wind generator set blade follower method and device based on unmanned plane |
| CN108537094A (en) * | 2017-03-03 | 2018-09-14 | 株式会社理光 | Image processing method, device and system |
| CN108664841A (en) * | 2017-03-27 | 2018-10-16 | 郑州宇通客车股份有限公司 | A kind of sound state object recognition methods and device based on laser point cloud |
| CN108731587A (en) * | 2017-04-14 | 2018-11-02 | 中交遥感载荷（北京）科技有限公司 | A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model |
| CN108829116A (en) * | 2018-10-09 | 2018-11-16 | 上海岚豹智能科技有限公司 | Barrier-avoiding method and equipment based on monocular cam |
| CN108961155A (en) * | 2018-07-13 | 2018-12-07 | 惠州市德赛西威汽车电子股份有限公司 | A kind of projective invariant bearing calibration of high-fidelity |
| CN108989686A (en) * | 2018-09-05 | 2018-12-11 | 深圳技威时代科技有限公司 | Captured in real-time device and control method based on humanoid tracking |
| CN109051321A (en) * | 2018-09-14 | 2018-12-21 | 山东上拓教育咨询有限公司 | A kind of fresh commodities circulating cases of the low temperature that intelligence follows automatically |
| CN109523592A (en) * | 2018-10-19 | 2019-03-26 | 天津大学 | A kind of interior flame localization method based on camera |
| CN109934873A (en) * | 2019-03-15 | 2019-06-25 | 百度在线网络技术（北京）有限公司 | Mark image acquiring method, device and equipment |
| CN110298293A (en) * | 2019-06-25 | 2019-10-01 | 重庆紫光华山智安科技有限公司 | One kind anti-wander away method, apparatus, readable storage medium storing program for executing and electric terminal |
| CN110342134A (en) * | 2019-07-23 | 2019-10-18 | 珠海市一微半导体有限公司 | A kind of garbage classification identifying system and its method based on binocular vision |
| CN110595443A (en) * | 2019-08-22 | 2019-12-20 | 苏州佳世达光电有限公司 | Projection device |
| CN110673607A (en) * | 2019-09-25 | 2020-01-10 | 优地网络有限公司 | Feature point extraction method and device in dynamic scene and terminal equipment |
| CN111583334A (en) * | 2020-05-26 | 2020-08-25 | 广东电网有限责任公司培训与评价中心 | Three-dimensional space positioning method, device and equipment for transformer substation personnel |
| CN111640300A (en) * | 2020-04-28 | 2020-09-08 | 武汉万集信息技术有限公司 | Vehicle detection processing method and device |
| CN112348493A (en) * | 2021-01-07 | 2021-02-09 | 北京电信易通信息技术股份有限公司 | Intelligent conference recording system and method |
| CN112819770A (en) * | 2021-01-26 | 2021-05-18 | 中国人民解放军陆军军医大学第一附属医院 | Iodine contrast agent allergy monitoring method and system |
| CN113077511A (en) * | 2020-01-06 | 2021-07-06 | 初速度（苏州）科技有限公司 | Multi-camera target matching and tracking method and device for automobile |
| CN113221909A (en) * | 2021-05-12 | 2021-08-06 | 佛山育脉科技有限公司 | Image processing method, image processing apparatus, and computer-readable storage medium |
| CN113327291A (en) * | 2020-03-16 | 2021-08-31 | 天目爱视（北京）科技有限公司 | Calibration method for 3D modeling of remote target based on continuous shooting |
| CN113688724A (en) * | 2021-08-24 | 2021-11-23 | 桂林电子科技大学 | Swimming pool drowning monitoring method based on binocular vision |
| CN113923420A (en) * | 2021-11-18 | 2022-01-11 | 京东方科技集团股份有限公司 | Area adjustment method and device, camera and storage medium |
| CN113965733A (en) * | 2021-12-07 | 2022-01-21 | 中国联合网络通信集团有限公司 | Binocular video monitoring method, system, computer equipment and storage medium |
| CN114267155A (en) * | 2021-11-05 | 2022-04-01 | 国能大渡河革什扎水电开发有限公司 | Geological disaster monitoring and early warning system based on video recognition technology |
| CN114283119A (en) * | 2021-12-02 | 2022-04-05 | 上海韦地科技集团有限公司 | Irradiation-resistant camera control system |
| CN114993244A (en) * | 2022-05-09 | 2022-09-02 | 深圳供电局有限公司 | Target ranging device and method for power transformation operation area |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103868460A (en) * | 2014-03-13 | 2014-06-18 | 桂林电子科技大学 | Parallax optimization algorithm-based binocular stereo vision automatic measurement method |
| CN103903279A (en) * | 2014-03-21 | 2014-07-02 | 上海大学 | Parallel tracking system and method based on bionic binocular vision onboard platform |
| CN104463906A (en) * | 2014-11-11 | 2015-03-25 | 广东中星电子有限公司 | Object tracking device and method |
-
2016
- 2016-01-27 CN CN201610054348.8A patent/CN105550670B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103868460A (en) * | 2014-03-13 | 2014-06-18 | 桂林电子科技大学 | Parallax optimization algorithm-based binocular stereo vision automatic measurement method |
| CN103903279A (en) * | 2014-03-21 | 2014-07-02 | 上海大学 | Parallel tracking system and method based on bionic binocular vision onboard platform |
| CN104463906A (en) * | 2014-11-11 | 2015-03-25 | 广东中星电子有限公司 | Object tracking device and method |
Cited By (46)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018014420A1 (en) * | 2016-07-21 | 2018-01-25 | 深圳曼塔智能科技有限公司 | Light-emitting target recognition-based unmanned aerial vehicle tracking control system and method |
| CN108087208A (en) * | 2016-11-21 | 2018-05-29 | 北京金风科创风电设备有限公司 | Wind generator set blade follower method and device based on unmanned plane |
| CN108090922A (en) * | 2016-11-21 | 2018-05-29 | 中国科学院沈阳计算技术研究所有限公司 | Intelligent Target pursuit path recording method |
| CN106447680B (en) * | 2016-11-23 | 2019-09-17 | 湖南华诺星空电子技术有限公司 | The object detecting and tracking method that radar is merged with vision under dynamic background environment |
| CN106447680A (en) * | 2016-11-23 | 2017-02-22 | 湖南华诺星空电子技术有限公司 | Method for radar and vision fused target detecting and tracking in dynamic background environment |
| CN108537094A (en) * | 2017-03-03 | 2018-09-14 | 株式会社理光 | Image processing method, device and system |
| CN108664841B (en) * | 2017-03-27 | 2021-05-11 | 郑州宇通客车股份有限公司 | Dynamic and static target object identification method and device based on laser point cloud |
| CN108664841A (en) * | 2017-03-27 | 2018-10-16 | 郑州宇通客车股份有限公司 | A kind of sound state object recognition methods and device based on laser point cloud |
| CN108731587A (en) * | 2017-04-14 | 2018-11-02 | 中交遥感载荷（北京）科技有限公司 | A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model |
| CN107122770B (en) * | 2017-06-13 | 2023-06-27 | 驭势（上海）汽车科技有限公司 | Multi-camera system, intelligent driving system, automobile, method and storage medium |
| CN107122770A (en) * | 2017-06-13 | 2017-09-01 | 驭势（上海）汽车科技有限公司 | Many mesh camera systems, intelligent driving system, automobile, method and storage medium |
| CN107367767A (en) * | 2017-06-27 | 2017-11-21 | 上海卓思智能科技股份有限公司 | A kind of vent cabinet window foreign matter detecting method and system and a kind of controller |
| CN107246696A (en) * | 2017-06-27 | 2017-10-13 | 上海卓思智能科技股份有限公司 | A kind of vent cabinet windowing area measuring method and system and a kind of controller |
| CN107773248A (en) * | 2017-09-30 | 2018-03-09 | 优视眼动科技（北京）有限公司 | Eye tracker and image processing method |
| CN107884767A (en) * | 2017-10-31 | 2018-04-06 | 暨南大学 | A kind of method of binocular vision system measurement ship distance and height |
| CN107992820A (en) * | 2017-11-29 | 2018-05-04 | 北京伟景智能科技有限公司 | Counter automatic selling method based on binocular vision |
| CN108961155A (en) * | 2018-07-13 | 2018-12-07 | 惠州市德赛西威汽车电子股份有限公司 | A kind of projective invariant bearing calibration of high-fidelity |
| CN108961155B (en) * | 2018-07-13 | 2023-06-27 | 惠州市德赛西威汽车电子股份有限公司 | High-fidelity fisheye lens distortion correction method |
| CN108989686A (en) * | 2018-09-05 | 2018-12-11 | 深圳技威时代科技有限公司 | Captured in real-time device and control method based on humanoid tracking |
| CN108989686B (en) * | 2018-09-05 | 2021-02-19 | 深圳技威时代科技有限公司 | Real-time shooting device based on human shape tracking and control method |
| CN109051321A (en) * | 2018-09-14 | 2018-12-21 | 山东上拓教育咨询有限公司 | A kind of fresh commodities circulating cases of the low temperature that intelligence follows automatically |
| CN108829116A (en) * | 2018-10-09 | 2018-11-16 | 上海岚豹智能科技有限公司 | Barrier-avoiding method and equipment based on monocular cam |
| CN109523592A (en) * | 2018-10-19 | 2019-03-26 | 天津大学 | A kind of interior flame localization method based on camera |
| CN109934873A (en) * | 2019-03-15 | 2019-06-25 | 百度在线网络技术（北京）有限公司 | Mark image acquiring method, device and equipment |
| CN110298293A (en) * | 2019-06-25 | 2019-10-01 | 重庆紫光华山智安科技有限公司 | One kind anti-wander away method, apparatus, readable storage medium storing program for executing and electric terminal |
| CN110342134A (en) * | 2019-07-23 | 2019-10-18 | 珠海市一微半导体有限公司 | A kind of garbage classification identifying system and its method based on binocular vision |
| CN110595443A (en) * | 2019-08-22 | 2019-12-20 | 苏州佳世达光电有限公司 | Projection device |
| CN110673607A (en) * | 2019-09-25 | 2020-01-10 | 优地网络有限公司 | Feature point extraction method and device in dynamic scene and terminal equipment |
| CN113077511A (en) * | 2020-01-06 | 2021-07-06 | 初速度（苏州）科技有限公司 | Multi-camera target matching and tracking method and device for automobile |
| CN113327291A (en) * | 2020-03-16 | 2021-08-31 | 天目爱视（北京）科技有限公司 | Calibration method for 3D modeling of remote target based on continuous shooting |
| CN113327291B (en) * | 2020-03-16 | 2024-03-22 | 天目爱视（北京）科技有限公司 | Calibration method for 3D modeling of remote target object based on continuous shooting |
| CN111640300A (en) * | 2020-04-28 | 2020-09-08 | 武汉万集信息技术有限公司 | Vehicle detection processing method and device |
| CN111640300B (en) * | 2020-04-28 | 2022-06-17 | 武汉万集信息技术有限公司 | Vehicle detection processing method and device |
| CN111583334A (en) * | 2020-05-26 | 2020-08-25 | 广东电网有限责任公司培训与评价中心 | Three-dimensional space positioning method, device and equipment for transformer substation personnel |
| CN111583334B (en) * | 2020-05-26 | 2023-03-14 | 广东电网有限责任公司培训与评价中心 | Three-dimensional space positioning method, device and equipment for transformer substation personnel |
| CN112348493A (en) * | 2021-01-07 | 2021-02-09 | 北京电信易通信息技术股份有限公司 | Intelligent conference recording system and method |
| CN112819770A (en) * | 2021-01-26 | 2021-05-18 | 中国人民解放军陆军军医大学第一附属医院 | Iodine contrast agent allergy monitoring method and system |
| CN113221909A (en) * | 2021-05-12 | 2021-08-06 | 佛山育脉科技有限公司 | Image processing method, image processing apparatus, and computer-readable storage medium |
| CN113688724A (en) * | 2021-08-24 | 2021-11-23 | 桂林电子科技大学 | Swimming pool drowning monitoring method based on binocular vision |
| CN113688724B (en) * | 2021-08-24 | 2023-03-24 | 桂林电子科技大学 | Swimming pool drowning monitoring method based on binocular vision |
| CN114267155A (en) * | 2021-11-05 | 2022-04-01 | 国能大渡河革什扎水电开发有限公司 | Geological disaster monitoring and early warning system based on video recognition technology |
| CN113923420A (en) * | 2021-11-18 | 2022-01-11 | 京东方科技集团股份有限公司 | Area adjustment method and device, camera and storage medium |
| CN114283119A (en) * | 2021-12-02 | 2022-04-05 | 上海韦地科技集团有限公司 | Irradiation-resistant camera control system |
| CN114283119B (en) * | 2021-12-02 | 2022-12-13 | 上海韦地科技集团有限公司 | Irradiation-resistant camera control system |
| CN113965733A (en) * | 2021-12-07 | 2022-01-21 | 中国联合网络通信集团有限公司 | Binocular video monitoring method, system, computer equipment and storage medium |
| CN114993244A (en) * | 2022-05-09 | 2022-09-02 | 深圳供电局有限公司 | Target ranging device and method for power transformation operation area |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105550670B (en) | 2019-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105550670A (en) | Target object dynamic tracking and measurement positioning method | |
| CN108731587A (en) | A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model | |
| CN103868460B (en) | Binocular stereo vision method for automatic measurement based on parallax optimized algorithm | |
| CN103955920B (en) | Binocular vision obstacle detection method based on three-dimensional point cloud segmentation | |
| CN102622767B (en) | Method for positioning binocular non-calibrated space | |
| CN104902246A (en) | Video monitoring method and device | |
| CN102447835A (en) | Non-blind-area multi-target cooperative tracking method and system | |
| CN105286871A (en) | Video processing-based body height measurement method | |
| CN104966062A (en) | Video monitoring method and device | |
| CN102819847A (en) | Method for extracting movement track based on PTZ mobile camera | |
| CN106600643B (en) | A kind of demographic method based on trajectory analysis | |
| CN110132226A (en) | The distance and azimuth angle measurement system and method for a kind of unmanned plane line walking | |
| CN106295657A (en) | A kind of method extracting human height's feature during video data structure | |
| CN103729620A (en) | Multi-view pedestrian detection method based on multi-view Bayesian network | |
| WO2022127181A1 (en) | Passenger flow monitoring method and apparatus, and electronic device and storage medium | |
| CN104786227B (en) | Drop switch based on robot for high-voltage hot-line work changes control system and method | |
| CN112669280A (en) | Unmanned aerial vehicle oblique aerial photography right-angle image control point target detection method based on LSD algorithm | |
| CN106709432B (en) | Human head detection counting method based on binocular stereo vision | |
| CN103533332A (en) | Image processing method for converting 2D video into 3D video | |
| CN110909571A (en) | High-precision face recognition space positioning method | |
| CN102194249B (en) | Water current modeling data capturing device with combination of infrared rays and visible light | |
| CN104858877B (en) | High-tension line drop switch changes the control method of control system automatically | |
| CN116580107A (en) | Cross-view multi-target real-time track tracking method and system | |
| CN109697428A (en) | Positioning system is identified based on the unmanned plane of RGB_D and depth convolutional network | |
| CN104933732A (en) | Method for detecting and tracking movement target based on omnidirectional vision of robot |
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 | ||
| 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: 20190712 Termination date: 20220127 |