CN102862574B - The method of vehicle active safety is realized based on smart mobile phone - Google Patents
The method of vehicle active safety is realized based on smart mobile phone Download PDFInfo
- Publication number
- CN102862574B CN102862574B CN201210358336.6A CN201210358336A CN102862574B CN 102862574 B CN102862574 B CN 102862574B CN 201210358336 A CN201210358336 A CN 201210358336A CN 102862574 B CN102862574 B CN 102862574B
- Authority
- CN
- China
- Prior art keywords
- road surface
- region
- lane mark
- mobile phone
- smart mobile
- 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)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses the method realizing vehicle active safety based on smart mobile phone, the method comprises the steps: (1) utilizes smart mobile phone to obtain vehicle front video image, and extracts road surface grey parameter; (2) lane mark deviation detection is carried out according to the road surface grey parameter extracted; (3) survey of preceding object quality testing is carried out according to the road surface grey parameter extracted; (4) result surveyed according to lane mark deviation detection and preceding object quality testing carries out danger early warning.This method utilizes smart mobile phone as the hardware platform of data processing, and lane mark extractive technique in combining image processing technology field and road obstacle extractive technique, vehicle whether run-off-road line can be detected in real time, and detect front and whether have obstacle, ensure the traffic safety of chaufeur.
Description
Technical field
The present invention relates to automobile active safety technology, be specifically related to the method realizing vehicle active safety based on smart mobile phone.
Background technology
Along with the development of society, the traffic safety problem in city is day by day serious.According to statistics, within 2011, the whole nation relates to altogether the road traffic accident 210812 of personal casualty, causes 62387 people's death, 237421 people injured, causes huge economic loss and personal casualty.Traditional vehicle safety means as: bumper/spoiler, safety strap, safety air bag etc. are all passive types, are called vehicle passive security technology, cannot avoid the generation of accident, can only reduce the loss that accident is brought.Therefore, during vehicle active safety technologies is more and more applied to Car design and is produced.
Existing vehicle active safety technologies, as ACC(cruise system), LDWS (Lane Departure Warning System) etc. needs extra sensor (as laser radar) to carry out perception to surrounding environment, these sensors are all that vehicle completes installation before dispatching from the factory, and the vehicle not installing sensor additional then cannot use vehicle active safety technologies after dispatching from the factory.Therefore, existing vehicle active safety technologies portability is poor.If vehicle active safety technologies can be implemented on other hardware platforms, depart from the bus of vehicle itself, so its portability can strengthen greatly.Intelligent mobile phone terminal has self independently sensor and treater, is a good hardware platform.
Image processing techniques is widely used in field of traffic, as Car license recognition, traffic information extraction etc. due to advantages such as its obtaining information amount are large, cost is low.It is equally also applied in vehicle active safety, as lane mark deviation detection etc.Image processing techniques in existing vehicle active safety technologies all relies on the dedicated imbedded system such as PC or DSP, and the arithmetic capability of these hardware platforms is powerful, can ensure the real-time of certain complicated algorithm.But, the arithmetic capability of smart mobile phone is relatively weak, and its treater also will process communication service, the existing vehicle active safety technologies based on image procossing is comparatively complicated again simultaneously, if therefore prior art is grafted directly to intelligent mobile phone terminal, the real-time of system can be caused greatly to reduce, and practicality is also had a greatly reduced quality.The lane detection technology utilizing image processing techniques to realize for existing and the result for retrieval of detection of obstacles technology as follows:
Patent publication No. CN 101608924A discloses a kind of based on the method for detecting lane lines that gray scale is estimated and Cascade H ough converts.First this invention is carried out gray scale to the sensitizing range of vehicle front road conditions original image I gathered and is estimated to be divided into vehicle shadow region, the non-mark region in road surface and pavement marker region (the vehicle body region containing front vehicles); Then Mathematical Morphology Method is adopted to estimate from road surface gray scale the boundary image obtaining pavement marker region the region divided; The boundary image of road pavement mark region carries out Hough transform to extract straight line characteristics of image wherein subsequently; Finally, the detection to lane mark is achieved by the search of road pavement mark edge vanishing point.The defect of this invention is, when utilizing Hough transform inspection vehicle diatom, can be subject to the impact at other strong edges, and its operand is larger.
Patent publication No. CN 102156979A discloses a kind of lane mark method for quick based on GrowCut, and it comprises: acquisition monitoring video the image selecting at least one frame are as key frame; Edge segmentation Seed Points is determined in described key frame images acceptance of the bid; GrowCut algorithm is utilized to carry out edge segmentation; Adopt median filter to the smoothing filtering of segmentation result; Again by smothing filtering result, carry out edge extracting; Edge extracts result again, carries out point half process; Again to point half result, adopt polynomial curve fitting method to carry out matching to lane mark, obtain final lane mark.The method is applicable to detect static background image, and cannot extract the lane mark in dynamic background, is not suitable for vehicle active safety technologies.
Patent publication No. CN 101447078A discloses a kind of method for obstacle segmentation and device.This inventive embodiment comprises: the consecutive frame image current frame image in sequence image and frame period being more than or equal to 1 frame is divided into identical some pieces, difference is done to the correspondence position image block of described current frame image and described consecutive frame image, and determines the difference block region in difference image; Calculate perpendicularity and the grain of each piecemeal corresponding with described difference block region in current frame image; By corresponding with described difference block region in current frame image and the piecemeal with perpendicularity and grain is defined as obstacle piecemeal; By described obstacle piecemeal according to connection generate rule barrier region.The shortcoming of this invention is, when splitting obstacle, need to travel through entire image, its operand is comparatively large, cannot realize on intelligent mobile phone terminal.
Patent publication No. CN 101419667A discloses the method for obstacle in a kind of recognition image, comprise: the image obtaining present frame and the before N frame nearest with present frame, obtained every two field picture is divided by identical mode, the block region after several divisions is all obtained for every two field picture; Calculate the moving obstacle confidence level in present frame and each piece region before N frame corresponding to nearest with present frame; According to described present frame and with present frame nearest before the moving obstacle confidence level in described each piece of region of N frame, determine in current frame image, whether each piece of region is obstacle successively; The obstacle in image is determined according to each piece of region.The defect of the method is, the image of its N frame before needing record and present frame nearest, required spatial cache is comparatively large, is not suitable for intelligent mobile phone terminal.
Summary of the invention
The present invention is directed to existing vehicle active safety technologies to need based on the existing sensor of vehicle, portable poor and operand large, the problem such as need the hardware platform of high request to support, and a kind of method realizing vehicle active safety based on smart mobile phone is provided.The method utilizes smart mobile phone as the hardware platform of data processing, and lane mark extractive technique in combining image processing technology field and road obstacle extractive technique, vehicle whether run-off-road line can be detected in real time, and detect front and whether have obstacle, ensure the traffic safety of chaufeur.
In order to achieve the above object, the present invention adopts following technical scheme:
Realize the method for vehicle active safety based on smart mobile phone, the method comprises the steps:
(1) utilize smart mobile phone to obtain vehicle front video image, and extract road surface grey parameter;
(2) lane mark deviation detection is carried out according to the road surface grey parameter extracted;
(3) survey of preceding object quality testing is carried out according to the road surface grey parameter extracted;
(4) result surveyed according to lane mark deviation detection and preceding object quality testing carries out danger early warning.
In preferred embodiment of the present invention, the process extracting road surface grey parameter in described step (1) is as follows:
(11) region is extracted for extracting corresponding road surface grey parameter in the region going restriction one piece in region to belong to road surface of the vehicle front video image display obtained as road surface grey parameter;
(13) the road surface average gray in grey parameter extraction region, road surface is extracted;
(14) the road surface gray scale maximum permission fluctuation difference in grey parameter extraction region, road surface is extracted;
(15) the road surface gray scale gradual change compensating factor in grey parameter extraction region, road surface is extracted.
Further, in described step (2), the process of lane mark deviation detection is as follows:
(21) set up white, yellow, gray color model: refer to according to extracted road surface grey parameter, dialogue, Huang, ash three kinds of colors carry out modeling, if the rgb value of a certain pixel meets a certain model, then think that this pixel is the color that this model characterizes;
(22) lane mark border initial point is searched: the border initial point of upwards search vehicle left and right sides lane mark from bottom the video image obtained;
(23) next lane mark boundary point is searched: the last lane mark boundary point according to searching sets up estimation range, then in this estimation range, searches for next lane mark boundary point; Carry out iteration execution, until lane mark boundary point cannot be found in estimation range;
(24) lane mark position analysis: the lane mark information according to detecting calculates its coefficient of deviation.
Further, the process of carrying out the survey of preceding object quality testing in described step (3) is as follows:
(32) region, road surface start line is set; In the region, road surface on video image display road surface, a cut-off rule is set, in order to the region, road surface in segmented video image and front cover;
(32) search for region, road surface: according to road surface grey parameter, from the start line of region, road surface, upwards search for region, road surface every row, if run into the region of non-road surface color, then stop search, and this point is designated as the road surface region separation point of these row;
(33) preceding object thing is analyzed: according to the road surface region separation point searched, and detect whether there is sunk area, if there is sunk area, it meets certain requirement, then this thinks that this region exists obstacle.
Further, the specific implementation process of described step (4) is as follows: if the lane mark coefficient of deviation detected exceedes a certain scope, then think that vehicle departs from, utilize smart mobile phone to send voice message; If when detecting that preceding object thing is in the danger early warning region set by user, then think that needs carry out keeping away barrier operation, utilize smart mobile phone to send voice message.
Technical scheme provided by the present invention, do not travel through full figure when it carries out image procossing, operand is little, can well be applied to intelligent mobile phone terminal.
This programme realizes based on common smart mobile phone, without the need to vehicle loading any sensor, the vehicle active safety technologies realized thus can be implanted into any vehicle very easily, can prevent lane mark from departing to a certain extent and bump against with preceding object thing, wide application, practical, with low cost.
Moreover the image processing techniques operand used due to this programme is little, can ensure the real-time of the system of embodiment.
Accompanying drawing explanation
The present invention is further illustrated below in conjunction with the drawings and specific embodiments.
Fig. 1 is the principle flow chart that the present invention realizes;
Fig. 2 is the schematic flow sheet that the present invention carries out road surface grey parameter extraction;
Mobile video image display schematic diagram when Fig. 3 is the invention process;
Fig. 4 is the schematic flow sheet that the present invention carries out lane mark deviation detection;
Fig. 5 is the schematic flow sheet that the present invention carries out the survey of preceding object quality testing;
Fig. 6 is the schematic flow sheet that the present invention carries out region, road surface;
Fig. 7 is the schematic flow sheet that the present invention carries out danger early warning.
Detailed description of the invention
The technological means realized to make the present invention, creation characteristic, reaching object and effect is easy to understand, below in conjunction with concrete diagram, setting forth the present invention further.
See Fig. 1, the present invention is based on the method that smart mobile phone realizes vehicle active safety, its specific implementation process is as follows:
Step 101: extract road surface grey parameter.
Concrete, first vehicle front video image is obtained, then in the region shown by image, set road surface grey parameter extract region, and extract road surface grey parameter from this region, comprise road surface average gray, road surface gray scale fluctuation difference and road surface gray scale gradual change compensating factor.
Step 102: according to the road surface grey parameter extracted, carry out lane mark deviation detection.
Concrete, first according to the road surface grey parameter of extraction, definition is white, yellow, gray color model, then finds lane mark border initial point, then finds next lane mark boundary point, is searching for the laggard moving traffic lane line position analysis of all lane mark boundary points.
Step 103: according to the road surface grey parameter extracted, carry out the survey of preceding object quality testing.
Concrete, first set region, road surface start line, then according to the road surface grey parameter extracted, from the start line of region, road surface, upwards search for region, road surface, then carry out road surface regional analysis and obtain preceding object thing information.
Step 104: survey according to lane mark deviation detection and preceding object quality testing, carry out danger early warning.
Concrete, if after carrying out lane mark deviation detection, find automotive run-off-road line, then send alarm sound to carry out danger early warning by smart mobile phone.Same, if after carrying out the survey of preceding object quality testing, find there is obstacle in front is closely, then send alarm sound to carry out danger early warning by smart mobile phone.
Based on above-mentioned principle, specific embodiment of the invention process is as follows:
The first step: extract road surface grey parameter
See Fig. 2, the leaching process that this example carries out road surface grey parameter is as follows:
Step 201: obtain vehicle front video image.
Vehicle front video image acquisition equipment is the post-positioned pick-up head of smart mobile phone, and the post-positioned pick-up head of smart mobile phone can the video image of real time shooting vehicle front, then this image transmitting is carried out subsequent treatment to the core processor of smart mobile phone.
Step 202: setting road surface grey parameter extracts region.
As shown in Figure 3, road surface grey parameter extracts 202a territory, district and refers to most possible one piece of region for road surface in the region 202c of image display, general above front cover 202b, the centre of two lane mark 202f, this region does not affect by front cover, be not subject to front vehicles impact, the intensity profile situation on road surface can be reflected preferably yet.The width in this region is 1/2nd of picture traverse, is highly 1/8th of picture altitude, this region center ofthe in the picture, and upper-lower position is in middle the latter half of image.
Step 203: extract road surface average gray.
Road surface average gray can reflect the average gray value on road surface.Its concrete extracting method is: extract in region at road surface grey parameter, extract the gray value being in the pixel of even number line and even column, and gray value is carried out descending sort, get rid of the pixel that gray value is front 15% and rear 15%, then the gray value of residual pixel point is added and can obtains road surface average gray divided by the number of remaining pixel again, easy for describing, be defined as I
aver.
Step 204: extract road surface gray scale fluctuation difference.
Road surface gray scale fluctuation difference reflects the maximum fluctuation amplitude of the road surface gray value that system allows, if the gray value of a certain pixel has exceeded this amplitude, then can think that this pixel does not belong to region, road surface.Its concrete extracting method is: the pixel value of pixel remaining during extraction road surface average gray and road surface average gray are done difference, calculate its sum of squares, then divided by the number of remaining pixel, the standard deviation of gray value is obtained after opening root again, finally using the twice of this standard deviation as road surface gray scale fluctuation difference, easy for describing, be defined as I
d.Its computing formula is as follows:
Step 205: extract road surface gray scale gradual change compensating factor.
Due to the impact of light, the gray value on the road surface in the vehicle front video image that camera collects not is invariable, and being has certain gradual change trend, and its trend is generally with the distance of car far away, its gray value is larger, and can be similar to and think linear change.If all adopt identical road surface average gray will cause certain error to weigh road surface gray scale to the overall situation, therefore need the error that road surface gray scale gradual change compensating factor brings to compensate gradual change.Its concrete extracting method is: calculate the gray value aviation value that road surface grey parameter extracts top and lowermost end a line in region respectively, top flat average and low side aviation value are subtracted each other, height again divided by road surface grey parameter extraction region just obtains road surface gray scale gradual change compensating factor, easy for describing, be defined as α.Its computing formula is as follows:
The compensation formula of road surface average gray is:
I '
aver=I
aver+ (pixel ordinate-bottom row ordinate) * α.
Second step: lane mark deviation detection
See Fig. 4, the implementation process that this example carries out lane mark deviation detection is as follows:
Step 301: definition is white, yellow, gray color model.
Lane mark on road surface has two kinds of colors, white and yellow, and the feature according to distribution of color is carried out lane detection, therefore needs first to define white, yellow, gray color model.
Shown in the color model of white is defined as follows:
Wherein R, G, B are respectively each channel value of RGB of certain pixel, and thresold1 is each interchannel relevance threshold of RGB, and I is the gray value of certain pixel, I '
averfor the road surface average gray value of this pixel position, I
dfor road surface gray scale fluctuation difference extracted before.If certain pixel meets above-listed inequality, then think that this pixel belongs to white lane mark.
Shown in the color model of yellow is defined as follows:
Wherein thresold3 is the minimal difference threshold value between R, G and channel B value.If certain pixel meets all above-listed inequality, then think that this pixel belongs to yellow lane mark.
Shown in the color model of grey is defined as follows:
If certain pixel meets above-listed inequality, then think that this pixel belongs to grey.
Step 302: find lane mark border initial point.
First the border initial point of left-lane line is searched for.The initial point of search is the bottom intermediate point of image, and the direction of search is from centre left, and from bottom upwards, the pixel of search need meet the following conditions: (1) this pixel and its left pixel point all meet white or yellow line color model; (2) on the right side of this pixel, two pixels all meet gray color model.If search the pixel meeting this condition, then it can be used as left-hand lane line boundary initial point; If search does not also search to image half place the pixel meeting above condition, then think that left side is without Road.
Then the border initial point of right lane line is searched for.The initial point of search is the bottom intermediate point of image, and the direction of search is from centre to the right, and from bottom upwards, the pixel of search need meet the following conditions: (1) this pixel and its right pixel point all meet white or yellow line color model; (2) on the left of this pixel, two pixels all meet gray color model.If search the pixel meeting this condition, then it can be used as right-hand lane line boundary initial point; If search does not also search to image half place the pixel meeting above condition, then think that right side is without Road.
Step 303: find next lane mark boundary point.
First set up next boundary point estimation range, this region is:
Wherein x
i-1, y
i-1be respectively the horizontal stroke, the ordinate that search a lane mark boundary point, and thresold1, thresold2 characterize estimation range, if thresold1 is less, thresold2 is larger, so operand will reduce, but likely there will be the situation of next boundary point not in estimation range, h is the height of vehicle front video image.Consider that the slope of lane mark is within the specific limits, if therefore meet thresold1 >=thresold2*2, so next lane mark boundary point is bound in estimation range.The first half due to image is sky, and therefore without the need to searching in this region, we are by searching for limitation in height at 3/4 place of picture altitude here, to reduce calculated amount.
After setting up next boundary point estimation range, can search for next boundary point in this region, its searching method is with step 302.If find lane mark boundary point in estimation range, then repeated execution of steps 303; If do not find lane mark boundary point in estimation range, so think the detection completed this lane mark.This part can use the mode of iteration to realize.
Step 304: lane mark position analysis.
May be there are following four kinds of situations in the testing result of two lane maries in left and right: (1) two lane mark all exists; (2) only there is one, left side; (3) only there is one, right side; (4) do not exist.Situation (4) is not considered whether it departs from, and for other three kinds of situations, needs to calculate its coefficient of deviation respectively.
For situation (1), two, left and right lane mark gets some boundary point pairs, namely the boundary point that two ordinates are identical, then the center-point abscissa of each boundary point pair is calculated, then the average of each center-point abscissa is calculated, this numerical representation method center abscissa of Road, be designated as W ', its computing formula is as follows:
For situation (2), some boundary points got by the left-hand lane line that only need detect at it, and the intermediate value then calculating its abscissa and image right side boundary can obtain W ', value, and its computing formula is as follows:
Wherein w is the width of vehicle front video image.
For situation (3), some boundary points got by the right-hand lane line that only need detect at it, and the intermediate value then calculating its abscissa and image left border can obtain W ' value, and its computing formula is as follows:
Finally calculate coefficient of deviation, coefficient of deviation β can use following formulae discovery:
3rd step, preceding object quality testing is surveyed
See Fig. 5, the implementation process that this example carries out the survey of preceding object quality testing is as follows:
Step 401: region, road surface start line is set.
See Fig. 3, because the picture seen in vehicle front video Figure 20 2c can be divided into front cover 202b and region, road surface 202d two parts, therefore preceding object quality testing is surveyed, there is no need to search for from image base, and should search for from the 202d of region, road surface, region, road surface start line 202e can be regarded as the cut-off rule of front cover region 202b and region, road surface 202d.User needs the angle of placing by changing mobile phone that this line should be in region, road surface in use, and in the provisioning process, the ordinate of this line is set to the ordinate that road surface grey parameter extracts sections bottom horizontal line.
Step 402: upwards search for region, road surface from the start line of region, road surface.
As shown in Figure 6, the basic thought that this example carries out searching for region, road surface is, for a certain row of image, upwards search for from the start line of region, road surface, if run into the pixel of non-road surface color, then stop search, and record stops coordinate when detecting, and this point is called road surface region separation point.Because the first half of image mostly is sky, so there is no necessity search, therefore, if search to picture altitude 3/4 time also do not find non-road surface colored pixels point, then stop search.Road surface color herein, refers to the Huang defined in step 301, white, ash three kinds of colors.In order to reduce operand, this example is not all searched for each row pixel of image, but several row in interval are once searched for.If interval is larger, that operand is less, but some less obstacle may be left in the basket.
Step 403: carry out road surface regional analysis and obtain preceding object thing information.
Region, road surface under normal circumstances there will not be concave shape, if find to there is concave shape after carrying out road surface regional analysis, then thinks and obstacle detected.The method of the detection of concave shape is: to there being search from the road surface region separation point on the left of image, if the difference of the height of the height of certain road surface region separation point and next road surface region separation point is greater than a certain threshold value, then think that this road surface region separation point is for depression left side initial point, if the difference of the height of the height of certain road surface region separation point and a upper road surface region separation point is greater than a certain threshold value, then think that this road surface region separation point is for depression right side initial point.If depression left side initial point is adjacent with depression right side initial point, and this coordinate of 2 meets following condition, then think to there is depression between these 2.
Wherein x
lwith x
rbe respectively the abscissa of the left and right initial point that caves in, y
lwith y
rbe respectively the ordinate of the left and right initial point that caves in, thresold1, for limiting the minimum width of depression, can get rid of the impact of little noise spot, and thresold2 is for limiting the diff-H of left and right sides initial point, if exceed this threshold value, then think that two initial points do not form depression about this.
4th step, dangerous forecast
See Fig. 7, the implementation process that this example carries out danger early warning is as follows:
Step 501: judge whether run-off-road line.
Coefficient of deviation β should meet-1< β < 1, if met
then think that vehicle departs from, otherwise think that vehicle departs from, concrete, if
then think that vehicle departs to the right, utilize smart mobile phone to send voice message, prompting chaufeur travels left; If
then think that vehicle departs from left, utilize smart mobile phone to send voice message, prompting chaufeur is to right travel.
Step 502: judge whether that barrier kept away by needs.
Because the obstacle detected not necessarily needs to carry out keeping away barrier, therefore, chaufeur can according to the driving habit of oneself, smart mobile phone arranges danger early warning region, if when obstacle occurs in this region, utilize smart mobile phone to send voice message, prompting chaufeur carries out keeping away barrier operation.
More than show and describe groundwork of the present invention, principal character and advantage of the present invention.The technical personnel of the industry should be understood; the present invention is not restricted to the described embodiments; what describe in above-described embodiment and specification sheets just illustrates principle of the present invention; without departing from the spirit and scope of the present invention; the present invention also has various changes and modifications, and these changes and improvements all fall in the claimed scope of the invention.Application claims protection domain is defined by appending claims and equivalent thereof.
Claims (4)
1. realize the method for vehicle active safety based on smart mobile phone, it is characterized in that, described method comprises the steps:
(1) utilize smart mobile phone to obtain vehicle front video image, and extract road surface grey parameter; The process wherein extracting road surface grey parameter is as follows:
(11) region is extracted for extracting corresponding road surface grey parameter in the region going restriction one piece in region to belong to road surface of the vehicle front video image display obtained as road surface grey parameter;
(12) the road surface average gray in grey parameter extraction region, road surface is extracted;
(13) the road surface gray scale maximum permission fluctuation difference in grey parameter extraction region, road surface is extracted;
(14) the road surface gray scale gradual change compensating factor in grey parameter extraction region, road surface is extracted, extracting method is: calculate the gray value aviation value that road surface grey parameter extracts top and lowermost end a line in region respectively, top flat average and low side aviation value are subtracted each other, more just obtains road surface gray scale gradual change compensating factor divided by the height in road surface grey parameter extraction region;
(2) lane mark deviation detection is carried out according to the road surface grey parameter extracted;
(3) survey of preceding object quality testing is carried out according to the road surface grey parameter extracted;
(4) result surveyed according to lane mark deviation detection and preceding object quality testing carries out danger early warning.
2. the method realizing vehicle active safety based on smart mobile phone according to claim 1, is characterized in that, in described step (2), the process of lane mark deviation detection is as follows:
(21) set up white, yellow, gray color model: refer to according to extracted road surface grey parameter, dialogue, Huang, ash three kinds of colors carry out modeling, if the rgb value of a certain pixel meets a certain model, then think that this pixel is the color that this model characterizes;
(22) lane mark border initial point is searched: the border initial point of upwards search vehicle left and right sides lane mark from bottom the video image obtained;
(23) next lane mark boundary point is searched: the last lane mark boundary point according to searching sets up estimation range, then in this estimation range, searches for next lane mark boundary point; Carry out iteration execution, until lane mark boundary point cannot be found in estimation range;
(24) lane mark position analysis: the lane mark information according to detecting calculates its coefficient of deviation.
3. the method realizing vehicle active safety based on smart mobile phone according to claim 1, is characterized in that, the process of carrying out the survey of preceding object quality testing in described step (3) is as follows:
(31) region, road surface start line is set; In the region, road surface on video image display road surface, a cut-off rule is set, in order to the region, road surface in segmented video image and front cover;
(32) search for region, road surface: according to road surface grey parameter, every row upwards search search region, road surface from the start line of region, road surface, if run into the pixel of non-road surface color, then stop search, and this point is designated as the road surface region separation point of these row;
(33) preceding object thing is analyzed: according to the road surface region separation point searched, and detect whether there is sunk area, if there is sunk area, it meets certain requirement, then this thinks that this region exists obstacle.
4. the method realizing vehicle active safety based on smart mobile phone according to claim 1, it is characterized in that, the specific implementation process of described step (4) is as follows: if the lane mark coefficient of deviation detected exceedes a certain scope, then think that vehicle departs from, utilize smart mobile phone to send voice message; If when detecting that preceding object thing is in the danger early warning region set by user, then think that needs carry out keeping away barrier operation, utilize smart mobile phone to send voice message.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210358336.6A CN102862574B (en) | 2012-09-21 | 2012-09-21 | The method of vehicle active safety is realized based on smart mobile phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210358336.6A CN102862574B (en) | 2012-09-21 | 2012-09-21 | The method of vehicle active safety is realized based on smart mobile phone |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102862574A CN102862574A (en) | 2013-01-09 |
| CN102862574B true CN102862574B (en) | 2015-08-19 |
Family
ID=47441781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210358336.6A Active CN102862574B (en) | 2012-09-21 | 2012-09-21 | The method of vehicle active safety is realized based on smart mobile phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102862574B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103465857A (en) * | 2013-09-17 | 2013-12-25 | 上海羽视澄蓝信息科技有限公司 | Mobile-phone-based active safety early-warning method for automobile |
| CN103617412B (en) * | 2013-10-31 | 2017-01-18 | 电子科技大学 | Real-time lane line detection method |
| CN103942852A (en) * | 2014-04-04 | 2014-07-23 | 广东翼卡车联网服务有限公司 | Travelling recording method and system based on mobile phone terminal |
| CN103871243A (en) * | 2014-04-16 | 2014-06-18 | 武汉欧普威科技有限公司 | Wireless vehicle management system and method based on active safety platform |
| CN104157160B (en) * | 2014-08-08 | 2016-08-17 | 中国联合网络通信集团有限公司 | Vehicle travel control method, device and vehicle |
| KR102224995B1 (en) * | 2014-10-28 | 2021-03-09 | 현대모비스 주식회사 | Apparatus and method for controlling outputting information about blind spots of vehicle using mobile device |
| CN105730330B (en) * | 2014-12-11 | 2018-09-28 | 财团法人车辆研究测试中心 | Driving safety system and its barrier screening technique |
| CN104537888A (en) * | 2014-12-25 | 2015-04-22 | 东风汽车公司 | Automobile collision alarming method based on mobile phone |
| CN104833364B (en) * | 2015-05-07 | 2018-05-18 | 深圳市爱民科技有限公司 | A kind of emergency route indicating means on bump course |
| CN106293032B (en) * | 2015-06-08 | 2021-09-24 | 北京三星通信技术研究有限公司 | Portable terminal device, and control method and apparatus thereof |
| CN105292087B (en) * | 2015-10-14 | 2019-03-12 | 南宁学院 | A kind of deviation correcting system and its correcting method based on wheel braking |
| CN105216793A (en) * | 2015-10-14 | 2016-01-06 | 南宁学院 | A kind of deviation correcting system based on electric power steering and correcting method thereof |
| CN106203268A (en) * | 2016-06-28 | 2016-12-07 | 成都之达科技有限公司 | vehicle state monitoring method based on infrared identification |
| CN106203267A (en) * | 2016-06-28 | 2016-12-07 | 成都之达科技有限公司 | Vehicle collision avoidance method based on machine vision |
| CN106791641A (en) * | 2016-12-16 | 2017-05-31 | 深圳安狗数据技术有限公司 | A kind of DAS (Driver Assistant System) and auxiliary driving method |
| CN106686165B (en) * | 2016-12-30 | 2018-08-17 | 维沃移动通信有限公司 | A kind of method and mobile terminal of road conditions detection |
| CN107826116A (en) * | 2017-10-31 | 2018-03-23 | 安徽融合智能科技有限公司 | The operating method of automobile active safety drive assist system based on mobile terminal |
| CN108284793A (en) * | 2018-01-10 | 2018-07-17 | 深圳市鑫汇达机械设计有限公司 | A kind of vehicle sub-controlling unit |
| CN108873896B (en) * | 2018-06-15 | 2021-07-02 | 驭势科技（北京）有限公司 | Lane line simulation method and device and storage medium |
| CN112131914B (en) * | 2019-06-25 | 2022-10-21 | 北京市商汤科技开发有限公司 | Lane line attribute detection method and device, electronic equipment and intelligent equipment |
| CN114074658A (en) * | 2020-08-14 | 2022-02-22 | 为昇科科技股份有限公司 | Early warning system and method for detecting approaching object |
| CN112613509A (en) * | 2020-12-25 | 2021-04-06 | 杭州智诺科技股份有限公司 | Railway wagon carriage number identification snapshot method and system |
| CN112987754B (en) * | 2021-04-14 | 2021-09-24 | 北京三快在线科技有限公司 | Unmanned equipment control method and device, storage medium and electronic equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1782668A (en) * | 2004-12-03 | 2006-06-07 | 曾俊元 | Method and device for preventing collison by video obstacle sensing |
| CN101016053A (en) * | 2007-01-25 | 2007-08-15 | 吉林大学 | Warning method and system for preventing collision for vehicle on high standard highway |
| CN101016052A (en) * | 2007-01-25 | 2007-08-15 | 吉林大学 | Warning method and system for preventing deviation for vehicle on high standard highway |
| CN101608924A (en) * | 2009-05-20 | 2009-12-23 | 电子科技大学 | A kind of method for detecting lane lines based on gray scale estimation and cascade Hough transform |
| CN101844542A (en) * | 2008-09-10 | 2010-09-29 | 财团法人交大思源基金会 | Intelligent driving assistant systems |
| WO2010111992A1 (en) * | 2009-03-30 | 2010-10-07 | Conti Temic Microelectronic Gmbh | Method and device for lane detection |
| CN102289948A (en) * | 2011-09-02 | 2011-12-21 | 浙江大学 | Multi-characteristic fusion multi-vehicle video tracking method under highway scene |
| CN102298693A (en) * | 2011-05-18 | 2011-12-28 | 浙江大学 | Expressway bend detection method based on computer vision |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4437714B2 (en) * | 2004-07-15 | 2010-03-24 | 三菱電機株式会社 | Lane recognition image processing device |
| WO2007077682A1 (en) * | 2005-12-28 | 2007-07-12 | Honda Motor Co., Ltd. | Vehicle and lane mark detection device |
| JP5016073B2 (en) * | 2010-02-12 | 2012-09-05 | 株式会社デンソー | White line recognition device |
-
2012
- 2012-09-21 CN CN201210358336.6A patent/CN102862574B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1782668A (en) * | 2004-12-03 | 2006-06-07 | 曾俊元 | Method and device for preventing collison by video obstacle sensing |
| CN101016053A (en) * | 2007-01-25 | 2007-08-15 | 吉林大学 | Warning method and system for preventing collision for vehicle on high standard highway |
| CN101016052A (en) * | 2007-01-25 | 2007-08-15 | 吉林大学 | Warning method and system for preventing deviation for vehicle on high standard highway |
| CN101844542A (en) * | 2008-09-10 | 2010-09-29 | 财团法人交大思源基金会 | Intelligent driving assistant systems |
| WO2010111992A1 (en) * | 2009-03-30 | 2010-10-07 | Conti Temic Microelectronic Gmbh | Method and device for lane detection |
| CN101608924A (en) * | 2009-05-20 | 2009-12-23 | 电子科技大学 | A kind of method for detecting lane lines based on gray scale estimation and cascade Hough transform |
| CN102298693A (en) * | 2011-05-18 | 2011-12-28 | 浙江大学 | Expressway bend detection method based on computer vision |
| CN102289948A (en) * | 2011-09-02 | 2011-12-21 | 浙江大学 | Multi-characteristic fusion multi-vehicle video tracking method under highway scene |
Non-Patent Citations (2)
| Title |
|---|
| Lane Detection using Color-Based segmentation;Kuo-Yu Chiu and Sheng-Fuu Lin;《IEEE》;20051231;706-711 * |
| 复杂路面环境下前方车辆识别目标函数的选取;马雷;《汽车工程》;20090430;第31卷(第4期);299-303 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102862574A (en) | 2013-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102862574B (en) | The method of vehicle active safety is realized based on smart mobile phone | |
| CN102509291B (en) | Pavement disease detecting and recognizing method based on wireless online video sensor | |
| CN108345822B (en) | Point cloud data processing method and device | |
| CN101303735B (en) | Method for detecting moving objects in a blind spot region of a vehicle and blind spot detection device | |
| CN106845547B (en) | A kind of intelligent automobile positioning and road markings identifying system and method based on camera | |
| CN103176185B (en) | Method and system for detecting road barrier | |
| US9141870B2 (en) | Three-dimensional object detection device and three-dimensional object detection method | |
| CN101739829B (en) | Video-based vehicle overspeed monitoring method and system | |
| US8339282B2 (en) | Security systems | |
| CN103488975B (en) | A kind of method that zebra crossing based on intelligent driving detect in real time | |
| CN103714538B (en) | road edge detection method, device and vehicle | |
| CN104657727A (en) | Lane line detection method | |
| Wu et al. | Adaptive vehicle detector approach for complex environments | |
| CN106951879A (en) | Multiple features fusion vehicle checking method based on camera and millimetre-wave radar | |
| CN105574552A (en) | Vehicle ranging and collision early warning method based on monocular vision | |
| CN103914688A (en) | Urban road obstacle recognition system | |
| CN102855759A (en) | Automatic collecting method of high-resolution satellite remote sensing traffic flow information | |
| CN110298300A (en) | A method of detection vehicle violation crimping | |
| CN104063882B (en) | Vehicle video speed measuring method based on binocular camera | |
| CN105825185A (en) | Early warning method and device against collision of vehicles | |
| CN102663357A (en) | Color characteristic-based detection algorithm for stall at parking lot | |
| CN102867417A (en) | Taxi anti-forgery system and taxi anti-forgery method | |
| CN102999749A (en) | Intelligent safety belt regulation violation event detecting method based on face detection | |
| CN105930855A (en) | Vehicle detection method based on deep convolution neural network | |
| CN102789686B (en) | Road traffic flow detecting method based on road surface brightness composite mode recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20181212 Address after: 215505 Jiangsu Changzhou high tech Industrial Development Zone, 333 Hu Shan Road 1 Tongji Science and Technology Square 2302 Patentee after: Suzhou Ainuoxin Information Technology Co., Ltd. Address before: Room 209D, 1121 North Zhongshan Road, Yangpu District, Shanghai, 200092 Patentee before: Shanghai Yongchang Information Technology Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| CP02 | Change in the address of a patent holder |
Address after: 215558 Tongji Science and Technology Plaza, 333 Hushan Road, Changshu High-tech Industrial Development Zone, Suzhou City, Jiangsu Province, 2302 Patentee after: Suzhou Ainuoxin Information Technology Co., Ltd. Address before: 215505 Jiangsu Changzhou high tech Industrial Development Zone, 333 Hu Shan Road 1 Tongji Science and Technology Square 2302 Patentee before: Suzhou Ainuoxin Information Technology Co., Ltd. |
|
| CP02 | Change in the address of a patent holder | ||
| CP03 | Change of name, title or address |
Address after: 201804 room 1, floor 2, No. 82, Lane 1221, Hanggui Road, Anting Town, Jiading District, Shanghai Patentee after: Shanghai ainuoweixin Intelligent Technology Co.,Ltd. Address before: 215558 Tongji Science and Technology Plaza, 333 Hushan Road, Changshu High-tech Industrial Development Zone, Suzhou City, Jiangsu Province, 2302 Patentee before: Suzhou Ainuoxin Information Technology Co.,Ltd. |
|
| CP03 | Change of name, title or address |