CN105957345B - Vehicle operation data processing method - Google Patents
Vehicle operation data processing method Download PDFInfo
- Publication number
- CN105957345B CN105957345B CN201610403920.7A CN201610403920A CN105957345B CN 105957345 B CN105957345 B CN 105957345B CN 201610403920 A CN201610403920 A CN 201610403920A CN 105957345 B CN105957345 B CN 105957345B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- mrow
- instrument board
- data
- accident
- 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.)
- Active
Links
Classifications
-
- 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
- G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
-
- 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
- G08G1/0125—Traffic data processing
-
- 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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
Abstract
The invention discloses vehicle operation data processing method, including:Instrument board template database is set up according to automobile model;The image information and car status information in front of the instrument board image information in vehicle travel process, vehicle traveling are obtained in real time;Vehicle-state is monitored, vehicle-state is abnormal if finding:Instrument board image information is extracted with instrument board template database compared with pair, identification vehicle speed data and instrument board indication signal data generate image and travelling data with reference to other information, carry out collecting preservation, and alert;If it was found that vehicle-state is traffic accident:The travelling data for relating to thing vehicle is extracted, accident generating process is reduced.The present invention connects the event occurred in driving conditions with corresponding instrument board image, and view data can judge as intuitively evidence, for judgement violating the regulations, accident responsibility.
Description
Technical field
The present invention relates to vehicle traveling record and driving conditions reduction technique field, and in particular to vehicle operation data processing
Method.
Background technology
When special event, accident or failure occur for automobile, the reason for recording special event, analysis accident or failure needs
Travelling data is obtained, but these travelling datas are being difficult to obtain afterwards.
Driver is when driving, and main focus is three aspects:Front visual angle, instrument board, well perceive vehicle
Transport condition (such as sound, vibrations).It is few to carry out collecting preservation with the information of this three aspect in existing system, it is used in combination
In driving conditions reduction.And the configured information of instrument board is more and not general, many drivers, which can only debate, knows conventional instruction
Data, to special fault cues often no concept or can ignore these prompting.
Other existing vehicle is often positioned with GPS positioning system, but because GPS positioning precision is relatively low, vehicle movement
During gps signal existence and stability problem so that continuity, the reliability of data are not high enough.
It is many in existing scheme to be used as reference by gathering electronic data, and electronic data can not turn into intuitively evidence;By
It is different in the species model of vehicle, can be with the traveling shape of accurate recording vehicle currently without a kind of simple common apparatus
State, complete evidence can not be preserved when there is accident or incident, the state of vehicle at that time can not be completely reproduced up.
The content of the invention
Goal of the invention:In view of the shortcomings of the prior art, the present invention provides a kind of vehicle operation data processing method, Neng Goushi
Driving conditions are now reappeared afterwards.
Technical scheme:Vehicle operation data processing method of the present invention, including following method:
(1) instrument board template database is set up according to automobile model;
(2) in real time obtain vehicle travel process in instrument board image information, vehicle traveling in front of image information and
Car status information;
(3) vehicle-state is monitored, vehicle strange happening part is abnormal state if finding:Extract instrument board image information and instrument board
Template database compared to pair, identification vehicle speed data and instrument board indication signal data, with reference to vehicle travel in front of image information,
Data after image and travelling data of the car status information generation with digital watermarking, extraction process carry out collecting preservation, and
To vehicle driver's alert;
(4) vehicle-state is monitored, vehicle strange happening part is traffic accident if finding:Crucial 2, vehicle in traffic accident is determined,
The instrument board image information and car status information of crucial vehicle are extracted, using the histogram analysis in data mining, is determined different
Event material time point, on the basis of material time point, in extraction traffic accident in material time point before involved vehicle traveling
The image information of side carries out the extracting and matching feature points based on SIFT algorithms, retrodicts to set out and makes trouble therefore the driving of each preceding vehicle
Data, recover accident panoramic view, reduce accident generating process.
Above-mentioned technical proposal is further improved, the information of the instrument board template data library storage includes instrument board picture,
Figure, position, indicating mode and the instruction implication of the indication signal of each in instrument board.
Further, the instrument board image information is imaged by being fixed at least one instrument board on steering wheel support base
Machine, which is shot, to be obtained;Image information in front of the vehicle traveling is shot by the forward direction visual angle camera on vehicle to be obtained;It is described
Car status information passes through the angular-rate sensor being arranged on vehicle, shock sensor, acceleration transducer, gyro sensors
One or more of device, speed pulse sensor are obtained.
Further, the instrument board image information in vehicle travel process is obtained in the step (2) in real time including as follows
Step:
(21) pending instrument board image is read, bianry image is translated into;
(22) remove area in bianry image too small, can be certainly non-pointer and the region of scale;
(23) position indicator hand of dial, white portion is expanded, and removes unrelated parameter;
(24) connected region border is searched, while retaining image, in case marking below;
(25) the largest connected region of pointer in all connected regions is found out;
(26) pointer angle is extracted with Radon algorithms, speed is obtained by the scale template of instrument board template database memory storage
Spend information;
(27) the connected region figure in largest connected region is deducted, identification falls back R, neutral gear N, automatic catch D, parking P letters
Breath, determines gear information;
(28) extract non-image with the storage of data forms after critical data, reduce the pressure of storage image.
Further, the digital watermarking includes instrument board analyze data, onboard sensor data, shooting time mark.
Further, the method for reduction accident generating process is as follows in the step (4):
First, extract and relate to the travelling data of thing vehicle in traffic accident and collect;
Second, by the instrument board image information in the travelling data of each vehicle, the image information in front of vehicle traveling
And car status information determines that accident corresponding time point occurs for each vehicle;
3rd, on the basis of the time point for occurring accident by each vehicle, retrodict set out make trouble thus before each vehicle driving
Track, while the vehicle-mounted image on correspondence time point is shown, the process that reduction accident occurs;
(31) it is distributed by the histogram approximate data in data mining, Decision Tree Inductive analyzes triggered time point;
(32) to relating to the image information I in thing vehicle in front of two crucial vehicle travelings1(x,y)、I2(x, y), carries out base
In the extracting and matching feature points of SIFT algorithms, recover accident panoramic view:Metric space is built using Gaussian convolution computing, it is high
This difference function D (x, y, σ) with two of constant multiplication factor k adjacent scalogram aberrations by being calculated:
D (x, y, σ)=(G (x, y, k σ)-G (x, y, σ)) * I (x, y)=L (x, y, k σ)-L (x, y, σ) (1)
Initial pictures obtain a series of metric spaces, i.e. Gauss yardstick (DOG) space through progressively Gaussian convolution computing, empty
Between extreme point detection key point be made up of the Local Extremum in DOG spaces, middle test point and it with yardstick 8 phases
Totally 26 points compare for adjoint point and corresponding 9 × 2 points of neighbouring yardstick, to ensure in metric space and two dimensional image space all
Detect extreme point;
The direction distribution of extreme point:For each sampled point L (x, y), calculate its gradient modulus value m (x, y) and direction (x,
Y) formula is
The generation of feature point description:Reference axis is rotated on the gradient direction of characteristic point, it is ensured that rotational invariance, led to
A characteristic point is described frequently with 16 son points, the gradient accumulated value per 8 directions in height point is then calculated, obtains feature
The characteristic vector of point description, is 4*4*8=128 dimensional vectors;Resulting characteristic vector has invariable rotary shape so can be with
I is extracted respectively1And I2Characteristic vector in two images carries out Feature Points Matching, is spliced into a width panoramic view;
4th, the process occurred according to the accident of live reduction and simulation modules exhibit determines the responsibility of accident.
Beneficial effect:Compared with prior art, advantages of the present invention:
1st, the event occurred in driving conditions is connected with corresponding instrument board image, view data can be as straight
The evidence of sight, for it is violating the regulations judge, accident responsibility judges etc.;
2nd, by the analysis to instrument board image, the indication signal data of instrument board can be recognized, when failures are detected,
Prompting driver that can be promptly and accurately;
3rd, by live reduction and simulation module it is accurate, driving conditions are intuitively reappeared, available for scene of a traffic accident reduction
With quick place's fix duty, vehicle trouble investigation and reproduction, vehicle peccancy evidence obtaining.
Brief description of the drawings
Fig. 1 is hardware layout figure of the invention;
Fig. 2 is flow chart of data processing figure of the present invention;
Fig. 3 is accident reduction process flow chart.
Embodiment
Technical solution of the present invention is described in detail below.
Embodiment 1:As shown in figure 1, setting instrument on the support base of steering wheel 5 according to the size and structure of meter panel of motor vehicle
Disk video camera 2, when instrument board is wider or during more dispersed distribution, an instrument disk video camera can not shoot complete instrument board figure
Picture, the instrument board image of different piece can be shot respectively using multiple instrument disk video cameras 2;To visual angle before being installed on vehicle
Camera 4;Angular-rate sensor 6, shock sensor 7, acceleration transducer 8, gyro sensor 9, speed are set on vehicle
Pulse transducer 10 is spent, angular-rate sensor 6 is arranged on the steering wheel 5 of vehicle, the rotational angle for detecting steering wheel 5,
Shock sensor 7 is used for the transport condition for detecting vehicle and abnormal shake, seismism, the collection vehicle of acceleration transducer 8
Acceleration information, the collection vehicle of gyro sensor 9 turn to data, the collection vehicle road speed of speed pulse sensor 10
Data, the position installation data processing module 3 of instrument disk video camera 2 and live reduction and simulation module, data processing on vehicle
Module loading has instrument board template database and provided with wireless communication module and alarm module, and instrument board template database is according to vapour
Vehicle number is set up, and the information of storage includes figure, position, the indicating mode of each indication signal in instrument board picture, instrument board
With instruction implication, wireless communication module is used for and mobile network base station or traffic control system base station communication, data processing module
By wireless communication module obtain base station provide present road information (such as speed limit restricted driving information, front road section traffic volume accident or
The information of road congestion) driver is given, and put on record to base station upload vehicular events data, event includes violating the regulations, accident, dashed forward
Hair failure such as vehicle casts anchor alarm.
Running data processing method is carried out based on above-mentioned setting:
(1) instrument board template database is set up according to automobile model;
(2) in real time obtain vehicle travel process in instrument board image information, vehicle traveling in front of image information and
Car status information;
Instrument board real-time image information acquisition algorithm:
1. pending instrument board image is read, bianry image is translated into;
2. area is too small, can affirm the region of non-pointer and scale in removal image;
3. it is positioning pointer, white portion is expanded, unrelated small articles (traveling milimeter number, ambient temperature) is removed in corrosion;
4. connected region border is searched, while retaining this figure, in case marking below;
5. (the largest connected region) of most probable pointer in all connected regions is found out;
6. pointer angle is extracted with Radon algorithms, velocity information is obtained by with the contrast of instrument board template database;
7. button removes the connected region figure in largest connected region, recognizes R, N, D, P information, determines gear information;
8. data forms are stored and non-image, the pressure of reduction storage image after extraction critical data;
(3) data processing module monitoring vehicle-state, if finding, vehicle-state is abnormal, and such as Vehicle Speed is abnormal, turn
To abnormal, braking exception, instrument board alarm:Instrument board image information is extracted with instrument board template database compared with pair, identification car
Fast data and instrument board indication signal data, instrument board can be divided into various in mechanical and two kinds of electronic type, mechanical instrument board
The position of indication signal is fixed, can there is the switching at difference in functionality interface in electronic instrument disk, and the content of display also has
Institute is different.Therefore, when setting up template for electronic instrument disk, the feature of each function interface should be defined first, then describe each
The implication of instrument display information in individual function interface.For example in mechanical instrument board speed shows the form of mostly pointer rotation,
And speed is then indirectly displayed as digital form by some in electronic instrument disk, the numeral or word in electronic instrument disk are believed
Breath should be identified using OCR technique, because the installation site of instrument disk video camera, angle can be present necessarily in each vehicle
Difference, it is therefore desirable to matched instrument board image with the progress of instrument board template by accuracy registration algorithm, calculates installation and misses
Difference, view data can be modified in actual moving process, in favor of the automatic identification to instrument board information, reduce fortune
Calculation amount;In combination with the image of the image information in front of vehicle traveling, car status information generation with digital watermarking and driving
Data, image includes the combination of instrument board image or instrument board image and forward direction multi-view image, and digital watermarking includes instrument board
Data after analyze data, onboard sensor data, shooting time mark, extraction process carry out Macro or mass analysis preservation and by report
Unit is warned to vehicle driver's alert;
(4) if finding, vehicle-state strange happening part is traffic accident, extracts two travelling datas for relating to thing vehicle, such as Fig. 2, figure
Shown in 3, on the basis of the time point that by each vehicle accident occurs for live reduction and simulation module, retrodict to set out and make trouble therefore each preceding car
Travelling data, while show correspondence time point on combination image, reduce accident generating process;Determine to close in traffic accident
2, key vehicle, extracts instrument board data and vehicle status data, and various data point are merged by the statistical method in data mining
Separate out triggered time point;Multi-section vehicle front view picture involved in the time point in traffic accident is extracted to carry out being based on SIFTF
Extracting and matching feature points, recover accident panoramic view.
Live reduction and simulation module is realized by the following method:
With reference to vehicle speed data and Vehicular turn data, the actual travel track of vehicle is simulated, and travel rail in display
The vehicle-mounted image on correspondence time point is shown while mark, so as to reproduce the traveling process of vehicle;
Wherein:Vehicle-mounted image is the group of two kinds of images of instrument board image or instrument board figure image and forward direction multi-view image
Close, can be than more comprehensively reappearing the scene that driver observes by the combination of above two image;
Vehicle speed data is extracted or provided by speed pulse sensor in instrument board image, and Vehicular turn data are passed by angle
Either gyro sensor provides or is analyzed and obtained to the image of viewpoint cameras to preceding sensor;
By the comparative analysis to the preceding consecutive image shot to viewpoint cameras, the object of reference information extracted in image,
It may determine that vehicle is in the state that straight-going state is in turning, it is possible to obtain the angle turned.
When using live reduction and simulation resume module traffic accident, it is realized by the following method:
First, obtain the travelling data for the data processing module for being related to vehicle in traffic accident and collect;
Second, determined according to the travelling data of each vehicle and accident corresponding time point occurs for each vehicle;
Can by before checking to multi-view image, the mutation of checking speed, control shock sensor, acceleration transducer
The methods such as data, determine that accident corresponding time point occurs in travelling data;
3rd, on the basis of the time point for occurring accident by each vehicle, retrodict set out make trouble thus before each vehicle driving
Track, while the vehicle-mounted image on correspondence time point is shown, the process that reduction accident occurs:
Various data analyses are merged by the statistical method in data mining and go out triggered time point:
Histogram is to carry out approximate data using branch mailbox to be distributed, and is a kind of data regularization form.According to time point, timing statisticses
Point feature is worth quantity.The data of strange happening part triggering are few event, so statistics with histogram is effective.Multi-class data determines strange happening
At part time point, final material time point is obtained using Decision Tree Inductive;
Assuming that vehicle 1 and 2 is crucial vehicle in traffic accident, the image in front of traveling photographed is I1(x,y)、I2(x,
Y), two width figures are subjected to characteristic matching with SIFT algorithms and complete panoramic view;
SIFT algorithms, scale invariant feature conversion (Scale-invariant feature transform, SIFT) is one
The algorithm for planting computer vision is used for detecting and describing the locality characteristic in image, and it finds extreme point in space scale, and
Its position, yardstick, rotational invariants are extracted, this algorithm was delivered by David Lowe in 1999, improve within 2004 and summarize,
This algorithm can be used for image mosaic.SIFT algorithm key steps are as follows:
1st, metric space is built:For the position of the stable key point of effective detection in metric space, Lowe is proposed
Difference of Gaussian convolution, difference of Gaussian function D (x, y, σ) can be by with two of constant multiplication factor k adjacent yardstick image differences
Calculate:
D (x, y, σ)=(G (x, y, k σ)-G (x, y, σ)) * I (x, y)=L (x, y, k σ)-L (x, y, σ) (1);
Initial pictures obtain a series of metric spaces, i.e. Gauss yardstick (DOG) space through progressively Gaussian convolution computing.
2nd, spatial extrema point is detected:
Key point is made up of the Local Extremum in DOG spaces, middle test point and its 8 consecutive points with yardstick
Totally 26 points compare 9 × 2 points corresponding with neighbouring yardstick, to ensure all to detect in metric space and two dimensional image space
To extreme point.
3rd, the direction distribution of extreme point:
For each sampled point L (x, y), the formula for calculating its gradient modulus value m (x, y) and direction (x, y) is
4th, the generation of feature point description:
Reference axis is rotated on the gradient direction of characteristic point, it is ensured that rotational invariance, generally retouched using 16 son points
State a characteristic point, the gradient accumulated value in 8 directions in then calculating per height point, obtain the feature of feature point description to
Amount, is 4*4*8=128 dimensional vectors;Resulting characteristic vector has invariable rotary shape so I can be extracted respectively1And I2Two width
Characteristic vector in image carries out Feature Points Matching, is spliced into final image I.
Finally, the process occurred according to the accident of live reduction and simulation modules exhibit determines the responsibility of accident.
As described above, although the present invention has been represented and described with reference to specific preferred embodiment, it must not be explained
For to the limitation of itself of the invention., can be right under the premise of the spirit and scope of the present invention that appended claims are defined are not departed from
Various changes can be made in the form and details for it.
Claims (5)
1. vehicle operation data processing method, it is characterised in that comprise the following steps:
(1) instrument board template database is set up according to automobile model;
(2) image information and vehicle in front of the instrument board image information in vehicle travel process, vehicle traveling are obtained in real time
Status information;
(3) vehicle-state is monitored, vehicle strange happening part is abnormal state if finding:Extract instrument board image information and instrument board template
Database recognizes vehicle speed data and instrument board indication signal data compared to, with reference to the image information in front of vehicle traveling, vehicle
Image and travelling data of the status information generation with digital watermarking, the data after extraction process collect preservation, and to car
Driver's alert;
(4) vehicle-state is monitored, vehicle strange happening part is traffic accident if finding:Crucial 2, vehicle in traffic accident is determined, is extracted
The instrument board image information and car status information of crucial vehicle, using the histogram analysis in data mining, determine strange happening part
Material time point, on the basis of material time point, in extraction traffic accident in material time point in front of involved vehicle traveling
Image information carries out the extracting and matching feature points based on SIFT algorithms, retrodicts to set out and makes trouble therefore the driving number of each preceding vehicle
According to recovery accident panoramic view reduces accident generating process;
The method of the reduction accident generating process is as follows:
First, extract and relate to the travelling data of thing vehicle in traffic accident and collect;
Second, by the instrument board image information in the travelling data of each vehicle, image information in front of vehicle traveling and
Car status information determines that accident corresponding time point occurs for each vehicle;
3rd, on the basis of the time point for occurring accident by each vehicle, retrodict set out make trouble thus before each vehicle wheelpath,
The vehicle-mounted image on correspondence time point, the process that reduction accident occurs are shown simultaneously;
(31) it is distributed by the histogram approximate data in data mining, Decision Tree Inductive analyzes triggered time point;
(32) to relating to the image information I in thing vehicle in front of two crucial vehicle travelings1(x,y)、I2(x, y), carries out being based on SIFT
The extracting and matching feature points of algorithm, recover accident panoramic view:Metric space, difference of Gaussian are built using Gaussian convolution computing
Function D (x, y, σ) with two of constant multiplication factor k adjacent scalogram aberrations by being calculated:
D (x, y, σ)=(G (x, y, k σ)-G (x, y, σ)) * I (x, y)=L (x, y, k σ)-L (x, y, σ) (1)
Initial pictures obtain a series of metric spaces, i.e. Gauss yardstick DOG spaces, spatial extrema through progressively Gaussian convolution computing
Point detection key point is made up of the Local Extremum in DOG spaces, middle test point and it with yardstick 8 consecutive points and
Totally 26 points compare corresponding 9 × 2 points of neighbouring yardstick, to ensure all to detect in metric space and two dimensional image space
Extreme point;
The direction distribution of extreme point:For each sampled point L (x, y), its gradient modulus value m (x, y) and direction θ (x, y) is calculated
Formula be:
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msqrt>
<mrow>
<msup>
<mrow>
<mo>(</mo>
<mi>L</mi>
<mo>(</mo>
<mi>x</mi>
<mo>+</mo>
<mn>1</mn>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
<mo>-</mo>
<mi>L</mi>
<mo>(</mo>
<mi>x</mi>
<mo>-</mo>
<mn>1</mn>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>+</mo>
<msup>
<mrow>
<mo>(</mo>
<mi>L</mi>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
<mo>-</mo>
<mi>L</mi>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>-</mo>
<mn>1</mn>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>&theta;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msup>
<mi>tan</mi>
<mrow>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msup>
<mrow>
<mo>(</mo>
<mi>L</mi>
<mo>(</mo>
<mrow>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
<mo>-</mo>
<mi>L</mi>
<mo>(</mo>
<mrow>
<mi>x</mi>
<mo>,</mo>
<mi>y</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
<mo>/</mo>
<mrow>
<mo>(</mo>
<mi>L</mi>
<mo>(</mo>
<mrow>
<mi>x</mi>
<mo>+</mo>
<mn>1</mn>
<mo>,</mo>
<mi>y</mi>
</mrow>
<mo>)</mo>
<mo>-</mo>
<mi>L</mi>
<mo>(</mo>
<mrow>
<mi>x</mi>
<mo>-</mo>
<mn>1</mn>
<mo>,</mo>
<mi>y</mi>
</mrow>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>2</mn>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
The generation of feature point description:Reference axis is rotated on the gradient direction of characteristic point, it is ensured that rotational invariance, using 16
Height point describes a characteristic point, then calculates the gradient accumulated value per 8 directions in height point, obtains feature point description
Characteristic vector, be 4*4*8=128 dimensional vectors;Resulting characteristic vector has invariable rotary shape so can extract respectively
I1And I2Characteristic vector in two images carries out Feature Points Matching, is spliced into a width panoramic view;
4th, the process occurred according to the accident of live reduction and simulation modules exhibit determines the responsibility of accident.
2. vehicle operation data processing method according to claim 1, it is characterised in that:The instrument board template database
The information of storage includes the figure of each indication signal in instrument board picture, instrument board, position, indicating mode and indicates implication.
3. vehicle operation data processing method according to claim 1, it is characterised in that:The instrument board image information is led to
Cross and be fixed at least one instrument disk video camera shooting acquisition on steering wheel support base;Image information in front of the vehicle traveling
Shot and obtained by the forward direction visual angle camera on vehicle;The car status information is passed by the angular speed being arranged on vehicle
One or more of sensor, shock sensor, acceleration transducer, gyro sensor, speed pulse sensor are obtained.
4. vehicle operation data processing method according to claim 1, it is characterised in that:Obtained in real time in the step (2)
The instrument board image information in vehicle travel process is taken to comprise the following steps:
(21) pending instrument board image is read, bianry image is translated into;
(22) remove area in bianry image too small, can be certainly non-pointer and the region of scale;
(23) position indicator hand of dial, white portion is expanded, and removes unrelated parameter;
(24) connected region border is searched, while retaining image, in case marking below;
(25) the largest connected region of pointer in all connected regions is found out;
(26) pointer angle is extracted with Radon algorithms, speed letter is obtained by the scale template of instrument board template database memory storage
Breath;
(27) the connected region figure in largest connected region is deducted, identification falls back R, neutral gear N, automatic transmission D, parking P information, really
Determine gear information;
(28) extract after critical data with data form storage images, reduce the pressure of storage image.
5. vehicle operation data processing method according to claim 1, it is characterised in that:The digital watermarking includes instrument
Disk analyze data, onboard sensor data, shooting time mark.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610403920.7A CN105957345B (en) | 2016-06-08 | 2016-06-08 | Vehicle operation data processing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610403920.7A CN105957345B (en) | 2016-06-08 | 2016-06-08 | Vehicle operation data processing method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105957345A CN105957345A (en) | 2016-09-21 |
CN105957345B true CN105957345B (en) | 2017-10-31 |
Family
ID=56908942
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610403920.7A Active CN105957345B (en) | 2016-06-08 | 2016-06-08 | Vehicle operation data processing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105957345B (en) |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107423681A (en) * | 2017-05-27 | 2017-12-01 | 上海骁达信息科技有限公司 | A kind of vehicle automatic identification method and system |
CN107547816A (en) * | 2017-08-11 | 2018-01-05 | 深圳市麦道微电子技术有限公司 | A kind of processing system for mixing car body real time information with realtime graphic |
CN107529045A (en) * | 2017-10-19 | 2017-12-29 | 南京中高知识产权股份有限公司 | A kind of 360 degree of full-view image systems of automobile and its method of work |
JP7043785B2 (en) * | 2017-10-25 | 2022-03-30 | 株式会社Ihi | Information generator |
CN108053520B (en) * | 2018-01-15 | 2020-11-10 | 陈世辉 | Intelligent processing system for traffic accident processing |
CN108833596B (en) * | 2018-08-05 | 2019-06-21 | 北京吉宝通科技发展有限公司 | A kind of circumference vibration alarming system based on Internet of Things |
CN109819224B (en) * | 2019-03-20 | 2020-12-01 | 安徽宽广科技有限公司 | Vehicle-mounted video monitoring system based on Internet of vehicles |
CN114283040A (en) * | 2019-12-04 | 2022-04-05 | 蚂蚁胜信(上海)信息技术有限公司 | Traffic accident processing method and device and electronic equipment |
CN113053099A (en) * | 2019-12-26 | 2021-06-29 | 北京万集智能网联技术有限公司 | Abnormal traffic incident processing method and device |
CN111310696B (en) * | 2020-02-26 | 2023-09-15 | 广州小鹏汽车科技有限公司 | Parking accident identification method and device based on analysis of abnormal parking behaviors and vehicle |
CN111917873A (en) * | 2020-07-31 | 2020-11-10 | 中国建设银行股份有限公司 | Accident processing system, method, device, terminal and storage medium |
CN113361413A (en) * | 2021-06-08 | 2021-09-07 | 南京三百云信息科技有限公司 | Mileage display area detection method, device, equipment and storage medium |
CN115643351A (en) * | 2022-10-12 | 2023-01-24 | 中国第一汽车股份有限公司 | Method and device for automatically generating electronic commemorative postcard by vehicle |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7587762B2 (en) * | 2002-08-09 | 2009-09-08 | Netscout Systems, Inc. | Intrusion detection system and network flow director method |
JP2005018575A (en) * | 2003-06-27 | 2005-01-20 | Sony Corp | Car navigation device |
CN102314772A (en) * | 2010-07-01 | 2012-01-11 | 北京中星微电子有限公司 | Vehicle over-speed detection method, vehicle over-speed detection device and vehicle over-speed alarm system |
CN202217329U (en) * | 2011-08-11 | 2012-05-09 | 浙江金刚汽车有限公司 | Vehicle driving recording system |
CN103824346B (en) * | 2014-02-17 | 2016-04-13 | 深圳市宇恒互动科技开发有限公司 | Driving recording and replay method and system |
-
2016
- 2016-06-08 CN CN201610403920.7A patent/CN105957345B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN105957345A (en) | 2016-09-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105957345B (en) | Vehicle operation data processing method | |
CN106097480A (en) | Vehicle operation data record system | |
CN102208012B (en) | Landscape coupling reference data generation system and position measuring system | |
KR102267335B1 (en) | Method for detecting a speed employing difference of distance between an object and a monitoring camera | |
CN104296756B (en) | Run the method and motor vehicle of motor vehicle | |
CN110298300B (en) | Method for detecting vehicle illegal line pressing | |
US20200240795A1 (en) | Road map generation system and road map generation method | |
US20100030540A1 (en) | System and method for reconstructing traffic accident | |
EP3751480A1 (en) | System and method for detecting on-street parking violations | |
JP4761156B2 (en) | Feature position recognition apparatus and feature position recognition method | |
US11475679B2 (en) | Road map generation system and road map generation method | |
CN104574954A (en) | Vehicle checking method and system based on free flow system as well as control equipment | |
CN109766867A (en) | Travel condition of vehicle determines method, apparatus, computer equipment and storage medium | |
CN103426211A (en) | Vehicle traveling condition analysis system, analysis apparatus and analysis method | |
CN112735164B (en) | Test data construction method and test method | |
CN112598743B (en) | Pose estimation method and related device for monocular vision image | |
JP6419260B1 (en) | Traffic information acquisition device, traffic information acquisition system, traffic information acquisition method, and traffic information acquisition program | |
CN110539748B (en) | Congestion car following system and terminal based on look-around | |
CN109345576B (en) | Vehicle running speed identification method and system | |
KR102414632B1 (en) | Method for determining the location of a fixed object using multiple observation information | |
KR20150031942A (en) | System for collecting vehicle information and method thereof | |
CN111126336A (en) | Sample collection method, device and equipment | |
CN111372051A (en) | Multi-camera linkage blind area detection method and device and electronic equipment | |
Hassan et al. | ChattSpeed: Toward a New Dataset for Single Camera Visual Speed Estimation for Urban Testbeds | |
CN112334357A (en) | Travel control device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |