CN110189377A - A kind of high precision speed-measuring method based on binocular stereo vision - Google Patents
A kind of high precision speed-measuring method based on binocular stereo vision Download PDFInfo
- Publication number
- CN110189377A CN110189377A CN201910431759.8A CN201910431759A CN110189377A CN 110189377 A CN110189377 A CN 110189377A CN 201910431759 A CN201910431759 A CN 201910431759A CN 110189377 A CN110189377 A CN 110189377A
- Authority
- CN
- China
- Prior art keywords
- speed
- camera
- binocular
- pixel
- binocular camera
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
- G06T7/85—Stereo camera calibration
-
- 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
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a kind of high precision speed-measuring methods based on binocular stereo vision, include the following steps: longitudinal binocular camera and the binocular camera that laterally tests the speed of testing the speed of (a) setting；(b) longitudinal binocular camera and binocular camera that laterally tests the speed of testing the speed is demarcated, to obtain the corresponding relationship of camera pixel point and actual size；(c) using the image of longitudinal test the speed binocular camera and lateral binocular camera acquisition moving object of testing the speed, pass through the characteristic point that SIFT Corner Detection obtains moving object；(d) the directly corresponding spacing of adjacent two field pictures is calculated separately using binocular parallax algorithm, the longitudinal speed and lateral velocity component of moving object is calculated according to frame per second；(e) range velocity component and lateral velocity component are synthesized in right amount, thus obtains the speed of moving object.The problems such as solving existing barrage ranging to need first to identify that testee just can be carried out range estimation, having ignored the speed amount and not high enough measurement accuracy of Some vehicles transverse shifting.
Description
Technical field
The invention belongs to computer vision fields, and in particular to a kind of high precision speed-measuring side based on binocular stereo vision
Method.
Background technique
If real time speed measuring can be carried out to the vehicle run in vehicle monitoring region, the velocity information that will acquire is passed in time
Traffic control department strictly investigates and prosecutes over-speed vehicles, it will the road safety situation in very big raising China.
In recent years, the computer vision development for starting from the 1950s is very rapid, it has merged image procossing, mould
The multiple subjects such as formula identification, computer technology, computer graphics, neuro-physiology, it is intended to the function of biological vision is simulated,
Sophisticated category is carried out to object and determines the position of object, shape.Guidance and computer graphical due to vision Frame Theory
Introducing, after entering 21 century, computer vision starts to be widely used in industrial environment.
As an important branch of computer vision, the research of stereoscopic vision is increasingly taken seriously.Stereoscopic vision master
If by simulating biological eyes since the difference at visual angle generates parallax, to determine the distance letter of object according to parallax
Breath includes mainly the contents such as camera calibration, feature extraction, Stereo matching, disparity computation, there is fairly perfect theoretical body
System.Stereoscopic vision can calculate separately out the range information of target object, Jin Erjing at former and later two moment in vision tests the speed
Really obtain the speed of service of target object.
Summary of the invention
The present invention provides a kind of high precision speed-measuring methods based on binocular stereo vision, solve existing barrage ranging
It needs first to identify that testee just can be carried out range estimation, has ignored the speed amount and measurement accuracy of Some vehicles transverse shifting
The problems such as not high enough.
The present invention can be achieved through the following technical solutions:
A kind of high precision speed-measuring method based on binocular stereo vision, it is characterised in that include the following steps:
(a) longitudinal binocular camera and the binocular camera that laterally tests the speed of testing the speed of setting；
(b) it tests the speed binocular camera and laterally the binocular camera that tests the speed is demarcated to the longitudinal direction, to obtain camera pixel point
With the corresponding relationship of actual size；
(c) it is tested the speed binocular camera and laterally the binocular camera that tests the speed obtains the image of moving object, and is led to using the longitudinal direction
Cross the characteristic point that SIFT Corner Detection obtains moving object；
(d) the directly corresponding spacing of adjacent two field pictures is calculated separately using binocular parallax algorithm, is calculated according to frame per second
The range velocity component and lateral velocity component of the moving object；
(e) range velocity component and lateral velocity component are synthesized in right amount, thus obtains the moving object
Speed.
Preferably, the longitudinal direction test the speed binocular camera and laterally test the speed binocular camera be equipped with holder and automation adjust
Device and algorithm.
Preferably, the moving object is automobile, and carries out judging whether to exceed the speed limit after obtaining speed, if hypervelocity
Then camera starts to carry out adpative exposure adjustment, acquires license plate image, character is cut and identified using CNN sliding window, extracts
License plate number is uploaded to cloud.
Preferably, the mobile object is automobile, and realizes test as follows:
Step S1, the preparation needed are camera calibration, and the ranging function to camera is needed before realizing speed measuring function
It can be carried out calibration.First from tailstock portion different location, three thinner scales are pasted in parallel, are then grabbed with the camera of the tailstock
A picture is clapped, according to the reality of pixel summation N1, N2, N3 and scale shared by the actual wheel edge of three scales on picture
Border length L1, L2, L3 are defined the corresponding actual size size Pi of pixel in image, definition mode Pi=
Li/Ni.This method is repeated several times and obtains the corresponding relationship of a series of every row pixel and actual size size, recycles
MATLAB fits every row pixel nonlinear approximation equation corresponding with actual size size and database, is relative displacement
Calculating provide reference.
Step S2: it firstly the need of the automobile for extracting movement from a frame image, uses mixed Gaussian background modeling and calculates
Method is completed.In order to obtain the corresponding pixel of a certain particle on automobile, the inspection of SIFT angle point need to be carried out to the automobile of extraction
Survey, the obtained pixel of SIFT Corner Detection be it is highly stable, convenient for long-term tracking.
Step S3: the corresponding spacing of every frame is measured using binocular vision.
Step S4: having taken the feature descriptor of SIFT angle point, later by the feature of the extreme point in two width or multiple image
Descriptor is matched, and the pixel that same particle is corresponding in different images in actual environment, i.e. same place are found.Greatly
The parallax that most Stereo Matching Algorithms are calculated all is some discrete specific integer values, can meet the precision being normally applied
It is required that.But in the relatively high occasion of some required precisions, as in accurate three-dimensionalreconstruction, it is necessary to after initial parallax acquisition
Parallax is refined using some measures, such as using the method for sub-pix segmentation.Optimized Matching feature pyramid cost letter
Number calculates refinement error and matches vehicle body pixel, calculates relative displacement.
Step S5: according to the nonlinear fitting relationship for the pixel correspondingly-sized size demarcated, processing module is utilized
Computing capability is to section (Nr, Nb) on integrated, obtained relative displacement Xr.In order to reduce calculation amount, Bus- Speed Monitoring is improved
Real-time, integrating range differential to pixel size can be utilized the corresponding size of pixel each in integrating range
It is cumulative to obtain rear car and the relative displacement X from vehicler, as shown in Figure 3.The time difference t=1/ frame of before and after frames is calculated according to frame per second
Laterally opposed vehicle velocity V is calculated according to equation V=X/t further according to speed and lateral displacement in rater.It can similarly calculate vertical
To opposite vehicle velocity Vy.(since camera frame per second is higher, the front cross frame time difference is shorter, and the vehicle speed of acquisition can be considered this time
Average speed in difference).
Step S6: calculated laterally opposed vehicle velocity VrWith longitudinally opposed vehicle velocity Vy, closed according to the vector of speed
At theorem, the actual vehicle speed of rear car is obtained, completes real time speed measuring function.
The beneficial technical effect of the present invention is as follows:
A kind of high precision speed-measuring method based on binocular stereo vision of the invention, using binocular holder, adaptive process monitoring
With dynamic compensating image visual angle.Install ribs additional, algorithm is adaptive, soft or hard combination, and slowing down concussion influences.By calculating left and right
Camera parallax, directly measurement spacing.Detection frame per second reaches 60FPS, and instantaneous velocity is more accurate.Horizontal, vertical ranging respectively, obtains essence
Quasi velosity vector.There can be a large amount of video camera to be monitored on existing highway, it is easy to realize under conditions of low cost
Good result.It only needs upgrade algorithm software that can greatly improve the precision to test the speed in the maintenance of system, it is high not to need purchase
Expensive hardware device carrys out upgrading.The image information of over-speed vehicles can also be can recorde while obtaining speed, side
Just law enfrocement official handles furious driving in accordance with the law.And for the installation of equipment not harsh requirement, it is easily installed.
Detailed description of the invention
Fig. 1 is the top view of the embodiment of the present invention；
Wherein, 1- vehicle, the left camera of 2-, the left camera of 3-, 4- preceding camera, 5- rear camera；
Fig. 2 is the mounting means of camera of the present invention and the schematic diagram for the process that tests the speed；
Fig. 3 is that relative displacement of the invention calculates schematic diagram；
Fig. 4 is a kind of high precision speed-measuring method flow chart of steps based on binocular stereo vision of the invention.
Specific embodiment
With reference to the accompanying drawing and the preferred embodiment specific embodiment that the present invention will be described in detail.
Such as Fig. 1, camera 2,3 is installed above vehicle 1, as the binocular camera that laterally tests the speed for carrying out cross to vehicle 1
It to testing the speed, keeps out the wind on the right side of vehicle 1 and camera 4,5 is installed, as longitudinal binocular camera that tests the speed for carrying out longitudinal direction to vehicle 1
It tests the speed, and then the violation vehicle of hypervelocity is captured.For having determined that the vehicle for hypervelocity, simultaneously using the cutting of CNN sliding window
It identifies character, extracts license plate number, be uploaded to cloud, so that traffic police retains evidence.
Such as Fig. 2, a kind of camera based on binocular stereo vision presses Zhang Fangfa.It is equipped with holder automation regulating device and
Algorithm avoids the time-consuming and error of manual adjustment, can a key holder is adjusted to horizontality.Holder fixed pin is to mounting rod
Issuable vibration, deformation precisely adjust imaging angle by servo motor, and the measurement distance of switch target arrives road plane
Distance.And using high frame per second high-resolution tilt angle instrument dynamic monitoring and compensating image visual angle.
The detailed process to test the speed is as shown in figure 4, be divided into following steps:
Step S1, the preparation needed are camera calibration, and the ranging function to camera is needed before realizing speed measuring function
It can be carried out calibration.First from tailstock portion different location, three thinner scales are pasted in parallel, are then grabbed with the camera of the tailstock
A picture is clapped, according to the reality of pixel summation N1, N2, N3 and scale shared by the actual wheel edge of three scales on picture
Border length L1, L2, L3 are defined the corresponding actual size size Pi of pixel in image, definition mode Pi=
Li/Ni.This method is repeated several times and obtains the corresponding relationship of a series of every row pixel and actual size size, recycles
MATLAB fits every row pixel nonlinear approximation equation corresponding with actual size size and database, is relative displacement
Calculating provide reference.
Step S2: it firstly the need of the automobile for extracting movement from a frame image, uses mixed Gaussian background modeling and calculates
Method is completed.In order to obtain the corresponding pixel of a certain particle on automobile, the inspection of SIFT angle point need to be carried out to the automobile of extraction
Survey, the obtained pixel of SIFT Corner Detection be it is highly stable, convenient for long-term tracking.
Step S3: the corresponding spacing of every frame is measured using binocular vision.
Step S4: having taken the feature descriptor of SIFT angle point, later by the feature of the extreme point in two width or multiple image
Descriptor is matched, and the pixel that same particle is corresponding in different images in actual environment, i.e. same place are found.Greatly
The parallax that most Stereo Matching Algorithms are calculated all is some discrete specific integer values, can meet the precision being normally applied
It is required that.But in the relatively high occasion of some required precisions, as in accurate three-dimensionalreconstruction, it is necessary to after initial parallax acquisition
Parallax is refined using some measures, such as using the method for sub-pix segmentation.Optimized Matching feature pyramid cost letter
Number calculates refinement error and matches vehicle body pixel, calculates relative displacement.
Step S5: according to the nonlinear fitting relationship for the pixel correspondingly-sized size demarcated, processing module is utilized
Computing capability is to section (Nf, Nb) on integrated, obtained relative displacement Xr.In order to reduce calculation amount, Bus- Speed Monitoring is improved
Real-time, integrating range differential to pixel size can be utilized the corresponding size of pixel each in integrating range
It is cumulative to obtain rear car and the relative displacement X from vehicler, as shown in Figure 3.The time difference t=1/ frame of before and after frames is calculated according to frame per second
Laterally opposed vehicle velocity V is calculated according to equation V=X/t further according to speed and lateral displacement in rater.It can similarly calculate vertical
To opposite vehicle velocity Vy.(since camera frame per second is higher, the front cross frame time difference is shorter, and the vehicle speed of acquisition can be considered this time
Average speed in difference).
Step S6: calculated laterally opposed vehicle velocity VrWith longitudinally opposed vehicle velocity Vy, closed according to the vector of speed
At theorem, the actual vehicle speed of rear car is obtained, completes real time speed measuring function.
Although specific embodiments of the present invention have been described above, it will be appreciated by those of skill in the art that these
It is merely illustrative of, without departing from the principle and essence of the present invention, a variety of changes can be made to these embodiments
It more or modifies, therefore, protection scope of the present invention is defined by the appended claims.
Claims (4)
1. a kind of high precision speed-measuring method based on binocular stereo vision, it is characterised in that include the following steps:
(a) longitudinal binocular camera and the binocular camera that laterally tests the speed of testing the speed of setting；
(b) it tests the speed binocular camera and laterally the binocular camera that tests the speed is demarcated to the longitudinal direction, to obtain camera pixel point and reality
The corresponding relationship of border size；
(c) it is tested the speed binocular camera and laterally the binocular camera that tests the speed obtains the image of moving object, and is passed through using the longitudinal direction
The characteristic point of SIFT Corner Detection acquisition moving object；
(d) the directly corresponding spacing of adjacent two field pictures is calculated separately using binocular parallax algorithm, is calculated according to frame per second described
The range velocity component and lateral velocity component of moving object；
(e) range velocity component and lateral velocity component are synthesized in right amount, thus obtains the speed of the moving object
Degree.
2. as described in claim 1 based on the high precision speed-measuring method of binocular stereo vision, which is characterized in that longitudinal survey
Fast binocular camera and the binocular camera that laterally tests the speed are equipped with holder and automation regulating device and algorithm.
3. as claimed in claim 1 or 2 based on the high precision speed-measuring method of binocular stereo vision, which is characterized in that the fortune
Animal body is automobile, and carries out judging whether to exceed the speed limit after obtaining speed, and camera starts to carry out adaptive if hypervelocity
Exposure adjustment, acquires license plate image, and character is cut and identified using CNN sliding window, extracts license plate number, is uploaded to cloud.
4. as described in claim 1 based on the high precision speed-measuring method of binocular stereo vision, which is characterized in that the motive objects
Body is automobile, and realizes test as follows:
Step S1, the preparation needed are camera calibration, needed before realizing speed measuring function to the distance measurement function of camera into
Rower is fixed, first from tailstock portion different location, pastes three thinner scales in parallel, then captures one with the camera of the tailstock
Picture, according to the physical length of pixel summation N1, N2, N3 and scale shared by the actual wheel edge of three scales on picture
L1, L2, L3 are defined the corresponding actual size size Pi of pixel in image, definition mode Pi=Li/Ni；It is more
The secondary this method that repeats obtains the corresponding relationship of a series of every row pixel and actual size size, and MATLAB is recycled to fit
Every row pixel nonlinear approximation equation corresponding with actual size size and database, provide ginseng for the calculating of relative displacement
It examines；
Step S2: it firstly the need of the automobile for extracting movement from a frame image, uses mixed Gaussian background modeling algorithm and comes
It completes；In order to obtain the corresponding pixel of a certain particle on automobile, SIFT Corner Detection, SIFT need to be carried out to the automobile of extraction
The obtained pixel of Corner Detection be it is highly stable, convenient for long-term tracking；
Step S3: the corresponding spacing of every frame is measured using binocular vision；
Step S4: having taken the feature descriptor of SIFT angle point, later describes the feature of the extreme point in two width or multiple image
Symbol is matched, and the pixel that same particle is corresponding in different images in actual environment, i.e. same place are found；It is most of vertical
The parallax that body matching algorithm calculates all is some discrete specific integer values, can meet the required precision being normally applied；But
In the relatively high occasion of some required precisions, as in accurate three-dimensionalreconstruction, it is necessary to be used after initial parallax acquisition
Measure refines parallax, such as using the method for sub-pix segmentation；Optimized Matching feature pyramid cost function calculates refinement
Error matches vehicle body pixel, calculates relative displacement；
Step S5: according to the nonlinear fitting relationship for the pixel correspondingly-sized size demarcated, the calculating of processing module is utilized
Ability is to section (Nf, Nb) on integrated, obtained relative displacement Xr；In order to reduce calculation amount, the real-time of Bus- Speed Monitoring is improved
Property, integrating range differential to pixel size can be added up using the corresponding size of pixel each in integrating range
To rear car with from the relative displacement X of vehicler, as shown in Figure 3；The time difference t=1/ frame per second of before and after frames, then root are calculated according to frame per second
According to speed and lateral displacement, laterally opposed vehicle velocity V is calculated according to equation V=X/tr: it can similarly calculate longitudinally opposed vehicle
Fast Vy；(since camera frame per second is higher, the front cross frame time difference is shorter, and the vehicle speed of acquisition can be considered flat in this section of time difference
Equal speed)；
Step S6: calculated laterally opposed vehicle velocity VrWith longitudinally opposed vehicle velocity Vy, fixed according to the Vector modulation of speed
Reason obtains the actual vehicle speed of rear car, completes real time speed measuring function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910431759.8A CN110189377A (en) | 2019-05-14 | 2019-05-14 | A kind of high precision speed-measuring method based on binocular stereo vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910431759.8A CN110189377A (en) | 2019-05-14 | 2019-05-14 | A kind of high precision speed-measuring method based on binocular stereo vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110189377A true CN110189377A (en) | 2019-08-30 |
Family
ID=67717394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910431759.8A Pending CN110189377A (en) | 2019-05-14 | 2019-05-14 | A kind of high precision speed-measuring method based on binocular stereo vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110189377A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322702A (en) * | 2019-07-08 | 2019-10-11 | 中原工学院 | A kind of Vehicular intelligent speed-measuring method based on Binocular Stereo Vision System |
| CN110513116A (en) * | 2019-09-30 | 2019-11-29 | 西南石油大学 | A kind of monitoring device and monitoring method of tunnel jacking construction guiding |
| CN112115889A (en) * | 2020-09-23 | 2020-12-22 | 成都信息工程大学 | Intelligent vehicle moving target detection method based on vision |
| CN117560481A (en) * | 2024-01-10 | 2024-02-13 | 浙江伽奈维医疗科技有限公司 | Binocular camera capable of reducing beam deformation and temperature dynamic compensation method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2575078A2 (en) * | 2011-09-30 | 2013-04-03 | Ricoh Company, Ltd. | Front vehicle detecting method and front vehicle detecting apparatus |
| CN106375706A (en) * | 2015-07-22 | 2017-02-01 | 宇龙计算机通信科技(深圳)有限公司 | Method and apparatus for velocity measurement of moving object by use of double pick-up heads, and terminal |
| CN107507245A (en) * | 2017-08-18 | 2017-12-22 | 南京阿尔特交通科技有限公司 | A kind of dynamic collecting method and system of vehicle follow gallop track |
| CN107705331A (en) * | 2017-10-30 | 2018-02-16 | 中原工学院 | A kind of automobile video frequency speed-measuring method based on multiple views video camera |
| CN108230748A (en) * | 2016-12-22 | 2018-06-29 | 柯美汽车零部件（上海）有限公司 | The low cost that a kind of view-based access control model perceives is overtaken other vehicles early warning system |
| CN108538060A (en) * | 2018-04-13 | 2018-09-14 | 上海工程技术大学 | A kind of intelligence based on vehicle-mounted back vision camera is broken rules and regulations monitoring method and system |
-
2019
- 2019-05-14 CN CN201910431759.8A patent/CN110189377A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2575078A2 (en) * | 2011-09-30 | 2013-04-03 | Ricoh Company, Ltd. | Front vehicle detecting method and front vehicle detecting apparatus |
| CN106375706A (en) * | 2015-07-22 | 2017-02-01 | 宇龙计算机通信科技(深圳)有限公司 | Method and apparatus for velocity measurement of moving object by use of double pick-up heads, and terminal |
| CN108230748A (en) * | 2016-12-22 | 2018-06-29 | 柯美汽车零部件（上海）有限公司 | The low cost that a kind of view-based access control model perceives is overtaken other vehicles early warning system |
| CN107507245A (en) * | 2017-08-18 | 2017-12-22 | 南京阿尔特交通科技有限公司 | A kind of dynamic collecting method and system of vehicle follow gallop track |
| CN107705331A (en) * | 2017-10-30 | 2018-02-16 | 中原工学院 | A kind of automobile video frequency speed-measuring method based on multiple views video camera |
| CN108538060A (en) * | 2018-04-13 | 2018-09-14 | 上海工程技术大学 | A kind of intelligence based on vehicle-mounted back vision camera is broken rules and regulations monitoring method and system |
Non-Patent Citations (3)
| Title |
|---|
| 博古斯拉夫·赛干内克等: "《三维计算机视觉技术和算法导论》", 31 October 2014, 国防工业出版社 * |
| 孙水发等: "《视频前景检测及其在水电工程监测中的应用》", 31 December 2014, 国防工业出版社 * |
| 张红梅等: "《水下导航定位技术》", 30 November 2010, 武汉大学出版社 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110322702A (en) * | 2019-07-08 | 2019-10-11 | 中原工学院 | A kind of Vehicular intelligent speed-measuring method based on Binocular Stereo Vision System |
| CN110513116A (en) * | 2019-09-30 | 2019-11-29 | 西南石油大学 | A kind of monitoring device and monitoring method of tunnel jacking construction guiding |
| CN112115889A (en) * | 2020-09-23 | 2020-12-22 | 成都信息工程大学 | Intelligent vehicle moving target detection method based on vision |
| CN112115889B (en) * | 2020-09-23 | 2022-08-30 | 成都信息工程大学 | Intelligent vehicle moving target detection method based on vision |
| CN117560481A (en) * | 2024-01-10 | 2024-02-13 | 浙江伽奈维医疗科技有限公司 | Binocular camera capable of reducing beam deformation and temperature dynamic compensation method |
| CN117560481B (en) * | 2024-01-10 | 2024-04-05 | 浙江伽奈维医疗科技有限公司 | Binocular camera capable of reducing beam deformation and temperature dynamic compensation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110322702B (en) | Intelligent vehicle speed measuring method based on binocular stereo vision system | |
| Fernández Llorca et al. | Vision‐based vehicle speed estimation: A survey | |
| US10565867B2 (en) | Detection and documentation of tailgating and speeding violations | |
| CN110189377A (en) | A kind of high precision speed-measuring method based on binocular stereo vision | |
| CN102612634B (en) | A calibration apparatus, a distance measurement system and a calibration method | |
| CN112700470B (en) | Target detection and track extraction method based on traffic video stream | |
| CN105711597B (en) | Front locally travels context aware systems and method | |
| Llorca et al. | Two-camera based accurate vehicle speed measurement using average speed at a fixed point | |
| EP3792660A1 (en) | Method, apparatus and system for measuring distance | |
| Kim et al. | Reliability verification of vehicle speed estimate method in forensic videos | |
| CN106370884A (en) | Vehicle speed measurement method based on binocular camera computer vision technology | |
| CN109747530A (en) | A kind of dual camera and millimeter wave merge automobile sensory perceptual system | |
| CN106978774A (en) | A kind of road surface pit automatic testing method | |
| CN107796373B (en) | Distance measurement method based on monocular vision of front vehicle driven by lane plane geometric model | |
| DE102013211930A1 (en) | Binocular broad base line object comparison method using a minimal cost flownet | |
| Gorodnichev et al. | On automated safety distance monitoring methods by stereo cameras | |
| CN112906777A (en) | Target detection method and device, electronic equipment and storage medium | |
| US20190180121A1 (en) | Detection of Objects from Images of a Camera | |
| CN107290738A (en) | A kind of method and apparatus for measuring front vehicles distance | |
| CN114495064A (en) | Monocular depth estimation-based vehicle surrounding obstacle early warning method | |
| CN111429756A (en) | Highway tunnel rear-end collision prevention early warning method based on video technology | |
| CN114419874A (en) | Target driving safety risk early warning method based on data fusion of roadside sensing equipment | |
| Sochor et al. | Brnocompspeed: Review of traffic camera calibration and comprehensive dataset for monocular speed measurement | |
| CN110023953A (en) | Information processing equipment, imaging device, apparatus control system, moving body, information processing method and computer program product | |
| CN106197382B (en) | A kind of vehicle-mounted single camera target dynamic distance measuring method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |