CN107886751A - A kind of car-mounted terminal real-time road early warning system - Google Patents
A kind of car-mounted terminal real-time road early warning system Download PDFInfo
- Publication number
- CN107886751A CN107886751A CN201711081209.5A CN201711081209A CN107886751A CN 107886751 A CN107886751 A CN 107886751A CN 201711081209 A CN201711081209 A CN 201711081209A CN 107886751 A CN107886751 A CN 107886751A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- road conditions
- image
- msup
- 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
Links
- 238000012545 processing Methods 0.000 claims abstract description 23
- 230000006837 decompression Effects 0.000 claims abstract description 19
- 230000006835 compression Effects 0.000 claims abstract description 9
- 238000007906 compression Methods 0.000 claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000006243 chemical reaction Methods 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 11
- 230000009466 transformation Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000004891 communication Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 2
- 230000011218 segmentation Effects 0.000 claims description 2
- 238000000638 solvent extraction Methods 0.000 claims description 2
- 230000001105 regulatory effect Effects 0.000 claims 1
- 230000017105 transposition Effects 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000008775 paternal effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 102000037983 regulatory factors Human genes 0.000 description 1
- 108091008025 regulatory factors Proteins 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- 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
- G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Atmospheric Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention provides a kind of car-mounted terminal real-time road early warning system, the system includes camera, image processing module, decompression module, cloud computing platform and terminal device, and the camera is used to capture current road conditions image;Described image processing module is used to be compressed road conditions image, store, and the road conditions image transmitting after compression is handled by wireless network to decompression module carries out decompression operations;The cloud computing platform is used to analyze and process the road conditions image that decompression obtains, and obtains current road condition data;The terminal device is used to provide the user current traffic information according to the road condition data that cloud computing platform provides, and facilitates user according to traffic information travel route planning;And terminal device connects with the prior-warning device on car, when front gets congestion, also early warning, prompting changing route actively can be sent to driver.
Description
Technical field
The present invention relates to automobile technical field, and in particular to a kind of car-mounted terminal real-time road early warning system.
Background technology
With rapid development of economy and the quickening of urbanization process, traffic conflict becomes increasingly conspicuous, and traffic jam issue is tight
The normal life of people is have impact on again.Traffic information is a kind of real-time Traffic Information, and it reflects specific region
Interior road traffic condition and Recent Changes trend.Traffic information can not only provide for vehicle supervision department's transporting traffic
Foundation, and provide help for human pilot and common traveler reasonable selection route.Therefore, traffic information is adopted
Collect, handle, report and issue the traffic jam issue that can aid in and solve current getting worse.
Daily camera needs to gather substantial amounts of real-time traffic information, by these videos take up space greatly, is passing
The problems such as Flux Loss is big, transmission speed is low, poor real occurs in defeated, inconvenience is brought to people's trip, it is how real
The problem of quickly transmission and timely processing are a urgent need to resolve now is carried out to these video informations.
The content of the invention
A kind of in view of the above-mentioned problems, the present invention is intended to provide car-mounted terminal real-time road early warning system.
The purpose of the present invention is realized using following technical scheme:
A kind of car-mounted terminal real-time road early warning system, it is characterized in that, including camera, image processing module, solution pressing mold
Block, cloud computing platform and terminal device, the camera are used to capture current traffic information, obtain current road conditions image;It is described
Image processing module is used to be compressed road conditions image, store, and by wireless network by the road conditions image transmitting after processing
Decompression operations are carried out to decompression module;Described cloud computing platform is used to carry out integrated management to the road conditions image that decompression obtains;
The terminal device is used to provide the user current traffic information, facilitates user according to traffic information travel route planning.
Beneficial effects of the present invention are:User can obtain the real-time road condition information of car-mounted terminal by this system, according to
Traffic information changes path in time, reduces traffic jam, and terminal device connects with the prior-warning device on car, when front occurs
During congestion, also actively early warning, prompting changing route can be sent to driver.
Brief description of the drawings
Using accompanying drawing, the invention will be further described, but the embodiment in accompanying drawing does not form any limit to the present invention
System, for one of ordinary skill in the art, on the premise of not paying creative work, can also be obtained according to the following drawings
Other accompanying drawings.
Fig. 1 is the frame construction drawing of the present invention.
Reference:
Camera 1;Image processing module 2;Decompression module 3;Cloud computing platform 4;Terminal device 5;Receiver 6;Centre
Manage unit 7;Memory cell 8;Wireless communication unit 9.
Embodiment
With reference to following application scenarios, the invention will be further described.
Referring to Fig. 1, a kind of car-mounted terminal real-time road early warning system of the present embodiment, it is characterized in that, including camera 1,
Image processing module 2, decompression module 3, cloud computing platform 4 and terminal device 5, the camera 1 are used to capture current road conditions letter
Breath, obtains current road conditions image;Described image processing module 2 is used to be compressed road conditions image, store, and passes through wireless network
Road conditions image transmitting after processing to decompression module 3 is carried out decompression operations by network;Described cloud computing platform 4 is used for decompressing
To road conditions image analyzed and processed, obtain current road condition data;The terminal device 5 is used for according to cloud computing platform 4
The road condition data of offer provides the user current traffic information, facilitates user according to traffic information travel route planning.
Preferably, a camera 1 is installed every 50 kilometers, the camera 1 is used for capturing road conditions image, and will on road
Capture obtained road conditions image and receiver 6 is sent to by wireless network.
Preferably, referring to Fig. 1, receiver 6 that described image processing module 2 includes being sequentially connected, CPU 7,
Memory cell 8, wireless communication unit 9.The receiver 6 is used to receive the road conditions image captured from camera;The center
The one side of processing unit 7 is used to carry out denoising to road conditions image, the random noise in road conditions image is removed, after obtaining denoising
Road conditions image, on the other hand be used for the road conditions image after denoising is compressed;The memory cell 8 is used for after storing compression
Road conditions image;The wireless communication unit 9 is used for the road conditions image transmitting after compression to decompression module.
Preferably, discrete cosine transform (DCT) is to carry out discrete cosine transform to image, and image is changed to from transform of spatial domain
Frequency domain, then the conversion coefficient in frequency domain is handled, then carry out inverse transformation and image is transformed into spatial domain from frequency domain, so as to reach
To the purpose for removing picture noise.
Preferably, non-lower sampling contourlet conversion passes through non-lower using pyramid (NSP) and the filter of non-lower sampling direction
Ripple device group (NSDFB) is completed, and NSP carried out accordingly except that can remove the down-sampling in LP decomposable processes, and to wave filter
Up-sampling, using binary channels non-lower sampling wave filter, complete similar multi-resolution decomposition;NSDFB is adopted by removing in DFB
Sample process, and corresponding up-sampling is carried out to wave filter to form, direction wave filter group has been carried out appropriate up-sampling it
Afterwards, it is possible to allow anisotropic filter preferably partly to cover on the band logical passband of tower wave filter, frequency alias can be overcome
Phenomenon, possess translation invariance.
Preferably, the CPU 7 is used to remove the random noise in road conditions image, and to the road conditions after denoising
Image carries out partitioning pretreatment, obtains image block, and carries out sparse transformation to image block, obtains a series of road conditions subgraph
Block, it is specially:
1) pretreatment early stage is carried out to the road conditions image of collection, removes the disturbing factor in road conditions image, obtain one newly
Road conditions image, wherein road conditions image size is N × N, using discrete cosine transform (DCT) method carry out early stage locate in advance
Reason;
2) with pixel x in new road conditions imageaCentered on set a size be R × R square area as inspection
Rope scope, using selecting pixel x in function pair range of searchbCarry out it is preselected, as F (Na,Nb) it is more than selected threshold tau
(xa), then the pixel is pixel xaSimilar pixel point, travel through range of search all pixels point, obtain pixel xaPhase
Like set Ua, its select function for:
Wherein, Na、NbRespectively with pixel xaAnd NbCentered on, size is M × M image block;μa、σaRespectively Na's
The average and standard deviation of gray value, μb、σbFor NbGray value average and standard deviation, σijFor NaAnd NbGray value association side
Difference, α, β and γ are regulatory factors, and meet alpha+beta+γ=1, C1、C2And C3For modifying factor, it is primarily to ensure selection
Function is significant, σFiAnd σFjRespectively NaAnd NbFrequency coefficient gray value standard deviation, σFijFor block of pixels NaAnd block of pixels
NbFrequency coefficient covariance;
3) similar set U is calculatedaInterior all similitude xaWeights size, and all pixels point in range of search is carried out
Weighted average, obtain pixel x to be estimatedaDenoising estimate, its weight computing formula is:
Pixel xaThe calculation formula of denoising estimate be:
Wherein,For block of pixels NaWith block of pixels NbGauss weighted euclidean distance, κ is road conditions figure
The standard deviation of the Gaussian kernel of picture, h are filter factor, and Ω is with pixel xaCentered on range of search, w (xa,xb) it is weights,
By with xaAnd xbCentered on, size is M × M image block NaAnd NbBetween similarity determine, and 0w (xa,xb)≤1, y (xa)
For pixel xaDenoising estimate, y (xb) it is range of search pixel xbDenoising estimate;
4) pixel in all new road conditions images is traveled through, show that the denoising of all pixels point in new road conditions image is estimated
Evaluation, the gray value of corresponding pixel points in new road conditions image is replaced with the denoising estimate being calculated, it is secondary so as to obtain
Pending road conditions image X after denoising;
5) pending road conditions image is carried out waiting big segmentation, obtains one group of subimage blockIts
In, xjThe column vector form of j-th of subimage block is represented, N is the size of road conditions image, and S is the size of subimage block, is adopted afterwards
Sparse transformation is carried out to subimage block with non-lower sampling contourlet conversion, obtains a series of road conditions subimage block
Beneficial effect:Denoising twice is first carried out to the road conditions image that camera 1 collects, can effectively remove road
Random noise in condition image, the identification of image after denoising is improved, meanwhile, in second denoising, weighed using function pair is selected
Weight coefficient is modified, and adaptively can filter out random noise.This reduces number for follow-up perception compression processing image
According to treating capacity, compression speed is improved, saves memory headroom.Converted using non-lower sampling contourlet to pending road
Condition image carries out sparse processing, and the principal character of road conditions image can be described with minimum information, saves memory headroom, improves
PDR.
Preferably, the CPU 7 is additionally operable to be observed projection to road conditions subimage block, obtains road conditions subgraph
As the sampled value of block, it is specially:
1) below equation group is used, produces one group of sequenceIts self-defined equation group is:
2) from sequenceIn equidistant one group of number successively, form a new sequence, the length of its new sequence is
S2;
3) new sequence is subjected to permutation and combination according to row order, obtains the projection matrix Φ that a size is S × SB, will project
Matrix is normalized, and obtains a new projection matrix ΦB';
4) Φ is usedB' is compressed processing to road conditions subimage block, obtains the sampled value of road conditions subimage block, is specially:
yj=ΦB' xj′
Wherein, yjFor the sampled value of j-th of road conditions subimage block;xj' it is j-th of road conditions subimage block.
Beneficial effect:Sampling observation projection is directly carried out to road conditions subimage block, this processing method is not only in storage
The calculation matrix φ of whole road conditions image need not be stored, and projection is observed to road conditions subimage block, greatly reduces fortune
Memory space is calculated, accelerates the compression speed of road conditions image;Need not wait for after whole sparse road conditions image is all measured again
Perform the encoding operation, each road conditions subimage block can individually be handled, it is ensured that real-time.
Preferably, the decompression module 3 is used to be iterated reconstruct to the sampled value of road conditions subimage block, obtains road conditions
The iterative value of image block, the iterative value for merging all road conditions subimage blocks can obtain complete reconstructed image, be specially:
1) utilizeThe primary iteration value of j-th of subimage block in road conditions image is obtained, wherein,
Represent the primary iteration value of j-th of subimage block, ΦB'TFor ΦB' transposed matrix, yjFor adopting for j-th road conditions subimage block
Sample value;
2) road conditions subimage block is filtered, obtainedWherein i is iterative algebra;To filter j-th obtained
The filtering image block of road conditions subimage block;
3) useProjection formula pairProjection calculating is carried out, is obtained
Wherein,For iteration j-th obtained of image block is projected afterwards i times;ΦB'TFor the projection matrix Φ of image blockB''s
Transposed matrix, yjFor the sampled value of j-th of road conditions subimage block;
4) projection is obtained using sparse matrix ψSparse transformation is carried out, and is obtained using function pair sparse transformation is screened
To conversion coefficient screened, if meet | w | >=τ, a new conversion coefficient is calculated using threshold function table, otherwise
By conversion coefficient zero setting, wherein τ is self-defined screening value, is finally obtained again through sparse inverse transformationSelf-defined screening function
For:
Wherein, sparse matrix ψ is a selected matrix,To screen function, w is conversion coefficient, wpFor conversion coefficient w
Paternal number, τ for convergence control coefrficient, f be conversion coefficient w estimate, σwFor variation coefficient w 3 × 3 neighborhood estimates
Edge mean square deviation,For the estimated value of j-th of the image block obtained through sparse inverse transformation;
5) iteration function pair is utilizedProcessing is iterated, is obtainedIterative valueIts iteration function defined
For:
6) residual error D is calculated(i)If | D(i+1)-D(i)|10-4, then exportOtherwise, step 2 is jumped to, is iterated to calculate,
Until meet | D(i+1)-D(i)|10-4, export iterative valueWherein, the calculation formula of residual error is:
7) sampled value of all image blocks is traveled through, and is iterated reconstruct, the iterative value of all image blocks is obtained, utilizes
The iterative value arrived realizes the reconstruct to view picture road conditions image, you can obtains the reconstructed image of road conditions image.
Claims (6)
1. a kind of car-mounted terminal real-time road early warning system, it is characterized in that, including camera, image processing module, decompression module,
Cloud computing platform and terminal device, the camera are used to capture current road conditions image;Described image processing module be used for pair
Road conditions image is compressed, stored, and the road conditions image transmitting after compression is handled by wireless network to decompression module is carried out
Decompression operations;The cloud computing platform is used to analyze and process the road conditions image that decompression obtains, and obtains current road conditions number
According to;The terminal device is used to provide the user current traffic information according to the road condition data that cloud computing platform provides, convenient
User is according to traffic information travel route planning.
2. a kind of car-mounted terminal real-time road early warning system according to claim 1, it is characterized in that, described image processing mould
Block includes receiver, CPU, memory cell and wireless communication unit, and the receiver, which is used to receive, comes from camera
The road conditions image of candid photograph;On the one hand the CPU is used to carry out denoising to road conditions image, remove road conditions image
In random noise, obtain the road conditions image after denoising, on the other hand be used for the road conditions image after denoising is compressed;It is described
Memory cell is used to store the road conditions image after compression;The wireless communication unit is used to arrive the road conditions image transmitting after compression
Decompression module.
3. a kind of car-mounted terminal real-time road early warning system according to claim 2, it is characterized in that, terminal device also with car
On early warning system be connected, when traffic jam occurs, early warning can be sent, remind user to adjust route, again path planning.
4. a kind of car-mounted terminal real-time road early warning system according to claim 3, it is characterized in that, the central processing list
Member is used to remove the random noise in road conditions image, and carries out partitioning pretreatment to the road conditions image after denoising, obtains image block,
Sparse transformation is carried out to image block afterwards, obtains a series of road conditions subimage block, is specially:
1) pretreatment early stage is carried out to the road conditions image of collection, removes the disturbing factor in road conditions image, obtain a new road
Condition image, wherein road conditions image size are N × N, and pretreatment early stage is carried out using discrete cosine transform (DCT) method;
2) with pixel x in new road conditions imageaCentered on set a size be R × R square area as retrieval model
Enclose, using selecting pixel x in function pair range of searchbCarry out it is preselected, as F (Na,Nb) it is more than selected threshold tau (xa), then
The pixel is pixel xaSimilar pixel point, travel through range of search all pixels point, obtain pixel xaSimilar set
Ua, its select function for:
<mrow>
<mi>F</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>N</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>N</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msup>
<mrow>
<mo>&lsqb;</mo>
<mfrac>
<mrow>
<msqrt>
<mrow>
<mn>2</mn>
<msub>
<mi>&mu;</mi>
<mi>a</mi>
</msub>
<msub>
<mi>&mu;</mi>
<mi>b</mi>
</msub>
</mrow>
</msqrt>
<mo>+</mo>
<msub>
<mi>C</mi>
<mn>1</mn>
</msub>
</mrow>
<mrow>
<msub>
<mi>&mu;</mi>
<mi>a</mi>
</msub>
<mo>+</mo>
<msub>
<mi>&mu;</mi>
<mi>b</mi>
</msub>
<mo>+</mo>
<msub>
<mi>C</mi>
<mn>1</mn>
</msub>
</mrow>
</mfrac>
<mo>&rsqb;</mo>
</mrow>
<mi>&alpha;</mi>
</msup>
<msup>
<mrow>
<mo>&lsqb;</mo>
<mfrac>
<mrow>
<msqrt>
<mrow>
<mn>2</mn>
<msub>
<mi>&sigma;</mi>
<mi>a</mi>
</msub>
<msub>
<mi>&sigma;</mi>
<mi>b</mi>
</msub>
</mrow>
</msqrt>
<mo>+</mo>
<msub>
<mi>C</mi>
<mn>2</mn>
</msub>
</mrow>
<mrow>
<msub>
<mi>&sigma;</mi>
<mi>a</mi>
</msub>
<mo>+</mo>
<msub>
<mi>&sigma;</mi>
<mi>b</mi>
</msub>
<mo>+</mo>
<msub>
<mi>C</mi>
<mn>2</mn>
</msub>
</mrow>
</mfrac>
<mo>&rsqb;</mo>
</mrow>
<mi>&beta;</mi>
</msup>
<msup>
<mrow>
<mo>&lsqb;</mo>
<mfrac>
<mrow>
<msqrt>
<mrow>
<mn>2</mn>
<msub>
<mi>&sigma;</mi>
<mrow>
<mi>F</mi>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
</mrow>
</msqrt>
<mo>+</mo>
<msub>
<mi>C</mi>
<mn>3</mn>
</msub>
</mrow>
<mrow>
<msub>
<mi>&sigma;</mi>
<mrow>
<mi>F</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>+</mo>
<msub>
<mi>&sigma;</mi>
<mrow>
<mi>F</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>+</mo>
<msub>
<mi>C</mi>
<mn>3</mn>
</msub>
</mrow>
</mfrac>
<mo>&rsqb;</mo>
</mrow>
<mi>&gamma;</mi>
</msup>
<mo>;</mo>
</mrow>
Wherein, Na、NbRespectively with pixel xaAnd NbCentered on, size is M × M image block;μa、σaRespectively NaGray scale
The average and standard deviation of value, μb、σbFor NbGray value average and standard deviation, σijFor NaAnd NbGray value covariance, α,
β and γ is regulatory factor, and meets alpha+beta+γ=1, C1、C2And C3For modifying factor, it is primarily to ensure that selection function has
Meaning, σFiAnd σFjRespectively NaAnd NbFrequency coefficient gray value standard deviation, σFijFor block of pixels NaWith block of pixels NbFrequency
The covariance of domain coefficient;
3) similar set U is calculatedaInterior all similitude xaWeights size, all pixels point in range of search is added afterwards
Weight average, obtain pixel x to be estimatedaDenoising estimate, its weight computing formula is:
<mrow>
<mi>w</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>x</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>F</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>N</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>N</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
<mo><</mo>
<mi>&tau;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>a</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>exp</mi>
<mrow>
<mo>(</mo>
<mo>-</mo>
<mfrac>
<mrow>
<mo>|</mo>
<mo>|</mo>
<mi>v</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>N</mi>
<mi>a</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mi>v</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>N</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>|</mo>
<msubsup>
<mo>|</mo>
<mrow>
<mn>2</mn>
<mo>,</mo>
<mi>&kappa;</mi>
</mrow>
<mn>2</mn>
</msubsup>
</mrow>
<msup>
<mi>h</mi>
<mn>2</mn>
</msup>
</mfrac>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>F</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>N</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>N</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>F</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>N</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>N</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<mi>&tau;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>a</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Pixel xaThe calculation formula of denoising estimate be:
<mrow>
<mi>y</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>a</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mrow>
<msub>
<mi>&Sigma;</mi>
<msub>
<mi>x</mi>
<mi>b</mi>
</msub>
</msub>
<mi>w</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>x</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<munder>
<mo>&Sigma;</mo>
<mrow>
<msub>
<mi>x</mi>
<mi>b</mi>
</msub>
<mo>&Subset;</mo>
<mi>&Omega;</mi>
</mrow>
</munder>
<mi>w</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>a</mi>
</msub>
<mo>,</mo>
<msub>
<mi>x</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>y</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>b</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
Wherein,For block of pixels NaWith block of pixels NbGauss weighted euclidean distance, κ is road conditions image
The standard deviation of Gaussian kernel, h are filter factor, and Ω is with pixel xaCentered on range of search, w (xa,xb) be weights, by with
xaAnd xbCentered on, size is M × M image block NaAnd NbBetween similarity determine, and 0≤w (xa,xb)≤1;y(xa) be
Pixel xaDenoising estimate, y (xb) it is range of search pixel xbDenoising estimate;
4) pixel in all new road conditions images is traveled through, draws the denoising estimation of all pixels point in new road conditions image
Value, the gray value of corresponding pixel points in new road conditions image is replaced with the denoising estimate being calculated, so as to obtain secondary go
Pending road conditions image X after making an uproar;
5) pending road conditions image is carried out waiting big segmentation, obtains one group of subimage blockWherein, xjRepresent
The column vector form of j-th of subimage block, N are the size of road conditions image, and S is the size of subimage block, afterwards using non-lower sampling
Contourlet conversion carries out sparse transformation to subimage block, obtains a series of road conditions subimage block
5. a kind of car-mounted terminal real-time road early warning system according to claim 4, it is characterized in that, the central processing list
Member is additionally operable to be observed projection to road conditions subimage block, obtains the sampled value of road conditions subimage block, is specially:
1) below equation group is used, produces one group of sequenceIts self-defined equation group is:
2) from sequenceIn equidistant one group of number successively, form a new sequence, the length of its new sequence is S2;
3) new sequence is subjected to permutation and combination according to row order, obtains the projection matrix Φ that a size is S × SB, by projection matrix
It is normalized, obtains a new projection matrix ΦB';
4) Φ is usedB' is compressed processing to road conditions subimage block, obtains the sampled value of road conditions subimage block, is specially:
yj=ΦB' xj′
Wherein, yjFor the sampled value of j-th of road conditions subimage block;xj' it is j-th of road conditions subimage block.
A kind of 6. car-mounted terminal real-time road early warning system according to claim 5, it is characterized in that the decompression module is used
Reconstruct is iterated in the sampled value to road conditions subimage block, obtains the iterative value of road conditions subimage block, merges all road conditions
The iterative value of image block can obtain complete reconstructed image, be specially:
1) utilizeThe primary iteration value of j-th of subimage block in road conditions image is obtained, wherein,Represent
The primary iteration value of j-th of subimage block, ΦB'TFor ΦB' transposed matrix, yjFor the sampled value of j-th of road conditions subimage block;
2) road conditions subimage block is filtered, obtainedWherein i is iterative algebra;To filter j-th obtained of road conditions
The filtering image block of subimage block;
3) useProjection formula pairProjection calculating is carried out, is obtained
Wherein,For iteration j-th obtained of image block is projected afterwards i times;ΦB'TFor the projection matrix Φ of image blockB' transposition
Matrix, yjFor the sampled value of j-th of road conditions subimage block;
4) projection is obtained using sparse matrix ψSparse transformation is carried out, and using screening what function pair sparse transformation obtained
Conversion coefficient is screened, if met | w | >=τ, a new conversion coefficient is calculated using threshold function table, otherwise will become
Coefficient zero setting is changed, wherein τ is self-defined screening value, is finally obtained again through sparse inverse transformationIt is self-defined screening function be:
<mrow>
<mo>&dtri;</mo>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<mi>w</mi>
<msqrt>
<mrow>
<msup>
<mi>w</mi>
<mn>2</mn>
</msup>
<mo>+</mo>
<msubsup>
<mi>w</mi>
<mi>p</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
<mo>-</mo>
<mi>&tau;</mi>
<mfrac>
<mrow>
<msqrt>
<mn>3</mn>
</msqrt>
<msup>
<mi>f</mi>
<mn>2</mn>
</msup>
</mrow>
<msub>
<mi>&sigma;</mi>
<mi>w</mi>
</msub>
</mfrac>
<mi>w</mi>
</mrow>
<msqrt>
<mrow>
<msup>
<mi>w</mi>
<mn>2</mn>
</msup>
<mo>+</mo>
<msubsup>
<mi>w</mi>
<mi>p</mi>
<mn>2</mn>
</msubsup>
</mrow>
</msqrt>
</mfrac>
</mtd>
<mtd>
<mrow>
<mrow>
<mo>|</mo>
<mi>w</mi>
<mo>|</mo>
</mrow>
<mo>&GreaterEqual;</mo>
<mi>&tau;</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>o</mi>
<mi>t</mi>
<mi>h</mi>
<mi>e</mi>
<mi>r</mi>
<mi>s</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
Wherein, sparse matrix ψ is a selected matrix,To screen function, w is conversion coefficient, wpFor conversion coefficient w father
Coefficient, τ for convergence control coefrficient, f be conversion coefficient w estimate, σwFor the edge of variation coefficient w 3 × 3 neighborhood estimates
Mean square deviation,For the estimated value of j-th of the image block obtained through sparse inverse transformation;
5) iteration function pair is utilizedProcessing is iterated, is obtainedIterative valueIts define iteration function be:
6) residual error D is calculated(i)If | D(i+1)-D(i)| < 10-4, then exportOtherwise, step 2 is jumped to, is iterated to calculate, directly
To satisfaction | D(i+1)-D(i)| < 10-4, export iterative valueWherein, the calculation formula of residual error is:
<mrow>
<msup>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<msqrt>
<mi>N</mi>
</msqrt>
</mfrac>
<mo>|</mo>
<mo>|</mo>
<msup>
<mi>X</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>-</mo>
<msup>
<mover>
<mover>
<mi>X</mi>
<mo>^</mo>
</mover>
<mo>^</mo>
</mover>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
</msup>
<mo>|</mo>
<msub>
<mo>|</mo>
<mn>2</mn>
</msub>
</mrow>
7) sampled value of all image blocks is traveled through, and is iterated reconstruct, the iterative value of all image blocks is obtained, utilizes what is obtained
Iterative value realizes the reconstruct to view picture road conditions image, you can obtains the reconstructed image of road conditions image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711081209.5A CN107886751B (en) | 2017-11-07 | 2017-11-07 | Real-time road condition early warning system of vehicle-mounted terminal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711081209.5A CN107886751B (en) | 2017-11-07 | 2017-11-07 | Real-time road condition early warning system of vehicle-mounted terminal |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107886751A true CN107886751A (en) | 2018-04-06 |
CN107886751B CN107886751B (en) | 2020-12-25 |
Family
ID=61779068
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711081209.5A Active CN107886751B (en) | 2017-11-07 | 2017-11-07 | Real-time road condition early warning system of vehicle-mounted terminal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107886751B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107692392A (en) * | 2017-11-07 | 2018-02-16 | 韦彩霞 | A kind of safety helment warning device for site safety operation |
CN108846372A (en) * | 2018-06-27 | 2018-11-20 | 肖鑫茹 | A kind of intelligent transportation system |
CN109017557A (en) * | 2018-06-20 | 2018-12-18 | 上海科世达-华阳汽车电器有限公司 | A kind of vehicle road condition detection system and autonomous driving vehicle |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101344991A (en) * | 2008-09-03 | 2009-01-14 | 华为技术有限公司 | Method, device and system for providing road information |
CN103489312A (en) * | 2013-09-22 | 2014-01-01 | 江苏大学 | Traffic flow information collection method based on image compression |
US8666180B2 (en) * | 2009-12-04 | 2014-03-04 | Stc.Unm | System and methods of compressed sensing as applied to computer graphics and computer imaging |
CN106448266A (en) * | 2016-10-27 | 2017-02-22 | 深圳市元征软件开发有限公司 | Vehicle driving warning method, vehicle driving warning device and vehicle driving warning system |
-
2017
- 2017-11-07 CN CN201711081209.5A patent/CN107886751B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101344991A (en) * | 2008-09-03 | 2009-01-14 | 华为技术有限公司 | Method, device and system for providing road information |
US8666180B2 (en) * | 2009-12-04 | 2014-03-04 | Stc.Unm | System and methods of compressed sensing as applied to computer graphics and computer imaging |
CN103489312A (en) * | 2013-09-22 | 2014-01-01 | 江苏大学 | Traffic flow information collection method based on image compression |
CN106448266A (en) * | 2016-10-27 | 2017-02-22 | 深圳市元征软件开发有限公司 | Vehicle driving warning method, vehicle driving warning device and vehicle driving warning system |
Non-Patent Citations (1)
Title |
---|
沈萍萍,余勤: "基于离散余弦变换的非局部均值图像去噪算法", 《计算机工程与设计》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107692392A (en) * | 2017-11-07 | 2018-02-16 | 韦彩霞 | A kind of safety helment warning device for site safety operation |
CN107692392B (en) * | 2017-11-07 | 2019-12-10 | 南京万和消防科技有限公司 | A safety helmet warning device for building site safety work |
CN109017557A (en) * | 2018-06-20 | 2018-12-18 | 上海科世达-华阳汽车电器有限公司 | A kind of vehicle road condition detection system and autonomous driving vehicle |
CN108846372A (en) * | 2018-06-27 | 2018-11-20 | 肖鑫茹 | A kind of intelligent transportation system |
Also Published As
Publication number | Publication date |
---|---|
CN107886751B (en) | 2020-12-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116342596B (en) | YOLOv5 improved substation equipment nut defect identification detection method | |
CN107886751A (en) | A kind of car-mounted terminal real-time road early warning system | |
CN113936256A (en) | Image target detection method, device, equipment and storage medium | |
CN110120024A (en) | Method, apparatus, equipment and the storage medium of image procossing | |
CN112287912A (en) | Deep learning-based lane line detection method and device | |
CN110766039B (en) | Muck truck transportation state identification method, medium, equipment and muck truck | |
CN104754340A (en) | Reconnaissance image compression method for unmanned aerial vehicle | |
CN111382808A (en) | Vehicle detection processing method and device | |
CN111738114B (en) | Vehicle target detection method based on anchor-free accurate sampling remote sensing image | |
CN112307853A (en) | Detection method of aerial image, storage medium and electronic device | |
CN116030074A (en) | Identification method, re-identification method and related equipment for road diseases | |
CN108132054A (en) | For generating the method and apparatus of information | |
CN111611918A (en) | Traffic flow data set acquisition and construction method based on aerial photography data and deep learning | |
CN110490822A (en) | The method and apparatus that image removes motion blur | |
CN103337175A (en) | Vehicle type recognition system based on real-time video steam | |
CN116665092A (en) | Method and system for identifying sewage suspended matters based on IA-YOLOV7 | |
CN115601236A (en) | Remote sensing image super-resolution reconstruction method based on characteristic information distillation network | |
CN117557780A (en) | Target detection algorithm for airborne multi-mode learning | |
CN115761552B (en) | Target detection method, device and medium for unmanned aerial vehicle carrying platform | |
CN111542006A (en) | Object identification method based on wireless radio frequency signal | |
CN116563748A (en) | Height measuring method and system for high-rise construction building | |
CN114821651B (en) | Pedestrian re-recognition method, system, equipment and computer readable storage medium | |
Singh et al. | Critical Electrical Infrastructure Segmentation in Arctic Conditions | |
CN110796003A (en) | Lane line detection method and device and electronic equipment | |
CN105260992A (en) | Traffic image denoising algorithm based on robust principal component decomposition and feature space reconstruction |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20201119 Address after: B603, 6th floor, building 1, No. 36, North Third Ring Road East, Dongcheng District, Beijing 100010 Applicant after: Mushroom car Union Information Technology Co.,Ltd. Address before: No. 44, people's East Road, Yulin, the Guangxi Zhuang Autonomous Region, the Guangxi Zhuang Autonomous Region Applicant before: Gong Tuting |
|
GR01 | Patent grant | ||
GR01 | Patent grant |