CN1506874A - Road accidence monitoring method - Google Patents
Road accidence monitoring method Download PDFInfo
- Publication number
- CN1506874A CN1506874A CNA031566227A CN03156622A CN1506874A CN 1506874 A CN1506874 A CN 1506874A CN A031566227 A CNA031566227 A CN A031566227A CN 03156622 A CN03156622 A CN 03156622A CN 1506874 A CN1506874 A CN 1506874A
- Authority
- CN
- China
- Prior art keywords
- gray
- vehicle
- scale value
- pixel
- accident
- 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.)
- Pending
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
Disclosed is a method for detecting an accident more swiftly and reliably. The method for detecting an accident obtains an image from a predetermined region on a road, determining whether there exists an accident or not depending on change transition of gray levels for pixels on a line type trap set in advance on the basis of the obtained image. At the moment, a real vehicle and a shadow could be discriminated with use of quantity of change and frequency of the gray levels on the line type trap.
Description
Technical field
The present invention relates to the method for monitoring accident on highway, relate more specifically to method with rapider and reliable mode monitoring accident on highway.
Background technology
According to dictionary, accident means and exists fault, and more particularly, accident means " fault of improper generation on highway, or the fault of the traffic capacity of all reduction highways are as traffic accident, vehicle trouble or parking, the barrier that falls and maintenance job ".
If this accident has taken place on highway, just need, but up to now promptly with the accident declaration traffic control center, these accidents are normally by knowing through the driver's of the vehicle of the scene of the accident report.
Therefore, the control of vehicle and dredge and incured loss through delay runs into very big difficulty in many vehicles are long-time.
Particularly, in the high country of the distribution cost resemble the Korea S, the generation of this accident can cause serious problem.
Consider these situations, the someone has proposed and can whether have the accidence monitoring method of accidents caused vehicle by monitoring the highway inspection recently.
Fig. 1 is the exemplary plot of the method that is used for the monitoring accident of explanation prior art.
That is, Fig. 1 has shown the screen-picture that obtains by the camera that is installed in the tunnel.The method of this monitoring accident proposes in Australia.
With reference to Fig. 1, from the screen highway three tracks 3 and walkway 1 are arranged, and, a plurality of box traps (box type trAP) 7 are arranged respectively on track 3 and walkway 1, a box trap 7 comprises a plurality of pixels simultaneously.To depend on the circumstances and what pixels a box trap comprises.
Generally, can moving by 7 monitorings of the box trap on screen, set 3 vehicles that move 5 and this vehicle 5 along the track.
That is, be included in along the track gray-scale value that picture that each pixel in the 3 a plurality of box traps that are provided with 7 have the relevant highway of taking from the camera (not shown) obtains.
At this moment, box trap 7 is detected the variation of gray-scale value in ground, unit one by one, thereby detect vehicle 5 and whether move or stop.
Camera is taken the highway image in real time and captured image is offered the traffic control center in real time.
Therefore, if box trap 7 is followed the trail of gray-scale value in ground, unit one by one, just can detect accident vehicle from the seasonal effect in time series angle.
At this moment, if gray-scale value changes in time, it is mobile that vehicle 5 is considered to, and when not variation in time of gray-scale value, vehicle 5 is considered to stop.
If detect accident, can produce predetermined alerting signal and take to tackle the measure of this accident by aforesaid program.
But the method that is used for the monitoring accident of prior art has been used the gray-scale value of a large amount of pixels for the detection accident, needs a large amount of calculation step when the detection accident, has consumed many times.
Simultaneously, the method that is used for the monitoring accident of prior art around environment do not have can differentiate exactly in the place that changes such as the tunnel vehicle on own route.
But and the tunnel is different, common surrounding environment each constantly all in the place that changes, the method for prior art will differentiate exactly that vehicle is inconvenient.
In other words, the shade of many non-vehicles such as shade tree or street lamp is usually arranged at the periphery of highway, or the shade of vehicle, these shadow falls are mapped on the track, thereby produce these are projected the problem that shade misidentification on the track is made vehicle.
Therefore, can not differentiate vehicle exactly, this may produce serious problem aspect the reliability of the method for this monitoring accident.
Summary of the invention
The objective of the invention is to solve the above problems at least and/or shortcoming, and the advantage that describes below is provided at least.
Therefore, an object of the present invention is by providing a kind of line style trap that utilizes to solve aforesaid problem with the method for the accident of mode detection faster.
Another object of the present invention provides and a kind ofly can prevent to differentiate mistakenly that vehicle improves the method for the detection accident of reliability by using gray value information.
By providing a kind of method that detects accident to realize aforesaid and other purpose and advantage, this method may further comprise the steps: the presumptive area from highway obtains an images; Calculate gray-scale value from the image that obtained corresponding to each pixel of predetermined line style trap; The variation conversion that reaches by calculating gray-scale value in the preset time section determines whether to exist accident.
According to another aspect of the present invention, the method for detection accident may further comprise the steps: the presumptive area from highway obtains an images; Calculate gray-scale value from the image that obtained corresponding to each pixel of predetermined line style trap; Use the variable quantity of the gray-scale value that calculates to catch vehicle; The gray-scale value that reaches by follow the tracks of institute's seizure vehicle in the preset time section determines whether to exist accident.
Description of drawings
Above-mentioned purpose of the present invention, feature and advantage will become clearer by following detailed description in conjunction with the accompanying drawings, wherein:
Fig. 1 is the exemplary plot of the screen of the accidence monitoring method of explanation prior art;
Fig. 2 is the exemplary plot of explanation according to the screen of the accidence monitoring method of the preferred embodiments of the present invention;
Fig. 3 a and Fig. 3 b are for showing the chart according to the frequency of the vehicle of the preferred embodiments of the present invention and non-vehicle.
Fig. 4 is the process flow diagram of explanation according to the accidence monitoring method of the preferred embodiments of the present invention.
Embodiment
Following detailed description is introduced the accidence monitoring method according to the preferred embodiments of the present invention with reference to the accompanying drawings.
Fig. 2 is the exemplary plot of explanation according to the screen of the accidence monitoring method of the preferred embodiments of the present invention.
Figure 2 shows that the screen of the image that demonstration obtains by the camera that is installed on the highway.
In this screen, there is track 14 both sides, and the centre is a central partition 16,14 somebody trade, the outside or buildings in the track, and simultaneously, vehicle 13 moves on track 14 respectively.
In addition, the vehicle 13 in also have moving and be incident upon shade 15 on each track 14 by sunlight.
On the screen of the image that obtains by aforementioned program, form predefined line style trap 11.
This line style trap 11 is arranged on each track 14 along the direct of travel of vehicle 13.Certainly, as long as trap 11 is positioned on the track 14, can just line style trap 11 be arranged on the track 14 regardless of the position in track 14.At this moment, line style trap 11 is configured to parallel with track 14 a queue of.
The present invention uses the gray-scale value detection accident of the relevant pixel of the line style trap 11 that is provided with in such a way, and at this moment, each pixel should cover line style trap 11, and promptly the gray-scale value of other pixel on line style trap 11 has not been excluded.
As mentioned above, according to prior art, setting be the box trap that comprises numerous pixels, and the gray-scale value of these pixels is used to the detection accident, therefore needs considerable arithmetic operation.
But the present invention only considers to be included in pixel on the line style trap 11 and the detection accident, so the mode of detection accident is quicker.
Although be arranged in the gray-scale value of the pixel of row on this line style trap 11 generally is accurate, because the gray-scale value of some other other pixel of factor possibility and out of true.
For addressing this problem, the present invention calculates the mean value of a certain pixel and the gray-scale value of the pixel of the predetermined number that is present in these pixel front and back, specify the representative gray-scale value of the pixel gray-scale value of this calculating, calculate in such a way at the representative gray-scale value that is arranged in all pixels on the line style trap 11 for relevant pixel.
For example, suppose that pixel 1, pixel 2, pixel 3, pixel 4, pixel 5, pixel 6, pixel 7 are included on the line style trap 11, and each pixel there is gray-scale value.Consider pixel 1, the mean value of each gray-scale value of calculating pixel 1 and pixel 2 also can be appointed as the gray-scale value that calculates the gray-scale value of pixel 1.
Equally, the mean value of each gray-scale value of calculating pixel 1, pixel 2, pixel 3, thereby the representative gray-scale value of pixel 2 is calculated.In this way, 7 representative gray-scale value is calculated from pixel 1 to pixel.
By the analysis to the variation conversion of each representative gray-scale value of calculating in such a way, associated vehicle 13 is identified, and can determine whether the vehicle of being discerned accident has taken place.
In Fig. 2, No. 3 vehicles are identified and have caused accident, thereby are identified black line style trap 19 of front and back mark of the vehicle 17 that has caused accident at this.
Calculate the representative gray-scale value of each pixel in such a way, the precision of gray-scale value can be further improved thus.
In addition, if the vehicle 13 in moving is identified by the line style trap 11 on the screen, corresponding to the mark 12 of relevant vehicle 13 by perpendicular to line style trap 11 marks.
Simultaneously, the shade 15 of vehicle 13 or the shade 15 of shade tree and street lamp are arranged on screen.
If these shades 15 may be taken shade as vehicle on the track by mistake.
For preventing this mistake, the present invention has the gray value information of predefined each vehicle and shade.
This gray value information is shown in Fig. 3 a and Fig. 3 b.
Here, Fig. 3 a has shown the gray value information of vehicle.Generally, vehicle has various lights by the many parts reflections that are present in vehicle itself, therefore exists various gray-scale values,, exist very wide gray-scale value scope from high to extremely low, thereby the frequency of each gray-scale value is relatively low that is.
Fig. 3 b has shown the gray value information of shade.Generally, shade has similar substantially gray-scale value in each zone, and therefore, to compare gray-value variation less with vehicle, but frequency is higher relatively.
Therefore, under the situation of vehicle, the amplitude broad of gray-value variation but frequency is relatively low, and under the situation of shade, the amplitude of gray-value variation is narrow but frequency is higher relatively.
If this gray value information is known in advance, can distinguish relevant gray-scale value by the more current gray value information that is identified as the zone of vehicle with predefined gray value information is real vehicle or shade.
Fig. 4 is the process flow diagram of explanation according to the accidence monitoring method of the preferred embodiments of the present invention.
With reference to Fig. 4, the first step uses camera to obtain an images (S21) from the presumptive area on the highway, and this camera is installed on the crossing of urban district or highway.
If obtained image in such a way, line style trap (S22) is set on the basis of the image that is obtained, at this moment,, also can set in advance the line style trap if camera obtains image from same presumptive area termly.
This line style trap preferably is arranged on the track abreast with the track.
In addition, the image that is obtained can pass through predefined screen display, and whether the operator also can accident take place by the visual visual detection vehicle that shows in such a way.
Certainly, the objective of the invention is to use the variation transition detection car accident of gray-scale value of the image of acquisition, rather than visual detection car accident in such a way.
At this moment, if the line style trap is set, calculate the gray-scale value (S23) corresponding to each pixel of the line style trap of setting, here, calculating is meant obtains the gray-scale value that drops on the pixel on the line style trap in the gray-scale value that obtains on the camera picture shot.
If calculate the gray-scale value of each pixel in such a way, be the precision of the gray-scale value that ensures each pixel, calculate the representative gray-scale value (S24) of each pixel of each presumptive area.
As mentioned above, calculate the mean value of a certain pixel and the pixel gray-scale value of the predetermined number that is present in these pixel front and back, and this mean value that calculates is designated as the representative gray-scale value of this pixel.
Consider next pixel, calculate the mean value of this next pixel and the pixel gray-scale value of the predetermined number that is present in these next pixel front and back in a similar manner, and the mean value that this calculates is appointed as the representative gray-scale value of this next pixel.By this method, the representative gray-scale value that is included in all pixels in the line style trap is calculated.
Catch vehicle (S25) by the variable quantity that uses the average gray value that calculates in such a way.That is, the analysis that is present in the gray-scale value of each pixel on the line style trap discloses, automobile storage point and do not have automobile storage point between gray-scale value be different.If a bit there is the gray-value variation of this mode in certain, just think that relevant point has automobile storage to exist.
If follow the trail of vehicle in such a way, the gray value information and the predefined gray value information of the vehicle of being caught compared, thereby determine whether it is real vehicle (S26).
Here, gray value information is represented the amplitude of variation and the frequency of gray-scale value.
As mentioned above, vehicle is different (with reference to Fig. 3 a and Fig. 3 b) with its gray value information of shade that is not vehicle.
Can determine by using this different gray value information whether the vehicle of current seizure is real vehicle.
Promptly, result according to the comparison of the gray value information of the vehicle that is captured and predefined gray value information, if the gray value information of the vehicle that is captured conforms to the gray value information of predefined vehicle, the vehicle that then is captured is confirmed to be real vehicle.
On the contrary, if the gray value information of the vehicle that is captured conforms to predefined shade gray value information, the vehicle that then is captured is confirmed to be shade.
If be confirmed to be real vehicle by step S26 vehicle, judge whether that then the vehicle of being caught has stopped the time of predetermined length (S27), this judgement can not change in the preset time section by the gray-scale value of checking the vehicle of whether being caught at an easy rate and carries out.
That is, if the gray-scale value of the vehicle of being caught does not change in the preset time section, the vehicle of being caught is considered to stop always, and the possibility that associated vehicle breaks down is very high.
On the contrary, if the gray-scale value of the vehicle of being caught continues to change in the preset time section, it is mobile that the vehicle of being caught is considered to, and vehicle may be normal vehicle.
To judge whether that in such a way time that vehicle has stopped predetermined length is the normal vehicle misidentification of temporary parking to be made to cause the mistake of the vehicle of accident in order preventing in advance, as press the vehicle that traffic lights are indicated temporary parking.
Therefore, if the vehicle that the result caught judged of S27 keeps stopping time of predetermined length set by step, the vehicle of being followed the trail of is identified and has caused accident (S28).
If vehicle is identified and has caused accident in such a way, cause that on screen the front and back of the vehicle 17 of accident form black line style trap 19 (with reference to Fig. 2).
Illustrate clearly that as the front this accidence monitoring method only uses the gray-scale value detection accident of the relevant pixel on the line style trap, thus rapider than the method detection accident of the box trap of prior art.
In addition, this accidence monitoring method uses the variable quantity and the frequency detection accident of gray-scale value, thereby has prevented from the shade misidentification is done the mistake of vehicle, may can reach high reliability on the detection accident.
Though the present invention shows with reference to its specific preferred embodiment and illustrates, it will be understood by those skilled in the art that and can carry out the change on various forms and the details and not break away from the spirit and scope of the present invention that appended claim is determined it.
Aforesaid embodiment and advantage are exemplary, should not be regarded as limitation of the present invention.Principle of the present invention can be applied on the equipment of other type at an easy rate.Instructions of the present invention is illustrative, does not limit the scope of the claims.Many replacements, modifications and variations are conspicuous for these those skilled in the art.In the claims, the bar item that device adds function is in order to cover the structure of the function that execution described here puies forward, and is not only the equivalence of structure and is the structure of equivalence.
Claims (15)
1, be used for the method for detection accident on the highway, may further comprise the steps:
Presumptive area from highway obtains an images;
Calculate gray-scale value from the image that obtained corresponding to each pixel of predetermined line style trap; And
Determine whether to exist accident by the variation conversion of calculating gray-scale value in the preset time section.
2, the method for claim 1 further is included in the step that shows the image that obtains on the display screen.
3, the method for claim 1 is characterized in that, this line style trap is arranged on the track.
4, the method for claim 1 is characterized in that, the gray-scale value that is calculated is the mean value of a certain pixel and the gray-scale value of the pixel of the predetermined number that is present in these pixel front and back.
5, the method for claim 1 is characterized in that, the line style trap of this setting comprises by the row arranged picture.
6, be used for the method for detection accident on the highway, may further comprise the steps:
Presumptive area from highway obtains an images;
Calculate gray-scale value from the image that obtained corresponding to each pixel of predetermined line style trap;
Use the variable quantity of the gray-scale value that calculates to catch vehicle; And
Determine whether to exist accident by the gray-scale value of in the preset time section, following the tracks of institute's seizure vehicle.
7, method as claimed in claim 6 further is included in the step that shows the image that obtains on the display screen.
8, method as claimed in claim 6 is characterized in that, this line style trap is arranged on the track.
9, method as claimed in claim 6 is characterized in that, the gray-scale value that is calculated is the mean value of a certain pixel and the gray-scale value of the pixel of the predetermined number that is present in these pixel front and back.
10, method as claimed in claim 6 is characterized in that, the line style trap of this setting comprises by the row arranged picture.
11, method as claimed in claim 6 further comprises by relatively being included in corresponding to the gray value information of gray value information in the line style trap of the vehicle of being followed the trail of and predefined real vehicles and determines whether a vehicle is real vehicle.
12, method as claimed in claim 11 is characterized in that, if gray value information is consistent with the gray value information of predefined real vehicles, the vehicle of being caught is confirmed to be real vehicle.
13, method as claimed in claim 11 is characterized in that, gray value information is the amplitude and the frequency of gray-value variation.
14, method as claimed in claim 11 is characterized in that, if the vehicle of being caught is confirmed to be real vehicle, corresponding to the vehicle of being confirmed the relevant vehicle on the display screen is done a mark.
15, method as claimed in claim 6 is characterized in that, if the gray-scale value of the vehicle of being caught changes at preset time Duan Wei, the vehicle of being caught is confirmed to be and has caused accident.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR79442/2002 | 2002-12-13 | ||
| KR1020020079442A KR20040051777A (en) | 2002-12-13 | 2002-12-13 | Noticing method of vehicle-trouble |
| KR79443/2002 | 2002-12-13 | ||
| KR1020020079443A KR20040051778A (en) | 2002-12-13 | 2002-12-13 | Noticing method of vehicle-trouble |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1506874A true CN1506874A (en) | 2004-06-23 |
Family
ID=34277785
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA031566227A Pending CN1506874A (en) | 2002-12-13 | 2003-09-05 | Road accidence monitoring method |
| CNA031571530A Pending CN1506919A (en) | 2002-12-13 | 2003-09-16 | Traffic accidence monitoring method |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA031571530A Pending CN1506919A (en) | 2002-12-13 | 2003-09-16 | Traffic accidence monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN1506874A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108597217A (en) * | 2018-05-07 | 2018-09-28 | 郑州市交通规划勘察设计研究院 | A kind of expressway traffic accident monitoring reminding method |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105160867B (en) * | 2015-08-20 | 2017-06-16 | 南京安通杰科技实业有限公司 | Traffic message Forecasting Methodology |
| US11302125B2 (en) | 2019-07-30 | 2022-04-12 | Bendix Commercial Vehicle Systems Llc | Information-enhanced off-vehicle event identification |
| CN113870564B (en) * | 2021-10-26 | 2022-09-06 | 安徽百诚慧通科技股份有限公司 | Traffic jam classification method and system for closed road section, electronic device and storage medium |
-
2003
- 2003-09-05 CN CNA031566227A patent/CN1506874A/en active Pending
- 2003-09-16 CN CNA031571530A patent/CN1506919A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108597217A (en) * | 2018-05-07 | 2018-09-28 | 郑州市交通规划勘察设计研究院 | A kind of expressway traffic accident monitoring reminding method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1506919A (en) | 2004-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104520913B (en) | Vehicle environment identification device | |
| JP4702426B2 (en) | Vehicle detection device, vehicle detection program, and light control device | |
| EP2057583B1 (en) | Automatic high beam control method | |
| US8913128B2 (en) | Image-based barrier detection and warning system and method thereof | |
| JP4993322B2 (en) | Identification and classification of light spots around the vehicle by a camera | |
| JP7358762B2 (en) | Road anomaly detection device, road anomaly detection method, and road anomaly detection program | |
| CN101469985A (en) | Single-frame image detection apparatus for vehicle queue length at road junction and its working method | |
| RU2619059C2 (en) | Device for detecting three-dimensional objects and method for detecting three-dimensional objects | |
| CN102332209A (en) | Automobile violation video monitoring method | |
| CN101075376A (en) | Intelligent video traffic monitoring system based on multi-viewpoints and its method | |
| CN101042802A (en) | Traffic information sensor and method and system for traffic information detecting | |
| GB2511612A (en) | Apparatus and method for detecting vehicle weave | |
| CN112258668A (en) | Method for detecting roadside vehicle parking behavior based on high-position camera | |
| CN1448886A (en) | Apparatus and method for measuring vehicle queue length | |
| US9665783B2 (en) | Night parking detection | |
| CN105654734A (en) | Data processing method and device for vehicle violations | |
| JP4980970B2 (en) | Image pickup means adjustment device and object detection device | |
| JP2020035362A (en) | Image processing device, axle quantity detection system, fee setting device, fee setting system and program | |
| CN104008649B (en) | Radar tracking is utilized quickly to find the system and method for runway exception parking reason | |
| CN103886753A (en) | System and method for quickly confirming reasons for abnormal parking at signal lamp control intersection | |
| JPWO2011055418A1 (en) | Distance measuring device and distance measuring method | |
| CN101034497A (en) | Method and device for implementing classification of vehicle and bicycle | |
| CN106627416A (en) | Method, device and system for detecting road type | |
| CN1506874A (en) | Road accidence monitoring method | |
| CN201203419Y (en) | Single frame image detection device of city road crossing vehicle queue length |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |