CN110458029A - Vehicle checking method and device in a kind of foggy environment - Google Patents
Vehicle checking method and device in a kind of foggy environment Download PDFInfo
- Publication number
- CN110458029A CN110458029A CN201910633933.7A CN201910633933A CN110458029A CN 110458029 A CN110458029 A CN 110458029A CN 201910633933 A CN201910633933 A CN 201910633933A CN 110458029 A CN110458029 A CN 110458029A
- Authority
- CN
- China
- Prior art keywords
- transmissivity
- image
- luminous intensity
- traffic
- dark areas
- 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
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 35
- 230000003044 adaptive effect Effects 0.000 claims abstract description 33
- 238000001514 detection method Methods 0.000 claims abstract description 32
- 239000003595 mist Substances 0.000 claims abstract description 23
- 230000008030 elimination Effects 0.000 claims abstract description 20
- 238000003379 elimination reaction Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 238000012876 topography Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000010287 polarization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000036186 satiety Effects 0.000 description 1
- 235000019627 satiety Nutrition 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to technical field of vehicle detection, and in particular to vehicle checking method and device in a kind of foggy environment, first acquisition traffic image, and gray processing processing is carried out to traffic image, using gray processing treated traffic image as gray level image；And then the maximum region of brightness value is obtained after being split to gray level image, using the average brightness value of all pixels point in the maximum region of brightness value as atmosphere luminous intensity, pixel in gray level image is divided into bright area and dark areas, and generate the adaptive transmissivity being made of the transmissivity of dark areas and the transmissivity of bright area, atmosphere luminous intensity and adaptive transmissivity are inputted into atmospherical scattering model, defogging processing is carried out to traffic image, using defogging treated traffic image as mist elimination image, finally by the vehicle image in algorithm of target detection identification mist elimination image, the present invention can improve defog effect, image after being avoided that defogging again is distorted, to improve the accuracy and speed of vehicle detection.
Description
Technical field
The present invention relates to technical field of vehicle detection, and in particular to vehicle checking method and dress in a kind of foggy environment
It sets.
Background technique
In recent years, due to environmental change, haze weather starts continually to occur.On the one hand, it in foggy environment, drives
Not only sight is unclear by the person of sailing, visual range is short, but also the row of front and back vehicle can not be correctly judged by automotive camera system
Situation is sailed, is easy to cause traffic accident；On the other hand, the reduction of atmospheric visibility also causes traffic monitoring camera real
When obtain effective condition of road surface, bring huge resistance to traffic guidance.Therefore, how to realize it is quick to traffic image,
Effective defogging has very important realistic meaning for vehicle driving, traffic monitoring and traffic guidance；How severe
Haze environment in realize that high quality defogging to image and the detection of the high-precision to vehicle are also one challenging
Business.
In conventional images processing and computer vision field, defogging algorithm can be divided into according to the quantity of input picture
Single image defogging and multiple image combine defogging.Multiple image combines defogging and mainly utilizes Same Scene not on the same day
The a series of images obtained under the conditions of gas finally obtains the image of sharpening by information such as analysis polarization, target depths.In
It obtains that this series of images is not only at high cost, difficulty is big in Same Scene, but also will receive the restriction of scene and weather condition, it is difficult
To be promoted in practice.
Currently, single image defogging makes substantial progress, above-mentioned Railway Project is not only avoided, and the image restored is clear
Clear Du Genggao.Single image defogging is the ill-posed problem for lacking constraint.Demisting process usually requires to introduce some hypothesis
Or prior information.The demisting and the details as much as possible in restoration scenario for realizing high quality are a challenging tasks.
Some advanced defogging algorithms can restore most of details of image, but the color that may result in image after defogging becomes different
Often.In addition, unreasonable transmission estimation also results in the satiety of certain regional areas in image and/or serious distortion.
In traffic scene, driver needs constantly to judge the condition of road surface in front during traveling.
In haze sky, due to running upon more mist on road, atmospheric visibility decline, this when, the judgment of driver also will
It decreases.Therefore, for the defogging of traffic image, defog effect is on the one hand improved, on the other hand then to be avoided as far as possible
There is serious distortion in image after mist, and the situation of this quick for driver, correct judgement road ahead has important meaning
Justice.
Summary of the invention
The purpose of the present invention is to provide the vehicle checking methods and device in a kind of foggy environment, can improve defogging effect
Fruit, and it is avoided that the image after defogging is distorted, to improve the accuracy and speed of vehicle detection.
To achieve the goals above, the present invention the following technical schemes are provided:
A kind of vehicle checking method in foggy environment, comprising:
Traffic image is acquired, and gray processing processing is carried out to the traffic image, by gray processing treated traffic image
As gray level image；
The maximum region of brightness value is obtained after being split to the gray level image, it will be in the maximum region of the brightness value
The average brightness value of all pixels point is as atmosphere luminous intensity；
Pixel in the gray level image is divided into bright area and dark areas, and generates the transmission by the dark areas
The adaptive transmissivity of the transmissivity of rate and bright area composition；
The atmosphere luminous intensity and adaptive transmissivity are inputted into atmospherical scattering model, defogging is carried out to the traffic image
Processing, using defogging treated traffic image as mist elimination image；
The vehicle image in the mist elimination image is identified by algorithm of target detection.
Further, it is described the gray level image is split after obtain the maximum region of brightness value, by the brightness value
The step of average brightness value of all pixels point is as atmosphere luminous intensity in maximum region, comprising:
The gray level image is averagely divided into 2 topographies, the atmosphere luminous intensity for comparing 2 topographies is big
It is small, choose the biggish topography of atmosphere luminous intensity and continue average segmentation and compare, divide by multiple averaging and relatively after obtain
The topography for obtaining atmosphere light maximum intensity, using the topography of the atmosphere light maximum intensity as the maximum area of brightness value
Domain；
The atmosphere light strength calculation formula of the topography are as follows:Wherein, A indicates atmosphere light intensity
Degree, I (v) indicate that gray level image, v ∈ { r, g, b }, r, g, b respectively indicate three Color Channels of red, green, blue, Ψ (v) expression office
Portion's image,Operation indicates to be averaged after the brightness value of all pixels point in topography Ψ (v) is summed
Value.
Further, the step of pixel by the traffic image is divided into bright area and dark areas, comprising:
The segmentation threshold k of gray level image is obtained using OTSU algorithm, using I (x) < k image-region as dark areas, is indicated
Region B is expressed as using I (x) >=k image-region as bright area for region D.
Further, described to generate the adaptive transmissivity being made of the transmissivity of the dark areas and the transmissivity of bright area
The step of, comprising:
It obtains estimating transmissivity according to atmospherical scattering model and the atmosphere luminous intensity；
Transmissivity of the transmissivity as dark areas is estimated using described, and according to transmissivity and the dark areas estimated
Transmissivity generates the transmissivity of bright area, using the transmissivity of the transmissivity of the dark areas and bright area as adaptive transmission
Rate.
It is further, described that the step of estimating transmissivity is obtained according to atmospherical scattering model and the atmosphere luminous intensity, comprising:
By the dark J of traffic image Jdark(x) it is defined asIn formula, Jc
(y) channel in tri- Color Channels of r, g, b of traffic image J is indicated, Ω (x) represents the side centered on pixel x
Shape region；
Wherein,The minimum Color Channel figure of traffic image J is sought in expression,It indicates to obtaining
Minimum Color Channel figure carries out mini-value filtering, Jdark→0；
It regard formula I (x)=J (x) t (x)+A (1-t (x)) as atmospherical scattering model, in formula, I (x) is traffic image
Luminous intensity, J (x) are the luminous intensity of gray level image, and t (x) is transmissivity；
Enable J (x)=Jdark(x) → 0 it, can be obtainedWherein, IcIt (y) is gray level image
Tri- Color Channels of r, g, b in a channel,As estimate transmissivity.
Further, described to estimate transmissivity of the transmissivity as dark areas for described, and according to it is described estimate transmissivity and
The transmissivity of the dark areas generates the transmissivity of bright area, using the transmissivity of the transmissivity of the dark areas and bright area as
The step of adaptive transmissivity, comprising:
Transmissivity of the transmissivity as dark areas is estimated using described, is enabledIt is then described dark
The transmissivity in region is t ' (x)=1-Idark(x), x ∈ D；
The transmissivity of bright area is set as a constant, is denoted asWherein, max [] expression is soughtMaximum value, ε is constant, then the transmissivity of the bright area are as follows:
Then the calculation formula of adaptive transmissivity is as follows:
Wherein, t (x) is adaptive transmissivity,
Further, the algorithm of target detection uses FasterR-CNN model.
A kind of vehicle detection apparatus in foggy environment, described device include: memory, processor and are stored in described
In memory and the computer program that can run on the processor, the processor execute the computer program and operate in
In the module of following device:
Traffic image processing module carries out gray processing processing for acquiring traffic image, and to the traffic image, will be grey
Degreeization treated traffic image is as gray level image；
Atmosphere luminous intensity generation module, for obtaining the maximum region of brightness value after being split to the gray level image,
Using the average brightness value of all pixels point in the maximum region of the brightness value as atmosphere luminous intensity；
Adaptive transmissivity generation module, for the pixel in the gray level image to be divided into bright area and dark space
Domain, and generate the adaptive transmissivity being made of the transmissivity of the dark areas and the transmissivity of the bright area；
Mist elimination image generation module, for the atmosphere luminous intensity and adaptive transmissivity to be inputted atmospherical scattering model,
Defogging processing is carried out to the traffic image, using defogging treated traffic image as mist elimination image；
Vehicle image identification module, for identifying the vehicle image in the mist elimination image by algorithm of target detection.
The beneficial effects of the present invention are: the present invention discloses the vehicle checking method and device in a kind of foggy environment, first
Traffic image is acquired, and gray processing processing is carried out to the traffic image, using gray processing treated traffic image as gray scale
Image；And then the maximum region of brightness value is obtained after being split to the gray level image, by the maximum region of the brightness value
The average brightness value of middle all pixels point as atmosphere luminous intensity, by the pixel in the gray level image be divided into bright area and
Dark areas, and the adaptive transmissivity being made of the transmissivity of the dark areas and the transmissivity of the bright area is generated, by institute
Atmosphere luminous intensity and adaptive transmissivity input atmospherical scattering model are stated, defogging processing is carried out to the traffic image, by defogging
Treated traffic image identifies the vehicle figure in the mist elimination image finally by algorithm of target detection as mist elimination image
Picture.The present invention can improve the image after defog effect and defogging and be distorted, to improve the essence of vehicle detection
Degree and speed.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow diagram of the vehicle checking method in a kind of foggy environment of the embodiment of the present invention；
Fig. 2 is the flow diagram of step of embodiment of the present invention S300；
Fig. 3 is the structural schematic diagram of the vehicle detection apparatus in a kind of foggy environment of the embodiment of the present invention.
Specific embodiment
Clear, complete description is carried out to technical solution of the present invention below in conjunction with attached drawing, it is clear that described implementation
Example is a part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill
Personnel are obtained without making creative work so other embodiments, belong to protection scope of the present invention.
With reference to Fig. 1, the embodiment of the present invention provides the vehicle checking method in a kind of foggy environment, comprising:
Step S100, traffic image is acquired, and gray processing processing is carried out to the traffic image, treated by gray processing
Traffic image is as gray level image.
Step S200, the maximum region of brightness value is obtained after being split to the gray level image, most by the brightness value
The average brightness value of all pixels point is as atmosphere luminous intensity in big region.
Step S300, the pixel in the gray level image is divided into bright area and dark areas, and generated by described dark
The adaptive transmissivity of the transmissivity of the transmissivity in region and bright area composition.
In the traffic scene of haze weather, since the position of automotive camera system and preventing road monitoring system is set, display
What is obtained on device is usually the foggy image comprising bright areas such as white construction, traffic guardrail and large area skies, at this time
Just it can not directly use the method based on dark channel prior to realize effective defogging to image, or pass through traditional vehicle detection
Device carries out accurate detection to the vehicle of traveling.It needs to distinguish bright area and dark areas for traffic image.
Step S400, the atmosphere luminous intensity and adaptive transmissivity are inputted into atmospherical scattering model, to the communication chart
As carrying out defogging processing, using defogging treated traffic image as mist elimination image.
Those of ordinary skill in the art understand that the atmospherical scattering model is the prior art.
Step S500, the vehicle image in the mist elimination image is identified by algorithm of target detection.
In the present embodiment, for the image in traffic scene, a kind of quickly and effectively image defogging side first proposed
Method, initial estimation of this method independent of depth information in scene, but image defogging is completed using only single image, no
The problems such as halation and color distortion after only effectively avoiding image defogging, efficiently solve the residual mist in edge and sky color
The problems such as distortion, and there is minimum time complexity, the image after can improving defog effect and defogging occurs
Distortion.And then the vehicle image in the mist elimination image is quickly identified by algorithm of target detection, further improve vehicle inspection
The accuracy and speed of survey has stronger robustness, can be applied to a variety of traffic scenes.
As advanced optimizing for the present embodiment, in order to preferably estimate the value of atmosphere luminous intensity, the step S200
Include:
The gray level image is averagely divided into 2 topographies, the atmosphere luminous intensity for comparing 2 topographies is big
It is small, choose the biggish topography of atmosphere luminous intensity and continue average segmentation and compare, averagely divide by n times and relatively after obtain
The topography of atmosphere light maximum intensity, using the topography of the atmosphere light maximum intensity as the maximum area of average brightness value
Domain；Wherein, [6,10] n ∈, preferred n=8.
The atmosphere light strength calculation formula of the topography are as follows:Wherein, A indicates atmosphere light intensity
Degree, I (v) indicate that gray level image, v ∈ { r, g, b }, r, g, b respectively indicate three Color Channels of red, green, blue, Ψ (v) expression office
Portion's image,Operation indicates to be averaged after the brightness value of all pixels point in topography Ψ (v) is summed.
As advanced optimizing for the present embodiment, the pixel in the traffic image is divided into the step S300
Bright area and dark areas include:
The segmentation threshold k of gray level image is obtained using OTSU algorithm, using I (x) < k image-region as dark areas, is indicated
Region B is expressed as using I (x) >=k image-region as bright area for region D.
With reference to Fig. 2, the transmission by the dark areas is generated as advanced optimizing for the present embodiment, in the step S300
The step of adaptive transmissivity of the transmissivity of rate and bright area composition, comprising:
Step S310, it obtains estimating transmissivity according to atmospherical scattering model and the atmosphere luminous intensity；
Step S320, transmissivity of the transmissivity as dark areas is estimated using described, and estimates transmissivity and institute according to described
The transmissivity for stating dark areas generates the transmissivity of bright area, using the transmissivity of the transmissivity of the dark areas and bright area as certainly
Adapt to transmissivity.
As advanced optimizing for the present embodiment, the step S310 includes:
By the dark J of traffic image Jdark(x) it is defined asIn formula,
Jc(y) channel in tri- Color Channels of r, g, b of traffic image J is indicated, Ω (x) is represented centered on pixel x
Square region.In one embodiment, the square region is the region that one piece of pixel is 15 × 15.
In above-mentioned formula,The minimum Color Channel figure of traffic image J is sought in expression,Indicate to
The minimum Color Channel figure arrived carries out mini-value filtering, according to dark primary priori concept, Jdark→0；
It regard formula I (x)=J (x) t (x)+A (1-t (x)) as atmospherical scattering model, in formula, I (x) is traffic image
Luminous intensity, J (x) are the luminous intensity of gray level image, and t (x) is transmissivity.The transmissivity is medium transmission function, the medium
The transmitted intensity that transmission function is used to describe acquisition objects in images accounts for the percentage of incident intensity, and the target of defogging is exactly
Atmosphere luminous intensity A and transmissivity t (x) are estimated, and restores the luminous intensity J of gray level image by the luminous intensity I (x) of traffic image
(x)。
Enable J (x)=Jdark(x) → 0, two are carried out later simultaneously divided by atmosphere luminous intensity on the both sides of atmospherical scattering model
Secondary minimum Value Operations, can be obtainedWherein, IcIt (y) is tri- face of r, g, b of gray level image
A channel in chrominance channel,As estimate transmissivity.
As advanced optimizing for the present embodiment, the step S320 includes:
Transmissivity of the transmissivity as dark areas is estimated using described, is enabledIt is then described dark
The transmissivity in region is t ' (x)=1-Idark(x), x ∈ D.
It, can be directly using estimating since the lower region of brightness meets dark channel prior theory in the present embodiment
The transmissivity of transmissivity formula calculating dark areas.Transmissivity formula is estimated since the transmissivity in the higher region of brightness is not available
It is calculated, therefore an easy calculation is provided by bond area B and is calculated, the transmissivity of this part is unified
It is set as a constant, the constant is related to region D.
The transmissivity of bright area is set as a constant, is denoted as ε max [t (u)], whereinAs guiding filtering
Device, max [] expression are soughtMaximum value, ε is constant, then the transmissivity of the bright area are as follows:
Then the calculation formula of adaptive transmissivity is as follows:
Wherein, t (x) is adaptive transmissivity,
The present embodiment estimates to refine transmissivity that adaptive repairs transmissivity by using wave filter
Just, the accuracy of image procossing is improved.
As advanced optimizing for the present embodiment, the algorithm of target detection uses FasterR-CNN model.
With reference to Fig. 3, the present embodiment also provides the vehicle detection apparatus in a kind of foggy environment, and described device includes: storage
Device, processor and storage in the memory and the computer program that can run on the processor, the processor
The computer program is executed to operate in the module of following device:
Traffic image processing module 100 carries out gray processing processing for acquiring traffic image, and to the traffic image,
Using gray processing treated traffic image as gray level image；
Atmosphere luminous intensity generation module 200, for obtaining the maximum area of brightness value after being split to the gray level image
Domain, using the average brightness value of all pixels point in the maximum region of the brightness value as atmosphere luminous intensity；
Adaptive transmissivity generation module 300, for the pixel in the gray level image to be divided into bright area and dark
Region, and generate the adaptive transmissivity being made of the transmissivity of the dark areas and the transmissivity of the bright area；
Mist elimination image generation module 400, for the atmosphere luminous intensity and adaptive transmissivity to be inputted atmospheric scattering mould
Type carries out defogging processing to the traffic image, using defogging treated traffic image as mist elimination image；
Vehicle image identification module 500, for identifying the vehicle image in the mist elimination image by algorithm of target detection.
Vehicle detection apparatus in the foggy environment can run on desktop PC, notebook, palm PC and
Cloud server etc. calculates in equipment.Vehicle detection apparatus in the foggy environment, the device that can be run may include, but not only
It is limited to, processor, memory.It will be understood by those skilled in the art that the example is only the vehicle detection in foggy environment
The example of device does not constitute the restriction to the vehicle detection apparatus in foggy environment, may include more more or less than example
Component, perhaps combining vehicle detection apparatus in certain components or different components, such as the foggy environment can be with
Including input-output equipment, network access equipment, bus etc..
Alleged processor can be central processing unit (Central-Processing-Unit, CPU), can also be it
His general processor, digital signal processor (Digital-Signal-Processor, DSP), specific integrated circuit
(Application-Specific-Integrated-Circuit, ASIC), ready-made programmable gate array (Field-
Programmable-Gate-Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng, the processor is the control centre of the vehicle detection apparatus running gear in the foggy environment, using various interfaces and
Vehicle detection apparatus in the entire foggy environment of connection can running gear various pieces.
The memory can be used for storing the computer program and/or module, and the processor is by operation or executes
Computer program in the memory and/or module are stored, and calls the data being stored in memory, described in realization
The various functions of vehicle detection apparatus in foggy environment.The memory can mainly include storing program area and storing data
Area, wherein storing program area can application program needed for storage program area, at least one function (such as sound-playing function,
Image player function etc.) etc.；Storage data area, which can be stored, uses created data (such as audio data, electricity according to mobile phone
Script for story-telling etc.) etc..In addition, memory may include high-speed random access memory, it can also include nonvolatile memory, such as
Hard disk, memory, plug-in type hard disk, intelligent memory card (Smart-Media-Card, SMC), secure digital (Secure-
Digital, SD) card, flash card (Flash-Card), at least one disk memory, flush memory device or other volatibility are solid
State memory device.
Although the description of the disclosure is quite detailed and especially several embodiments are described, it is not
Any of these details or embodiment or any specific embodiments are intended to be limited to, but should be considered as is by reference to appended
A possibility that claim provides broad sense in view of the prior art for these claims explanation, to effectively cover the disclosure
Preset range.In addition, the disclosure is described with inventor's foreseeable embodiment above, its purpose is to be provided with
Description, and those equivalent modifications that the disclosure can be still represented to the unsubstantiality change of the disclosure still unforeseen at present.
Claims (8)
1. the vehicle checking method in a kind of foggy environment characterized by comprising
Acquire traffic image, and gray processing processing carried out to the traffic image, using gray processing treated traffic image as
Gray level image；
The maximum region of brightness value is obtained after being split to the gray level image, will be owned in the maximum region of the brightness value
The average brightness value of pixel is as atmosphere luminous intensity；
Pixel in the gray level image is divided into bright area and dark areas, and generate by the dark areas transmissivity and
The adaptive transmissivity of the transmissivity composition of the bright area；
The atmosphere luminous intensity and adaptive transmissivity are inputted into atmospherical scattering model, the traffic image is carried out at defogging
Reason, using defogging treated traffic image as mist elimination image；
The vehicle image in the mist elimination image is identified by algorithm of target detection.
2. the vehicle checking method in a kind of foggy environment according to claim 1, which is characterized in that described to the ash
Degree image obtains the maximum region of brightness value after being split, and all pixels point in the maximum region of the brightness value is averaged
The step of brightness value is as atmosphere luminous intensity, comprising:
The gray level image is averagely divided into 2 topographies, compares the atmosphere luminous intensity size of 2 topographies,
The biggish topography of atmosphere luminous intensity is chosen to continue average segmentation and compare, divide by multiple averaging and relatively after obtain it is big
The maximum topography of gas luminous intensity, using the topography of the atmosphere light maximum intensity as the maximum region of brightness value；
The atmosphere light strength calculation formula of the topography are as follows:Wherein, A indicates atmosphere luminous intensity, I
(v) indicate that gray level image, v ∈ { r, g, b }, r, g, b respectively indicate three Color Channels of red, green, blue, Ψ (v) indicates Local map
Picture,Operation indicates to be averaged after the brightness value of all pixels point in topography Ψ (v) is summed.
3. the vehicle checking method in a kind of foggy environment according to claim 2, which is characterized in that described by the friendship
The step of pixel in logical image is divided into bright area and dark areas, comprising:
Area is expressed as using I (x) < k image-region as dark areas using the segmentation threshold k that OTSU algorithm obtains gray level image
Domain D is expressed as region B using I (x) >=k image-region as bright area.
4. the vehicle checking method in a kind of foggy environment according to claim 3, which is characterized in that the generation is by institute
The step of adaptive transmissivity that the transmissivity of the transmissivity and bright area of stating dark areas forms, comprising:
It obtains estimating transmissivity according to atmospherical scattering model and the atmosphere luminous intensity；
Transmissivity of the transmissivity as dark areas is estimated using described, and according to the transmission for estimating transmissivity and the dark areas
Rate generates the transmissivity of bright area, using the transmissivity of the transmissivity of the dark areas and bright area as adaptive transmissivity.
5. the vehicle checking method in a kind of foggy environment according to claim 4, which is characterized in that described according to atmosphere
Scattering model and the atmosphere luminous intensity obtain the step of estimating transmissivity, comprising:
By the dark J of traffic image Jdark(x) it is defined asIn formula, Jc(y) table
Show a channel in tri- Color Channels of r, g, b of traffic image J, Ω (x) represents the squared region centered on pixel x
Domain；
Wherein,The minimum Color Channel figure of traffic image J is sought in expression,It indicates to obtained minimum
Color Channel figure carries out mini-value filtering, Jdark→0；
It regard formula I (x)=J (x) t (x)+A (1-t (x)) as atmospherical scattering model, in formula, I (x) is the light intensity of traffic image
Degree, J (x) are the luminous intensity of gray level image, and t (x) is transmissivity；
Enable J (x)=Jdark(x) → 0 it, can be obtainedWherein, IcBe (y) r of gray level image,
G, a channel in tri- Color Channels of b,As estimate transmissivity.
6. the vehicle checking method in a kind of foggy environment according to claim 5, which is characterized in that it is described will be described pre-
Estimate transmissivity of the transmissivity as dark areas, and bright area is generated according to the transmissivity for estimating transmissivity and the dark areas
Transmissivity, using the transmissivity of the transmissivity of the dark areas and bright area as the step of adaptive transmissivity, comprising:
Transmissivity of the transmissivity as dark areas is estimated using described, is enabledThe then dark areas
Transmissivity be t ' (x)=1-Idark(x), x ∈ D；
The transmissivity of bright area is set as a constant, is denoted asWherein, max [] expression is soughtMost
Big value, ε is constant, then the transmissivity of the bright area are as follows:
X ∈ B, u ∈ D, then the calculation formula of adaptive transmissivity is as follows:Wherein, t (x) is adaptive transmissivity,ε∈(0,1)。
7. the vehicle checking method in a kind of foggy environment according to claim 6, which is characterized in that the target detection
Algorithm uses FasterR-CNN model.
8. the vehicle detection apparatus in a kind of foggy environment, which is characterized in that described device include: memory, processor and
The computer program that can be run in the memory and on the processor is stored, the processor executes the computer
Program operates in the module of following device:
Traffic image processing module carries out gray processing processing for acquiring traffic image, and to the traffic image, by gray processing
Treated traffic image is as gray level image；
Atmosphere luminous intensity generation module, for obtaining the maximum region of brightness value after being split to the gray level image, by institute
The average brightness value of all pixels point in the maximum region of brightness value is stated as atmosphere luminous intensity；
Adaptive transmissivity generation module, for the pixel in the gray level image to be divided into bright area and dark areas, and
Generate the adaptive transmissivity being made of the transmissivity of the dark areas and the transmissivity of the bright area；
Mist elimination image generation module, for the atmosphere luminous intensity and adaptive transmissivity to be inputted atmospherical scattering model, to institute
It states traffic image and carries out defogging processing, using defogging treated traffic image as mist elimination image；
Vehicle image identification module, for identifying the vehicle image in the mist elimination image by algorithm of target detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910633933.7A CN110458029B (en) | 2019-07-15 | 2019-07-15 | Vehicle detection method and device in foggy environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910633933.7A CN110458029B (en) | 2019-07-15 | 2019-07-15 | Vehicle detection method and device in foggy environment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110458029A true CN110458029A (en) | 2019-11-15 |
| CN110458029B CN110458029B (en) | 2022-12-20 |
Family
ID=68482849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910633933.7A Active CN110458029B (en) | 2019-07-15 | 2019-07-15 | Vehicle detection method and device in foggy environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110458029B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523493A (en) * | 2020-04-27 | 2020-08-11 | 东南数字经济发展研究院 | Target detection algorithm for foggy weather image |
| CN113076997A (en) * | 2021-03-31 | 2021-07-06 | 南昌欧菲光电技术有限公司 | Lens band fog identification method, camera module and terminal equipment |
| CN113554872A (en) * | 2021-07-19 | 2021-10-26 | 昭通亮风台信息科技有限公司 | Detection early warning method and system for traffic intersection and curve |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110188775A1 (en) * | 2010-02-01 | 2011-08-04 | Microsoft Corporation | Single Image Haze Removal Using Dark Channel Priors |
| CN105787904A (en) * | 2016-03-25 | 2016-07-20 | 桂林航天工业学院 | Adaptive global dark channel prior image dehazing method for bright area |
| CN107301623A (en) * | 2017-05-11 | 2017-10-27 | 北京理工大学珠海学院 | A kind of traffic image defogging method split based on dark and image and system |
| CN108830803A (en) * | 2018-05-17 | 2018-11-16 | 昆明理工大学 | A kind of traffic video image defogging optimization algorithm |
| CN109785262A (en) * | 2019-01-11 | 2019-05-21 | 闽江学院 | Image defogging method based on dark channel prior and adaptive histogram equalization |
-
2019
- 2019-07-15 CN CN201910633933.7A patent/CN110458029B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110188775A1 (en) * | 2010-02-01 | 2011-08-04 | Microsoft Corporation | Single Image Haze Removal Using Dark Channel Priors |
| CN105787904A (en) * | 2016-03-25 | 2016-07-20 | 桂林航天工业学院 | Adaptive global dark channel prior image dehazing method for bright area |
| CN107301623A (en) * | 2017-05-11 | 2017-10-27 | 北京理工大学珠海学院 | A kind of traffic image defogging method split based on dark and image and system |
| CN108830803A (en) * | 2018-05-17 | 2018-11-16 | 昆明理工大学 | A kind of traffic video image defogging optimization algorithm |
| CN109785262A (en) * | 2019-01-11 | 2019-05-21 | 闽江学院 | Image defogging method based on dark channel prior and adaptive histogram equalization |
Non-Patent Citations (1)
| Title |
|---|
| 邱东芳等: "透射率和大气光自适应估计的暗通道去雾", 《计算机应用》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111523493A (en) * | 2020-04-27 | 2020-08-11 | 东南数字经济发展研究院 | Target detection algorithm for foggy weather image |
| CN113076997A (en) * | 2021-03-31 | 2021-07-06 | 南昌欧菲光电技术有限公司 | Lens band fog identification method, camera module and terminal equipment |
| CN113076997B (en) * | 2021-03-31 | 2023-01-03 | 南昌欧菲光电技术有限公司 | Lens band fog identification method, camera module and terminal equipment |
| CN113554872A (en) * | 2021-07-19 | 2021-10-26 | 昭通亮风台信息科技有限公司 | Detection early warning method and system for traffic intersection and curve |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110458029B (en) | 2022-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107301623B (en) | Traffic image defogging method and system based on dark channel and image segmentation | |
| Galdran et al. | Enhanced variational image dehazing | |
| CN109919879B (en) | Image defogging method based on dark channel prior and bright channel prior | |
| CN106846263B (en) | Based on the image defogging method for merging channel and sky being immunized | |
| CN106548461B (en) | Image defogging method | |
| CN110458029A (en) | Vehicle checking method and device in a kind of foggy environment | |
| CN108765342A (en) | A kind of underwater image restoration method based on improvement dark | |
| CN110211067B (en) | Defogging method for UUV visible light image on offshore surface | |
| CN106780380A (en) | A kind of image defogging method and system | |
| CN103985091A (en) | Single image defogging method based on luminance dark priori method and bilateral filtering | |
| CN102768760A (en) | Quick image dehazing method on basis of image textures | |
| CN105913390B (en) | A kind of image defogging method and system | |
| CN106127715A (en) | A kind of image defogging method and system | |
| CN102831591A (en) | Gaussian filter-based real-time defogging method for single image | |
| CN104182943B (en) | A kind of single image defogging method capable merging human-eye visual characteristic | |
| CN105139347A (en) | Polarized image defogging method combined with dark channel prior principle | |
| CN104504703A (en) | Welding spot color image segmentation method based on chip element SMT (surface mounting technology) | |
| CN103020921A (en) | Single image defogging method based on local statistical information | |
| CN107085830B (en) | Single image defogging method based on propagation filtering | |
| CN105023246B (en) | A kind of image enchancing method based on contrast and structural similarity | |
| CN109118440A (en) | Single image to the fog method based on transmissivity fusion with the estimation of adaptive atmosphere light | |
| CN105913391B (en) | A kind of defogging method can be changed Morphological Reconstruction based on shape | |
| CN108629750A (en) | A kind of night defogging method, terminal device and storage medium | |
| CN106780362B (en) | Road video defogging method based on dichromatic reflection model and bilateral filtering | |
| CN106504216B (en) | Single image to the fog method based on Variation Model |
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 |