CN107884767A - A kind of method of binocular vision system measurement ship distance and height - Google Patents

A kind of method of binocular vision system measurement ship distance and height Download PDF

Info

Publication number
CN107884767A
CN107884767A CN201711050019.7A CN201711050019A CN107884767A CN 107884767 A CN107884767 A CN 107884767A CN 201711050019 A CN201711050019 A CN 201711050019A CN 107884767 A CN107884767 A CN 107884767A
Authority
CN
China
Prior art keywords
ship
coordinate
camera
dimensional
image
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
Application number
CN201711050019.7A
Other languages
Chinese (zh)
Inventor
刘新东
何信
周曙
张新征
邓义军
罗志维
凌政
王晓俊
周正昊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangzhou Wangping Information Technology Co ltd
Jinan University
Original Assignee
Guangzhou Wangping Information Technology Co ltd
Jinan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangzhou Wangping Information Technology Co ltd, Jinan University filed Critical Guangzhou Wangping Information Technology Co ltd
Priority to CN201711050019.7A priority Critical patent/CN107884767A/en
Publication of CN107884767A publication Critical patent/CN107884767A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
    • G01S11/12—Systems for determining distance or velocity not using reflection or reradiation using electromagnetic waves other than radio waves
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
    • G01B11/06—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness for measuring thickness ; e.g. of sheet material
    • G01B11/0608—Height gauges

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Measurement Of Optical Distance (AREA)

Abstract

The invention discloses a kind of method of binocular vision system measurement ship distance and height, comprise the following steps：Step 1:Establish two and be used for the high binocular vision system of ship ranging survey；Step 2:Single camera calibration, step 3：Two cameras obtain left images and pre-processed respectively, so as to obtain on profile and contoured interior two dimensional image pixel point coordinates；Step 4：Binocular corrects, and calculates the parallax that target point is formed on the image of left and right two；Step 5：Stereo matching, obtain disparity map；Step 6：Three-dimensional reconstruction；Can be three-dimensional world coordinate by the two-dimensional points coordinate re-projection in image；Step 7：Ship is calculated to the distance of bridge；Step 8：The coordinate that y values are minimum and maximum in three-dimensional coordinate point is extracted, the height of ship is calculated.The present invention can monitor whether ship exceedes bridge regulation headway, prevent the accident of ship from colliding bridge from occurring, assure the safety for life and property of the people to the distance of bridge, detection ship height in real time.

Description

A kind of method of binocular vision system measurement ship distance and height
Technical field
The present invention relates to measure ship distance and height more particularly to a kind of binocular vision system in bridge collision prevention system Method.
Background technology
China's rivers are numerous, and increasing crossing-fiver bridge construction has to China's economic development greatly facilitates effect. But at the same time, often as the multiple area of maritime accidents forecast, bring huge people because bridge zone waters navigation environment is complicated The loss of member, economy, environment etc..The origin cause of formation of ship collision Bridge Accidents is varied, but often caused by bridge itself Caused by " genetic prerequisite " is insufficient, and ship operation is improper.
The old bridge having is in design, the waterway classification design standard of the satisfaction such as yardstick, anticollision at that time, due to navigation channel liter Level, larger-sized vessel, these old bridges can not meet the navigation demand of present ship.With the development of larger-sized vessel, bridge Beam Clearance dimension deficiency, ship superelevation, super drinking water, the problems such as navigating by water getting worse violating the regulations, not only brought to maritime sector supervision Huge pressure, navigation safety also receive serious threat.
Domestic ship ranging at present and superelevation detection technique more fall behind, mostly with the ranging height-finding system of Introduced From Abroad Based on equipment, and research and utilization ultrasonic wave and laser ranging survey height are had focused largely on, but ultrasonic ranging height-finding system is uncomfortable For windy occasion, and laser ranging height-finding system needs to use generating laser, laser pickoff and a variety of optical sensings Device, more sensor cause that its manufacture difficulty is larger, and cost is higher, and due to needing to keep the clean of optical system, Bridge zone open space working service is very inconvenient.A kind of sensor-based system using single-mode reality simultaneously is designed in consideration of it, needing badly The measurement of ship distance and height in existing bridge-collision-avoidance early warning system.
At present the method for main limitation ship superelevation be according to the situation of the tonnage of ship, according to water level information, with reference to The information such as the experience of navigation, real-time headroom, general restriction, but often due to supervision difficulty are carried out to the tonnage for ship of opening the navigation or air flight It is big, it is specified that uncertainty, and the demand of local economic development can not make rational mandatory provision to ship.Ship Oceangoing ship can will be by ship operation personnel rule of thumb by bridge main, and environmental information at that time makes their own and judged whether Pass through bridge.But some ship operations personnel leave things to chance psychology, and it is different per bridge block head room, plus external ship Terrain environment is not known about, the main reason for often generation into accident.
The content of the invention
It is an object of the invention to provide a kind of method of binocular vision system measurement ship distance and height, in ship During nearly bridge, it can monitor whether ship exceedes bridge regulation navigation to the distance of bridge, detection ship height in real time Highly, prevent the accident of ship from colliding bridge from occurring, assure the safety for life and property of the people, reduce system cost, reduce system and adjust Examination and maintenance difficulties.
What the present invention was achieved through the following technical solutions：
A kind of method of binocular vision system measurement ship distance and height, comprises the following steps：
Step 1:Establish two binocular vision systems being made up of two independent camera systems, the binocular vision system Surveyed for ship ranging high；
Step 2:Single camera calibration, draw relative between the inner parameter of each camera and two cameras Position；
Step 3：Two cameras obtain left images and pre-processed respectively, obtain the elementary contour information of ship, So as to obtain on profile and contoured interior two dimensional image pixel point coordinates；
Step 4：Binocular corrects, and matches target point corresponding picture point on left images, calculates target point left and right two The parallax formed on image；
Step 5：Stereo matching, according to parallax, obtain disparity map；
Step 6：Three-dimensional reconstruction；After getting parallax data, with reference to calibrating parameters and the two-dimensional image of ship contour area Vegetarian refreshments coordinate, you can by the two-dimensional points coordinate re-projection in image be three-dimensional world coordinate；
Step 7：The profile of ship is detected in two field picture, the pixel two dimension on ship profile and in profile is obtained and sits Data are marked, two-dimensional coordinate is converted into three-dimensional coordinate, using principle of triangulation, calculates ship to the distance of bridge, must Range data is arrived.
Step 8：Detect the profile of ship in two field picture, obtain in two field picture on ship profile and the pixel of profile quilt Point two-dimensional coordinate data, is converted to three-dimensional coordinate by two-dimensional coordinate, extracts the seat that y values are minimum and maximum in three-dimensional coordinate point Mark, minimum y-coordinate value subtract the height that maximum y-coordinate value obtains ship.
Further, two binocular vision systems in step 1 include two focal lengths and other inner parameter all sames Video camera, two video cameras are by optical axis is parallel to each other, X-axis mutually coincides, is placed along X-axis at a distance of T position.
Further, step 2 is based on gridiron pattern standardization, the internal reference of each camera and outer ginseng is calculated, so as to obtain Camera focal length f and imaging origin cx, cy, and eliminate radial distortion and tangential distortion and the generation for obtaining demarcating thing of camera Boundary's coordinate, binocular camera, which is calibrated, draws the inner parameter of each camera, and the relative position between two cameras, i.e., with On the basis of left camera, spin matrix R, translation vector t of the right camera relative to left camera.
Further, step 3 passes through the synchronous crawl water of two cameras of binocular vision system for being installed on bridge or so Face image, the collection to ship information is realized, gray proces, Gaussian Blur pre-treatment step are carried out to image, existing for removal Noise；Contour detecting is carried out to ship, obtains the elementary contour information of ship, is obtained on profile and contoured interior two dimensional image picture Vegetarian refreshments coordinate.
Further, the two images that step 4 will be eliminated after distortion using epipolar-line constraint, which are strictly gone, to be corresponded to so that two width Image to polar curve just in the same horizontal line so that any point and its pair on another piece image on piece image Should put just necessarily has identical line number, and only need to carry out linear search in the row can match corresponding points, calculate target point and exist The parallax formed on the image of left and right two.
Beneficial effects of the present invention：The invention provides a kind of side of binocular vision system measurement ship distance and height Method, utilize the binocular camera for being installed on bridge, it is only necessary to measure by capturing the picture frame can containing ship information For ship apart from the distance of bridge and the height of ship, the measuring method is quick, timely, efficient, can be good at meeting in real time Property demand, play early warning effect to bridge zone ship navigation, help safety of ship to pass through, play and ensure traffic safety and people's people's livelihood Order property safety effect.
Brief description of the drawings
Fig. 1 is Binocular Stereo Vision System imaging schematic diagram of the present invention；
Fig. 2 is the operation principle flow chart of binocular vision system of the present invention；
Fig. 3 is video camera imaging schematic diagram of the present invention；
Fig. 4 is image coordinate system schematic diagram of the present invention；
Fig. 5 is epipolar-line constraint schematic diagram of the present invention；
Fig. 6 is principle of triangulation figure of the present invention；
Fig. 7 is ship elevation carrection schematic diagram of the present invention.
Embodiment
The present invention is described in detail below in conjunction with accompanying drawing and specific embodiment, below will herein with the signal of the present invention The present invention is described in detail with reference to accompanying drawing and specific embodiment, is used for solving with the illustrative examples of the present invention and explanation herein The present invention is released, but it is not as a limitation of the invention.
Concrete scheme is as follows：
(1) according to the requirement of bridge collision prevention system, two independent camera systems are chosen, structure is applied to ship ranging Survey the binocular vision system of high request.It is mutually equal by optical axis using two focal lengths and the video camera of other inner parameter all sames Row, X-axis mutually coincide, place just composition said system at a distance of T position along X-axis.Wherein focal length of camera and two shootings The distance between machine T (i.e. baseline length) has considerable influence to measurement accuracy.
(2) single camera calibration：Based on gridiron pattern standardization, the internal reference of each camera and outer ginseng are calculated, so as to Obtain camera focal length f and imaging origin cx, cy, and eliminate camera radial distortion and tangential distortion and obtain demarcate thing World coordinates.Binocular camera calibrates the inner parameter that not only draw each camera, it is also necessary to is measured by demarcation Between two cameras relative position (i.e. on the basis of left camera, right camera relative to left camera spin matrix R, translation vector t).
(3) acquisition and pretreatment of image：It is same by the binocular vision system that is installed on bridge or so two cameras Step crawl Surface Picture, realizes the collection to ship information.The pre-treatment steps such as gray proces, Gaussian Blur are carried out to image, Noise existing for removal；Contour detecting is carried out to ship, obtains the elementary contour information of ship, is obtained on profile and contoured interior Two dimensional image pixel point coordinates.
(4) binocular corrects：The parallax that Three dimensional Targets point is formed on the image of left and right two is calculated, is first had to this Point two corresponding Pixel matchings on left images get up.However, on two-dimensional space match corresponding points be it is very time-consuming, In order to reduce matching hunting zone, using epipolar-line constraint so that the matching of corresponding points is reduced to linear search by two-dimensional search.Binocular School is precisely in order to strictly row is corresponding the two images after distorting are eliminated so that two images to polar curve just in same water On horizontal line, any point just necessarily has identical line number with its corresponding points on another piece image on such piece image, Linear search only, which need to be carried out, in the row can match corresponding points.
(5) Stereo matching：Binocular measuring principle is parallax of the utilization space same point in two camera imaging planes Size calculates the three-dimensional coordinate of spatial point.Complete the measurement of space coordinates point it may first have to know spatial point in two image planes Picpointed coordinate, this two picture point be referred to as match picture point.In actually measuring, the image to be obtained from left and right cameras recovers empty Between the three-dimensional coordinate put, the picpointed coordinate of matching, referred to as Stereo matching must be found in two images, so as to obtain parallax numbers According to obtaining disparity map.
(6) three-dimensional reconstruction：After getting parallax data, 3D point cloud is generated by three-dimensional reconstruction, 3D point cloud is 2D images Three-dimensional coordinate corresponding to pixel coordinate.
(7) ranging：The profile of ship is detected in two field picture, obtains the pixel two dimension on ship profile and in profile Coordinate data, two-dimensional coordinate is converted into three-dimensional coordinate, using principle of triangulation, calculates ship to the distance of bridge, just Range data is obtained.
(8) survey high：Detect the profile of ship in two field picture, obtain in two field picture on ship profile and the picture of profile quilt Vegetarian refreshments two-dimensional coordinate data, two-dimensional coordinate is converted into three-dimensional coordinate, extracts the seat that y values are minimum and maximum in three-dimensional coordinate point Mark, minimum y-coordinate value subtract the height that maximum y-coordinate value obtains ship.
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples The present invention is further elaborated.All steps of the present embodiment are all base Microsoft Visual Studio exploitations What environment and OpenCV function libraries were completed.
Fig. 1 is the principle of Binocular Stereo Vision System imaging.If two camera horizons are placed, o, o ' it is respectively two shootings The centre of perspectivity of machine, sensor coordinates origin are set to o points, spatial point p (x, y, the z) picpointed coordinates in two video camera image planes point Wei not P1(X1,Y1),P2(X2,Y2).If centre of perspectivity's spacing of two video cameras is T, the effective focal length f of video camera1,f2.Object point p The angle in xoz planes to the measurement angles of two centres of perspectivity being respectively α, β, two camera optical axis and x-axis is α0,β0Picture point P1,P2Horizontal field of view angle to the video camera centre of perspectivity is γ1,γ2, the angle of visual field of vertical direction is θ1,θ2。
From measurement of coordinates mathematical modeling, three-dimensional coordinate of the spatial point in Binocular Stereo Vision System coordinate system is not only Depending on its imaging point coordinate value in two image planes, the structural parameters T, α of sensor are additionally depended on0,β0,f1,f2。
With x, y, z coordinate direction of principal axis measurement error denotation coordination measurement composition error, then have：
Use ψiThe transmission function of each factor error is represented, it is defined as：
In formula, k takes x, y, z respectively；I represents T, α0,β0,f1,f2,X1,X2,Y1,Y2；FkDenotation coordination function is in each coordinate Projected on axle；δiIt is factor i error deltaiTo the component of global error contribution.
Binocular Stereo Vision System design process：
According to information such as the ship target distance and object height of required measurement and channel spans, sensor is primarily determined that The measurement depth of field and baseline length T size and the unilateral effective angle field range of video camera.Available field of view should be controlled in error Carry-over factor changes in the range of less angular field, i.e., the angle of visual field of video camera should be controlled within 30 °, within preferably 20 °.
According to the unilateral effective angle field range of video camera, the angle of camera optical axis and baseline is obtained by following formula:
Wherein, Φ is the angle of camera optical axis and baseline, and γ horizontal field of view angle, θ is vertical field of view angle.
3. according to the required precision of measurand, the scope that measurement error Δ is allowed is determined, is located at maximum field of view and is Δmax, estimate the size of image procossing error, the lens focus f of video camera obtained by following formula：
Wherein, δ represents the error of algorithm extraction picpointed coordinate, ψmaxRepresent the maximum of error transfer function.
4. determine video camera CCD form.According to horizontal angular field γmaxIt is big that CCD image planes sizes are calculated with lens focus f It is small, select suitable video camera.I.e.：
X=f*tan γmax
5. a pair calculating focal length is standardized, select closest to the standard focal length camera lens for calculating focal length.
6. after video camera and camera lens size is determined, final baseline length T size is determined by following formula：
It is such as undesirable, restart again from the first step, untill meeting to require.
Due to mutually being restricted between sensor construction parameter, when above theoretical design requirements can not be met, Ying Gen Traded off these structural parameters according to measurand and measuring environment.
Camera calibration：
In the ideal case, the imaging model of video camera is regarded as national forest park in Xiaokeng, as shown in figure 3, (Ow- xwywzw) it is world coordinate system, (Oc-xcyczc) it is camera coordinate system, (O1- xy) it is image physical coordinates system, (O-uv) is figure As pixel coordinate system.P is any point in three-dimensional scenic, P and video camera photocentre O line and video camera imaging plane phase Hand over, its intersection point is Pu.From object point P to imaging point PuBetween relation can be represented by following coordinate transform：
Transformational relation between camera coordinate system and world coordinate system can be by spin matrix R and translation matrix t expressions, can It is expressed as：
By similar triangles transformation of scale, the conversion between image physical coordinates system and camera coordinate system can be obtained Relation：
Such as Fig. 4, if principal point O1It is (u in the coordinate of image pixel coordinates system0,v0), each pixel is in image physical coordinates Physical size in system in x-axis and y-axis direction is respectively dx and dy, then image pixel coordinates system under image physical coordinates system with sitting Target transformational relation is as follows：
With reference to above formula, coordinate (xs of certain object point P under world coordinate system in three-dimensional scenic can be obtainedw,yw,zw) with its Relation between the coordinate (u, v) of subpoint Pu in video camera imaging plane is as follows：
Wherein, M is 3 × 4 matrixes, referred to as projection matrix；F be video camera focal length, fxFor Jiao of the video camera in x directions Away from fyFor focal length of the video camera in y directions, (u0,v0) be video camera principal point, M1 is completely by fx、fy、u0、v0Determine, be referred to as taking the photograph The inner parameter of camera；M2 determines by spin matrix R and translation vector t, referred to as the external parameter of video camera；XwExist for spatial point Homogeneous coordinates under world coordinate system.
Seek to obtain intrinsic parameter M by camera calibration1With outer parameter M2。
Using gridiron pattern standardization, i.e., a calibrated reference (chessboard paper) is placed in front of camera, camera obtains The image of chessboard paper.In order to obtain relatively good demarcation effect, it is necessary to convert the direction of chessboard paper, camera is set to obtain not Tongfang To 10~20 groups of chessboard paper image, thus calculate the Intrinsic Matrix of camera, spin matrix, translation matrix.So as to right Radial distortion existing for camera and tangential distortion are corrected.
Fig. 2 is the operation principle flow that binocular vision system measures ship distance and height in a kind of bridge collision prevention system Figure, is comprised the following steps that：
The acquisition and pretreatment of image：
Surface Picture is synchronously captured by the camera of left and right two being installed on bridge, realizes and ship information is adopted Collection.Gray proces and Gaussian Blur processing, noise existing for removal are carried out to the image collected；Edge inspection is carried out to image Survey, then contour detecting is carried out to image, obtain the elementary contour information of ship, obtain on profile and contoured interior two dimensional image picture Vegetarian refreshments coordinate.
Binocular corrects：
Target point caused parallax in the camera image of left and right two is calculated, first has to match the point in left and right figure As upper corresponding picture point.It is very time-consuming due to matching corresponding picture point on 2d, in order to reduce matching hunting zone, utilize Epipolar-line constraint causes the matching of corresponding points to be reduced to linear search by two-dimensional search.As shown in figure 5, pl、prTo be corresponded on left images Picture point, then pl、pr、P、ol、orOn with a pair of polar planes, the plane and the intersection of two planes of delineation are respectively polar curve l1 With polar curve l2, then plIn l1On, prIn l2On.Obviously, no matter and plIt is closely remote that corresponding P, which is, and its projection in right image is total In l2On, vice versa, projections of the P in left image is always in polar curve l1On.So to any point plOnly need at it Polar curve on find corresponding points pr, here it is epipolar-line constraint.
And the effect of binocular correction seeks to the two images after elimination distortion strictly to go to correspond to so that two images To polar curve, any point and its corresponding points on another piece image are just just in the same horizontal line, on such piece image Necessarily there is identical line number, only need to carry out linear search in the row can match corresponding points, be convenient to the solid of next step Matching.
Binocular ranging：
Binocular ranging is that most difficult and most critical a step, its result directly influence three-dimensional survey in binocular stereo vision The accuracy and speed of amount.Its principle is the same target point of matching in left and right camera imaging pair, by matching corresponding picture Point, calculates parallax, obtains disparity map.
Three-dimensional reconstruction：
After getting parallax data, with reference to calibrating parameters and the two-dimensional pixel point coordinates of ship contour area, you can will Two-dimensional points coordinate re-projection in image is three-dimensional world coordinate, and re-projection matrix is as follows：
Above formula removes c 'rOuter all parameters both are from left image, c 'rIt is x coordinate of the principal point in right image, T is left and right The centre-to-centre spacing of two cameras, cxWith cyFor the offset of origin in the coordinate system and three-dimensional coordinate system of left and right image plane.If key light Line intersects at infinity, then cx=c 'r, and the item in the lower right corner is 0, the given homogeneous point of a two dimension regards with what it was associated Poor d, it is possible to by this spot projection into three-dimensional：
Three-dimensional coordinate is exactly (X/W, Y/W, Z/W).
In practical operation, based on above-mentioned principle, need to only utilize cvStereoRectify () function in Opencv to Re-projection matrix Q is calculated after going out camera interior and exterior parameter, then re-projection matrix and disparity map disparityImage are inputted It can obtain into function reProjectImageTo3D () on ship profile and the three-dimensional coordinate number of each pixel of contoured interior According to.
Measurement of the ship to bridge distance：
Ship to bridge distance ask for process be actually in three-dimensional reconstruction Z coordinate ask for process,
In order to describe ship range measurement principle in detail, provide binocular stereo vision and measure ship to the basic of bridge distance Schematic diagram, such as Fig. 6, point P be ship z coordinate in camera coordinate system it is minimum a bit, OlWith OrIt is binocular stereo vision respectively The photocentre of two cameras of left and right in system, point P is p in the imaging point of left and right cameral、pr, xlAnd xrRespectively point P is in left and right The abscissa being imaged in two images, f are camera focal length, and T is the centre-to-centre spacing of two cameras in left and right, plTo prDistance be Parallax d, z is the depth information to be tried to achieve, then by Δ pplprSimilar in appearance to Δ pOlOr, proportional relation:
So as to：
I.e.：
Distance of the target point to imaging plane as available from the above equation:
Wherein d=xl-xrReferred to as parallax.
The measurement of ship height：
After ship profile carries out three-dimensional reconstruction on the ship profile that seeks out and each pixel of ship contoured interior In three-dimensional coordinate, such as Fig. 7, the minimum coordinate points p of y-coordinate is searched out1(X1,Y1,Z1) the coordinate points p maximum with y-coordinate2(X2, Y2,Z2), because two cameras are that parallel shelf is located on bridge, and the origin of regulation world coordinate system is the light of left camera The heart, i.e. world coordinate system overlap with left camera coordinate system, therefore only the y values of minimum need to be subtracted into maximum using what is above drawn Y values, so that it may try to achieve the height of ship.I.e.：H=Y1-Y2。
The invention provides ship distance in a kind of bridge collision prevention system and the binocular vision system of elevation carrection, peace is utilized Binocular camera loaded on bridge, it is only necessary to measure ship apart from bridge by capturing the picture frame can containing ship information The distance of beam and the height of ship, the measuring method is quick, timely, efficient, can be good at meeting the needs of real-time, right Bridge zone ship navigation plays early warning effect, helps safety of ship to pass through, and plays ensuring traffic safety and people life property safety Effect.The system realizes the measurement of ship distance and height only with a kind of sensor simultaneously simultaneously, and algorithm is simple, is designed to This is relatively low, reduces the difficulty of multi-sensor cooperation debugging and multisensor error correction.
For those skilled in the art, without departing from the inventive concept of the premise, this hair Bright embodiment is not limited to examples detailed above, and in the specific implementation process of the present invention, embodiment can be with Some deductions or optimization are made, these are deduced or optimization should all be considered as belonging to protection scope of the present invention.

Claims (5)

  1. A kind of 1. method of binocular vision system measurement ship distance and height, it is characterised in that：Comprise the following steps：
    Step 1:Two binocular vision systems being made up of two independent camera systems are established, the binocular vision system is used for Ship ranging is surveyed high；
    Step 2:Single camera calibration, draw the relative position between the inner parameter of each camera and two cameras Put；
    Step 3：Two cameras obtain left images and pre-processed respectively, obtain the elementary contour information of ship, so as to Obtain profile on and contoured interior two dimensional image pixel point coordinates；
    Step 4：Binocular corrects, and matches target point corresponding picture point on left images, calculates target point in the image of left and right two The parallax of upper formation；
    Step 5：Stereo matching, according to parallax, obtain disparity map；
    Step 6：Three-dimensional reconstruction；After getting parallax data, with reference to calibrating parameters and the two-dimensional image vegetarian refreshments of ship contour area Coordinate, you can by the two-dimensional points coordinate re-projection in image be three-dimensional world coordinate；
    Step 7：The profile of ship is detected in two field picture, obtains the pixel two-dimensional coordinate number on ship profile and in profile According to two-dimensional coordinate is converted into three-dimensional coordinate, using principle of triangulation, ship is calculated to the distance of bridge, just obtains Range data；
    Step 8：Detect the profile of ship in two field picture, obtain in two field picture on ship profile and the pixel two of profile quilt Dimension coordinate data, two-dimensional coordinate is converted into three-dimensional coordinate, extracts the coordinate that y values are minimum and maximum in three-dimensional coordinate point, most Small y-coordinate value subtracts the height that maximum y-coordinate value obtains ship.
  2. 2. the method for a kind of binocular vision system measurement ship distance and height according to claim 1, it is characterised in that： Two binocular vision systems in step 1 include two focal lengths and the video camera of other inner parameter all sames, two video cameras By optical axis is parallel to each other, X-axis mutually coincides, is placed along X-axis at a distance of T position.
  3. 3. the method for a kind of binocular vision system measurement ship distance and height according to claim 1, it is characterised in that： Step 2 is based on gridiron pattern standardization, calculates the internal reference of each camera and outer ginseng, so as to obtain camera focal length f and imaging Origin cx, cy, and eliminate radial distortion and tangential distortion and the world coordinates for obtaining demarcating thing of camera, binocular camera Calibration draws the inner parameter of each camera, with the relative position between two cameras, i.e., right on the basis of left camera Spin matrix R, translation vector t of the camera relative to left camera.
  4. 4. the method for a kind of binocular vision system measurement ship distance and height according to claim 1, it is characterised in that： Step 3 is realized to ship by the synchronous crawl Surface Pictures of two cameras of binocular vision system for being installed on bridge or so The collection of information, gray proces, Gaussian Blur pre-treatment step, noise existing for removal are carried out to image；Road wheel is entered to ship Exterior feature detection, obtains the elementary contour information of ship, obtains on profile and contoured interior two dimensional image pixel point coordinates.
  5. 5. the method for a kind of binocular vision system measurement ship distance and height according to claim 1, it is characterised in that： Step 4 will eliminate the two images after distortion using epipolar-line constraint, and strictly row is corresponding so that two images it is lucky to polar curve In the same horizontal line, so that any point just necessarily has phase with its corresponding points on another piece image on piece image Same line number, only need to carry out linear search in the row can match corresponding points, calculate target point shape on the image of left and right two Into parallax.
CN201711050019.7A 2017-10-31 2017-10-31 A kind of method of binocular vision system measurement ship distance and height Pending CN107884767A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711050019.7A CN107884767A (en) 2017-10-31 2017-10-31 A kind of method of binocular vision system measurement ship distance and height

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711050019.7A CN107884767A (en) 2017-10-31 2017-10-31 A kind of method of binocular vision system measurement ship distance and height

Publications (1)

Publication Number Publication Date
CN107884767A true CN107884767A (en) 2018-04-06

Family

ID=61783218

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711050019.7A Pending CN107884767A (en) 2017-10-31 2017-10-31 A kind of method of binocular vision system measurement ship distance and height

Country Status (1)

Country Link
CN (1) CN107884767A (en)

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108682186A (en) * 2018-04-25 2018-10-19 中国科学院长春光学精密机械与物理研究所 A kind of early warning system and method for bridge collision accident
CN108830905A (en) * 2018-05-22 2018-11-16 苏州敏行医学信息技术有限公司 The binocular calibration localization method and virtual emulation of simulating medical instrument cure teaching system
CN109360238A (en) * 2018-09-25 2019-02-19 广东国地规划科技股份有限公司 A kind of field data acquisition method and system based on binocular ranging
CN109686104A (en) * 2018-12-31 2019-04-26 武汉静磁栅机电制造有限公司 A kind of vehicle super-high detection device
CN109781016A (en) * 2019-03-13 2019-05-21 江西五十铃汽车有限公司 A kind of method of the high performances such as real-time measurement automobile
CN109813278A (en) * 2019-02-26 2019-05-28 北京中科慧眼科技有限公司 Ranging model modification method, distance measuring method, device and automated driving system
CN109903327A (en) * 2019-03-04 2019-06-18 西安电子科技大学 A kind of object dimension measurement method of sparse cloud
CN110207650A (en) * 2019-05-31 2019-09-06 重庆迪星天科技有限公司 Automobile-used highway height-limiting frame height measurement method and device based on binocular vision
CN110220471A (en) * 2019-06-27 2019-09-10 广西柳钢东信科技有限公司 A kind of molten metal trough erosion degree detection method
CN110751077A (en) * 2019-10-15 2020-02-04 武汉大学 Optical remote sensing picture ship detection method based on component matching and distance constraint
CN111197973A (en) * 2020-01-13 2020-05-26 上海眼控科技股份有限公司 Sea surface distance acquisition method and device, computer equipment and storage medium
CN111210468A (en) * 2018-11-22 2020-05-29 中移（杭州）信息技术有限公司 Image depth information acquisition method and device
CN111273763A (en) * 2018-12-05 2020-06-12 扩展知识公司 Augmented reality implementation method and computer system for implementing augmented reality
CN111709633A (en) * 2020-06-09 2020-09-25 吉林大学 Method, device and equipment for determining collision risk degree and storable medium
CN111986517A (en) * 2020-08-26 2020-11-24 珠海大横琴科技发展有限公司 Ship anomaly detection method and device and storage medium
CN112001962A (en) * 2020-07-24 2020-11-27 北京驭光科技发展有限公司 Ship height measuring and calculating method and system
CN112161997A (en) * 2020-09-28 2021-01-01 南京工程学院 Online precise visual measurement method and system for three-dimensional geometric dimension of semiconductor chip pin
CN112288853A (en) * 2020-10-29 2021-01-29 字节跳动有限公司 Three-dimensional reconstruction method, three-dimensional reconstruction device, and storage medium
CN112541951A (en) * 2020-11-13 2021-03-23 国网浙江省电力有限公司舟山供电公司 Monitoring system and monitoring method for preventing ship from hooking off cross-sea overhead power line
CN112686938A (en) * 2020-12-29 2021-04-20 重庆大学 Electric transmission line clear distance calculation and safety warning method based on binocular image ranging
CN112697072A (en) * 2020-12-03 2021-04-23 湖北文理学院 Precision self-checking road surface three-dimensional texture measuring method and system and storage medium
CN112955781A (en) * 2018-10-24 2021-06-11 Ids地质雷达有限公司 Photogrammetry system for locating geological radar data on a survey scene
CN113534737A (en) * 2021-07-15 2021-10-22 中国人民解放***箭军工程大学 PTZ (Pan/Tilt/zoom) dome camera control parameter acquisition system based on multi-view vision
CN114178282A (en) * 2021-11-12 2022-03-15 国能铁路装备有限责任公司 Brake beam cleaning production line, identification and positioning system, device and method
CN114322785A (en) * 2021-12-01 2022-04-12 浙江省送变电工程有限公司 Power transmission line multi-bundle conductor sag detection device and detection and adjustment method
CN115147723A (en) * 2022-07-11 2022-10-04 武汉理工大学 Inland ship identification and distance measurement method, system, medium, equipment and terminal
CN115265382A (en) * 2022-07-05 2022-11-01 浙江海莱云智科技有限公司 Ship height measurement and identification method based on video image
CN115797459A (en) * 2022-08-29 2023-03-14 南京航空航天大学 Binocular vision system distance measurement method with arbitrary focal length combination
CN117589064A (en) * 2024-01-19 2024-02-23 中理检验有限公司 Automatic wood measuring method and system

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS57196105A (en) * 1981-05-28 1982-12-02 Toshiba Corp Three-dimentional distance measuring method
CN102867304A (en) * 2012-09-04 2013-01-09 南京航空航天大学 Method for establishing relation between scene stereoscopic depth and vision difference in binocular stereoscopic vision system
CN103407407A (en) * 2013-08-28 2013-11-27 沈阳工业大学 Automobile safety distance warning device and method based on binocular stereo vision
CN203616781U (en) * 2013-12-10 2014-05-28 广州市地下铁道总公司 Binocular three-dimensional reconstruction visual system based on slide guide rails
CN105550670A (en) * 2016-01-27 2016-05-04 兰州理工大学 Target object dynamic tracking and measurement positioning method
CN105913410A (en) * 2016-03-03 2016-08-31 华北电力大学（保定） Long-distance moving object height measurement apparatus and method based on machine vision

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS57196105A (en) * 1981-05-28 1982-12-02 Toshiba Corp Three-dimentional distance measuring method
CN102867304A (en) * 2012-09-04 2013-01-09 南京航空航天大学 Method for establishing relation between scene stereoscopic depth and vision difference in binocular stereoscopic vision system
CN103407407A (en) * 2013-08-28 2013-11-27 沈阳工业大学 Automobile safety distance warning device and method based on binocular stereo vision
CN203616781U (en) * 2013-12-10 2014-05-28 广州市地下铁道总公司 Binocular three-dimensional reconstruction visual system based on slide guide rails
CN105550670A (en) * 2016-01-27 2016-05-04 兰州理工大学 Target object dynamic tracking and measurement positioning method
CN105913410A (en) * 2016-03-03 2016-08-31 华北电力大学（保定） Long-distance moving object height measurement apparatus and method based on machine vision

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
崔伟 等: "基于双目视觉的三维测量技术研究", 《第二十七届全国振动与噪声应用学术会议论文集》 *
韩慧妍: "基于双目立体视觉的三维模型重建方法研究", 《中国博士学位论文全文数据库 信息科技辑（月刊）》 *
黄椰 等: "基于双目立体视觉的船舶轨迹跟踪算法研究", 《计算机科学》 *

Cited By (48)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108682186B (en) * 2018-04-25 2021-05-18 中国科学院长春光学精密机械与物理研究所 Early warning system and method for ship bridge collision accident
CN108682186A (en) * 2018-04-25 2018-10-19 中国科学院长春光学精密机械与物理研究所 A kind of early warning system and method for bridge collision accident
CN108830905A (en) * 2018-05-22 2018-11-16 苏州敏行医学信息技术有限公司 The binocular calibration localization method and virtual emulation of simulating medical instrument cure teaching system
CN109360238B (en) * 2018-09-25 2020-05-12 广东国地规划科技股份有限公司 Field data acquisition method and system based on binocular ranging
CN109360238A (en) * 2018-09-25 2019-02-19 广东国地规划科技股份有限公司 A kind of field data acquisition method and system based on binocular ranging
CN112955781A (en) * 2018-10-24 2021-06-11 Ids地质雷达有限公司 Photogrammetry system for locating geological radar data on a survey scene
CN111210468A (en) * 2018-11-22 2020-05-29 中移（杭州）信息技术有限公司 Image depth information acquisition method and device
CN111210468B (en) * 2018-11-22 2023-07-11 中移（杭州）信息技术有限公司 Image depth information acquisition method and device
CN111273763A (en) * 2018-12-05 2020-06-12 扩展知识公司 Augmented reality implementation method and computer system for implementing augmented reality
CN109686104B (en) * 2018-12-31 2020-12-11 武汉静磁栅机电制造有限公司 Vehicle superelevation detection method
CN109686104A (en) * 2018-12-31 2019-04-26 武汉静磁栅机电制造有限公司 A kind of vehicle super-high detection device
CN109813278A (en) * 2019-02-26 2019-05-28 北京中科慧眼科技有限公司 Ranging model modification method, distance measuring method, device and automated driving system
CN109813278B (en) * 2019-02-26 2021-09-17 北京中科慧眼科技有限公司 Ranging model correction method, ranging method and device and automatic driving system
CN109903327A (en) * 2019-03-04 2019-06-18 西安电子科技大学 A kind of object dimension measurement method of sparse cloud
CN109781016A (en) * 2019-03-13 2019-05-21 江西五十铃汽车有限公司 A kind of method of the high performances such as real-time measurement automobile
CN109781016B (en) * 2019-03-13 2024-04-16 江西五十铃汽车有限公司 Method for measuring high performance of automobile and the like in real time
CN110207650A (en) * 2019-05-31 2019-09-06 重庆迪星天科技有限公司 Automobile-used highway height-limiting frame height measurement method and device based on binocular vision
CN110220471A (en) * 2019-06-27 2019-09-10 广西柳钢东信科技有限公司 A kind of molten metal trough erosion degree detection method
CN110220471B (en) * 2019-06-27 2023-12-08 广西柳钢东信科技有限公司 Method for detecting erosion degree of molten metal tank
CN110751077A (en) * 2019-10-15 2020-02-04 武汉大学 Optical remote sensing picture ship detection method based on component matching and distance constraint
CN110751077B (en) * 2019-10-15 2022-07-05 武汉大学 Optical remote sensing picture ship detection method based on component matching and distance constraint
CN111197973A (en) * 2020-01-13 2020-05-26 上海眼控科技股份有限公司 Sea surface distance acquisition method and device, computer equipment and storage medium
CN111709633A (en) * 2020-06-09 2020-09-25 吉林大学 Method, device and equipment for determining collision risk degree and storable medium
CN111709633B (en) * 2020-06-09 2022-09-06 吉林大学 Method, device and equipment for determining collision risk degree and storable medium
CN112001962B (en) * 2020-07-24 2021-03-05 北京驭光科技发展有限公司 Ship height measuring and calculating method and system
CN112001962A (en) * 2020-07-24 2020-11-27 北京驭光科技发展有限公司 Ship height measuring and calculating method and system
CN111986517A (en) * 2020-08-26 2020-11-24 珠海大横琴科技发展有限公司 Ship anomaly detection method and device and storage medium
CN112161997A (en) * 2020-09-28 2021-01-01 南京工程学院 Online precise visual measurement method and system for three-dimensional geometric dimension of semiconductor chip pin
CN112288853B (en) * 2020-10-29 2023-06-20 字节跳动有限公司 Three-dimensional reconstruction method, three-dimensional reconstruction device, and storage medium
CN112288853A (en) * 2020-10-29 2021-01-29 字节跳动有限公司 Three-dimensional reconstruction method, three-dimensional reconstruction device, and storage medium
CN112541951A (en) * 2020-11-13 2021-03-23 国网浙江省电力有限公司舟山供电公司 Monitoring system and monitoring method for preventing ship from hooking off cross-sea overhead power line
CN112697072A (en) * 2020-12-03 2021-04-23 湖北文理学院 Precision self-checking road surface three-dimensional texture measuring method and system and storage medium
CN112697072B (en) * 2020-12-03 2023-04-25 湖北文理学院 Precision self-checking pavement three-dimensional texture measuring method and system and storage medium
CN112686938B (en) * 2020-12-29 2024-03-15 重庆大学 Power transmission line clear distance calculation and safety alarm method based on binocular image ranging
CN112686938A (en) * 2020-12-29 2021-04-20 重庆大学 Electric transmission line clear distance calculation and safety warning method based on binocular image ranging
CN113534737B (en) * 2021-07-15 2022-07-19 中国人民解放***箭军工程大学 PTZ (Pan/Tilt/zoom) dome camera control parameter acquisition system based on multi-view vision
CN113534737A (en) * 2021-07-15 2021-10-22 中国人民解放***箭军工程大学 PTZ (Pan/Tilt/zoom) dome camera control parameter acquisition system based on multi-view vision
CN114178282A (en) * 2021-11-12 2022-03-15 国能铁路装备有限责任公司 Brake beam cleaning production line, identification and positioning system, device and method
CN114178282B (en) * 2021-11-12 2023-10-10 国能铁路装备有限责任公司 Brake beam cleaning production line, identification positioning system, device and method
CN114322785A (en) * 2021-12-01 2022-04-12 浙江省送变电工程有限公司 Power transmission line multi-bundle conductor sag detection device and detection and adjustment method
CN115265382A (en) * 2022-07-05 2022-11-01 浙江海莱云智科技有限公司 Ship height measurement and identification method based on video image
CN115265382B (en) * 2022-07-05 2023-08-22 浙江海莱云智科技有限公司 Ship height measurement and identification method based on video image
CN115147723A (en) * 2022-07-11 2022-10-04 武汉理工大学 Inland ship identification and distance measurement method, system, medium, equipment and terminal
US11948344B2 (en) 2022-07-11 2024-04-02 Wuhan University Of Technology Method, system, medium, equipment and terminal for inland vessel identification and depth estimation for smart maritime
CN115797459B (en) * 2022-08-29 2024-02-13 南京航空航天大学 Binocular vision system ranging method with arbitrary focal length combination
CN115797459A (en) * 2022-08-29 2023-03-14 南京航空航天大学 Binocular vision system distance measurement method with arbitrary focal length combination
CN117589064A (en) * 2024-01-19 2024-02-23 中理检验有限公司 Automatic wood measuring method and system
CN117589064B (en) * 2024-01-19 2024-03-29 中理检验有限公司 Automatic wood measuring method and system

Similar Documents

Publication Publication Date Title
CN107884767A (en) A kind of method of binocular vision system measurement ship distance and height
CN107389026B (en) A kind of monocular vision distance measuring method based on fixed point projective transformation
WO2021223368A1 (en) Target detection method based on vision, laser radar, and millimeter-wave radar
CN106681353B (en) The unmanned plane barrier-avoiding method and system merged based on binocular vision with light stream
CN110349221A (en) A kind of three-dimensional laser radar merges scaling method with binocular visible light sensor
CN105841688B (en) A kind of ship auxiliary anchors alongside the shore method and system
CN103424112B (en) A kind of motion carrier vision navigation method auxiliary based on laser plane
CN110148169A (en) A kind of vehicle target 3 D information obtaining method based on PTZ holder camera
CN110230998B (en) Rapid and precise three-dimensional measurement method and device based on line laser and binocular camera
CN107316325A (en) A kind of airborne laser point cloud based on image registration and Image registration fusion method
CN105913410A (en) Long-distance moving object height measurement apparatus and method based on machine vision
CN111046776A (en) Mobile robot traveling path obstacle detection method based on depth camera
CN110119698A (en) For determining the method, apparatus, equipment and storage medium of Obj State
CN111091076B (en) Tunnel limit data measuring method based on stereoscopic vision
CN104034305A (en) Real-time positioning method based on monocular vision
CN108362205A (en) Space ranging method based on fringe projection
CN108180888A (en) A kind of distance detection method based on rotating pick-up head
CN110443819A (en) A kind of track detection method and device of monorail train
McGuire et al. Smart iceberg management system–rapid iceberg profiling system
CN108919319A (en) Sea island reef satellite image Pillarless caving localization method and system
CN105424059B (en) Wide baseline near infrared camera position and orientation estimation method
Tournaire et al. Towards a sub-decimetric georeferencing of groundbased mobile mapping systems in urban areas: Matching ground-based and aerial-based imagery using roadmarks
CN117092659A (en) System and method for jointly measuring ship height by laser imaging radar and camera
Chenchen et al. A camera calibration method for obstacle distance measurement based on monocular vision
Boerner et al. Brute force matching between camera shots and synthetic images from point clouds

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
RJ01 Rejection of invention patent application after publication

Application publication date: 20180406

RJ01 Rejection of invention patent application after publication