CN102980562B - System and method for scene of traffic accident image processing - Google Patents

System and method for scene of traffic accident image processing Download PDF

Info

Publication number
CN102980562B
CN102980562B CN201210461033.7A CN201210461033A CN102980562B CN 102980562 B CN102980562 B CN 102980562B CN 201210461033 A CN201210461033 A CN 201210461033A CN 102980562 B CN102980562 B CN 102980562B
Authority
CN
China
Prior art keywords
image
scene
subpoint
traffic accident
straight line
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.)
Expired - Fee Related
Application number
CN201210461033.7A
Other languages
Chinese (zh)
Other versions
CN102980562A (en
Inventor
邱钺
刘汉冬
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.)
Beijing Billion Profit Information Technology Co ltd
Original Assignee
Cs & S Delineate Information Technology Co Ltd
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 Cs & S Delineate Information Technology Co Ltd filed Critical Cs & S Delineate Information Technology Co Ltd
Priority to CN201210461033.7A priority Critical patent/CN102980562B/en
Publication of CN102980562A publication Critical patent/CN102980562A/en
Application granted granted Critical
Publication of CN102980562B publication Critical patent/CN102980562B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a system and method for scene of a traffic accident image processing. A scene of a traffic accident image processing system comprises an imaging device and an image processing device, wherein the image processing device and the imaging device have data transmission path. The image processing device comprises the following steps: obtaining the relationship between the coordinate system in the image and the world coordinate system; obtaining the coordinate in the world coordinate system of the point S waiting to be measured in the image; obtaining the expression of the datum line in the world coordinate system in the image; selecting a plurality of subpoints in the datum line in the image and when the straight line formed by one of the subpoints and point S is perpendicular to the datum line in the world coordinate system, then recording the selected subpoint T; identifying the line between S and T in the image. The system and method for scene of the traffic accident image processing is applied to the traffic accident site disposal and can improve efficiency and precision.

Description

Scene of traffic accident image treatment system and method
Technical field
The present invention relates to a kind of image processing techniques, particularly utilize the technology that image is measured image Scene.
Background technology
In traffic accident survey processes, a critical step measures the scene of a traffic accident, such as, measure the data such as position, direction of accident vehicle.To the understanding of accident generating process and the confirmation of responsibility of accident, there is important function by measuring the critical data obtained.
Generally need to carry out temporary transient closing to scene, to retain on-the-spot reset condition as far as possible to the measurement of correlation of the scene of a traffic accident.This in-site measurement Problems existing is: close the on-the-spot public vehicles that affect and pass through, thus cause secondary to block up; Because Measuring Time is longer, easily produce the reset condition that unexpected influence factor destorys the scene in this course, make to be difficult to realize the accurate identification to the accident fact.For this reason, people introduce image measurement technology.Image measurement technology in the application mode of scene of a traffic accident process is: quick obtaining image scene at the scene, comprise the scale of known geometric data in image, the coordinate system obtained like this on image just can set up relation by scale and on-the-spot world coordinate system.After the image obtaining q.s, just can recover to be open to traffic, relevant in-site measurement can by realizing the image scene measurement obtained.The data obtained after measuring image scene can form traffic accident situ map (hereinafter referred to as scene photo), to record on-the-spot critical data.Traffic accident treatment professional analyzes traffic accident according to scene photo.
The scene of a traffic accident process of above-mentioned introducing image measurement technology can improve in-situ processing efficiency, reduces the traffic impact to public vehicles.But from the angle of traffic accident treatment professional, the information embodied with the traffic accident situ map that special symbol is drawn is more abstract, cannot express details factors in the scene of a traffic accident in detail further.If can identify critical data and boost line in image at the scene, the wheel of such as accident vehicle to the roadside vertical line as datum line, then, except the data measured, can also take into account other on-the-spot key elements to traffic accident identification simultaneously.But this demand faces a problem: by calculating the elements such as world coordinate system next one point or line to the data in image, return in image process at inverse, due to reasons such as lens distortions, inverse algorithm exists comparatively big error.Element under world coordinate system does not correctly identify out by inverse algorithm in the picture.To the vertical line in roadside, this problem is described for wheel: by the calculating to scene of traffic accident image, world coordinate system can be obtained get off and take turns to the vertical line in roadside, but when this vertical line inverse is returned image coordinate system, vertical line there will be problem that can not be crossing with roadside (datum line), or intersected, but the point that hangs down has deviation.Such result can make traffic accident treatment professional occur severe deviations to accident diagnoses.
Summary of the invention
In order to solve the problem that the element under world coordinate system correctly can not be identified in the picture in prior art, the invention provides a kind of scene of traffic accident image treatment system and method.
Technical scheme of the present invention is as follows:
Scene of traffic accident image treatment system, comprises imaging device and has the image processing apparatus of data transmission path with described imaging device, and described imaging device generates scene of traffic accident image; Put the scale of known geometric data in the described scene of a traffic accident, described scale occurs in described image;
Described image processing apparatus performs following steps:
A, with described scale for intermediary, obtain the relation between coordinate system and world coordinate system in described image;
B, obtain the coordinate of some S under world coordinate system to be measured in described image;
C, obtain the expression formula of datum line under world coordinate system in described image;
D, on datum line described in described image, select some subpoints, when judging that under world coordinate system one of described subpoint and the straight line that S is formed are vertical with described datum line, then this subpoint is recorded as T;
E, in described image, identify the line of S and T.
Preferably, judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straightway with one of described subpoint respectively, the subpoint that wherein the shortest straightway is corresponding is T.
Preferably, judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straight line or straightway with one of described subpoint respectively, calculate the angle of described straight line or straightway and described datum line, when described angle is right angle, then this straight line or described subpoint corresponding to straightway are T.
Preferably, on described datum line, in step D, select the step of some subpoints to comprise: to be pivot through the straight line of S with S, and rotate with predetermined progress, described straight line formation crossing with described datum line crosspoint when rotating to a position; Described crosspoint is described subpoint.
Preferably, judge that the step of described straight line or straightway and described datum line angle comprises: the trigonometric function value calculating described straight line or straightway and described datum line angle, when the difference of the described trigonometric function value the obtained trigonometric function value corresponding to right angle is in accuracy rating, then think that this angle is right angle.
Preferably, step e also comprises: the length value identifying the line of S and T in described image.
Scene of traffic accident image processing method, puts the scale of known geometric data in the described scene of a traffic accident, described scale occurs in described image; Also comprise the steps:
A, with described scale for intermediary, obtain the relation between coordinate system and world coordinate system in described image;
B, obtain the coordinate of some S under world coordinate system to be measured in described image;
C, obtain the expression formula of datum line under world coordinate system in described image;
D, on datum line described in described image, select some subpoints, when judging that under world coordinate system one of described subpoint and the straight line that S is formed are vertical with described datum line, then this subpoint is recorded as T;
E, in described image, identify the line of S and T.
Preferably, in described scene of traffic accident image processing method, judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straightway with one of described subpoint respectively, the subpoint that wherein the shortest straightway is corresponding is T.
Preferably, in described scene of traffic accident image processing method, judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straight line or straightway with one of described subpoint respectively, calculate the angle of described straight line or straightway and described datum line, when described angle is right angle, then this straight line or described subpoint corresponding to straightway are T.
Preferably, in described scene of traffic accident image processing method, on described datum line, the step of some subpoints is selected to comprise in step D: be pivot through the straight line of S with S, and rotate with predetermined progress, described straight line formation crossing with described datum line crosspoint when rotating to a position; Described crosspoint is described subpoint.
Technique effect of the present invention:
Technical scheme of the present invention, by datum line in the picture being selected some subpoints (i.e. the subpoint of S on datum line), next under world coordinate system, judge that whether the line of subpoint and the point to be measured selected is perpendicular to described datum line, if vertical, then on datum line, corresponding point is exactly the subpoint T (forward projection point) sought.Subpoint can be selected with comparalive ease in the picture, in other words can the scope at clear and definite forward projection point place relatively easily.Under actual coordinates, the subpoint selected is verified, thus can forward projection point be determined.Now clear and definite forward projection point position in the picture, therefore can realize making respective identification in the picture smoothly.The present invention can avoid inverse from world coordinate system to return image process, also just avoids the mark on image caused due to inverse algorithm and occurs the problem of deviation achieving object of the present invention.
Accompanying drawing explanation
Fig. 1 is the structure principle chart of scene of traffic accident image treatment system of the present invention.
Fig. 2 is the flow chart that in Fig. 1, image processing apparatus performs step.
Fig. 3 is the flow chart of the example selecting T point.
Fig. 4 is the flow chart of another example selecting T point.
Detailed description of the invention
Main word in the present invention is described.
Image (in) coordinate system, after referring to picked-up on-the-spot formation image, the coordinate system of two dimensional surface in image.
World coordinate system, referring to on-the-spot actual coordinate system, is three-dimensional system of coordinate in essence, but under this application-specific scene of traffic accident treatment, the meaningfully plane coordinate system on road surface.The relation between road surface plane coordinate system and image coordinate system can be obtained, thus realize the conversion between two coordinate systems.
Below in conjunction with accompanying drawing, technical scheme of the present invention is described in detail.
Fig. 1 display be the composition structure of scene of traffic accident image treatment system of the present invention, the image processing apparatus comprising imaging device and be connected with described imaging device.Imaging device can be the imaging device that digital camera etc. is commonly used, and image processing apparatus can be the computer-processing equipments such as PC, and device has image processing program.Imaging device carries out imaging to the scene of a traffic accident, and the image formed is transferred to image processing apparatus by data transmission path (such as data wire).
Fig. 2 shows the main flow of image processing apparatus process scene of traffic accident image, below describes in detail.
First, the relation in described image between coordinate system and world coordinate system is obtained.If desired reach the object to on-the-spot field survey by the measurement of image, then must opening relationships between image coordinate system and world coordinate system, set up conversion relation in other words, so that the measurement realized in the picture can be converted into on-the-spot real data.The relation obtaining image coordinate system and world coordinate system has multiple existing method, takes a single example be described at this.In the process of imaging at the scene, put the scale of the geometric datas such as known dimensions at the scene, scale also occurs in the picture.Because the geometric data of scale is known, scale therefore can be utilized as intermediary, set up the transformation relation of image coordinate system and world coordinate system.
Second step, obtains the coordinate of some S under world coordinate system to be measured in described image.Usually the point measured can clearly be needed in the picture, such as automotive wheel and ground contact points.And then need to mark the point to be measured vertical line to datum line (such as highway sideline).By this vertical line, traffic accident treatment professional just can carry out crucial judgement to traffic accident.The image of usual utilization routine, to the direct algorithms of reality, can obtain the coordinate of point to be measured under world coordinate system in image.
3rd step, obtains the expression formula of datum line under world coordinate system in described image.The datum line of specifying in the picture, by carrying out normal operation method process to two points on datum line, obtaining the coordinate under world coordinate system, thus can set up the expression formula of the datum lines such as linear equation.
4th step, determines forward projection point.In this step, by rough mode datum line in the picture obtains some candidate points, subpoint T is there is in these candidate points, namely the line of T and S is perpendicular to datum line, T is the orthographic projection of S on datum line in other words, symbol T refers in particular to this point exactly in the present invention, is main calculation results.Obtain the coordinate of these candidate points under world coordinate system, utilize candidate point and S to set up straight line respectively, calculate the angle of this straight line and datum line, when angle is right angle, then this candidate point is the line of T, S and T is the vertical line of datum line.
5th step, identifies the line of S and T in the picture.Determine T point in the 4th step after, corresponding candidate point can be found accordingly in the drawings, identify the line of S and T in the picture, judge intuitively from image to facilitate traffic accident treatment professional.Preferably, under world coordinate system, calculate the distance of S and T, and this distance is marked in the picture.
In the scene of a traffic accident, if the length at scene is comparatively large, and the Limited information that an image holds, the information at the larger scene of length is not comprised by an image.Multiple images can be obtained on this length direction for this reason.Scene of traffic accident image treatment system of the present invention multiple images to above-mentioned acquisition process.Coordinate system due to image all can be set up with world coordinate system and contact, and scene of traffic accident image treatment system of the present invention obtains the common interface point of adjacent two images, and namely this interface point all occurs in adjacent two images.Scene of traffic accident image treatment system so by described common interface point by the coordinate system of adjacent two images connect, the critical data of multiple Image Acquisition connects the most at last.When obtaining field sketch, the critical data in multiple images described can embody a concentrated reflection of in a sketch, is convenient to the foundation that traffic accident treatment professional has an entirety to judge.
Fig. 3 and Fig. 4 further illustrate details for the two kinds of methods how selecting T point.
First also be the point determining roughly some candidates in the method for Fig. 3 display, under world coordinate system, then calculate the length of these candidate points and S line section respectively, the candidate point of wherein the shortest line section correspondence is T.This method make use of a little the shortest to straight line vertical line section principle.Certainly, the application of this method is in order to ensure precision, and the density of candidate point should be reasonable.
Fig. 4 shows the another kind of method selecting T point.
First, be that the center of circle forms a virtual circle in the picture with S.This virtual radius of a circle can be arbitrary dimension.
Secondly, a radius of described imaginary circles is that pivot rotates once with S, and the angle of rotation sets according to the requirement of final result precision.Now this radius is formed or represents vectorial V.
3rd, under world coordinate system, try to achieve trigonometric function (such as cosine) value of angle between vectorial V and datum line in previous step.
4th, judge whether the trigonometric function value (such as cosine value) that " the 3rd " obtains meets to the corresponding trigonometric function value (such as cosine value) at right angle.This needs the absolute value of both calculating difference, and calculates relative error, sees and whether meets required precision.If do not meet required precision, just think the corresponding trigonometric function value not meeting right angle, the line of this point and S is not orthogonal to datum line in other words, then forward second step to and continue to perform (radius continues to rotate once).
5th, if meet required precision, just can think the corresponding trigonometric function value meeting right angle, the crosspoint of vectorial V place straight line and datum line is T.
Method of the present invention then follows the example shown in Fig. 2 to Fig. 4, carries out detailed description above, repeats no more.
It should be noted that and the foregoing is only preferred embodiment of the present invention, not thereby limit scope of patent protection of the present invention, the present invention can also adopt equivalent technologies to replace.Therefore the equivalence change that all utilizations description of the present invention and diagramatic content are done, or directly or indirectly apply to other correlative technology fields and be all in like manner all contained in the scope that the present invention contains.

Claims (10)

1. scene of traffic accident image treatment system, comprises imaging device and has the image processing apparatus of data transmission path with described imaging device, it is characterized in that:
Described imaging device generates scene of traffic accident image; Put the scale of known geometric data in the described scene of a traffic accident, described scale occurs in described image;
Described image processing apparatus performs following steps:
A, with described scale for intermediary, obtain the relation between coordinate system and world coordinate system in described image;
B, obtain the coordinate of some S under world coordinate system to be measured in described image;
C, obtain the expression formula of datum line under world coordinate system in described image;
D, on datum line described in described image, select some subpoints, when judging that under world coordinate system one of described subpoint and the straight line that S is formed are vertical with described datum line, then this subpoint is recorded as T;
E, in described image, identify the line of S and T.
2. scene of traffic accident image treatment system according to claim 1, it is characterized in that: judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straightway with one of described subpoint respectively, the subpoint that wherein the shortest straightway is corresponding is T.
3. scene of traffic accident image treatment system according to claim 1, it is characterized in that: judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straight line or straightway with one of described subpoint respectively, calculate the angle of described straight line or straightway and described datum line, when described angle is right angle, then this straight line or described subpoint corresponding to straightway are T.
4. scene of traffic accident image treatment system according to claim 3, it is characterized in that: on described datum line, in step D, select the step of some subpoints to comprise: be pivot through the straight line of S with S, and rotate with predetermined progress, described straight line formation crossing with described datum line crosspoint when rotating to a position; Described crosspoint is described subpoint.
5. scene of traffic accident image treatment system according to claim 3 or 4, it is characterized in that: judge that the step of described straight line or straightway and described datum line angle comprises: the trigonometric function value calculating described straight line or straightway and described datum line angle, when the difference of the described trigonometric function value the obtained trigonometric function value corresponding to right angle is in accuracy rating, then think that this angle is right angle.
6. scene of traffic accident image treatment system according to claim 5, is characterized in that: step e also comprises: the length value identifying the line of S and T in described image.
7. scene of traffic accident image processing method, is characterized in that: the scale putting known geometric data in the described scene of a traffic accident, and described scale occurs in described image; Also comprise the steps:
A, with described scale for intermediary, obtain the relation between coordinate system and world coordinate system in described image;
B, obtain the coordinate of some S under world coordinate system to be measured in described image;
C, obtain the expression formula of datum line under world coordinate system in described image;
D, on datum line described in described image, select some subpoints, when judging that under world coordinate system one of described subpoint and the straight line that S is formed are vertical with described datum line, then this subpoint is recorded as T;
E, in described image, identify the line of S and T.
8. scene of traffic accident image processing method according to claim 7, it is characterized in that: judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straightway with one of described subpoint respectively, the subpoint that wherein the shortest straightway is corresponding is T.
9. scene of traffic accident image processing method according to claim 7, it is characterized in that: judge in step D that described subpoint and the straight line that S is formed are vertical with described datum line and comprise the steps: that S forms straight line or straightway with one of described subpoint respectively, calculate the angle of described straight line or straightway and described datum line, when described angle is right angle, then this straight line or described subpoint corresponding to straightway are T.
10. scene of traffic accident image processing method according to claim 9, it is characterized in that: on described datum line, in step D, select the step of some subpoints to comprise: be pivot through the straight line of S with S, and rotate with predetermined progress, described straight line formation crossing with described datum line crosspoint when rotating to a position; Described crosspoint is described subpoint.
CN201210461033.7A 2012-11-15 2012-11-15 System and method for scene of traffic accident image processing Expired - Fee Related CN102980562B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210461033.7A CN102980562B (en) 2012-11-15 2012-11-15 System and method for scene of traffic accident image processing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210461033.7A CN102980562B (en) 2012-11-15 2012-11-15 System and method for scene of traffic accident image processing

Publications (2)

Publication Number Publication Date
CN102980562A CN102980562A (en) 2013-03-20
CN102980562B true CN102980562B (en) 2015-03-25

Family

ID=47854803

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210461033.7A Expired - Fee Related CN102980562B (en) 2012-11-15 2012-11-15 System and method for scene of traffic accident image processing

Country Status (1)

Country Link
CN (1) CN102980562B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103400498B (en) * 2013-08-14 2015-12-09 中国人民解放军第三军医大学第三附属医院 Based on scene of the accident drawing generating method and the generation system of electronic chart and data acquisition system (DAS)

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1055342C (en) * 1995-10-10 2000-08-09 王秀美 Computer-aided traffic accident site surveying system
JPH10267621A (en) * 1997-03-24 1998-10-09 Komatsu Ltd Apparatus and method for measurement of height of object
JP4102324B2 (en) * 2004-03-29 2008-06-18 株式会社トプコン Surveying data processing system, surveying data processing program, and electronic map display device
CN101033966A (en) * 2007-04-12 2007-09-12 上海交通大学 Photography measurement method of scene of a traffic accident
CN101319895B (en) * 2008-07-17 2011-08-31 上海交通大学 Hand-hold traffic accident fast on-site coordinate machine
CN202057328U (en) * 2011-04-12 2011-11-30 清华大学 Vehicle-mounted scale-free traffic accident scene quick surveying system based on binocular vision

Also Published As

Publication number Publication date
CN102980562A (en) 2013-03-20

Similar Documents

Publication Publication Date Title
US11443444B2 (en) Interior photographic documentation of architectural and industrial environments using 360 panoramic videos
Rashidi et al. Generating absolute-scale point cloud data of built infrastructure scenes using a monocular camera setting
KR20190090393A (en) Lane determining method, device and storage medium
EP1993072A1 (en) Method for comparison of 3D computer model and as-built situation of an industrial plant
CN109544623A (en) The measurement method and device in vehicle damage region
CN106767854B (en) Mobile device, garage map forming method and system
Gikas et al. A novel geodetic engineering method for accurate and automated road/railway centerline geometry extraction based on the bearing diagram and fractal behavior
Osman et al. 3D accident reconstruction using low-cost imaging technique
CN104280036A (en) Traffic information detection and positioning method, device and electronic equipment
Gerke Using horizontal and vertical building structure to constrain indirect sensor orientation
Vakili et al. Single–camera vehicle speed measurement using the geometry of the imaging system
CN103914521B (en) Street view image storage method and device based on mixed tile pyramids
Pauls et al. Can we trust our maps? An evaluation of road changes and a dataset for map validation
Dai et al. Three-dimensional modeling of site elements by analytically processing image data contained in site photos
Kang et al. An automatic mosaicking method for building facade texture mapping using a monocular close-range image sequence
CN112017262B (en) Pavement marker generation method and device, storage medium and electronic equipment
CN104166995A (en) Harris-SIFT binocular vision positioning method based on horse pace measurement
WO2016157802A1 (en) Information processing apparatus, information processing system, information processing method, and storage medium
CN102980562B (en) System and method for scene of traffic accident image processing
Terpstra et al. Using multiple photographs and USGS LiDAR to improve photogrammetric accuracy
EP3865982B1 (en) Augmented viewing of a scenery and subsurface infrastructure
Tsai et al. Vision-based roadway geometry computation
CN109345576B (en) Vehicle running speed identification method and system
CN103606321B (en) Based on video location and the road test judgment method of digital modeling combined technology
CN114743395A (en) Signal lamp detection method, device, equipment and medium

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CP01 Change in the name or title of a patent holder

Address after: 101200 Beijing Binhe Industrial Development Zone, Pinggu

Patentee after: Beijing billion profit information technology Co.,Ltd.

Address before: 101200 Beijing Binhe Industrial Development Zone, Pinggu

Patentee before: CS & S DELINEATE INFORMATION TECHNOLOGY Co.,Ltd.

CP01 Change in the name or title of a patent holder
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20150325

CF01 Termination of patent right due to non-payment of annual fee