CN205644979U - Dizzy system of traffic signals red light dim light based on embedded system - Google Patents
Dizzy system of traffic signals red light dim light based on embedded system Download PDFInfo
- Publication number
- CN205644979U CN205644979U CN201521064349.8U CN201521064349U CN205644979U CN 205644979 U CN205644979 U CN 205644979U CN 201521064349 U CN201521064349 U CN 201521064349U CN 205644979 U CN205644979 U CN 205644979U
- Authority
- CN
- China
- Prior art keywords
- embedded system
- red light
- processing unit
- image
- red
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Processing Of Color Television Signals (AREA)
- Image Processing (AREA)
Abstract
The utility model discloses a dizzy system of traffic signals red light dim light based on embedded system, including high definition digital camera, embedded system image processing unit and computer, high definition digital camera and computer link, embedded system image processing unit and computer link, embedded system image processing unit in to gained image the light halo around the red light handle, the red light halo that this scheme can produce the traffic red light in the video that obtains and image carries out area detection and effectively weakens filtering even, has improved the image quality in control field, carries out the preliminary treatment of image for methods such as follow -up signal lamp state judgement.
Description
Technical field
This utility model relates to the Intelligent traffic video monitoring image recognition in field and image processing techniques, particularly relates to a kind of traffic signal red light dim light based on embedded system and swoons system.
Background technology
In field of video monitoring, the application of high-definition digital video camera is more and more extensive, and the picture quality of high-definition digital video camera also becomes more and more important.Owing to sensor, camera lens etc. affect, when the signal lights in monitoring scene lights, the surrounding of lamp can scatter the light that a circle is close with signal lamp color, imaging the most just defines halation, such as when the red light in signal lights lights, the surrounding of red light there will be a circle red pixel, and these halation can cover the information around signal lights, and it is unclear that worse situation also can make the shape of signal lights thicken.The halation phenomenon that traffic light produce, if can be inhibited or remove completely, can not only effectively promote picture quality, and can help to the follow-up image processing algorithms such as the judgement of signal lights state.
The halation size that signal lights presents under different depth of exposures is the most different, when exposing less, halation around signal lights can not exist or the least, when exposure slowly increases, halation is also with becoming big, when exposing the biggest, around signal lights, itself is probably due to cross quick-fried and present white, is around halation the most entirely.So the extrinsic factor causing halation to vary in size is mainly the exposure size of camera, exposing the biggest, halation also can be the biggest.The difference of depth of exposure will make red light present different states, is often divided into normal overexposure, moderate overexposure, three kinds of situations of serious overexposure.It is redness that normal condition red light phase shows as red light region, has faint halation phenomenon, and this faint halation allows not remove；It is the most yellow that moderate overexposure mainly shows as red light Regional Red, it is divided into two kinds of situations: situation one is alternate for red light region reddish yellow, external margin has faint red halation, and in now requiring red light region, redness can not be removed as halation by mistake, and the faint halation in edge allows not remove；Situation two is complete yellow for red light region, and external margin has relatively strong red halation, is coated with red effect for not affecting later stage red light, and now halation must weaken and removes the most completely.Serious overexposure mainly shows as red light region and turns white, and there is relatively strong red halation at edge, and now red halation needs to remove completely.But the technology of present stage the most well utilizes the difference of overexposure degree even to remove to carry out effectively weakening of traffic red light halation.
Utility model content
For above-mentioned technological deficiency, the utility model proposes a kind of traffic signal red light dim light based on embedded system and swoon system and method.
In order to solve above-mentioned technical problem, the technical solution of the utility model is as follows:
A kind of traffic signal red light dim light based on embedded system is swooned system, including high-definition digital camera, embedded system graphics processing unit and computer, described high-definition digital camera is connected with computer, described embedded system graphics processing unit is connected with computer, and in the described embedded system graphics processing unit image to being obtained, the halation around red light processes.
Further, described high-definition digital camera and traffic light are separately positioned on two end supports of traffic route mouth, and the same straight line of horizontal direction.
Further, when traffic light red light is lighted, computer controls high-definition digital camera and shoots, and weaken by the halation around red light in the embedded system graphics processing unit image to being obtained or remove, and the image after processing shows on computers, or carry out printout by printing device.
Further, described embedded system graphics processing unit is FPGA graphics processing unit.
A kind of traffic signal red light based on embedded system subtracts corona method, comprises the steps:
11) on the image that the shooting of high-definition digital video camera obtains, carry out the drafting of GLOW INCLUSION AREA, be set to area-of-interest；
12) in this area-of-interest, red, yellow and white pixel point are added up by statistics with histogram method, be exposed the differentiation of degree according to the proportionate relationship of reddish yellow white three-color pixel point quantity；
13) depth of exposure obtained according to step 12) carry out dim light swoon intensity level Sr arrange；
14) locking to valid pixel, as long as meeting H component according to the characteristic in hsv color space to be redness between 0 ~ 30 and 300 ~ 360, the red pixel in this interval is the valid pixel subtracting in GLOW INCLUSION AREA；
15) find meet the red pixel of condition after in RGB color, carry out the removal of halation.
Further, in step 12), the step to the differentiation of depth of exposure includes: when red pixel number is more than total pixel number 80 percent, and when yellow pixel number is less than total pixel number 10, it is determined that for normal overexposure；When red pixel number is more than total pixel number 90 percent plus yellow pixel number sum, it is determined that for moderate overexposure；When white pixel number is more than total pixel number 90 percent plus red pixel number sum, it is determined that for serious overexposure.
Further, in step 13) Sr arrange comprise the steps: when signal lights cross quick-fried degree belong to normal time, then explanation current demand signal lamp without halation, directly return；When being moderate overexposure, saturated yellow pixel quantity in statistics area-of-interest, when saturated yellow pixel quantity is less than some, then it is judged to moderate overexposure situation one, the most directly returns, do not go halation to process, otherwise Sr value is then set to initial value 1；When for serious overexposure, Sr value is then set to initial value 2, and initial value 1 is more than initial value 2.
Further, in step 15), the removal step of halation includes: under the conditions of the red pixel searched out, its green channel G, blue channel B are compared, take higher value and be assigned to green channel G, and judge whether green channel G now is more than the red channel R of 2/3rds times, if more than, represent that red channel R, green channel G numerical value are nearer, Sr value reduction is set to initial value 3, otherwise put Sr value before then using, this Sr value is multiplied by green channel G numerical value now and is assigned to R, and initial value 3 is less than initial value 2.
The beneficial effects of the utility model are: the red halation that traffic red light can be produced in the video obtained and image by this programme carries out region detection effectively weakening and even filters, improve the picture quality in monitoring field, carry out the pretreatment of image for methods such as follow-up signal lamp condition adjudgement.Distinguishing exposure intensity by statistics with histogram method according to reddish yellow white three-color pixel proportion, actual workable, empirical tests effect is notable, and need not increase new hardware processing link, it is achieved method is simple and reliable easy.The locking of effective pixel points is carried out according to hsv color space hue H, reliable and stable, especially prominent to the capacity of resisting disturbance of changeable environment, carry out halo-reduced even removing in area-of-interest, quickening system processing speed, the system work efficiency that improves, time saving and energy saving.The system and method that dim light based on embedded system is dizzy, later maintenance is convenient, and software upgrading is convenient, beneficially datumization management, price economy.
Accompanying drawing explanation
Fig. 1 is the structural representation of this utility model system；
Fig. 2 is the workflow diagram that this utility model is swooned based on embedded system dim light；
Fig. 3 is method flow diagram of the present utility model；
Fig. 4 is the design sketch before and after this utility model traffic image dim light is swooned.
Detailed description of the invention
Below in conjunction with the drawings and specific embodiments, this utility model is described further.
As a example by this programme red light in signal lights, according to red eye feature under different conditions, analytic induction signal lights distribution of color under each state, and to doing, further dim light is dizzy to be processed according to this rule, the method can process the problem of signal red light halation under most of scene.
As shown in Figure 1, a kind of traffic signal red light dim light based on embedded system is swooned system, including high-definition digital camera, embedded system graphics processing unit and computer, described high-definition digital camera and traffic light are separately positioned on two end supports of traffic route mouth, and the same straight line of horizontal direction, described high-definition digital camera is connected with computer, described embedded system graphics processing unit is connected with computer, and in the described embedded system graphics processing unit image to being obtained, the halation around red light processes.When traffic light red light is lighted, computer controls high-definition digital camera and shoots, and weaken by the halation around red light in the embedded system graphics processing unit image to being obtained or remove, and the image after processing shows on computers, or carrying out printout by printing device, described embedded system graphics processing unit can be FPGA graphics processing unit or apply on DM8127 process chip.
Such as Fig. 2, shown in Fig. 3, a kind of traffic light based on embedded system subtract corona method, comprise the steps:
1) carrying out the drafting of GLOW INCLUSION AREA on the image that the shooting of high-definition digital video camera obtains, be set to area-of-interest, represent with yellow tag line on image, subsequent step all completes in area-of-interest, to reduce data volume, improves image processing efficiency；
2) in area-of-interest, red, yellow and white pixel point are added up by statistics with histogram method, be exposed the differentiation of degree according to the proportionate relationship of reddish yellow white three-color pixel point quantity.When red pixel number is more than total pixel number 80 percent, and when yellow pixel number is less than total pixel number 10, it is determined that for normal overexposure；When red pixel number is more than total pixel number 90 percent plus yellow pixel number sum, it is determined that for moderate overexposure；When white pixel number is more than total pixel number 90 percent plus red pixel number sum, it is determined that for serious overexposure.
3) according to differentiate depth of exposure carry out dim light swoon intensity level Sr arrange, this dim light intensity level Sr that swoons is a parameter, be the value after being multiplied with green channel G be assigned to red channel R reach lower red color channel value R a parameter, named dim light is swooned intensity level, this value is the biggest represent that red channel R subtracts the fewest, it is the most that relatively red halation retains, specifically comprise the following steps that when the normal overexposure of signal lights, when i.e. Flag flag bit is 0, directly return, do not go halation to process；When for moderate overexposure, when Flag flag bit is 1, carry out the statistics of yellow pixel point in area-of-interest, where it is determined that if condition is yellow saturated pixel count less than 50 percent yellow pixel point sum, then be judged to moderate overexposure situation one, the most directly return, halation is not gone to process, otherwise the Sr1 parameter value of Sr initial value 1 is then set to 1.2, this can allow amendment for the empirical value later stage, and general span is between 0 ~ 2；When for serious overexposure, when Flag flag bit is 2, the Sr1 parameter value of Sr initial value 2 is then set to 1.1.
4) locking of valid pixel.Carrying out the traversal of red pixel in area-of-interest according to tone H component in hsv color space, red pixel meets tone H component between 0 ~ 30 and 300 ~ 360.Red pixel in this interval is the valid pixel subtracting in GLOW INCLUSION AREA.
5) the weakening or remove of red halation: carry out the removal of redness halation after finding the red pixel meeting condition in RGB color.In order to prevent green channel G, blue channel B too small, and causing removing halation back scenery is completely black situation, to green channel G, blue channel B process.Under the conditions of the red pixel searched out, to its green channel G, blue channel B compares, take higher value and be assigned to green channel G, and judge whether green channel G now is more than the red channel R value of 2/3rds times, if more than, represent red channel R, green channel G numerical value is nearer, the Sr2 parameter value of Sr initial value 3 is reduced and is set to 1.0, otherwise the Sr1 parameter value of put Sr initial value 1 before then using, the Sr2 parameter of i.e. Sr initial value 3 is equal to the Sr1 parameter value of Sr initial value 1, the Sr2 parameter value of this Sr initial value 3 is multiplied by green channel G passage numerical value now, and do an amplitude limit between 0 to original R value, value after amplitude limit is assigned to red channel R.
Such then complete that the redness in video monitoring image area-of-interest is halo-reduced even to be filtered, it is illustrated in figure 4 traffic image and is swooned the design sketch before and after algorithm by the dim light that application this programme proposes.Hurdle, the left side one is the original image of video monitoring, can substantially observe around red light with the presence of relatively large red halation, has masked many image informations, it is impossible to the picture materials such as clear environment of observation background, and recovers to impact to red light；One hurdle, the right is through swooning the image after system based on embedded system dim light, the discovery that can be perfectly clear is in yellow line region, i.e. in area-of-interest, the picture material covered by halation and details have obtained a certain degree of recovery and reduction, on the premise of effectively removing red halation, do not affect the unlapped region of image halation, recover to lay good basis to later stage red light.
The above is only preferred implementation of the present utility model; it should be pointed out that, for those skilled in the art, without departing from the concept of the premise utility; can also make some improvements and modifications, these improvements and modifications also should be regarded as in this utility model protection domain.
Claims (3)
1. a traffic signal red light dim light based on embedded system is swooned system, it is characterized in that, including high-definition digital camera, embedded system graphics processing unit and computer, described high-definition digital camera is connected with computer, described embedded system graphics processing unit is connected with computer, and in the described embedded system graphics processing unit image to being obtained, the halation around red light processes;When traffic light red light is lighted, computer controls high-definition digital camera and shoots, and weaken by the halation around red light in the embedded system graphics processing unit image to being obtained or remove, and the image after processing shows on computers, or carry out printout by printing device.
A kind of traffic signal red light dim light based on embedded system the most according to claim 1 is swooned system, it is characterised in that described high-definition digital camera and traffic light are separately positioned on two end supports of traffic route mouth, and the same straight line of horizontal direction.
A kind of traffic signal red light dim light based on embedded system the most according to claim 1 is swooned system, it is characterised in that described embedded system graphics processing unit is FPGA graphics processing unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201521064349.8U CN205644979U (en) | 2015-12-18 | 2015-12-18 | Dizzy system of traffic signals red light dim light based on embedded system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201521064349.8U CN205644979U (en) | 2015-12-18 | 2015-12-18 | Dizzy system of traffic signals red light dim light based on embedded system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN205644979U true CN205644979U (en) | 2016-10-12 |
Family
ID=57079265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201521064349.8U Active CN205644979U (en) | 2015-12-18 | 2015-12-18 | Dizzy system of traffic signals red light dim light based on embedded system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN205644979U (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427639A (en) * | 2015-12-18 | 2016-03-23 | 杭州中威电子股份有限公司 | System and method for halo weakening of traffic signal red light based on embedded system |
| CN108932696A (en) * | 2017-05-26 | 2018-12-04 | 杭州海康威视数字技术股份有限公司 | The Halation inhibition method and device of signal lamp |
| CN109557109A (en) * | 2018-12-29 | 2019-04-02 | 中国肉类食品综合研究中心 | Freeze the detection method and device of meat packed state |
-
2015
- 2015-12-18 CN CN201521064349.8U patent/CN205644979U/en active Active
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427639A (en) * | 2015-12-18 | 2016-03-23 | 杭州中威电子股份有限公司 | System and method for halo weakening of traffic signal red light based on embedded system |
| CN108932696A (en) * | 2017-05-26 | 2018-12-04 | 杭州海康威视数字技术股份有限公司 | The Halation inhibition method and device of signal lamp |
| CN108932696B (en) * | 2017-05-26 | 2020-11-27 | 杭州海康威视数字技术股份有限公司 | Signal lamp halo suppression method and device |
| CN109557109A (en) * | 2018-12-29 | 2019-04-02 | 中国肉类食品综合研究中心 | Freeze the detection method and device of meat packed state |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108765443B (en) | Sign enhancement processing method for self-adaptive color threshold segmentation | |
| CN107730481B (en) | Traffic signal lamp image processing method and traffic signal lamp image processing device | |
| CN106780438A (en) | Defects of insulator detection method and system based on image procossing | |
| CN102496282B (en) | Traffic intersection signal light state identification method based on RGB color transformation | |
| CN205644979U (en) | Dizzy system of traffic signals red light dim light based on embedded system | |
| CN107742301B (en) | Image classification-based power transmission line image processing method under complex background | |
| CN105812674B (en) | A kind of color correct method, monitoring method and the device of signal lamp | |
| CN104715239A (en) | Vehicle color identification method based on defogging processing and weight blocking | |
| CN105427639B (en) | A kind of traffic signals red light dim light based on embedded system is swooned system and method | |
| CN103488987A (en) | Video-based method and device for detecting traffic lights | |
| CN107679495A (en) | A kind of detection method of transmission line of electricity periphery activity engineering truck | |
| CN104253976A (en) | Surveillance camera filtering system and method | |
| CN109166111B (en) | Electronic ink screen defect detection method and system | |
| CN102521564A (en) | Method for identifying tea leaves based on colors and shapes | |
| Wonghabut et al. | Traffic light color identification for automatic traffic light violation detection system | |
| Wang et al. | A piecewise-based contrast enhancement framework for low lighting video | |
| CN107507142A (en) | Signal lamp image processing method | |
| CN106600569A (en) | Signal lamp color effect enhancement processing method and apparatus thereof | |
| CN108711160B (en) | Target segmentation method based on HSI (high speed input/output) enhanced model | |
| CN104574377A (en) | Method for correcting yellow cast of red lamp of electronic police | |
| CN113537037A (en) | Pavement disease identification method, system, electronic device and storage medium | |
| CN107067734A (en) | A kind of urban signal controlling intersection vehicles are detained peccancy detection method | |
| KR102398874B1 (en) | Apparatus and method for separating foreground from background | |
| Yang et al. | A license plate segmentation algorithm based on MSER and template matching | |
| CN115006901B (en) | Automatic control method and system for intelligent backwashing filter |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200318 Address after: 310000 room 702, 1819 Xixing Road, Xixing street, Binjiang District, Hangzhou City, Zhejiang Province Patentee after: Hangzhou Zhongwei Digital Technology Co., Ltd Address before: 310012, Zhejiang building, No. 20, Xihu District, Zhejiang, Hangzhou, Wensanlu Road, 17 Patentee before: OB TELECOM ELECTRONICS Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Room 702, 1819 Xixing Road, Xixing street, Binjiang District, Hangzhou, Zhejiang 310000 Patentee after: Hangzhou Zhongwei Digital Technology Co.,Ltd. Address before: Room 702, 1819 Xixing Road, Xixing street, Binjiang District, Hangzhou, Zhejiang 310000 Patentee before: Hangzhou Zhongwei Digital Technology Co.,Ltd. |