CN108875640A - A kind of end-to-end unsupervised scene can traffic areas cognitive ability test method - Google Patents

A kind of end-to-end unsupervised scene can traffic areas cognitive ability test method Download PDF

Info

Publication number
CN108875640A
CN108875640A CN201810638850.2A CN201810638850A CN108875640A CN 108875640 A CN108875640 A CN 108875640A CN 201810638850 A CN201810638850 A CN 201810638850A CN 108875640 A CN108875640 A CN 108875640A
Authority
CN
China
Prior art keywords
scene
traffic areas
road surface
cognitive ability
test
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810638850.2A
Other languages
Chinese (zh)
Other versions
CN108875640B (en
Inventor
赵祥模
刘占文
沈超
高涛
樊星
董鸣
林杉
徐江
连心雨
陈婷
王润民
张凡
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.)
Changan University
Original Assignee
Changan 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 Changan University filed Critical Changan University
Priority to CN201810638850.2A priority Critical patent/CN108875640B/en
Publication of CN108875640A publication Critical patent/CN108875640A/en
Application granted granted Critical
Publication of CN108875640B publication Critical patent/CN108875640B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0463—Neocognitrons
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Traffic Control Systems (AREA)
  • Holo Graphy (AREA)

Abstract

The invention discloses a kind of end-to-end unsupervised scene can traffic areas cognitive ability test method, by the way that information of road surface is shown in test zone ground, true information of road surface can not only be reappeared, additionally it is possible to which road environment abundant is shown using this kind of mode；It is shown in test zone ground using information of road surface, tested test system can be made in test, obtains the identical detection environment of practical drive test；Using the method for holographic technique, object on reproduction road surface that can be true, three-dimensional makes to test environment closer to true environment；Obtained in such a way that varying environment is tested under different roadway scene infomation detection different scenes can traffic areas, can in all directions assessment can traffic areas detection system cognitive ability；So as to it is comprehensive embody can traffic areas detection system cognitive ability；Test method of the invention can provide a kind of effective, low-risk test and evaluation means for unmanned intelligent vehicle before practical drive test.

Description

A kind of end-to-end unsupervised scene can traffic areas cognitive ability test method
Technical field
The invention belongs to visual scene processing technology fields, and in particular to the end-to-end unsupervised scene of one kind can traffic areas The test method of cognitive ability.
Background technique
In order to promote the development of unmanned intelligent vehicle technology, need that unmanned intelligent vehicle is allowed to carry out under true road conditions Test, the problems of detection the relevant technologies.However, even if unmanned intelligent vehicle is allowed to be surveyed under true road conditions Examination, bring security risk and testing cost are also huge.Therefore, many methods are by the way of off-line test to can Traffic areas cognitive ability is tested and is assessed, although testing cost can be saved, these methods all use single width figure Picture or short-sighted frequency directly test cognitive algorithm, and test flexibility is limited by data set, and is not comprehensively considered Unmanned intelligent vehicle itself visual perception system configuration.It is thus impossible to comprehensively, really react unmanned intelligent vehicle to field Scape can traffic areas cognition ability.
As deep learning is theoretical and the development of technology, the road surface drivable region detection method based on deep learning are more next It is more perfect, but the surface conditions under real scene are extremely complex, then perfect detection method, always have consideration less than the case where, Therefore, it is necessary to a kind of close to real scene and can be realized the test mode of extreme environment scene, for testing cognitive algorithm Ability.
Summary of the invention
The purpose of the present invention is to provide a kind of end-to-end unsupervised scene can traffic areas cognitive ability test method, With overcome the deficiencies in the prior art.
In order to achieve the above objectives, the present invention adopts the following technical scheme that：
A kind of end-to-end unsupervised scene can traffic areas cognitive ability test method, include the following steps：
Information of road surface is shown to road plane viewing area, and carried out etc. than restoring to present by step 1)；
Step 2) constructs road surface line holographic projections area information using line holographic projections film, and obtaining that basic road surface is noiseless can Traffic areas r00；
Step 3) generates the scene of different complexities using holographic imaging by setting different scenes parameter；
Step 4), using under different roadway scene infomation detection different scenes can traffic areas be rx；X is different road surfaces The combination of scene information；
Step 5), definition can traffic areas cognitive ability be Sx
Wherein, x ∈ Ω={ 00,11,22,33,44,55,12,13,14,15,23,25,123,124,125 }, k are difference Test scene number under scenario parameters, i are different scenes coefficient, αxFor scene brightness coefficient, βxFor rain coefficient, γxFor scene mist Change coefficient, σxFor snowization coefficient, mxFor vehicle number, nxFor pedestrian's number, rROIxFor be detected can traffic areas with manually mark Can traffic areas friendship and ratio,
Wherein rxIndicate different scenes under be detected can traffic areas, RxIndicate that is manually marked under different scenes leads to Row region；
Step 6), calculate scene can under each scene of traffic areas cognitive ability can traffic areas detection average ability：
Wherein, N indicates the number in Ω；
Step 7), calculate scene can traffic areas cognitive ability under each scene can traffic areas detection stability：
SSD=∑x∈Ω(SAVG-Sx)2。
Further, it in step 1), is carried out etc. using the graphic display device plane information that satisfies the need than restoring and presenting.
Further, the display unit in the graphic display device, is placed in test site ground, and display area is not low In true road surface size.
Further, line holographic projections area information includes roadway scene information in step 3), specifically includes people, vehicle, rain, mist With snow holographic model.
Further, different scenes parameter includes scene brightness α, scene rain factor beta, scene atomization coefficient gamma, scene Snowization factor sigma, the vehicle number m under scene, pedestrian's number n under scene.
Further, specifically, in step 3), vehicle, people, rain, mist, snow line holographic projections are loaded respectively on basic road surface The test of model row, obtaining detection road surface is r11、r22、r33、r44、r55。
Further, vehicle and pedestrian, vehicle and rain, vehicle and mist, vehicle and snow, people are loaded respectively on basic road surface With rain, people and mist, people and snow line holographic projections model test, obtain detection can traffic areas be r12、r13、r14、r15、r23、 r24、r25。
Further, load vehicle simultaneously on basic road surface, pedestrian, rain line holographic projections model are tested, examined Survey can traffic areas be r123；Load vehicle simultaneously on basic road surface, pedestrian, mist line holographic projections model are tested, examined Survey can traffic areas be r124；On basic road surface simultaneously load vehicle, pedestrian, snow line holographic projections model test, examined Survey can traffic areas be r125。
Compared with prior art, the invention has the following beneficial technical effects：
End-to-end unsupervised scene of the invention a kind of can traffic areas cognitive ability test method, by by information of road surface It is shown in test zone ground, true information of road surface can not only be reappeared, additionally it is possible to be shown using this kind of mode rich Rich road environment；It is shown in test zone ground using information of road surface, tested test system can be made in test, is obtained practical The identical detection environment of drive test；Using the method for holographic technique, object on reproduction road surface that can be true, three-dimensional makes to test Environment is closer to true environment；Different roadway scene infomation detection difference fields are obtained in such a way that varying environment is tested Under scape can traffic areas, can in all directions assessment can traffic areas detection system cognitive ability；So as to synthesis Now can traffic areas detection system cognitive ability；Test method of the invention can be unmanned intelligent vehicle in practical drive test A kind of effective, low-risk test and evaluation means are provided before.
Detailed description of the invention
Fig. 1 is testing scheme schematic diagram of the present invention.
Fig. 2 is test scene example of the present invention.
In figure, 1 is road plane viewing area, and 2 be holographic imaging face, and 3 be holographic scene area, and 4 be line holographic projections, and 5 be tested Try battery limits；A is original scene figure, and b is adjustment luminance graph 1, and c is adjustment luminance graph 2, and d is rain scene figure, and e is atomization field Jing Tu, f are snowization scene figure.
Specific embodiment
The invention will be described in further detail with reference to the accompanying drawing：
As shown in Figure 1 and Figure 2, a kind of scene end to end of view-based access control model can traffic areas cognitive ability test method, Include the following steps：
Information of road surface is shown to road plane viewing area, and carried out etc. than restoring to present by step 1)；
Step 2) constructs road surface line holographic projections area information using line holographic projections film, and obtaining that basic road surface is noiseless can Traffic areas r00；
Step 3) generates the scene of different complexities using holographic imaging by setting different scenes parameter；
Step 4), using under different roadway scene infomation detection different scenes can traffic areas be rx；X is different road surfaces The combination of scene information；
Step 5), definition can traffic areas cognitive ability be Sx
Wherein, x ∈ Ω={ 00,11,22,33,44,55,12,13,14,15,23,25,123,124,125, }, k are not With test scene number under scenario parameters, i is different scenes coefficient, αxFor scene brightness coefficient, βxFor rain coefficient, γxFor scene It is atomized coefficient, σxFor snowization coefficient, mxFor vehicle number, nxFor pedestrian's number, rROIxIt can traffic areas and artificial mark for what is be detected Can traffic areas friendship and ratio,
Wherein rxIndicate different scenes under be detected can traffic areas, RxIndicate that is manually marked under different scenes leads to Row region；
Step 6), calculate scene can under each scene of traffic areas cognitive ability can traffic areas detection average ability：
Wherein, N indicates the number in Ω；
Step 7), calculate scene can traffic areas cognitive ability under each scene can traffic areas detection stability：
SSD=Σx∈Ω(SAVG-Sx)2。
In step 1), carried out etc. using the graphic display device plane information that satisfies the need than restoring and presenting；
Display unit in the graphic display device, is placed in test site ground, and display area is not less than true road Face size；
Line holographic projections area information includes roadway scene information in step 3), and it is holographic to specifically include people, vehicle, rain, mist and snow Model；
Different scenes parameter includes scene brightness α, scene rain factor beta, and scene is atomized coefficient gamma, scene snowization factor sigma, Vehicle number m under scene, pedestrian's number n under scene；
Specifically, loading vehicle, people, rain, mist, snow line holographic projections model row survey respectively on basic road surface in step 3) Examination, obtaining detection road surface is r11、r22、r33、r44、r55；
Loaded respectively on basic road surface vehicle and pedestrian, vehicle and rain, vehicle and mist, vehicle and snow, people and rain, people with Mist, people and snow line holographic projections model test, obtain detection can traffic areas be r12、r13、r14、r15、r23、r24、r25；
Load vehicle simultaneously on basic road surface, pedestrian, rain line holographic projections model are tested, obtaining detection can FOH Domain is r123；
Load vehicle simultaneously on basic road surface, pedestrian, mist line holographic projections model are tested, obtaining detection can FOH Domain is r124；
On basic road surface simultaneously load vehicle, pedestrian, snow line holographic projections model test, obtain detection can FOH Domain is r125。
End-to-end unsupervised scene of the invention a kind of can traffic areas Cognitive Aptitude Test and appraisal procedure, by difference Under roadway scene infomation detection different scenes can traffic areas be rx, capability comparison is carried out, can be regarded to based on different images Feel solution unmanned intelligent vehicle scene can traffic areas cognitive ability carry out test and evaluation, can be driven for nobody It sails intelligent vehicle and a kind of effective, low-risk test and evaluation means is provided before practical drive test.

Claims (8)

1. a kind of end-to-end unsupervised scene can traffic areas cognitive ability test method, which is characterized in that including following step Suddenly：
Information of road surface is shown to road plane viewing area, and carried out etc. than restoring to present by step 1)；
Step 2) constructs road surface line holographic projections area information using line holographic projections film, and obtains that basic road surface is noiseless to pass through Region r00；
Step 3) generates the scene of different complexities using holographic imaging by setting different scenes parameter；
Step 4), using under different roadway scene infomation detection different scenes can traffic areas be rx；X is different roadway scenes The combination of information；
Step 5), definition can traffic areas cognitive ability be Sx
Wherein, x ∈ Ω={ 00,11,22,33,44,55,12,13,14,15,23,25,123,124,125 }, k are different scenes Test scene number under parameter, i are different scenes coefficient, αxFor scene brightness coefficient, βxFor rain coefficient, γxIt is atomized for scene and is Number, σxFor snowization coefficient, mxFor vehicle number, nxFor pedestrian's number, rROIxFor be detected can traffic areas lead to what is manually marked The friendship in row region and ratio,
Wherein rxIndicate different scenes under be detected can traffic areas, RxWhat is manually marked under expression different scenes can FOH Domain；
Step 6), calculate scene can under each scene of traffic areas cognitive ability can traffic areas detection average ability：
Wherein, N indicates the number in Ω；
Step 7), calculate scene can traffic areas cognitive ability under each scene can traffic areas detection stability：
SSD=∑x∈Ω(SAVG-Sx)2。
2. the end-to-end unsupervised scene of one kind according to claim 1 can traffic areas cognitive ability test method, It is characterized in that, in step 1), is carried out etc. using the graphic display device plane information that satisfies the need than restoring and presenting.
3. the end-to-end unsupervised scene of one kind according to claim 2 can traffic areas cognitive ability test method, It is characterized in that, the display unit in the graphic display device, be placed in test site ground, display area is not less than true road Face size.
4. the end-to-end unsupervised scene of one kind according to claim 1 can traffic areas cognitive ability test method, It is characterized in that, line holographic projections area information includes roadway scene information in step 3), and it is holographic to specifically include people, vehicle, rain, mist and snow Model.
5. the end-to-end unsupervised scene of one kind according to claim 1 can traffic areas cognitive ability test method, It is characterized in that, different scenes parameter includes scene brightness α, scene rain factor beta, scene atomization coefficient gamma, scene snowization coefficient σ, the vehicle number m under scene, pedestrian's number n under scene.
6. the end-to-end unsupervised scene of one kind according to claim 1 can traffic areas cognitive ability test method, It is characterized in that, specifically, loading vehicle, people, rain, mist, snow line holographic projections model row survey respectively on basic road surface in step 3) Examination, obtaining detection road surface is r11、r22、r33、r44、r55。
7. the end-to-end unsupervised scene of one kind according to claim 1 can traffic areas cognitive ability test method, It is characterized in that, loads vehicle and pedestrian, vehicle and rain, vehicle and mist, vehicle and snow, people and rain, people respectively on basic road surface With mist, people and snow line holographic projections model test, obtain detection can traffic areas be r12、r13、r14、r15、r23、r24、r25。
8. the end-to-end unsupervised scene of one kind according to claim 1 can traffic areas cognitive ability test method, It is characterized in that, load vehicle simultaneously on basic road surface, pedestrian, rain line holographic projections model are tested, obtaining detection can pass through Region is r123；Load vehicle simultaneously on basic road surface, pedestrian, mist line holographic projections model are tested, obtaining detection can pass through Region is r124；On basic road surface simultaneously load vehicle, pedestrian, snow line holographic projections model test, obtain detection can pass through Region is r125。
CN201810638850.2A 2018-06-20 2018-06-20 Method for testing cognitive ability of passable area in end-to-end unsupervised scene Active CN108875640B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810638850.2A CN108875640B (en) 2018-06-20 2018-06-20 Method for testing cognitive ability of passable area in end-to-end unsupervised scene

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810638850.2A CN108875640B (en) 2018-06-20 2018-06-20 Method for testing cognitive ability of passable area in end-to-end unsupervised scene

Publications (2)

Publication Number Publication Date
CN108875640A true CN108875640A (en) 2018-11-23
CN108875640B CN108875640B (en) 2022-04-05

Family

ID=64340071

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810638850.2A Active CN108875640B (en) 2018-06-20 2018-06-20 Method for testing cognitive ability of passable area in end-to-end unsupervised scene

Country Status (1)

Country Link
CN (1) CN108875640B (en)

Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102156977A (en) * 2010-12-22 2011-08-17 浙江大学 Vision-based road detection method
CN103528585A (en) * 2013-09-26 2014-01-22 中北大学 Path planning method of passable area divided at unequal distance
CN104408757A (en) * 2014-11-07 2015-03-11 吉林大学 Method and system for adding haze effect to driving scene video
CN104933708A (en) * 2015-06-07 2015-09-23 浙江大学 Barrier detection method in vegetation environment based on multispectral and 3D feature fusion
CN105266982A (en) * 2014-07-18 2016-01-27 南京专创知识产权服务有限公司 Intelligent wheel chair
CN105550665A (en) * 2016-01-15 2016-05-04 北京理工大学 Method for detecting pilotless automobile through area based on binocular vision
CN105807762A (en) * 2015-01-21 2016-07-27 福特全球技术公司 Autonomous driving refined in virtual environments
CN106657926A (en) * 2017-01-23 2017-05-10 高艳云 Information transmission system based on intelligent terminal
CN106627456A (en) * 2015-07-20 2017-05-10 国网河南省电力公司南阳供电公司 Tyre puncture prevention device
CN107284379A (en) * 2017-07-28 2017-10-24 合肥芯福传感器技术有限公司 AR optical projection systems and projecting method for vehicle safe driving
CN206619301U (en) * 2017-03-06 2017-11-07 公安部交通管理科学研究所 Strengthen the safe driving simulated training educational system of special efficacy with environment
CN107871418A (en) * 2017-12-27 2018-04-03 吉林大学 It is a kind of to be used to evaluate the man-machine experiment porch for driving reliability altogether
CN207466645U (en) * 2017-10-25 2018-06-08 李炳午 Safe driving assistant system

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102156977A (en) * 2010-12-22 2011-08-17 浙江大学 Vision-based road detection method
CN103528585A (en) * 2013-09-26 2014-01-22 中北大学 Path planning method of passable area divided at unequal distance
CN105266982A (en) * 2014-07-18 2016-01-27 南京专创知识产权服务有限公司 Intelligent wheel chair
CN104408757A (en) * 2014-11-07 2015-03-11 吉林大学 Method and system for adding haze effect to driving scene video
CN105807762A (en) * 2015-01-21 2016-07-27 福特全球技术公司 Autonomous driving refined in virtual environments
CN104933708A (en) * 2015-06-07 2015-09-23 浙江大学 Barrier detection method in vegetation environment based on multispectral and 3D feature fusion
CN106627456A (en) * 2015-07-20 2017-05-10 国网河南省电力公司南阳供电公司 Tyre puncture prevention device
CN105550665A (en) * 2016-01-15 2016-05-04 北京理工大学 Method for detecting pilotless automobile through area based on binocular vision
CN106657926A (en) * 2017-01-23 2017-05-10 高艳云 Information transmission system based on intelligent terminal
CN206619301U (en) * 2017-03-06 2017-11-07 公安部交通管理科学研究所 Strengthen the safe driving simulated training educational system of special efficacy with environment
CN107284379A (en) * 2017-07-28 2017-10-24 合肥芯福传感器技术有限公司 AR optical projection systems and projecting method for vehicle safe driving
CN207466645U (en) * 2017-10-25 2018-06-08 李炳午 Safe driving assistant system
CN107871418A (en) * 2017-12-27 2018-04-03 吉林大学 It is a kind of to be used to evaluate the man-machine experiment porch for driving reliability altogether

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
朱长军等著: "《大学物理学 下 第2版》", 31 December 2017, 朱长军等著 *

Also Published As

Publication number Publication date
CN108875640B (en) 2022-04-05

Similar Documents

Publication Publication Date Title
US10607089B2 (en) Re-identifying an object in a test image
Daniel et al. Representational validity of landscape visualizations: the effects of graphical realism on perceived scenic beauty of forest vistas
CN108537743A (en) A kind of face-image Enhancement Method based on generation confrontation network
CN107679556A (en) The zero sample image sorting technique based on variation autocoder
WO2011047924A1 (en) Method for determining the pose of a camera and for recognizing an object of a real environment
US20230026811A1 (en) System and method for removing haze from remote sensing images
CN110033411A (en) The efficient joining method of highway construction scene panoramic picture based on unmanned plane
Pan et al. Multi‐source information art painting fusion interactive 3d dynamic scene virtual reality technology application research
Halmaoui et al. Contrast restoration of road images taken in foggy weather
CN114089786A (en) Autonomous inspection system based on unmanned aerial vehicle vision and along mountain highway
CN110503092B (en) Improved SSD monitoring video target detection method based on field adaptation
You et al. Dmrvisnet: Deep multihead regression network for pixel-wise visibility estimation under foggy weather
Hou et al. Factors affecting the performance of 3D thermal mapping for energy audits in a district by using infrared thermography (IRT) mounted on unmanned aircraft systems (UAS)
CN109285187A (en) A kind of farthest visible point detecting method based on traffic surveillance videos image
CN108875640A (en) A kind of end-to-end unsupervised scene can traffic areas cognitive ability test method
Chen et al. From depth-aware haze generation to real-world haze removal
Srivastava et al. Color correction for object tracking across multiple cameras
CN115373416A (en) Intelligent inspection method for railway power through line
Chen et al. A novel monocular calibration method for underwater vision measurement
Zhao et al. Visibility video detection with dark channel prior on highway
CN114120236A (en) Method for identifying and positioning low-altitude target
Zhao et al. Nodes2STRNet for structural dense displacement recognition by deformable mesh model and motion representation
Paulin et al. Person localization and distance determination using the raycast method
Savaria Novel Use of Neural Networks to Identify and Detect Electrical Infrastructure Performance
Sim A proposed method for disaggregating census data using object-oriented image classification and GIS

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant