CN105857180A - Hazy weather vehicle driving auxiliary system and method - Google Patents

Hazy weather vehicle driving auxiliary system and method Download PDF

Info

Publication number
CN105857180A
CN105857180A CN201610298881.9A CN201610298881A CN105857180A CN 105857180 A CN105857180 A CN 105857180A CN 201610298881 A CN201610298881 A CN 201610298881A CN 105857180 A CN105857180 A CN 105857180A
Authority
CN
China
Prior art keywords
image
mist elimination
processing module
module
vehicle
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201610298881.9A
Other languages
Chinese (zh)
Other versions
CN105857180B (en
Inventor
许皓
李刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangxi University
Original Assignee
Guangxi University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangxi University filed Critical Guangxi University
Priority to CN201610298881.9A priority Critical patent/CN105857180B/en
Publication of CN105857180A publication Critical patent/CN105857180A/en
Application granted granted Critical
Publication of CN105857180B publication Critical patent/CN105857180B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R1/00—Optical viewing arrangements; Real-time viewing arrangements for drivers or passengers using optical image capturing systems, e.g. cameras or video systems specially adapted for use in or on vehicles
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
    • B60Q9/00—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling
    • B60Q9/008—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling for anti-collision purposes
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
    • B60R2300/30—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of image processing
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
    • B60R2300/80—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement
    • B60R2300/8053—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement for bad weather conditions or night vision

Abstract

The invention discloses a hazy weather vehicle driving auxiliary system and method, and relates to the field of vehicle auxiliary driving. The system comprises an image collecting device, an image processing module, a display module and a warning module which are installed on a vehicle; the image collecting device sends collected video information of the position in front of the vehicle to the image processing module, and defogging processing is conducted on all frames of original images in the video information by the image processing module to obtain defogging images of corresponding frames; the display module installed on a center console of the vehicle receives and displays the defogging images; the image processing module conducts filtering processing on the corresponding defogging image of each frame and conducts differential processing on the defogging images of every two adjacent frames, and coming vehicle moving position information in the defogging images is obtained through analysis; the image processing module starts the warning module according to the moving position information. According to the system and method, a driver can be assisted in clearly seeing the environments in front of and behind the vehicle, and the functions of dynamically tracking a coming vehicle and assisting or warning the driver to drive properly are achieved.

Description

Haze sky vehicle drive assist system and method thereof
Technical field
The present invention relates to vehicle assistant drive field, particularly the present invention relates to by the dark algorithm of field of machine vision and based on The dynamic tracking algorithm of Kalman filtering applies in the problem of solution vehicle security drive, belongs to innovation based on image procossing fortune Use technical field.
Background technology
Haze problem has become as a serious environmental problem of the puzzlement society common people, and it not only affects the healthy of the masses, Simultaneously because its low visibility, the safety to car steering produces deep effect.Under haze weather the highest due to visibility, Hindering occurs in driver's sight line, and causes vehicle accident to take place frequently.This kind of device of the most urgent requirement can car assisted driver be observed Situation outside car, to make correct driving judgement, the generation of minimizing accident.
Summary of the invention
The goal of the invention of the present invention is, for the problems referred to above, it is provided that a kind of haze sky vehicle drive assist system, by figure and Algorithm process knows that the information of sending a car at vehicle front or rear, auxiliary human pilot see the environment before and after the vehicle under haze clearly, and Having the light that dynamic tracking is sent a car, auxiliary or alerting drivers are correctly driven, and improve the safety driven.
For reaching above-mentioned purpose, the technical solution adopted in the present invention is: haze sky vehicle drive assist system, including being arranged on Image collecting device, image processing module, display module and alarm module on vehicle, described image collecting device will collect The video information of vehicle front be sent in image processing module, described image processing module is to the every frame in described video information Original image carries out mist elimination process and obtains the mist elimination image of corresponding frame；The described display module being arranged on vehicle centre console receives And show described mist elimination image；Every frame mist elimination image is filtered processing by described image processing module, and going adjacent two frames Mist image carries out difference processing, analyzes the mobile location information obtaining sending a car in mist elimination image；Described image processing module is according to institute Stating mobile location information and start described alarm module, described alarm module sends corresponding auxiliary or warning information to driver.
When being traffic accident accident occurred frequently in haze sky, in haze sky, the observable field range of driver is the least, and car light Irradiation distance also shortens because of weather impact, and such driver finds to send a car or often only have the shortest response time when barrier. Technique scheme assist driver can accomplish in haze sky by building the DAS (Driver Assistant System) of a vehicle drive Safe driving.By the video image of image acquisition device to vehicle front in this system, and entered by image processing module Row mist elimination processes, and so can obtain video image the most clearly, to assist the situation in driver's clear view front；Meanwhile, Image processing module can also change according to the pixel of frame video image every in video image, analyzes to obtain whether there is phase in image Vehicle to motion, and remind driver evade accordingly or carry out corresponding operating accordingly.Here aid system be based on The inventive application of image procossing, carries out mist elimination process including based on algorithm to the image having haze, to obtain image clearly, Reference is observed for driver；Meanwhile, continue through the image after mist elimination is filtered, difference processing, and therefrom analyze Vehicle location and then remind driver to make corresponding action by voice or other modes.Compared to existing technology, system is simplified Circuit, has broken away from that sensor in traditional aid system is many, system complex, installation is complicated, function is the most not enough, not only provides The visual field is used for assisting driver clearly, but also has the autotracking function sending a car monitoring range, it is possible to more quickly find Send a car, remind driver to make a response earlier, thus the generation of minimizing accident.
Further, the every frame original image in described video information is gone by described image processing module based on dark algorithm Mist processes.Video image can quickly be processed by dark algorithm, improves the clear figure of video image, arrives and remove haze mist elimination Function.
Further, described dark algorithm steps is: 1) sweep the original image rectangular module of the L × L being odd number Retouch, add up the grey scale pixel value in this module；2) extract the grey scale pixel value of minimum, then substitute with minimum pixel gray value Whole grey scale pixel values in this module；3) step 1 is repeated) and step 2), whole original image is scanned, just can obtain Dark channel image to a secondary dark；4) described original image is used to subtract each other with the pixel value of dark channel image one to one, Obtain the described mist elimination image after mist elimination.A kind of quickly dark algorithm is provided here, optimizes algorithm steps, it is possible to quickly Go haze to process original image, reflect the image information of vehicle front in real time, it is to avoid Computer Vision postpones.
Further, every frame mist elimination image is filtered processing by described image processing module based on Kalman filtering algorithm.Karr Graceful filtering algorithm can carry out real-time renewal and process to the data of collection in worksite, reduces the noise effect following the tracks of target, for rear Continuous process provides safeguard.
Further, described difference processing is that the grey scale pixel value of two width mist elimination images of consecutive frame is entered by described image processing module Row subtracts each other, and chooses and subtract each other the region of variation that rear pixel value difference is relatively large in former frame mist elimination image, and from described region of variation In thus analyze and obtain mist elimination image is sent a car mobile location information.The grey scale pixel value of the light part sent a car in the picture and dusk The grey scale pixel value of dark background is different, when opposite automobile is during motion, in the image of shooting each frame out, The position of light is all different, as long as position there occurs skew, front and back two two field pictures subtract each other and just can detect that part There is bigger change in pixel value, thus can detect whether opposite sends a car.
Further, described difference processing is that the pixel one_to_one corresponding of adjacent former frame mist elimination image is subtracted by described image processing module Go the pixel of a later frame mist elimination image, and in former frame mist elimination image, mark the variation zone more than 150 of the pixel value difference after subtracting each other Territory；And analyze the mobile location information obtaining sending a car in mist elimination image.The grey scale pixel value of general picture is from 0 to 255, Wherein white is 255, and black is 0.The light sent a car in terms of pixel grey scale tends to white, and background pixel gray scale here tends to Lycoperdon polymorphum Vitt or black, at this moment arrange both differences more than 150, it is possible to distinguish the light sent a car very well and move the difference brought, Other interference factors are got rid of, it is achieved the autotracking to vehicle location.
Further, described image processing module obtains meeting the mobile message of the light sent a car according to region of variation shape profile； And analyze the mobile location information obtaining sending a car in mist elimination image.In mist elimination image, mark satisfactory part, wherein send a car Light position change cause the part of mark to be generally falcate or circle, can obtain accordingly sending a car in mist elimination image Position.
Further, described image collector is set to high-definition camera, and be arranged on vehicle front windshield and rear seat windscreen it Under console on.Sending a car on direction before and after vehicle is monitored, it is possible to preferably auxiliary driver drives.
Further, described alarm module is LED display module or voice reminder module.LED display module can be by setting The LED putting different colours carries out different situation warnings；Voice reminder module arranges multiple voice information, for different Driving situation, sends different voice messages.
Based on technique scheme, a kind of haze sky vehicle drive assisting method, follow the steps below,
1) start this system, the video information of the vehicle front that described image acquisition device arrives in haze sky, and be sent to figure As in processing module；
2) described image processing module carries out mist elimination process and obtains going of corresponding frame the every frame original image in described video information Mist image, and described mist elimination image is sent in the display module on vehicle centre console；
3) every frame mist elimination image is filtered processing by described image processing module, and the image of adjacent two frames is carried out difference processing, Analyze and obtain image comes the mobile location information of car light；
4) described image processing module starts described alarm module according to described mobile location information, and described alarm module is to driver Send corresponding auxiliary or warning information.
Can improve, by aid system, the safety coefficient that vehicle is driven in haze sky, this is quite a lot of one monitor the visual field remote, The passenger side sensitive, accurate, assists driver understand in advance in haze sky or tackle emergency situations, especially because the visual field not foot straps React various situations not in time.Changing householder method can apply on various vehicles, the suitability is strong；Without vehicle is carried out Transformation, removes and installs detection equipment；So greatly reduce the cost of use.
Owing to using technique scheme, the method have the advantages that
The present invention compared with prior art, simplifies hardware circuit, solves the driving environment monitoring problem of complexity；Supervise to external world Devices such as using the most traditional sensor has been broken away from survey field, turns to use photographic head and the algorithm of image procossing, so can make The precision of the environment monitoring the external world under haze weather is higher, and most image processing work is all complete by processor chips Becoming, hardware circuit is the simplest, has the most also simplified the burden that automobile is overall.Meanwhile, the height of dynamic tracking algorithm Sensitive property is easier to find to sending a car than human eye, thus reminds driver to make a response earlier, and there have safe driving to be the best Place.
Accompanying drawing explanation
Fig. 1 is operation principle flow chart of the present invention.
Fig. 2 is difference processing computing schematic diagram of the present invention.
Detailed description of the invention
The present invention includes four parts: image collecting device, image processing module, display module and alarm module.
Image capture module is vehicular high-definition camera, and photographic head is arranged on the console under the forward and backward windshield of automobile, It is used for shooting automobile forward-and-rearward situation.The video image of shooting is sent in image processing module by image capture module.
Image processing module is image processing module based on FPGA, processes image capture module and shoots the haze image come, logical Cross dark algorithm the road image that there is haze is carried out process to obtain mist elimination image clearly, be then output to vehicle-carrying display screen Middle for driver's reference.When sending a car in opposite, image processing module is sent a car by Kalman filtering algorithm dynamic tracking, analyzes Obtain the relative position information sent a car.
Display module is vehicle-carrying display screen, is arranged on console.Display module shows after image processing module mist elimination processes Video image.
Alarm module is vehicle-mounted voice equipment, based on the image processing module analysis result to relative position of sending a car, defeated to driver Go out different warnings or the sound bite of auxiliary driving.Alarm module can also be LED warning lamp simultaneously, in being arranged on before automobile On control platform, it is used for reminding driver.
Here operation logic and mode to system are described further: the video of vehicle front that image collecting device will collect Information is sent in image processing module, and image processing module carries out mist elimination process to the every frame original image in described video information And obtain the mist elimination image of corresponding frame.The described display module being arranged on vehicle centre console receives and shows described mist elimination image. Every frame mist elimination image is filtered processing by described image processing module, and the mist elimination image of adjacent two frames is carried out difference processing, Analyze the mobile location information obtaining sending a car in mist elimination image.Image processing module starts described police according to described mobile location information Showing module, alarm module sends corresponding auxiliary or warning information to driver.By image acquisition device to vehicle in system The video image in front, and carry out mist elimination process by image processing module, so can obtain video image the most clearly, To assist the situation in driver's clear view front；Meanwhile, image processing module can also be according to frame video figure every in video image The pixel change of picture, analyzes and obtains whether existing in image the vehicle of relative motion, and reminds driver to advise accordingly accordingly Keep away or carry out corresponding operating.Here aid system is inventive application based on image procossing, including based on algorithm to unclear The image having haze carry out mist elimination process, to obtain image clearly, observe reference for driver；Meanwhile, continue through Image after mist elimination is filtered, difference processing, i.e. can be therefrom to analyze vehicle location and then by voice or other modes Driver is reminded to make corresponding action.
As it is shown in figure 1, when driver is at haze sky starting automobile, first have to the load conditions before and after observation automobile, that The most first to open vehicle-mounted front camera and rear camera, photographic head is placed on below the windshield of automobile, is then aligned with forward-and-rearward road surface. After photographic head is filmed the broad image before and after automobile, it is sent to image processing module, image processing module prestores Image semantic classification program based on dark and dynamic tracking algorithm based on Kalman filtering, when the picture sent is through dark After channel image pretreatment, reach road conditions image clearly, export back in the display screen in driver's cabin, at this moment driver Just can see load conditions the most clearly.After image is apparent from, find whether front has through dynamic tracking algorithm again Motor vehicles is to sailing, if be detected that motor line approaches, the LED warning lamp in driver's cabin will be lighted, and driver sees Light to LED and just have watchful effect, thus carefully handle vehicle starting, thus the generation avoided traffic accident.
Certainly, image capture module can also be to install the position recorder on vehicle or other video equipments here.Display mould The video display equipment that block can be mounted on console or facilitate driver to watch, such as, be erected at the large-size screen monitors intelligence of console Mobile phone；Alarm module can be combined with display module, or is combined with vehicle-mounted stereo set.Image processing module can be base Processing module unit in single-chip microcomputer, FPGA or DSP etc., naturally it is also possible to be vehicle-mounted computer equipment.
In above-mentioned, image processing module carries out mist elimination process based on dark algorithm to the every frame original image in described video information. Video image can quickly be processed by dark algorithm, improves the clear figure of video image, arrives the function removing haze mist elimination.
In computer vision and computer graphics, can approximate abstract for " having mist ", camera bat for more fuzzy image The effect gone out is similar between camera and material object across one layer of thick fog.
Narasimhan etc. propose (the Monochrome atmospheric of the atmospherical scattering model under " haze " according to theory analysis scattering model):
I (x)=A ρ (x) e-βd(x)+A(1-e-βd(x)) (1)
In formula (1), for space coordinates, represent sky brightness (Skylight), for scene albedo, for the scene depth of field, for Atmospheric scattering coefficient.
In fuzzy outdoor picture, at least some pixel of a passage has relatively low pixel value, and this rule is applicable to Remove any region outside sky.That is, in fuzzy image, in certain local area, always there are some the darkest Composition.Thus release the definition of image dark channel:
J d a r k ( x ) = m i n c ∈ { r , g , b } ( m i n y ∈ Ω ( x ) ( J c ( y ) ) ) - - - ( 2 )
Some Color Channel representing image in formula (2).Represent a rectangular area centered by pixel, rectangular area Corresponding size is.Represent the minima of pixel, the namely dark of image in this rectangular area, also known as above-mentioned system Meter rule is the preferential rule of dark of image.
Here dark algorithm steps is:
1) the original image rectangular module of the L × L being odd number is scanned, adds up the grey scale pixel value in this module.
2) extract the grey scale pixel value of minimum, then substitute the whole grey scale pixel values in this module with minimum pixel gray value.
3) step 1 is repeated) and step 2), whole original image is scanned, just can obtain the dark figure of a secondary dark Picture.
4) use described original image to subtract each other with the pixel value of dark channel image one to one, obtain the described mist elimination figure after mist elimination Picture.A kind of quickly dark algorithm is provided here, optimizes algorithm steps, it is possible to quickly go haze to process original image, Reflect the image information of vehicle front in real time, it is to avoid Computer Vision postpones.
Here it is to be understood that the dynamic tracking algorithm cardinal principle of Kalman filtering algorithm or Kalman filtering is to pass through Kalman The video image of shooting is filtered by filtering algorithm, obtains new video image, then uses frame difference method to isolate motion mesh Mark, thus realize dynamic tracking.Kalman filtering algorithm can carry out real-time renewal and process to the data of collection in worksite, subtracts Follow the tracks of the noise effect of target less, provide safeguard for subsequent treatment.
In above-mentioned, difference processing is that the grey scale pixel value of two width mist elimination images of consecutive frame is subtracted each other by described image processing module, And choose former frame mist elimination image subtracted each other the region of variation that rear pixel value difference is relatively large, and from described region of variation thus point Analyse the mobile location information that obtains sending a car in mist elimination image.
The grey scale pixel value of the light part sent a car in the picture and the grey scale pixel value of dim background are different, when opposite vapour Car is during motion, and in the image of shooting each frame out, the position of light is all different, as long as position occurs Skew, front and back two two field pictures subtract each other the change that just can detect that the pixel value generation of that part is bigger, and thus to detect right Whether face sends a car.
The image of adjacent two frames is carried out difference, and sets a threshold value, if the result of difference more than the pixel of threshold value is exactly Moving target.As in figure 2 it is shown, for the video sequence shot time static, first to fk(x, y) frame picture and fk+1(x, y) frame Picture carries out smoothing denoising, then does frame difference method process, i.e. uses fk(x, y) frame picture deducts fk+1(x, y) frame picture, obtain two Value image.It is formulated as:
D k ( x , y ) = 1 | f k ( x , y ) - f k - 1 ( x , y ) | ≥ T 0 | f k ( x , y ) - f k - 1 ( x , y ) | ≤ T
T is the threshold value set, Dk(x, y) binary image obtained for detection.
Particularly, T is 150, and the pixel one_to_one corresponding of adjacent former frame mist elimination image is deducted a later frame by image processing module The pixel of mist image, and in former frame mist elimination image mark subtract each other after pixel value difference more than 150 region of variation be 1, its His zone marker is 0.Image processing module obtains meeting the mobile message of the light sent a car according to region of variation shape profile；And Analyze the mobile location information obtaining sending a car in mist elimination image.The grey scale pixel value of general picture is from 0 to 255, Qi Zhongbai Color is 255, and black is 0.The light sent a car in terms of pixel grey scale tend to white, and background pixel gray scale here tend to Lycoperdon polymorphum Vitt or Black, at this moment arranges both differences more than 150, it is possible to distinguish the light sent a car very well and move the difference brought, by other Interference factor is got rid of, it is achieved the autotracking to vehicle location.The district that in mist elimination image, the light change location sent a car of mark causes Territory, generally falcate or circle, can obtain the position sending a car in mist elimination image accordingly.And analysis obtains mist elimination image In the mobile location information sent a car.
Based on technique scheme, also disclosed herein is a kind of preferred embodiment: image collecting device includes being arranged in vehicle Drive recorder, the area of visual field at drive recorder respectively corresponding vehicle front and rear is arranged, and is used for shooting vehicle front-and rear-view Frequently information send video information to image processing module, image processing module is to be arranged on vehicle centre console by installing rack Intelligence large-screen mobile phone, mobile phone is provided with for the every frame original image stating in video information is carried out mist elimination process based on helping secretly Road algorithm and every frame mist elimination image is filtered process Kalman filtering algorithm, the mist elimination image of adjacent two frames is carried out difference The track algorithm processed.Mist elimination image is shown by mobile phone by mobile phone screen, observes road conditions video the most clearly for driver Image, i.e. display module are mobile phone screen.Simultaneously mobile phone continue through the image after mist elimination is filtered, difference processing, from In analyze vehicle location and remind driver to make corresponding action by the speech ciphering equipment of mobile phone.Here by by smart mobile phone Processing module, screen, speech ciphering equipment as image processing module, display module and alarm module, i.e. use existing hardware i.e. Whole aid system can be realized: have only to here the respective algorithms of image processing module in the application is installed on mobile phone, and even Connect vehicle-mounted video monitoring device, both can complete the installation of the application aid system；Its installation is the simplest, application has more suitable The property used；Can significantly simplify the burden that automobile is overall, efficiently improve drive safety simultaneously.
Based on technique scheme, base open a kind of haze sky vehicle drive assisting method, follows the steps below here,
1) start this system, the video information of the vehicle front that described image acquisition device arrives in haze sky, and be sent to figure As in processing module.
2) described image processing module carries out mist elimination process and obtains going of corresponding frame the every frame original image in described video information Mist image, and described mist elimination image is sent in the display module on vehicle centre console.
3) every frame mist elimination image is filtered processing by described image processing module, and the image of adjacent two frames is carried out difference processing, Analyze and obtain image comes the mobile location information of car light.
4) described image processing module starts described alarm module according to described mobile location information, and described alarm module is to driver Send corresponding auxiliary or warning information.
Can improve, by aid system, the safety coefficient that vehicle is driven in haze sky, this is quite a lot of one monitor the visual field remote, The passenger side sensitive, accurate, assists driver understand in advance in haze sky or tackle emergency situations, especially because the visual field not foot straps React various situations not in time.Changing householder method can apply on various vehicles, the suitability is strong；Without vehicle is carried out Transformation, removes and installs detection equipment；So greatly reduce the cost of use.
Compared with prior art, the invention provides what a kind of haze weather automobile guide instrument based on machine vision auxiliary was reached Provide the benefit that: the present invention makes every effort to go to solve complicated driving environment monitoring problem, relative to existing skill with the hardware circuit simplified Art, the present invention has broken away from devices such as using the most traditional sensor, has turned to use photographic head and image in the field of monitoring to external world The algorithm processed, so can make the precision monitoring the environment in the external world under haze weather higher, and most image procossing work Being all to be completed by processor chips, hardware circuit is the simplest, has the most also simplified the burden that automobile is overall.With Time, the hypersensitivity of dynamic tracking algorithm can be easier to find to sending a car than human eye, thus reminds driver to make a response earlier, Safe driving is had great benefit.
Described above is the detailed description for the preferable possible embodiments of the present invention, but embodiment is not limited to the special of the present invention Profit application range, the equal change completed under the technical spirit suggested by all present invention or modification change, the present invention all should be belonged to Contained the scope of the claims.

Claims (10)

1. haze sky vehicle drive assist system, including the image collecting device being arranged on vehicle, image processing module, display Module and alarm module, it is characterised in that: the video information of the vehicle front collected is sent to figure by described image collecting device As in processing module, described image processing module the every frame original image in described video information is carried out mist elimination process and obtain right Answer the mist elimination image of frame；The described display module being arranged on vehicle centre console receives and shows described mist elimination image；Described image Every frame mist elimination image is filtered processing by processing module, and the mist elimination image of adjacent two frames is carried out difference processing, and analysis obtains The mobile location information sent a car in mist elimination image；Described image processing module starts described warning mould according to described mobile location information Block, described alarm module sends corresponding auxiliary or warning information to driver.
Haze sky the most according to claim 1 vehicle drive assist system, it is characterised in that: described image processing module base In dark algorithm, the every frame original image in described video information is carried out mist elimination process.
Haze sky the most according to claim 2 vehicle drive assist system, it is characterised in that described dark algorithm steps For: 1) the original image rectangular module of the L × L being odd number is scanned, add up the grey scale pixel value in this module；2) Extract the grey scale pixel value of minimum, then substitute the whole grey scale pixel values in this module with minimum pixel gray value；3) repeat Step 1) and step 2), whole original image is scanned, obtains dark channel image；4) described original image and is used The pixel value of the dark channel image of one correspondence subtracts each other, and obtains the described mist elimination image after mist elimination.
Haze sky the most according to claim 1 vehicle drive assist system, it is characterised in that: described image processing module base It is filtered processing to every frame mist elimination image in Kalman filtering algorithm.
Haze sky the most according to claim 1 vehicle drive assist system, it is characterised in that: described difference processing is described The grey scale pixel value of two width mist elimination images of consecutive frame is subtracted each other by image processing module, and chooses phase in former frame mist elimination image Subtract the region of variation that rear pixel value difference is relatively large, and the movement thus analysis obtains sending a car in mist elimination image from described region of variation Positional information.
6. the haze sky vehicle drive assist system will not stated according to claim 5, it is characterised in that: described difference processing is The pixel one_to_one corresponding of adjacent former frame mist elimination image is deducted the pixel of a later frame mist elimination image by described image processing module, and The pixel value difference after mark subtracts each other in the former frame mist elimination image region of variation more than 150；And analysis obtains sending a car in mist elimination image Mobile location information.
Haze sky the most according to claim 6 vehicle drive assist system, it is characterised in that: described image processing module root Obtain meeting the mobile message of the light sent a car according to region of variation shape profile；And analyze the mobile position obtaining sending a car in mist elimination image Confidence ceases.
Haze sky the most according to claim 1 vehicle drive assist system, it is characterised in that: described image collector is set to High-definition camera, and be arranged on the console under vehicle front windshield and rear seat windscreen.
Haze sky the most according to claim 1 vehicle drive assist system, it is characterised in that: described alarm module is LED Display module or voice reminder module.
Haze sky the most according to claim 1 vehicle drive assisting method, it is characterised in that: follow the steps below,
1) start this system, the video information of the vehicle front that described image acquisition device arrives in haze sky, and be sent to figure As in processing module；
2) described image processing module carries out mist elimination process and obtains going of corresponding frame the every frame original image in described video information Mist image, and described mist elimination image is sent in the display module on vehicle centre console；
3) every frame mist elimination image is filtered processing by described image processing module, and the image of adjacent two frames is carried out difference processing, Analyze and obtain image comes the mobile location information of car light；
4) described image processing module starts described alarm module according to described mobile location information, and described alarm module is to driver Send corresponding auxiliary or warning information.
CN201610298881.9A 2016-05-09 2016-05-09 Haze sky vehicle drive assist system and its method Active CN105857180B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610298881.9A CN105857180B (en) 2016-05-09 2016-05-09 Haze sky vehicle drive assist system and its method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610298881.9A CN105857180B (en) 2016-05-09 2016-05-09 Haze sky vehicle drive assist system and its method

Publications (2)

Publication Number Publication Date
CN105857180A true CN105857180A (en) 2016-08-17
CN105857180B CN105857180B (en) 2018-06-01

Family

ID=56630484

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610298881.9A Active CN105857180B (en) 2016-05-09 2016-05-09 Haze sky vehicle drive assist system and its method

Country Status (1)

Country Link
CN (1) CN105857180B (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106971627A (en) * 2017-05-22 2017-07-21 佛山市神风航空科技有限公司 A kind of avoidance prior-warning device and determination methods
CN108515909A (en) * 2018-04-04 2018-09-11 京东方科技集团股份有限公司 A kind of automobile head-up-display system and its barrier prompt method
CN108528339A (en) * 2017-03-01 2018-09-14 京东方科技集团股份有限公司 A kind of display system and its display methods and vehicle
CN108594818A (en) * 2018-04-27 2018-09-28 深圳市商汤科技有限公司 Intelligent driving control method, intelligent vehicle-carried equipment and system
CN109747540A (en) * 2017-11-07 2019-05-14 意法半导体股份有限公司 The method of integrated driving person's auxiliary system, corresponding system, circuit, external member
CN110341595A (en) * 2019-06-26 2019-10-18 南京信息职业技术学院 A kind of door control system of automobile
CN111199169A (en) * 2018-11-16 2020-05-26 北京微播视界科技有限公司 Image processing method and device
CN111277729A (en) * 2020-02-26 2020-06-12 新疆大学 Video image processing method and device and electronic equipment
CN111428646A (en) * 2020-03-25 2020-07-17 东风汽车集团有限公司 Driver take-over guarantee method and system based on edge cloud computing
CN112140998A (en) * 2020-10-12 2020-12-29 华南师范大学 Vision enhancement defogging system for assisting motor vehicle driving

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5926164A (en) * 1988-09-30 1999-07-20 Aisin Seiki Kabushiki Kaisha Image processor
CN103065332A (en) * 2013-01-21 2013-04-24 信帧电子技术（北京）有限公司 Detection method and device for rapid movement behavior of pedestrian in foggy weather
CN103077501A (en) * 2012-12-30 2013-05-01 信帧电子技术（北京）有限公司 Road target marking method and device
CN103198458A (en) * 2013-04-10 2013-07-10 成都国腾电子技术股份有限公司 Haze image dark channel rapid computing method
CN105512623A (en) * 2015-12-02 2016-04-20 吉林大学 Foggy-day driving visual enhancement and visibility early warning system and method based on multiple sensors

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5926164A (en) * 1988-09-30 1999-07-20 Aisin Seiki Kabushiki Kaisha Image processor
CN103077501A (en) * 2012-12-30 2013-05-01 信帧电子技术（北京）有限公司 Road target marking method and device
CN103065332A (en) * 2013-01-21 2013-04-24 信帧电子技术（北京）有限公司 Detection method and device for rapid movement behavior of pedestrian in foggy weather
CN103198458A (en) * 2013-04-10 2013-07-10 成都国腾电子技术股份有限公司 Haze image dark channel rapid computing method
CN105512623A (en) * 2015-12-02 2016-04-20 吉林大学 Foggy-day driving visual enhancement and visibility early warning system and method based on multiple sensors

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
王莹: ""基于交通视频的车辆鲁棒检测和快速跟踪方法的研究"", 《工程科技II辑》 *
鱼朋波: ""高速公路机器视觉测速方法研究"", 《信息科技辑》 *

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108528339A (en) * 2017-03-01 2018-09-14 京东方科技集团股份有限公司 A kind of display system and its display methods and vehicle
CN106971627A (en) * 2017-05-22 2017-07-21 佛山市神风航空科技有限公司 A kind of avoidance prior-warning device and determination methods
CN109747540A (en) * 2017-11-07 2019-05-14 意法半导体股份有限公司 The method of integrated driving person's auxiliary system, corresponding system, circuit, external member
CN109747540B (en) * 2017-11-07 2023-02-21 意法半导体股份有限公司 Method for integrating a driver assistance system, corresponding system, circuit, kit
CN108515909B (en) * 2018-04-04 2021-04-20 京东方科技集团股份有限公司 Automobile head-up display system and obstacle prompting method thereof
CN108515909A (en) * 2018-04-04 2018-09-11 京东方科技集团股份有限公司 A kind of automobile head-up-display system and its barrier prompt method
WO2019192418A1 (en) * 2018-04-04 2019-10-10 京东方科技集团股份有限公司 Automobile head-up display system and obstacle prompting method thereof
US11087151B2 (en) 2018-04-04 2021-08-10 Boe Technology Group Co., Ltd. Automobile head-up display system and obstacle prompting method thereof
CN108594818A (en) * 2018-04-27 2018-09-28 深圳市商汤科技有限公司 Intelligent driving control method, intelligent vehicle-carried equipment and system
CN111199169A (en) * 2018-11-16 2020-05-26 北京微播视界科技有限公司 Image processing method and device
CN110341595A (en) * 2019-06-26 2019-10-18 南京信息职业技术学院 A kind of door control system of automobile
CN111277729A (en) * 2020-02-26 2020-06-12 新疆大学 Video image processing method and device and electronic equipment
CN111428646A (en) * 2020-03-25 2020-07-17 东风汽车集团有限公司 Driver take-over guarantee method and system based on edge cloud computing
CN111428646B (en) * 2020-03-25 2023-01-24 东风汽车集团有限公司 Driver take-over guarantee method and system based on edge cloud computing
CN112140998A (en) * 2020-10-12 2020-12-29 华南师范大学 Vision enhancement defogging system for assisting motor vehicle driving

Also Published As

Publication number Publication date
CN105857180B (en) 2018-06-01

Similar Documents

Publication Publication Date Title
CN105857180A (en) Hazy weather vehicle driving auxiliary system and method
CN1928494B (en) Driving support apparatus
JP5210994B2 (en) Image display device for vehicle
CN105459907B (en) Vehicle security system, blind area monitoring method and device
CN109636924B (en) Vehicle-mounted multi-mode augmented reality system based on real road condition information three-dimensional modeling
CN102371944B (en) Driver vision support system and vehicle including the system
EP3367365A1 (en) Parking support information display method and parking support device
WO2014109016A1 (en) Vehicle periphery display device
CN109109748A (en) A kind of pedestrian's identification early warning system for blind area on the right side of heavy motor truck
CN1834583A (en) Vehicle periphery monitoring apparatus
JP2010044561A (en) Monitoring device to be mounted on vehicle
CN112896159A (en) Driving safety early warning method and system
CN108482367A (en) A kind of method, apparatus and system driven based on intelligent back vision mirror auxiliary
CN111351474B (en) Vehicle moving target detection method, device and system
JP2008285105A (en) Information display device
CN114930264A (en) Method for remote-controlled driving of a motor vehicle comprising a remote-controlled operator, computer program product and remote-controlled driving system
DE102012200762A1 (en) Method for signaling traffic condition in environment of vehicle, involves recording surrounding area of vehicle using sensor, and indicating recognized sensitive object on display arranged in rear view mirror housing of vehicle
CN115489536B (en) Driving assistance method, system, equipment and readable storage medium
JP2009126493A (en) Obstacle detection device
JP2017212633A (en) Imaging apparatus, imaging display method and imaging display program
CN113276772B (en) Automobile electronic exterior rearview mirror system and control method
CN107399274B (en) Image superposition method
JP7057074B2 (en) On-board unit and driving support device
CN113246859B (en) Electronic rearview mirror with driving auxiliary system warning function
DE102019109491A1 (en) DATA PROCESSING DEVICE, MONITORING SYSTEM, WECKSYSTEM, DATA PROCESSING METHOD AND DATA PROCESSING PROGRAM

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