WO2018103408A1 - 一种无人机勘查交通事故现场的航拍图像采集方法及系统 - Google Patents

一种无人机勘查交通事故现场的航拍图像采集方法及系统 Download PDF

Info

Publication number
WO2018103408A1
WO2018103408A1 PCT/CN2017/102221 CN2017102221W WO2018103408A1 WO 2018103408 A1 WO2018103408 A1 WO 2018103408A1 CN 2017102221 W CN2017102221 W CN 2017102221W WO 2018103408 A1 WO2018103408 A1 WO 2018103408A1
Authority
WO
WIPO (PCT)
Prior art keywords
shooting
drone
scene
traffic accident
scheme
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.)
Ceased
Application number
PCT/CN2017/102221
Other languages
English (en)
French (fr)
Inventor
李熙莹
陈思嘉
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.)
Sun Yat Sen University
Original Assignee
Sun Yat Sen 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 Sun Yat Sen University filed Critical Sun Yat Sen University
Priority to US15/763,435 priority Critical patent/US10783794B2/en
Publication of WO2018103408A1 publication Critical patent/WO2018103408A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G5/00—Traffic control systems for aircraft
    • G08G5/70—Arrangements for monitoring traffic-related situations or conditions
    • G08G5/72—Arrangements for monitoring traffic-related situations or conditions for monitoring traffic
    • G08G5/723—Arrangements for monitoring traffic-related situations or conditions for monitoring traffic from the aircraft
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
    • G01C11/02—Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures
    • G01C11/025—Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures by scanning the object
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/0094—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots involving pointing a payload, e.g. camera, weapon, sensor, towards a fixed or moving target
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/012—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from other sources than vehicle or roadside beacons, e.g. mobile networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/20—Image signal generators
    • H04N13/204—Image signal generators using stereoscopic image cameras
    • H04N13/207—Image signal generators using stereoscopic image cameras using a single two-dimensional [2D] image sensor
    • H04N13/221—Image signal generators using stereoscopic image cameras using a single two-dimensional [2D] image sensor using the relative movement between cameras and objects

Definitions

  • the invention relates to the field of UAV application, in particular to an aerial image acquisition method and system for a UAV to survey a traffic accident scene.
  • the traditional road traffic accident site survey method is mainly to record the geometric information of the accident site by the traffic police through the simple method of measuring distance, observation and judgment, manual drawing, etc., to achieve the purpose of survey.
  • These traditional methods have the disadvantages of low efficiency, poor precision, and inability to perform secondary verification on the accident site, which cannot meet the increasing demands of people.
  • the method of reproducing the traffic accident scene based on the real scene can provide a strong judgment basis for liability judgment, cause analysis, insurance claims, and judicial identification, and generate more social value. Therefore, the use of various auxiliary exploration techniques to form a recurring scene of the accident scene has become a development trend of road traffic accident site investigation.
  • road traffic accident site surveys can be divided into two-dimensional surveys and three-dimensional surveys.
  • the traffic elements at the scene of the accident will not be on the same plane, and often cannot meet the requirements of the two-dimensional exploration method, resulting in a large error in the two-dimensional exploration method; and the three-dimensional exploration method can effectively reduce and substantially eliminate such errors.
  • 3D exploration can be divided into traffic accident site mapping technology and traffic accident 3D reconstruction technology.
  • Traffic accident site mapping technology refers to the investigation and record of the information and status of on-site traffic elements based on the actual situation on the site, presented in the scene of the accident scene, focusing only on the actual results, without inferring the cause and process of the accident.
  • Traffic accident site mapping technology can be divided into monocular camera survey method and multi-camera survey method according to the number of on-site surveying and mapping cameras.
  • the traffic accident site mapping technology still needs further improvement in measurement accuracy.
  • the three-dimensional reconstruction technology of the traffic accident scene includes three-dimensional reconstruction of the scene and process animation simulation process.
  • the three-dimensional reconstruction methods of traditional traffic accident scenes include manual measurement methods, three-dimensional laser scanning methods, and photogrammetry. It is also possible to use aerial aerial vehicles to shoot road conditions from high altitude.
  • the traditional manual measurement method's drawing modeling takes a long time and the effect deviates greatly from the actual situation. It can only be used for scene reappearing, and many details cannot be displayed.
  • Both the 3D laser scanning method and the photogrammetry method require the equipment to be able to surround the accident scene measurement, the moving efficiency is low, and the corresponding equipment is relatively bulky, generally loaded on the measuring vehicle, and the surrounding measurement cannot be performed in most scenes.
  • the existing aerial photography is mostly taken by helicopter, and there are defects such as high shooting height, inconspicuous object characteristics, and poor modeling effect, which are not suitable for on-site reconstruction of road traffic accidents.
  • drones used to film the scene, but it is only a simple on-site scanning at a certain height.
  • the overshoot profile of the field ie, the live map
  • the overshoot profile of the field is not yet able to display the live scene in three dimensions.
  • the object of the present invention is to provide an aerial image acquisition method for a UAV surveying traffic accident scene with good maneuverability, high efficiency, comprehensiveness and high precision.
  • Another object of the present invention is to provide an aerial image acquisition system for a UAV surveying traffic accident scene with good maneuverability, high efficiency, comprehensiveness and high precision.
  • An aerial image acquisition method for a UAV to survey a traffic accident scene comprising the following steps:
  • the corresponding low-altitude shooting scheme of the traffic accident scene drone includes, but is not limited to, the global range "bow"-shaped patrol vertical overhead shooting scheme And a combined shooting scheme, wherein the combined shooting scheme is formed by superimposing a global range "bow” shaped tour vertical vertical shooting scheme and a partial stereoscopic scene multi-level surrounding oblique shooting scheme;
  • the scene of the traffic accident is taken, and the aerial image sequence of the traffic accident scene is obtained.
  • the step of selecting a corresponding low-altitude shooting plan of the traffic accident on-site drone according to whether it is necessary to perform three-dimensional on-site reconstruction or on-site animation simulation includes:
  • step S11 determining whether it is necessary to perform a three-dimensional on-site reconstruction or on-site animation simulation to select a corresponding traffic accident, and if so, proceed to step S13; otherwise, proceed to step S12;
  • the step of selecting and calculating the shooting parameters of the drone according to the selected low-altitude shooting scheme of the drone includes:
  • step S22 Determine whether the selected low-altitude low-altitude shooting scheme is a combined shooting solution, and if yes, execute step S23; otherwise, perform a traffic accident scene shooting operation;
  • step S21 includes:
  • the shooting time T of the to-be-shot area and the number of captured images S are calculated according to the shooting interval T 0 , the flying speed V 0 , and the horizontal course interval D 0 , and the shooting time T is calculated as:
  • the calculation formula of the number S of captured images is: Where a and b are the length of the route side and the length of the horizontal side of the area to be photographed, respectively. Round up the symbol.
  • step S23 includes:
  • the step of performing a traffic accident scene shooting according to the selected low-altitude shooting scheme of the drone and the shooting parameters of the drone, and obtaining the aerial image sequence of the traffic accident scene includes:
  • step S32 Determine whether the selected low-altitude low-altitude shooting scheme is a combined shooting solution, and if yes, execute step S33; otherwise, end the collecting process and take a photo group at a horizontal interval of the overhead viewing angle as a sequence of aerial image of the traffic accident scene;
  • the multi-level surround tilt shooting scheme of the partial stereo scene firstly operates the drone to fly above the target object, and then the target object is taken in the framing center as the shooting principle according to the four heights H 1 , H 2 , H 3 and H 4 and four surrounding radii R 1 , R 2 , R 3 and R 4 are tilted around the target in order from high to low and from inside to outside, resulting in 4 heights 4 4 sets of photos with a radius.
  • An aerial image acquisition system for a drone surveying a traffic accident scene comprising:
  • the shooting plan selection module is configured to select a corresponding low-altitude shooting plan of the traffic accident on-site drone according to whether it is necessary to perform three-dimensional on-site reconstruction or live animation simulation, and the low-altitude shooting plan of the traffic accident on-site drone includes but not limited to the global scope “bow” a glyph vertical vertical panning scheme and a combined shooting scheme, wherein the combined shooting scheme is formed by superimposing a global range "bow" shaped tour vertical vertical shooting plan and a partial stereoscopic scene multi-level surrounding oblique shooting plan;
  • the parameter calibration module is configured to select and calculate the shooting parameters of the drone according to the selected low-altitude shooting scheme of the drone;
  • the on-site shooting module is used for scene shooting of a traffic accident according to the selected low-altitude shooting scheme of the drone and the shooting parameters of the drone, and obtains an aerial image sequence of the scene of the traffic accident.
  • parameter calibration module includes:
  • the first calibration unit is configured to select and calculate a shooting parameter of the UAV of the global range "bow" type vertical vertical shooting scheme
  • a first determining unit configured to determine whether the selected low-altitude low-altitude shooting scheme is a combined shooting scheme, and if yes, go to the second calibration unit; otherwise, jump to the live shooting module;
  • the second calibration unit is configured to select and calculate the shooting parameters of the drone shooting scheme of the partial stereo scene multi-level surround tilt shooting scheme.
  • the second calibration unit includes:
  • a setting subunit for setting four heights H 1 , H 2 , H 3 and H 4 of the drone trajectory, corresponding four surrounding radii R 1 , R 2 , R 3 and R 4 , and corresponding Four flight speeds V 1 , V 2 , V 3 and V 4 ;
  • a first calculating subunit for calculating four shooting angles ⁇ 1 of the camera of the drone based on the heights H 1 , H 2 , H 3 and H 4 and the respective surrounding radii R 1 , R 2 , R 3 and R 4 , ⁇ 2 , ⁇ 3 and ⁇ 4 ;
  • Second calculating sub-unit according to the height H 1, H 2, H 3 and H 4, around the respective radii R 1, R 2, R 3 and R 4, and corresponding flight speed V 1, V 2, V 3 and V 4 calculate the four shooting intervals D 1 , D 2 , D 3 and D 4 of the drone.
  • the live shooting module includes:
  • a first shooting unit configured to perform a traffic accident scene shooting according to a global range "bow"-shaped patrol vertical overhead shooting scheme and a shooting parameter of the drone, and obtain a horizontal interval photographing photo group of a top view angle, the global range "bow" shape
  • the aerial trajectory of the drone is photographed at a fixed height "bow"-shaped horizontal interval, in the drone
  • the angle of the camera is vertically downward, and the speed of the drone is kept at a constant speed;
  • a second determining unit configured to determine whether the selected low-altitude low-altitude shooting plan is a combined shooting plan, and if yes, go to the second shooting unit, otherwise, end the collecting process and take the photo group as a traffic at a horizontal interval of the overhead view Aerial image sequence of the accident scene;
  • the second photographing unit is configured to perform a scene shooting of a traffic accident according to a multi-level surround tilt shooting scheme of the partial stereo scene and a shooting parameter of the drone, obtain a photo group with different radii of different heights, and take a photo group of the horizontal interval of the overhead view and
  • the photo groups of different heights and different radii are superimposed as the aerial image sequence of the traffic accident scene, and the multi-level surround tilt shooting scheme of the partial stereo scene firstly operates the drone to fly above the target object, and then the target object is taken in the framing center as the shooting principle.
  • the beneficial effects of the method of the present invention include: selecting a corresponding low-altitude shooting scheme of a traffic accident scene drone according to whether three-dimensional on-site reconstruction or on-site animation is required, and shooting parameters of the drone according to the selected low-altitude shooting scheme of the drone.
  • the selection and calculation and the steps of shooting the scene of the traffic accident according to the selected low-altitude shooting scheme of the drone and the shooting parameters of the drone, the drone is used to collect the aerial image sequence, and the convenience of on-site forensics of the traffic accident is taken into account.
  • the low-altitude shooting scheme of the drone scene on the scene of the traffic accident includes a global range "bow"-shaped patrol vertical overhead shooting scheme and a combined shooting scheme, which can not only be photographed globally through the global range "bow"-shaped patrol vertical overhead shooting scheme.
  • the scene map of the angle of view, and the multi-level surround tilt shooting scheme of the partial stereo scene in the combined shooting scheme can obtain the surrounding shooting image sequences of different angles at different angles, realizing the comprehensive survey of the multi-angle of the scene of the traffic accident, for the three-dimensional reconstruction or animation simulation Accident scene offer Sufficient data, more comprehensive, and greatly improved the efficiency and maneuverability of the evidence collection on the scene of the accident; the low-altitude shooting program of the drone was used to capture the aerial image, the shooting height was low, and the characteristics of the helicopter aerial object were not obvious and built. The defect with poor mold effect has higher precision.
  • the beneficial effects of the system of the invention are: including a shooting plan selection module, a parameter calibration module and a live shooting module, using a drone to perform aerial image sequence collection, taking into account the convenience and mobility of on-site forensics of traffic accidents; traffic accidents
  • the on-site UAV low-altitude shooting solution includes a global range "bow"-shaped touring vertical overhead shooting scheme and a combined shooting scheme, which can not only obtain a bird's-eye view of the bird's-eye view through the global range "bow” shape, but also obtain a bird's-eye view of the bird's eye view, and
  • the multi-level surround shooting images of different angles and different levels can be obtained by the multi-level surround tilt shooting scheme of the partial stereo scene in the combined shooting scheme, and the comprehensive survey of the multi-angle of the scene of the traffic accident is realized, which provides a three-dimensional reconstruction or animation simulation scene of the accident scene.
  • FIG. 1 is an overall flow chart of an aerial image acquisition method for a UAV surveying a traffic accident scene according to the present invention
  • FIG. 2 is a schematic diagram of an aerial trajectory and a camera angle of a combined shooting scheme of the unmanned aerial vehicle of the present invention
  • FIG. 3 is a top view of an aerial trajectory of the combined shooting scheme of the unmanned aerial vehicle of the present invention.
  • FIG. 4 is a schematic diagram of a trajectory of a global "bow"-shaped horizontal interval vertical overshoot of the UAV of the present invention
  • FIG. 5 is a top view of a trajectory of a UAV that performs a global "bow"-shaped horizontal interval vertical panning according to the present invention
  • FIG. 6 is a schematic diagram of a trajectory of a local multi-angle circumferential tilt shooting of the drone of the present invention
  • FIG. 7 is a top plan view of a trajectory of a local multi-angle circumferential tilting shooting of the drone of the present invention.
  • an aerial image capturing method for a UAV to survey a traffic accident scene includes the following steps:
  • the corresponding low-altitude shooting scheme of the traffic accident scene drone includes, but is not limited to, the global range "bow"-shaped patrol vertical overhead shooting scheme And a combined shooting scheme, wherein the combined shooting scheme is formed by superimposing a global range "bow” shaped tour vertical vertical shooting scheme and a partial stereoscopic scene multi-level surrounding oblique shooting scheme;
  • the scene of the traffic accident is taken, and the aerial image sequence of the traffic accident scene is obtained.
  • 3D on-site reconstruction or live animation simulation there is no need for 3D on-site reconstruction or live animation simulation, which means that only a scene view with a bird's-eye view is needed.
  • 3D on-site reconstruction or live animation simulation is required, which means that the scene map with the top view angle and the surrounding image sequence with different angles at different angles are needed.
  • a combined shooting scheme is needed: both the global range "bow" shape and the vertical vertical shooting scheme are required.
  • the captured photo group performs low-precision and rapid modeling, and requires a high-precision 3D field reconstruction or high-precision live animation simulation of the surrounding captured image sequence of the partial stereo scene multi-level surround tilt shooting scheme.
  • the drone of the present invention is preferably an existing quadrotor unmanned aerial vehicle carrying a camera.
  • the step of selecting a corresponding traffic accident on-site UAV low-altitude shooting plan according to whether three-dimensional on-site reconstruction or on-site animation is required is performed, including:
  • step S11 determining whether it is necessary to perform a three-dimensional on-site reconstruction or on-site animation simulation to select a corresponding traffic accident, and if so, proceed to step S13; otherwise, proceed to step S12;
  • the shooting parameters of the drone according to the selected low-altitude shooting scheme of the drone include:
  • step S22 Determine whether the selected low-altitude low-altitude shooting scheme is a combined shooting solution, and if yes, execute step S23; otherwise, perform a traffic accident scene shooting operation;
  • step S21 includes:
  • the shooting time T of the to-be-shot area and the number of captured images S are calculated according to the shooting interval T 0 , the flying speed V 0 , and the horizontal course interval D 0 , and the shooting time T is calculated as:
  • the calculation formula of the number S of captured images is: Where a and b are the length of the route side and the length of the horizontal side of the area to be photographed, respectively. Round up the symbol.
  • the step S23 includes:
  • the step of performing a traffic accident scene shooting according to the selected low-altitude shooting scheme of the drone and the shooting parameters of the drone, and obtaining the aerial image sequence of the traffic accident scene includes:
  • step S32 Determine whether the selected low-altitude low-altitude shooting scheme is a combined shooting solution, and if yes, execute step S33; otherwise, end the collecting process and take a photo group at a horizontal interval of the overhead viewing angle as a sequence of aerial image of the traffic accident scene;
  • the multi-level surround tilt shooting scheme of the partial stereo scene firstly operates the drone to fly above the target, and then the target is in the framing center as the shooting principle, according to the four heights H 1 , H 2 , H 3 and H 4 and four surrounding radii R 1 , R 2 , R 3 and R 4 are tilted around the target in order from high to low and from inside to outside, resulting in 4 heights 4 4 sets of photos with a radius.
  • the fixed-height "bow”-shaped horizontal interval patrol is taken, which means that the drone is fixed in height, the trajectory is “bow” shape, and the horizontal interval is fixed (to ensure the image coincidence degree requirement on the horizontally spaced route), and the tour is carried out.
  • Shooting is carried out.
  • an aerial image acquisition system for a UAV to survey a traffic accident scene includes:
  • the shooting plan selection module is configured to select a corresponding low-altitude shooting plan of the traffic accident on-site drone according to whether it is necessary to perform three-dimensional on-site reconstruction or live animation simulation, and the low-altitude shooting plan of the traffic accident on-site drone includes but not limited to the global scope “bow” a glyph vertical vertical panning scheme and a combined shooting scheme, wherein the combined shooting scheme is formed by superimposing a global range "bow" shaped tour vertical vertical shooting plan and a partial stereoscopic scene multi-level surrounding oblique shooting plan;
  • the parameter calibration module is configured to select and calculate the shooting parameters of the drone according to the selected low-altitude shooting scheme of the drone;
  • the on-site shooting module is used for scene shooting of a traffic accident according to the selected low-altitude shooting scheme of the drone and the shooting parameters of the drone, and obtains an aerial image sequence of the scene of the traffic accident.
  • the parameter calibration module includes:
  • the first calibration unit is configured to select and calculate a shooting parameter of the UAV of the global range "bow" type vertical vertical shooting scheme
  • a first determining unit configured to determine whether the selected low-altitude low-altitude shooting scheme is a combined shooting scheme, and if yes, go to the second calibration unit; otherwise, jump to the live shooting module;
  • the second calibration unit is configured to select and calculate the shooting parameters of the drone shooting scheme of the partial stereo scene multi-level surround tilt shooting scheme.
  • the second calibration unit includes:
  • a first calculating subunit for calculating four shooting angles ⁇ 1 of the camera of the drone based on the heights H 1 , H 2 , H 3 and H 4 and the respective surrounding radii R 1 , R 2 , R 3 and R 4 , ⁇ 2 , ⁇ 3 and ⁇ 4 ;
  • Second calculating sub-unit according to the height H 1, H 2, H 3 and H 4, around the respective radii R 1, R 2, R 3 and R 4, and corresponding flight speed V 1, V 2, V 3 and V 4 calculate the four shooting intervals D 1 , D 2 , D 3 and D 4 of the drone.
  • the live shooting module includes:
  • a first shooting unit configured to perform a traffic accident scene shooting according to a global range "bow"-shaped patrol vertical overhead shooting scheme and a shooting parameter of the drone, and obtain a horizontal interval photographing photo group of a top view angle, the global range "bow" shape
  • the aerial trajectory of the drone is photographed at a fixed height "bow"-shaped horizontal interval.
  • the angle of the camera in the drone is vertically downward, and the traveling speed of the drone is kept at a constant speed;
  • a second determining unit configured to determine whether the selected low-altitude low-altitude shooting plan is a combined shooting plan, and if yes, go to the second shooting unit, otherwise, end the collecting process and take the photo group as a traffic at a horizontal interval of the overhead view Aerial image sequence of the accident scene;
  • the second photographing unit is configured to perform a scene shooting of a traffic accident according to a multi-level surround tilt shooting scheme of the partial stereo scene and a shooting parameter of the drone, obtain a photo group with different radii of different heights, and take a photo group of the horizontal interval of the overhead view and
  • the photo groups of different heights and different radii are superimposed as the aerial image sequence of the traffic accident scene, and the multi-level surround tilt shooting scheme of the partial stereo scene firstly operates the drone to fly above the target object, and then the target object is taken in the framing center as the shooting principle.
  • the existing mature site survey methods mainly include a method of splicing a vertical aerial photograph and a method of modeling a large area with multi-lens tilt photography.
  • the two methods need to perform aerial photography at a higher altitude and the same plane. Poor maneuverability, low efficiency, insufficient comprehensiveness and low precision.
  • this embodiment combines the basic characteristics of a four-rotor aerial vehicle (a type of unmanned aerial vehicle) and the forensic method for the reconstruction of a three-dimensional scene on an accident site, and proposes an aerial image acquisition of a drone surveying traffic accident scene.
  • the method has developed different aerial trajectories and shooting schemes for the global and local at the ultra-low altitude, and the aerial photography exploration method is realized by the drone, which greatly improves the forensic efficiency and mobility of the accident scene, and the ground close-range photography method is adopted.
  • the advantages are extended to the air, enabling multi-view comprehensive exploration, and can be combined with computer software systems to achieve rapid application of accident scene maps, 3D field reconstruction or on-site animation simulation.
  • the method can also utilize the constraints such as stereo calibration in the scene to improve the accuracy of the three-dimensional reconstruction measurement, as follows:
  • the results of the traditional method of stereo calibration or / and on-site ground personnel are used to accurately constrain the spatial relationship in the modeling results, and after setting reasonable constraints and confirming that the error reaches an acceptable range,
  • the spatial measurement results of the modal scene plot the scene of the accident without the need to continue to block the continuous measurement at the scene.
  • the subsequent three-dimensional field modeling or scene animation simulation and other accident analysis operations may be performed according to the collected aerial image sequence.
  • the overall shooting plan design goal should be that the global top view should meet the integrity of the accident site plan.
  • Important three-dimensional targets (such as accident vehicles or major traffic elements) need to meet multi-angle observation requirements after 3D reconstruction.
  • the calibration must be modeled.
  • the scene has the actual reference meaning (that is, the measured value of the reconstructed scene after the reasonable proportion constraint on the calibration object needs to have the actual reference value compared with the ground survey result). Therefore, the shooting scheme of the embodiment is designed as two basic shooting schemes: one is a global range "bow" shaped tour vertical vertical shooting scheme, and the other is a partial stereoscopic scene multi-level surrounding oblique shooting scheme.
  • the two basic shooting schemes are superimposed to constitute the combined shooting scheme of the present invention, and the shooting mode and camera shooting angle are as shown in Figs. 2 and 3, respectively.
  • the two basic shooting schemes are described in detail below.
  • the shooting mode is shown in Figures 4 and 5.
  • the aerial trajectory of the quadrotor UAV aerial camera is shot at a fixed height "bow"-shaped horizontal interval.
  • the camera angle is vertically downward, and the quadrotor is not.
  • the traveling speed of the man-machine aerial camera is kept at a constant speed.
  • the image sequence obtained in this way is mainly related to the flying height, the horizontal course interval, the heading flight speed and the shooting interval.
  • the need for flying height H 0, flight speed V 0, and the photographing interval T 0 D 0 level of these intervals are set.
  • the flying height H 0 and the shooting interval T 0 are important setting parameters.
  • a 20mm fixed-focus wide-angle lens with a viewing angle of about 94° should be selected.
  • the angle of view can be used. Simplified to 90° for estimation, and the corresponding shooting parameter selection and calculation process specifically includes:
  • the ground projection length corresponding to the long side of the aerial photograph is 2H 0 or 20m
  • the ground projection length corresponding to the short side of the aerial photograph is 1.5H 0 or 15m.
  • the aerial trajectory is shown in Figures 6 and 7, and the shooting mode is as follows: operating the four-rotor drone aerial vehicle to fly Above the target (including the vehicle and the front checkerboard calibration, etc.), determine the center point; then use the target in the framing center as the shooting principle, according to the four heights H 1 , H 2 , H 3 and H 4 , and four Surrounding the radii R 1 , R 2 , R 3 and R 4 , the tilting is performed in order from high to low and from the inside to the outside.
  • the angle of the camera (ie, the angle between the camera's optical axis and the vertical line of the ground: ⁇ 1 , ⁇ 2 , ⁇ 3 , and ⁇ 4 ) is also different depending on the height of the aerial trajectory and the surrounding radius.
  • the shooting intervals D 1 , D 2 , D 3 and D 4 also need to be designed according to the design flying height H 1 , H 2 , H 3 and H 4 during the field operation, around the radii R 1 , R 2 , R 3 and R 4 and the design
  • the flight speeds V 1 , V 2 , V 3 and V 4 are determined.
  • the general shooting process of the multi-level surround tilt shooting scheme of the partial stereo scene is: using the quadrotor drone aerial camera to perform surround shooting around the calibration object, and four sets of photos are obtained according to four heights and four radii.
  • the present embodiment can also use the quadrotor UAV aerial camera to capture four groups of photos in the target area (ie, the vehicle and the front checkerboard area). The lowest one of the heights captures a set of photos at a larger radius, that is, the partial stereo scene multi-level surround tilt shooting scheme can obtain five sets of photos according to five heights and five radii.
  • the present invention has the following advantages:
  • UAV is used for image sequence collection, taking into account convenience and maneuverability, which can realize comprehensive survey of multiple angles of view, providing sufficient data for 3D reconstruction scene of the accident, greatly improving the efficiency and maneuvering of the forensic scene. Sex.
  • the combined shooting scheme gives the global service and the trajectory scheme during partial shooting. That is, the overhead shooting adopts the “bow”-shaped track scanning method, and the partial shooting adopts the multi-height and multi-radius circumferential surrounding track scanning scheme, which is more comprehensive.
  • the image sequence collection method proposed by the present invention can be flexibly applied: if only a scene view of a bird's-eye view is required, then only a group of photographs taken by the global range "bow"-shaped patrol vertical overhead plan is selected for low-precision rapid modeling.
  • the model thus generated can ensure the modeling effect of the bird's-eye view, and the contour and line information can basically meet the requirements of the scene drawing; if high-precision 3D field reconstruction or high-precision live animation is required, a combined shooting scheme is adopted.
  • the results can be measured in combination with the traditional method of the stereo calibration or/and the ground personnel, and the spatial relationship in the modeling result can be accurately constrained, and after the reasonable constraint is set and the error is confirmed to be within an acceptable range,
  • the accident scene map can be drawn according to the spatial measurement results of the modeling scene, without the need to continue to block the continuous measurement on the site, which is more convenient.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Multimedia (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Signal Processing (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Image Analysis (AREA)
  • Traffic Control Systems (AREA)
  • Image Processing (AREA)
  • Processing Or Creating Images (AREA)

Abstract

一种无人机勘查交通事故现场的航拍图像采集方法及系统，方法包括根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，交通事故现场无人机低空拍摄方案包括但不限于全局范围"弓"字形巡回垂直俯拍方案以及组合拍摄方案，组合拍摄方案由全局范围"弓"字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。具有机动性好、效率高、全面和精度高的优点，可广泛应用于无人机应用领域。

Description

一种无人机勘查交通事故现场的航拍图像采集方法及系统 技术领域
本发明涉及无人机应用领域，尤其是一种无人机勘查交通事故现场的航拍图像采集方法及系统。
背景技术
传统的道路交通事故现场勘查方法主要是由交警通过量具测距、观察判断、人工绘图等简易方法对事故现场几何信息进行记录，实现勘测目的。这些传统方法存在效率低、精度差、无法对事故现场进行二次验证等缺陷，满足不了人们日益增长的高要求。而基于实景对交通事故现场的重现方法可以对责任判定、原因分析、保险理赔、司法鉴定提供有力的判断依据，产生更多的社会价值。因此，采用各种辅助勘查技术以形成可以重现的事故现场景象，成为道路交通事故现场勘查的发展趋势。
按照获取信息的层次不同，可以将道路交通事故现场勘查分为二维勘查和三维勘查。实际应用中事故现场的交通要素不会处于同一平面上，往往无法满足二维勘查法的要求，导致二维勘查法产生较大误差；而三维勘查法能够有效减小并基本消除这种误差，减少不同交通要素因高度原因对测量结果的影响。三维勘查按照技术手段和呈现方式可以分为交通事故现场测绘技术和交通事故三维重建技术。
交通事故现场测绘技术是指，基于现场实际情况，对现场交通要素的信息和状况进行勘查记录，以事故现场图方式呈现，只关注现实结果，而不推断事故原因和过程。交通事故现场测绘技术根据现场测绘相机的数量，又可分为单目相机勘测法和多目相机勘测法，但由于交通事故现场的不确定性和对相机进行实时多次标定的难度较大，交通事故现场测绘技术在测量精度上仍需要进一步改进。
而交通事故现场的三维重建技术包括场景三维重构和过程动画模拟过程。传统交通事故现场的三维重建方法包括人工测量方法、三维激光扫描法和摄影测量等，还可以利用航拍飞行器从高空拍摄路面情况。传统的人工测量方法的绘图建模耗时久且效果与实际情况偏差较大，仅能用于场景的概况重现，很多细节无法展现。三维激光扫描法和摄影测量法均要求设备能够围绕事故现场环绕测量，移动效率偏低，且相应的设备比较笨重，一般要装载在测量车上，在多数现场无法实施环绕测量。现有的航空拍摄多采用直升机拍摄，存在拍摄高度高、物体特征不明显、建模效果较差等不适用于道路交通事故现场重建的缺陷。此外，目前也有采用无人机来拍摄现场的研究，但其只是进行一定高度的简单现场扫描式拍摄，只能获取现 场的俯拍概貌(即现场图)，尚无法三维显示现场实景。
发明内容
为解决上述技术问题，本发明的目的在于：提供一种机动性好、效率高、全面和精度高的无人机勘查交通事故现场的航拍图像采集方法。
本发明的另一目的在于：提供一种机动性好、效率高、全面和精度高的无人机勘查交通事故现场的航拍图像采集系统。
本发明所采取的技术方案是：
一种无人机勘查交通事故现场的航拍图像采集方法，包括以下步骤：
根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，所述交通事故现场无人机低空拍摄方案包括但不限于全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，所述组合拍摄方案由全局范围“弓”字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；
根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；
根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。
进一步，所述根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案这一步骤，其包括：
S11、判断是否需要进行三维现场重建或现场动画模拟选择相应的交通事故，若是，则执行步骤S13，反之，则执行步骤S12；
S12、选择全局范围“弓”字形巡回垂直俯拍方案作为交通事故现场无人机低空拍摄方案；
S13、选择组合拍摄方案作为交通事故现场无人机低空拍摄方案。
进一步，所述根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算这一步骤，其包括：
S21、对全局范围“弓”字形巡回垂直俯拍方案无人机的拍摄参数进行选择和计算；
S22、判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则执行步骤S23，反之，则执行交通事故现场拍摄操作；
S23、对局部立体场景多层次环绕倾斜拍摄方案无人机的拍摄参数进行选择和计算。
进一步，所述步骤S21包括：
设定无人机航拍器的高度H0和拍摄间隔T0；
根据无人机航拍器的高度H0、相机的视角以及航拍图像的长短边比例分别计算航拍图 像的长边在地面上的投影长度和航拍图像的短边在地面上的投影长度；
根据高度H0、拍摄间隔T0和航向方向上前后两幅图像重合度要求计算无人机航拍器的飞行速度V0；
根据高度H0和水平间隔航线上图像重合度要求计算水平航线间隔D0；
根据拍摄间隔T0、飞行速度V0和水平航线间隔D0计算待拍摄区域的拍摄时间T以及拍摄图像数量S，所述拍摄时间T的计算公式为：
Figure PCTCN2017102221-appb-000001
所述拍摄图像数量S的计算公式为：
Figure PCTCN2017102221-appb-000002
其中，a和b分别为待拍摄区域的航线边长度和水平边长度，
Figure PCTCN2017102221-appb-000003
为向上取整符号。
进一步，所述步骤S23包括：
设定无人机航轨迹的四个高度H1、H2、H3和H4，相应的四种环绕半径R1、R2、R3和R4，以及相应的四种飞行速度V1、V2、V3和V4；
根据高度H1、H2、H3和H4以及相应的环绕半径R1、R2、R3和R4计算无人机的相机的四种拍摄角度θ1、θ2、θ3和θ4；
根据高度H1、H2、H3和H4，相应的环绕半径R1、R2、R3和R4，以及相应的飞行速度V1、V2、V3和V4计算无人机的四种拍摄间隔D1、D2、D3和D4。
进一步，所述根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列这一步骤，其包括：
S31、根据全局范围“弓”字形巡回垂直俯拍方案和无人机的拍摄参数进行交通事故现场拍摄，得到俯视视角的水平间隔拍摄照片组，所述全局范围“弓”字形巡回垂直俯拍方案中无人机的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，无人机中相机的角度垂直向下，无人机的行进速度保持匀速；
S32、判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则执行步骤S33，反之，则结束采集流程并以俯视视角的水平间隔拍摄照片组作为交通事故现场的航拍图像序列；
S33、根据局部立体场景多层次环绕倾斜拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到不同高度不同半径的照片组，并将俯视视角的水平间隔拍摄照片组和不同高度不同半径的照片组叠加为交通事故现场的航拍图像序列，所述局部立体场景多层次环绕倾斜拍摄方案先操作无人机飞到目标物上方，然后以目标物在取景中心为拍摄原则，按照四个高度 H1、H2、H3和H4以及四种环绕半径R1、R2、R3和R4，以从高到低和从里向外的顺序环绕目标物倾斜拍摄，得到4个高度4个半径的4组照片。
本发明所采取的另一技术方案是：
一种无人机勘查交通事故现场的航拍图像采集系统，包括：
拍摄方案选择模块，用于根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，所述交通事故现场无人机低空拍摄方案包括但不限于全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，所述组合拍摄方案由全局范围“弓”字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；
参数标定模块，用于根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；
现场拍摄模块，用于根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。
进一步，所述参数标定模块包括：
第一标定单元，用于对全局范围“弓”字形巡回垂直俯拍方案无人机的拍摄参数进行选择和计算；
第一判断单元，用于判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则转至第二标定单元，反之，则跳至现场拍摄模块；
第二标定单元，用于对局部立体场景多层次环绕倾斜拍摄方案无人机的拍摄参数进行选择和计算。
进一步，所述第二标定单元包括：
设定子单元，用于设定无人机航轨迹的四个高度H1、H2、H3和H4，相应的四种环绕半径R1、R2、R3和R4，以及相应的四种飞行速度V1、V2、V3和V4；
第一计算子单元，用于根据高度H1、H2、H3和H4以及相应的环绕半径R1、R2、R3和R4计算无人机的相机的四种拍摄角度θ1、θ2、θ3和θ4；
第二计算子单元，用于根据高度H1、H2、H3和H4，相应的环绕半径R1、R2、R3和R4，以及相应的飞行速度V1、V2、V3和V4计算无人机的四种拍摄间隔D1、D2、D3和D4。
进一步，所述现场拍摄模块包括：
第一拍摄单元，用于根据全局范围“弓”字形巡回垂直俯拍方案和无人机的拍摄参数进行交通事故现场拍摄，得到俯视视角的水平间隔拍摄照片组，所述全局范围“弓”字形巡回垂直俯拍方案中无人机的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，无人机中 相机的角度垂直向下，无人机的行进速度保持匀速；
第二判断单元，用于判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则转至第二拍摄单元，反之，则结束采集流程并以俯视视角的水平间隔拍摄照片组作为交通事故现场的航拍图像序列；
第二拍摄单元，用于根据局部立体场景多层次环绕倾斜拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到不同高度不同半径的照片组，并将俯视视角的水平间隔拍摄照片组和不同高度不同半径的照片组叠加为交通事故现场的航拍图像序列，所述局部立体场景多层次环绕倾斜拍摄方案先操作无人机飞到目标物上方，然后以目标物在取景中心为拍摄原则，按照四个高度H1、H2、H3和H4以及四种环绕半径R1、R2、R3和R4，以从高到低和从里向外的顺序环绕目标物倾斜拍摄，得到4个高度4个半径的4组照片。
本发明的方法的有益效果是：包括根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算以及根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄的步骤，采用了无人机来进行航拍图像序列采集，兼顾了交通事故现场取证的便利性和机动性；交通事故现场无人机低空拍摄方案包括全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，不仅能通过全局范围“弓”字形巡回垂直俯拍方案进行全局拍摄来获得俯视视角的现场图，而且能通过组合拍摄方案中的局部立体场景多层次环绕倾斜拍摄方案获取不同角度不同层次的环绕拍摄图像序列，实现了交通事故现场多视角的综合勘查，为三维重建或动画模拟事故现场实景提供了充分的数据，更加全面，也大大提升了对事故现场的取证效率和机动性；采用了无人机低空拍摄方案来进行航拍图像采集，拍摄高度低，克服了直升机航空拍摄物体特征不明显和建模效果不好的缺陷，精度更高。
本发明的系统的有益效果是：包括拍摄方案选择模块、参数标定模块和现场拍摄模块，采用了无人机来进行航拍图像序列采集，兼顾了交通事故现场取证的便利性和机动性；交通事故现场无人机低空拍摄方案包括全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，不仅能通过全局范围“弓”字形巡回垂直俯拍方案进行全局拍摄来获得俯视视角的现场图，而且能通过组合拍摄方案中的局部立体场景多层次环绕倾斜拍摄方案获取不同角度不同层次的多组环绕拍摄图像，实现了交通事故现场多视角的综合勘查，为三维重建或动画模拟事故现场实景提供了充分的数据，更加全面，也大大提升了对事故现场的取证效率和机动性；采用了无人机低空拍摄方案来进行航拍图像采集，拍摄高度低，克服了直升机航空拍摄物体特征不明显和建模效果不好的缺陷，精度更高。
附图说明
图1为本发明一种无人机勘查交通事故现场的航拍图像采集方法的整体流程图；
图2为本发明无人机组合拍摄方案的航拍轨迹与相机角度示意图；
图3为本发明无人机组合拍摄方案的航拍轨迹俯视图；
图4为本发明无人机进行全局“弓”字形水平间隔垂直俯拍的轨迹示意图；
图5为本发明无人机进行全局“弓”字形水平间隔垂直俯拍的轨迹俯视图；
图6为本发明无人机进行局部多角度环绕倾斜拍摄的轨迹示意图；
图7为本发明无人机进行局部多角度环绕倾斜拍摄的轨迹俯视图。
具体实施方式
参照图1，一种无人机勘查交通事故现场的航拍图像采集方法，包括以下步骤：
根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，所述交通事故现场无人机低空拍摄方案包括但不限于全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，所述组合拍摄方案由全局范围“弓”字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；
根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；
根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。
其中，不需要进行三维现场重建或现场动画模拟，表示只需要俯视视角的现场图，此时只需要选择全局范围“弓”字形巡回垂直俯拍方案拍摄的照片组进行低精度快速建模，这样生成的模型可以确保鸟瞰视角的建模效果，轮廓和线条信息基本可以满足现场图绘制要求。需要进行三维现场重建或现场动画模拟，表示同时需要俯视视角的现场图和不同角度不同层次的环绕拍摄图像序列，此时需要采用组合拍摄方案：既需要全局范围“弓”字形巡回垂直俯拍方案拍摄的照片组进行低精度快速建模，又需要局部立体场景多层次环绕倾斜拍摄方案的环绕拍摄图像序列进行高精度三维现场重建或高精度现场动画模拟。
本发明的无人机优选现有的四旋翼无人航拍飞行器，该飞行器携带有相机。
进一步作为优选的实施方式，所述根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案这一步骤，其包括：
S11、判断是否需要进行三维现场重建或现场动画模拟选择相应的交通事故，若是，则执行步骤S13，反之，则执行步骤S12；
S12、选择全局范围“弓”字形巡回垂直俯拍方案作为交通事故现场无人机低空拍摄方案；
S13、选择组合拍摄方案作为交通事故现场无人机低空拍摄方案。
进一步作为优选的实施方式，所述根据选择的无人机低空拍摄方案对无人机的拍摄参数 进行选择和计算这一步骤，其包括：
S21、对全局范围“弓”字形巡回垂直俯拍方案无人机的拍摄参数进行选择和计算；
S22、判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则执行步骤S23，反之，则执行交通事故现场拍摄操作；
S23、对局部立体场景多层次环绕倾斜拍摄方案无人机的拍摄参数进行选择和计算。
进一步作为优选的实施方式，所述步骤S21包括：
设定无人机航拍器的高度H0和拍摄间隔T0；
根据无人机航拍器的高度H0、相机的视角以及航拍图像的长短边比例分别计算航拍图像的长边在地面上的投影长度和航拍图像的短边在地面上的投影长度；
根据高度H0、拍摄间隔T0和航向方向上前后两幅图像重合度要求计算无人机航拍器的飞行速度V0；
根据高度H0和水平间隔航线上图像重合度要求计算水平航线间隔D0；
根据拍摄间隔T0、飞行速度V0和水平航线间隔D0计算待拍摄区域的拍摄时间T以及拍摄图像数量S，所述拍摄时间T的计算公式为：
Figure PCTCN2017102221-appb-000004
所述拍摄图像数量S的计算公式为：
Figure PCTCN2017102221-appb-000005
其中，a和b分别为待拍摄区域的航线边长度和水平边长度，
Figure PCTCN2017102221-appb-000006
为向上取整符号。
进一步作为优选的实施方式，所述步骤S23包括：
设定无人机航轨迹的四个高度H1、H2、H3和H4，相应的四种环绕半径R1、R2、R3和R4，以及相应的四种飞行速度V1、V2、V3和V4；
根据高度H1、H2、H3和H4以及相应的环绕半径R1、R2、R3和R4计算无人机的相机的四种拍摄角度θ1、θ2、θ3和θ4；
根据高度H1、H2、H3和H4，相应的环绕半径R1、R2、R3和R4，以及相应的飞行速度V1、V2、V3和V4计算无人机的四种拍摄间隔D1、D2、D3和D4。
进一步作为优选的实施方式，所述根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列这一步骤，其包括：
S31、根据全局范围“弓”字形巡回垂直俯拍方案和无人机的拍摄参数进行交通事故现场 拍摄，得到俯视视角的水平间隔拍摄照片组，所述全局范围“弓”字形巡回垂直俯拍方案中无人机的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，无人机中相机的角度垂直向下，无人机的行进速度保持匀速；
S32、判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则执行步骤S33，反之，则结束采集流程并以俯视视角的水平间隔拍摄照片组作为交通事故现场的航拍图像序列；
S33、根据局部立体场景多层次环绕倾斜拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到不同高度不同半径的照片组，并将俯视视角的水平间隔拍摄照片组和不同高度不同半径的照片组叠加为交通事故现场的航拍图像序列，所述局部立体场景多层次环绕倾斜拍摄方案先操作无人机飞到目标物上方，然后以目标物在取景中心为拍摄原则，按照四个高度H1、H2、H3和H4以及四种环绕半径R1、R2、R3和R4，以从高到低和从里向外的顺序环绕目标物倾斜拍摄，得到4个高度4个半径的4组照片。
其中，呈固定高度“弓”字形水平间隔巡回进行拍摄，是指无人机高度固定，轨迹为“弓”字形，且水平间隔固定(保证满足水平间隔航线上的图像重合度要求)，巡回进行拍摄。
参照图1，一种无人机勘查交通事故现场的航拍图像采集系统，包括：
拍摄方案选择模块，用于根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，所述交通事故现场无人机低空拍摄方案包括但不限于全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，所述组合拍摄方案由全局范围“弓”字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；
参数标定模块，用于根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；
现场拍摄模块，用于根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。
进一步作为优选的实施方式，所述参数标定模块包括：
第一标定单元，用于对全局范围“弓”字形巡回垂直俯拍方案无人机的拍摄参数进行选择和计算；
第一判断单元，用于判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则转至第二标定单元，反之，则跳至现场拍摄模块；
第二标定单元，用于对局部立体场景多层次环绕倾斜拍摄方案无人机的拍摄参数进行选择和计算。
进一步作为优选的实施方式，所述第二标定单元包括：
设定子单元，用于设定无人机航轨迹的四个高度H1、H2、H3和H4，相应的四种环绕半 径R1、R2、R3和R4，以及相应的四种飞行速度V1、V2、V3和V4；
第一计算子单元，用于根据高度H1、H2、H3和H4以及相应的环绕半径R1、R2、R3和R4计算无人机的相机的四种拍摄角度θ1、θ2、θ3和θ4；
第二计算子单元，用于根据高度H1、H2、H3和H4，相应的环绕半径R1、R2、R3和R4，以及相应的飞行速度V1、V2、V3和V4计算无人机的四种拍摄间隔D1、D2、D3和D4。
进一步作为优选的实施方式，所述现场拍摄模块包括：
第一拍摄单元，用于根据全局范围“弓”字形巡回垂直俯拍方案和无人机的拍摄参数进行交通事故现场拍摄，得到俯视视角的水平间隔拍摄照片组，所述全局范围“弓”字形巡回垂直俯拍方案中无人机的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，无人机中相机的角度垂直向下，无人机的行进速度保持匀速；
第二判断单元，用于判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则转至第二拍摄单元，反之，则结束采集流程并以俯视视角的水平间隔拍摄照片组作为交通事故现场的航拍图像序列；
第二拍摄单元，用于根据局部立体场景多层次环绕倾斜拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到不同高度不同半径的照片组，并将俯视视角的水平间隔拍摄照片组和不同高度不同半径的照片组叠加为交通事故现场的航拍图像序列，所述局部立体场景多层次环绕倾斜拍摄方案先操作无人机飞到目标物上方，然后以目标物在取景中心为拍摄原则，按照四个高度H1、H2、H3和H4以及四种环绕半径R1、R2、R3和R4，以从高到低和从里向外的顺序环绕目标物倾斜拍摄，得到4个高度4个半径的4组照片。
下面结合说明书附图和具体实施例对本发明作进一步解释和说明。
实施例一
现有成熟的现场勘查方法主要包括对俯视垂直航拍照片进行拼接的方法以及搭载多镜头倾斜摄影对大区域进行航拍建模的方法，这两种方法需要在较高的高度和同一平面进行航拍，机动性差、效率低、不够全面和精度低。为此，本实施例结合四旋翼航拍飞行器(无人机的一种)的基本特点和以事故现场三维场景重建为目的的取证方式，提出了一种无人机勘查交通事故现场的航拍图像采集方法，该方法在超低空针对全局和局部制定了不同的航拍轨迹和拍摄方案，利用无人机实现了空中摄影勘查方法，大大提升了对事故现场的取证效率和机动性，将地面近景摄影方法的优点扩展到空中，实现了多视角的综合勘查，并能结合计算机软件系统实现快速生成事故现场图、进行三维现场重建或现场动画模拟的应用目标。此外，该方法还可以利用场景中的立体标定物等约束来提高三维重建测量值的精度，具体做法为：结 合立体标定物或/和现场地面人员的传统方法测量结果，对建模结果中的空间关系进行准确的约束，而在设定合理约束并确认误差达到可接受的范围内后，即可根据建模场景的空间测量结果绘制事故现场图，而不需要继续封锁现场持续测量。本实施例交通事故现场的航拍图像序列采集结束后，可根据采集的航拍图像序列进行后续的三维现场建模或现场动画模拟等事故分析操作。
整体的拍摄方案设计目标应当是全局俯视图满足事故现场平面勘查上的完整性，重要的立体目标(如事故车辆或主要交通元素)需要在三维重建之后满足多角度观察需要，标定物必须在建模场景中具有实际的参照意义(即对标定物进行合理比例约束之后对重建场景的测量值与地面勘查结果相比，需具有实际参考价值)。因此，本实施例的拍摄方案设计为两种基本拍摄方案：一是全局范围“弓”字形巡回垂直俯拍方案，二是局部立体场景多层次环绕倾斜拍摄方案。这两种基本拍摄方案叠加在一起就构成了本发明的组合拍摄方案，其拍摄方式和相机拍摄角度分别如图2和3所示。下面对这两种基本拍摄方案进行详细说明。
(一)全局范围“弓”字形巡回垂直俯拍方案。
对于全局范围拍摄来说，其拍摄方式如图4和5所示，四旋翼无人机航拍器的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，相机角度垂直向下，四旋翼无人机航拍器的行进速度保持匀速。直线前进时，航向上前后两张照片的重叠度(即overlap ratio)Ofa应当保持50％以上(即Ofa>＝50％)，相邻两条航带即旁向照片重叠度Olr应当保持60％以上(即Olr>＝60％)。由于飞行高度较低，这种方式获得的图像序列主要与飞行高度、水平航线间隔、航向飞行速度和拍摄间隔相关。为了满足航向上前后图像50％以上重合和水平方向60％以上重合，需要对飞行高度H0、飞行速度V0、拍摄间隔T0和水平航线间隔D0进行设置。从四旋翼无人机航拍器的实际操作可行性来说，飞行高度H0和拍摄间隔T0是重要的设置参数。而对于四旋翼无人机航拍器的相机来说，宜选用视角约为94°的20mm定焦广角镜头，考虑到GPS误差、四旋翼无人机航拍器抖动和航线飘移等实际情况，可将视角简化为90°来估算，而相应的拍摄参数选择和计算过程具体包括：
(1)设定飞行高度H0＝10m，拍摄间隔T0＝2s。
(2)若航拍相片长短边比例4:3，视角90°按照计算，则航拍相片长边对应的地面投影长度为2H0即20m，航拍相片短边对应的地面投影长度为1.5H0即15m。
(3)为了保证航向方向上前后图像重合度＞50％，假设航向未有明显飘移，四旋翼无人机航拍器沿直线向前飞行，则前后两幅图像在航向上需要有0.75H0以上的重合，由于拍摄间 隔T0＝2s，则可限制四旋翼无人机航拍器的飞行速度V0为：
Figure PCTCN2017102221-appb-000007
(4)为了保证水平间隔航线上图像重合度＞60％，假设四旋翼无人机航拍器都在同一水平线上拍摄，则其水平航线间隔D0应满足：0<D0≤1.2H0＝12m。
(5)对于一块航线边长度为a，水平边长度为b，范围为a*b的待拍摄区域来说，其拍摄时间
Figure PCTCN2017102221-appb-000008
而拍摄间隔T0＝2s，故其拍摄图像数量
Figure PCTCN2017102221-appb-000009
假设a＝96m，b＝48m，则有:
Figure PCTCN2017102221-appb-000010
(二)局部立体场景多层次环绕倾斜拍摄方案
对于以三维场景重现(即三维建模或动画模拟)为目标的局部立体场景拍摄方案，其航拍轨迹如图6和7所示，其拍摄方式为：操作四旋翼无人机航拍器飞到目标物(包括车辆和车前棋盘格标定物等)上方，确定中心点；然后以目标物在取景中心为拍摄原则，按照四个高度H1、H2、H3和H4，以及四种环绕半径R1、R2、R3和R4，以从高到低、从里向外的顺序环绕倾斜拍摄。根据航拍轨迹高度和环绕半径的不同，相机的角度(即相机光轴与地面垂直线之间所夹的角度：θ1、θ2、θ3和θ4)也有区别。拍摄间隔D1、D2、D3和D4也需要根据现场作业时的设计飞行高度H1、H2、H3和H4，环绕半径R1、R2、R3和R4和设计飞行速度V1、V2、V3和V4而定。为了达到现场重建的视觉要求，可设定每个高度至少需要采集32张图像，以满足三维重建的要求。
局部立体场景多层次环绕倾斜拍摄方案的一般拍摄流程为：使用四旋翼无人机航拍器围绕标定物进行环绕拍摄，按照四个高度四种半径，得到四组照片。而为了提升目标区域的建模效果，本实施例在使用四旋翼无人机航拍器对目标区域(即车辆和车前棋盘格标定物区域)环绕拍摄到4组照片后，还可以对这四个高度中最低的一个高度按照一种更大的半径多拍摄一组照片，也就是说，局部立体场景多层次环绕倾斜拍摄方案可以按照四个高度五种半径，得到五组照片。
设定H1＝11m，R1＝5m；H2＝9m，R2＝7m；H3＝7m，R3＝9m；H4＝5m，R4＝5m；H4＝5m，R5＝11m；则经过筛选的图片数量如表1所示：
表1 目标物体与目标区域的拍摄照片数量
Figure PCTCN2017102221-appb-000011
与现有技术相比，本发明具有以下优点：
1)采用无人机进行图像序列采集，兼顾了便利性和机动性，可以实现多视角的综合勘查，为三维重建事故现场景象提供了充分的数据，大大提升了对事故现场的取证效率和机动性。
2)组合拍摄方案给出了全局服务和局部拍摄时的航迹方案，即俯拍采用“弓”字形航迹扫描方式，局部拍摄采用多高度多半径的圆周环绕航迹扫描方案，更加全面。
3)本发明提出的图像序列采集方法可以进行灵活应用：如果只需要俯视视角的现场图，那么只需要选择全局范围“弓”字形巡回垂直俯拍方案拍摄的照片组进行低精度快速建模，这样生成的模型可以确保鸟瞰视角的建模效果，轮廓和线条信息基本可以满足现场图绘制要求；如果需要进行高精度三维现场重建或高精度现场动画模拟，则采用组合拍摄方案。
4)可结合立体标定物或/和现场地面人员的传统方法测量结果，对建模结果中的空间关系进行准确的约束，而在设定合理约束并确认误差达到可接受的范围内后，即可根据建模场景的空间测量结果绘制事故现场图，而不需要继续封锁现场持续测量，更加便捷。
以上是对本发明的较佳实施进行了具体说明，但本发明并不限于所述实施例，熟悉本领域的技术人员在不违背本发明精神的前提下还可做作出种种的等同变形或替换，这些等同的变形或替换均包含在本申请权利要求所限定的范围内。

Claims (10)

  1. 一种无人机勘查交通事故现场的航拍图像采集方法，其特征在于：包括以下步骤：
    根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，所述交通事故现场无人机低空拍摄方案包括但不限于全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，所述组合拍摄方案由全局范围“弓”字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；
    根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；
    根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。
  2. 根据权利要求1所述的一种无人机勘查交通事故现场的航拍图像采集方法，其特征在于：所述根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案这一步骤，其包括：
    S11、判断是否需要进行三维现场重建或现场动画模拟选择相应的交通事故，若是，则执行步骤S13，反之，则执行步骤S12；
    S12、选择全局范围“弓”字形巡回垂直俯拍方案作为交通事故现场无人机低空拍摄方案；
    S13、选择组合拍摄方案作为交通事故现场无人机低空拍摄方案。
  3. 根据权利要求1所述的一种无人机勘查交通事故现场的航拍图像采集方法，其特征在于：所述根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算这一步骤，其包括：
    S21、对全局范围“弓”字形巡回垂直俯拍方案无人机的拍摄参数进行选择和计算；
    S22、判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则执行步骤S23，反之，则执行交通事故现场拍摄操作；
    S23、对局部立体场景多层次环绕倾斜拍摄方案无人机的拍摄参数进行选择和计算。
  4. 根据权利要求3所述的一种无人机勘查交通事故现场的航拍图像采集方法，其特征在于：所述步骤S21包括：
    设定无人机航拍器的高度H0和拍摄间隔T0；
    根据无人机航拍器的高度H0、相机的视角以及航拍图像的长短边比例分别计算航拍图像的长边在地面上的投影长度和航拍图像的短边在地面上的投影长度；
    根据高度H0、拍摄间隔T0和航向方向上前后两幅图像重合度要求计算无人机航拍器的飞行速度V0；
    根据高度H0和水平间隔航线上图像重合度要求计算水平航线间隔D0；
    根据拍摄间隔T0、飞行速度V0和水平航线间隔D0计算待拍摄区域的拍摄时间T以及拍摄图像数量S，所述拍摄时间T的计算公式为：
    Figure PCTCN2017102221-appb-100001
    所述拍摄图像数量S的计算公式为：
    Figure PCTCN2017102221-appb-100002
    其中，a和b分别为待拍摄区域的航线边长度和水平边长度，
    Figure PCTCN2017102221-appb-100003
    为向上取整符号。
  5. 根据权利要求3或4所述的一种无人机勘查交通事故现场的航拍图像采集方法，其特征在于：所述步骤S23包括：
    设定无人机航轨迹的四个高度H1、H2、H3和H4，相应的四种环绕半径R1、R2、R3和R4，以及相应的四种飞行速度V1、V2、V3和V4；
    根据高度H1、H2、H3和H4以及相应的环绕半径R1、R2、R3和R4计算无人机的相机的四种拍摄角度θ1、θ2、θ3和θ4；
    根据高度H1、H2、H3和H4，相应的环绕半径R1、R2、R3和R4，以及相应的飞行速度V1、V2、V3和V4计算无人机的四种拍摄间隔D1、D2、D3和D4。
  6. 根据权利要求5所述的一种无人机勘查交通事故现场的航拍图像采集方法，其特征在于：所述根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列这一步骤，其包括：
    S31、根据全局范围“弓”字形巡回垂直俯拍方案和无人机的拍摄参数进行交通事故现场拍摄，得到俯视视角的水平间隔拍摄照片组，所述全局范围“弓”字形巡回垂直俯拍方案中无人机的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，无人机中相机的角度垂直向下，无人机的行进速度保持匀速；
    S32、判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则执行步骤S33，反之，则结束采集流程并以俯视视角的水平间隔拍摄照片组作为交通事故现场的航拍图像序列；
    S33、根据局部立体场景多层次环绕倾斜拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到不同高度不同半径的照片组，并将俯视视角的水平间隔拍摄照片组和不同高度不同半径的照片组叠加为交通事故现场的航拍图像序列，所述局部立体场景多层次环绕倾斜拍摄方案先操作无人机飞到目标物上方，然后以目标物在取景中心为拍摄原则，按照四个高度H1、H2、H3和H4以及四种环绕半径R1、R2、R3和R4，以从高到低和从里向外的顺序环绕目标物倾斜拍摄，得到4个高度4个半径的4组照片。
  7. 一种无人机勘查交通事故现场的航拍图像采集系统，其特征在于：包括：
    拍摄方案选择模块，用于根据是否需要进行三维现场重建或现场动画模拟选择相应的交通事故现场无人机低空拍摄方案，所述交通事故现场无人机低空拍摄方案包括但不限于全局范围“弓”字形巡回垂直俯拍方案以及组合拍摄方案，所述组合拍摄方案由全局范围“弓”字形巡回垂直俯拍方案和局部立体场景多层次环绕倾斜拍摄方案叠加而成；
    参数标定模块，用于根据选择的无人机低空拍摄方案对无人机的拍摄参数进行选择和计算；
    现场拍摄模块，用于根据选择的无人机低空拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到交通事故现场的航拍图像序列。
  8. 根据权利要求7所述的一种无人机勘查交通事故现场的航拍图像采集系统，其特征在于：所述参数标定模块包括：
    第一标定单元，用于对全局范围“弓”字形巡回垂直俯拍方案无人机的拍摄参数进行选择和计算；
    第一判断单元，用于判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则转至第二标定单元，反之，则跳至现场拍摄模块；
    第二标定单元，用于对局部立体场景多层次环绕倾斜拍摄方案无人机的拍摄参数进行选择和计算。
  9. 根据权利要求8所述的一种无人机勘查交通事故现场的航拍图像采集系统，其特征在于：所述第二标定单元包括：
    设定子单元，用于设定无人机航轨迹的四个高度H1、H2、H3和H4，相应的四种环绕半径R1、R2、R3和R4，以及相应的四种飞行速度V1、V2、V3和V4；
    第一计算子单元，用于根据高度H1、H2、H3和H4以及相应的环绕半径R1、R2、R3和R4计算无人机的相机的四种拍摄角度θ1、θ2、θ3和θ4；
    第二计算子单元，用于根据高度H1、H2、H3和H4，相应的环绕半径R1、R2、R3和R4，以及相应的飞行速度V1、V2、V3和V4计算无人机的四种拍摄间隔D1、D2、D3和D4。
  10. 根据权利要求9所述的一种无人机勘查交通事故现场的航拍图像采集系统，其特征在于：所述现场拍摄模块包括：
    第一拍摄单元，用于根据全局范围“弓”字形巡回垂直俯拍方案和无人机的拍摄参数进行交通事故现场拍摄，得到俯视视角的水平间隔拍摄照片组，所述全局范围“弓”字形巡回垂直俯拍方案中无人机的航拍轨迹呈固定高度“弓”字形水平间隔巡回进行拍摄，无人机中 相机的角度垂直向下，无人机的行进速度保持匀速；
    第二判断单元，用于判断选择的无人机低空拍摄方案是否为组合拍摄方案，若是，则转至第二拍摄单元，反之，则结束采集流程并以俯视视角的水平间隔拍摄照片组作为交通事故现场的航拍图像序列；
    第二拍摄单元，用于根据局部立体场景多层次环绕倾斜拍摄方案和无人机的拍摄参数进行交通事故现场拍摄，得到不同高度不同半径的照片组，并将俯视视角的水平间隔拍摄照片组和不同高度不同半径的照片组叠加为交通事故现场的航拍图像序列，所述局部立体场景多层次环绕倾斜拍摄方案先操作无人机飞到目标物上方，然后以目标物在取景中心为拍摄原则，按照四个高度H1、H2、H3和H4以及四种环绕半径R1、R2、R3和R4，以从高到低和从里向外的顺序环绕目标物倾斜拍摄，得到4个高度4个半径的4组照片。
PCT/CN2017/102221 2016-12-09 2017-09-19 一种无人机勘查交通事故现场的航拍图像采集方法及系统 Ceased WO2018103408A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/763,435 US10783794B2 (en) 2016-12-09 2017-09-19 Aerial image acquisition method and system for investigating traffic accident site by unmanned aerial vehicle

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201611129001.1 2016-12-09
CN201611129001.1A CN106767706B (zh) 2016-12-09 2016-12-09 一种无人机勘查交通事故现场的航拍图像采集方法及系统

Publications (1)

Publication Number Publication Date
WO2018103408A1 true WO2018103408A1 (zh) 2018-06-14

Family

ID=58874907

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/102221 Ceased WO2018103408A1 (zh) 2016-12-09 2017-09-19 一种无人机勘查交通事故现场的航拍图像采集方法及系统

Country Status (3)

Country Link
US (1) US10783794B2 (zh)
CN (1) CN106767706B (zh)
WO (1) WO2018103408A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200265725A1 (en) * 2017-09-07 2020-08-20 Guangzhou Xaircraft Technology Co., Ltd Method and Apparatus for Planning Navigation Region of Unmanned Aerial Vehicle, and Remote Control
CN113206958A (zh) * 2021-04-30 2021-08-03 成都睿铂科技有限责任公司 一种航线拍摄方法

Families Citing this family (40)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106767706B (zh) 2016-12-09 2019-05-14 中山大学 一种无人机勘查交通事故现场的航拍图像采集方法及系统
DE112017006840B4 (de) * 2017-01-16 2023-11-02 Fujitsu Limited Informationsverarbeitungsprogramm, Informationsverarbeitungsverfahren und Informationsverarbeitungsvorrichtung
CN107144264A (zh) * 2017-06-19 2017-09-08 北京理工大学 一种用于固定翼无人机采集高清路面图像的航空摄影方法
WO2019000404A1 (zh) * 2017-06-30 2019-01-03 深圳市大疆创新科技有限公司 控制终端和无人机及其控制方法
CN109389642A (zh) * 2017-08-04 2019-02-26 惠州市阿图达机电有限公司 视觉系统对机器人的标定方法、系统和具有存储功能装置
JP7022559B2 (ja) * 2017-10-17 2022-02-18 株式会社トプコン 無人航空機の制御方法および無人航空機の制御用プログラム
US10689110B2 (en) * 2018-02-12 2020-06-23 Wipro Limited Method and system for performing inspection and maintenance tasks of three-dimensional structures using drones
TWI688502B (zh) * 2018-02-14 2020-03-21 先進光電科技股份有限公司 用於警告車輛障礙物的設備
CN108776953A (zh) * 2018-06-22 2018-11-09 理光软件研究所（北京）有限公司 提高航空摄影具体位置分辨率的方法以及系统
CN109064555B (zh) * 2018-08-06 2023-06-06 百度在线网络技术（北京）有限公司 3d建模的方法、装置和存储介质
CN109211199B (zh) * 2018-08-21 2021-09-03 浙江量度智能科技有限公司 无人机套耕航线方法和系统
CN110874699B (zh) * 2018-08-31 2024-02-09 杭州海康机器人股份有限公司 记录物品的物流信息方法、装置及系统
EP3776485B1 (en) * 2018-09-26 2022-01-26 Coherent Logix, Inc. Any world view generation
CN109238240B (zh) * 2018-10-22 2021-01-08 武汉大势智慧科技有限公司 一种顾及地形的无人机倾斜摄影方法及其摄影系统
KR20210102889A (ko) * 2018-11-21 2021-08-20 광저우 엑스에어크래프트 테크놀로지 씨오 엘티디 측량 및 매핑 표본점의 계획 방법, 장치, 제어 단말기 및 저장 매체
CN109960281A (zh) * 2019-04-17 2019-07-02 深圳市道通智能航空技术有限公司 环绕飞行的控制方法、装置、终端及存储介质
CN109976370B (zh) * 2019-04-19 2022-09-30 深圳市道通智能航空技术股份有限公司 立面环绕飞行的控制方法、装置、终端及存储介质
WO2021035731A1 (zh) * 2019-08-30 2021-03-04 深圳市大疆创新科技有限公司 无人飞行器的控制方法、装置及计算机可读存储介质
WO2021056411A1 (zh) * 2019-09-27 2021-04-01 深圳市大疆创新科技有限公司 航线调整方法、地面端设备、无人机、系统和存储介质
GB2599840B (en) * 2020-01-03 2023-02-08 Mobileye Vision Technologies Ltd Vehicle navigation with occluded pedestrians
CN111553904B (zh) * 2020-04-29 2022-11-22 厦门大学 基于无人机的区域人数统计方法及系统
CN113875222B (zh) * 2020-07-16 2023-11-24 深圳市大疆创新科技有限公司 拍摄控制方法和装置、无人机及计算机可读存储介质
CN111913005B (zh) * 2020-08-28 2021-02-23 西华大学 车辆坠崖车速快速计算系统及计算方法
CN112146629A (zh) * 2020-09-24 2020-12-29 武汉大学 一种多角度贴近摄影航迹与姿态规划方法
CN112669465B (zh) * 2020-12-02 2024-03-15 海能达通信股份有限公司 场景转换方法、电子设备及存储介质
CN112698661B (zh) * 2021-03-22 2021-08-24 成都睿铂科技有限责任公司 一种飞行器的航测数据采集方法、装置、系统及存储介质
WO2022205210A1 (zh) * 2021-03-31 2022-10-06 深圳市大疆创新科技有限公司 拍摄方法、装置及计算机可读存储介质，终端设备
WO2022205208A1 (zh) * 2021-03-31 2022-10-06 深圳市大疆创新科技有限公司 拍摄方法、装置、计算机可读存储介质和终端设备
CN113867410B (zh) * 2021-11-17 2023-11-03 武汉大势智慧科技有限公司 一种无人机航拍数据的采集模式识别方法和系统
CN114428510B (zh) * 2022-01-27 2024-04-19 成都睿铂科技有限责任公司 环绕航线修正方法及系统
CN114485572B (zh) * 2022-02-25 2024-08-02 韶关市测绘研究院有限公司 一种降误差的无人机测绘系统
CN114777744B (zh) * 2022-04-25 2024-03-08 中国科学院古脊椎动物与古人类研究所 一种古生物领域的地质测量方法、装置及电子设备
CN114926983A (zh) * 2022-05-11 2022-08-19 中国地质大学（武汉） 一种面向交通事故应急的多尺度综合感知方法
CN115512056B (zh) * 2022-09-13 2026-03-10 浙江虎穴科技有限公司 基于无人机的场景三维重建方法、装置及设备
CN116045917B (zh) * 2022-11-17 2024-04-26 西北工业大学 一种航空图像获取精度测试系统及测试方法
CN118484024B (zh) * 2024-04-30 2024-11-05 广东警官学院(广东省公安司法管理干部学院) 一种用于交通事故现场的无人机航拍参数计算方法
CN119359956A (zh) * 2024-09-30 2025-01-24 中铁建工集团有限公司 基于消费级无人机倾斜摄影建模方法
CN119319944B (zh) * 2024-12-16 2025-03-11 深圳市博坦智能有限公司 飞行器的拍摄重现方法及飞行器
CN119990320B (zh) * 2025-01-24 2026-01-30 国家市场监督管理总局缺陷产品召回技术中心 一种事故还原的方法、装置、电子设备及存储介质
CN120563665B (zh) * 2025-05-28 2026-03-17 广东警官学院(广东省公安司法管理干部学院) 基于无人机的交通事故现场快速制图误差修正方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120075461A1 (en) * 2009-03-27 2012-03-29 Qifeng Yu Ground-based videometrics guiding method for aircraft landing or unmanned aerial vehicles recovery
CN104714557A (zh) * 2015-03-26 2015-06-17 清华大学 无人机定点环绕飞行的控制方法
CN105045279A (zh) * 2015-08-03 2015-11-11 余江 一种利用无人飞行器航拍自动生成全景照片的系统及方法
CN105629980A (zh) * 2015-12-23 2016-06-01 深圳速鸟创新科技有限公司 一种单相机倾斜摄影三维建模系统
CN105721751A (zh) * 2016-03-28 2016-06-29 中国人民解放军第三军医大学第三附属医院 一种夜间交通事故现场信息采集方法
CN106767706A (zh) * 2016-12-09 2017-05-31 中山大学 一种无人机勘查交通事故现场的航拍图像采集方法及系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7642953B2 (en) * 2007-07-19 2010-01-05 The Boeing Company Method and apparatus for three dimensional tomographic image reconstruction of objects
US9075415B2 (en) * 2013-03-11 2015-07-07 Airphrame, Inc. Unmanned aerial vehicle and methods for controlling same
US10250821B2 (en) * 2013-11-27 2019-04-02 Honeywell International Inc. Generating a three-dimensional model of an industrial plant using an unmanned aerial vehicle
EP3065042B1 (en) * 2015-02-13 2018-11-07 LG Electronics Inc. Mobile terminal and method for controlling the same
CN104881037A (zh) * 2015-04-01 2015-09-02 广州天翔航空科技有限公司 植保无人机的喷药方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120075461A1 (en) * 2009-03-27 2012-03-29 Qifeng Yu Ground-based videometrics guiding method for aircraft landing or unmanned aerial vehicles recovery
CN104714557A (zh) * 2015-03-26 2015-06-17 清华大学 无人机定点环绕飞行的控制方法
CN105045279A (zh) * 2015-08-03 2015-11-11 余江 一种利用无人飞行器航拍自动生成全景照片的系统及方法
CN105629980A (zh) * 2015-12-23 2016-06-01 深圳速鸟创新科技有限公司 一种单相机倾斜摄影三维建模系统
CN105721751A (zh) * 2016-03-28 2016-06-29 中国人民解放军第三军医大学第三附属医院 一种夜间交通事故现场信息采集方法
CN106767706A (zh) * 2016-12-09 2017-05-31 中山大学 一种无人机勘查交通事故现场的航拍图像采集方法及系统

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200265725A1 (en) * 2017-09-07 2020-08-20 Guangzhou Xaircraft Technology Co., Ltd Method and Apparatus for Planning Navigation Region of Unmanned Aerial Vehicle, and Remote Control
CN113206958A (zh) * 2021-04-30 2021-08-03 成都睿铂科技有限责任公司 一种航线拍摄方法
CN113206958B (zh) * 2021-04-30 2023-06-09 成都睿铂科技有限责任公司 一种航线拍摄方法

Also Published As

Publication number Publication date
CN106767706A (zh) 2017-05-31
US10783794B2 (en) 2020-09-22
US20200066169A1 (en) 2020-02-27
CN106767706B (zh) 2019-05-14

Similar Documents

Publication Publication Date Title
WO2018103408A1 (zh) 一种无人机勘查交通事故现场的航拍图像采集方法及系统
CN107514993B (zh) 基于无人机的面向单体建筑建模的数据采集方法及系统
WO2021004312A1 (zh) 一种基于双目立体视觉系统的车辆智能测轨迹方法
CN108335337B (zh) 一种正射影像图的生成方法及装置
CN110319772B (zh) 基于无人机的视觉大跨度测距方法
CN113850126A (zh) 一种基于无人机的目标检测和三维定位方法和系统
CN107492069B (zh) 基于多镜头传感器的图像融合方法
CN110799921A (zh) 拍摄方法、装置和无人机
CN112949478A (zh) 基于云台相机的目标检测方法
CN114359406B (zh) 自动对焦双目摄像头的标定、3d视觉及深度点云计算方法
CN104363438B (zh) 全景立体影像制作方法
JPH0554128A (ja) 写真測定を使用する自動ビデオ映像データベースの生成
CN109655065A (zh) 一种无人机五航线规划方法及装置
JP2009217524A (ja) 都市景観の３次元動画生成および閲覧システム
CN109859269B (zh) 岸基视频辅助定位无人机大范围流场测量方法及装置
CN104079916A (zh) 一种全景三维视觉传感器及使用方法
JP4418857B1 (ja) 路線の３次元動画生成用画像取得システム
CN111243021A (zh) 基于多组合相机的车载视觉定位方法、系统及存储介质
CN117710810A (zh) 一种基于无人机和神经网络的桥梁病害区域的三维定位方法及系统
CN112334853A (zh) 航线调整方法、地面端设备、无人机、系统和存储介质
CN106251357A (zh) 基于虚拟现实与视觉定位系统
CN113362265B (zh) 一种低成本的无人机正射影像快速地理拼接方法
CN108195359B (zh) 空间数据的采集方法及系统
CN106846385A (zh) 基于无人机的多传感遥感影像匹配方法、装置和系统
CN116222592B (zh) 一种基于多源数据的高精地图生成方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17879238

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17879238

Country of ref document: EP

Kind code of ref document: A1