CN111126237A - Safe vehicle distance detection method based on machine vision - Google Patents
Safe vehicle distance detection method based on machine vision Download PDFInfo
- Publication number
- CN111126237A CN111126237A CN201911314890.2A CN201911314890A CN111126237A CN 111126237 A CN111126237 A CN 111126237A CN 201911314890 A CN201911314890 A CN 201911314890A CN 111126237 A CN111126237 A CN 111126237A
- Authority
- CN
- China
- Prior art keywords
- distance
- vehicle
- safe
- license plate
- vehicle distance
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 63
- 238000012549 training Methods 0.000 claims abstract description 32
- 238000003384 imaging method Methods 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims abstract description 12
- 206010039203 Road traffic accident Diseases 0.000 claims abstract description 6
- 230000008859 change Effects 0.000 claims abstract description 6
- 238000013178 mathematical model Methods 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 18
- 238000013528 artificial neural network Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 230000000903 blocking effect Effects 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000037081 physical activity Effects 0.000 description 1
- 230000035484 reaction time Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/24—Reminder alarms, e.g. anti-loss alarms
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0965—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages responding to signals from another vehicle, e.g. emergency vehicle
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/123—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams
- G08G1/133—Traffic control systems for road vehicles indicating the position of vehicles, e.g. scheduled vehicles; Managing passenger vehicles circulating according to a fixed timetable, e.g. buses, trains, trams within the vehicle ; Indicators inside the vehicles or at stops
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a machine vision-based safe vehicle distance detection method, which comprises the following steps: a) building a training material library; b) manually labeling the picture processed in the step a); c) obtaining a license plate detection model; d) establishing a vehicle distance detection mathematical model according to a similar triangle principle in the camera imaging process; e) obtaining a vehicle distance detection model; f) establishing a piecewise function algorithm of the real-time relative speed and distance of the vehicle; g) algorithmic model deployment. The safe vehicle distance detection method based on the machine vision obtains the distance of the front vehicle through the pictures containing the license plate of the front vehicle, obtains the relative speed through the change of the vehicle distance in the two adjacent pictures, and sends out a warning signal when the relative speed is not matched with the vehicle distance so as to avoid traffic accidents caused by long-time fatigue driving and distraction of a driver.
Description
Technical Field
The invention relates to a safe vehicle distance detection method, in particular to a safe vehicle distance detection method based on machine vision.
Background
In the long-distance transportation process, a driver inevitably has the phenomenon of distraction and even fatigue driving. According to the related research of occupational safety and health, the problems of impaired enthusiasm, inattention, decreased thinking judgment capability and physical activity capability and the like often occur when people work in a fatigue state, so that the safety distance between people and a front vehicle cannot be judged in time, the best reaction time is missed, and the front vehicle is collided to cause a traffic accident. According to the research report of original Daimler-Benz automobile manufacturing company in Germany, the following results are shown: in a dangerous situation, if the driver is given a half second of response time ahead, 30% of rear-end accidents, 50% of road surface related information accidents, and 60% of head-on collisions can be reduced, respectively. In order to improve the safety performance of drivers of logistics vehicles and ensure safe operation, the invention provides an effective method for automatically detecting the vehicle distance by a machine vision technology. The camera installed on the vehicle is used for shooting images of the visual angle of a driver, the images are transmitted to the image processor in real time, warning signals are sent to the driver in a dangerous driving state possibly causing collision accidents, and the accident probability caused by long-time fatigue driving and distraction of the driver is reduced.
Disclosure of Invention
In order to overcome the defects of the technical problems, the invention provides a safe vehicle distance detection method based on machine vision.
The invention discloses a machine vision-based safe vehicle distance detection method, which is characterized by comprising the following steps of:
a) establishing a training material library, collecting pictures of front trucks and small-sized vehicles under various driving scenes by utilizing an early warning device arranged on the vehicles as training materials, wherein the collected pictures are used for ensuring that license plates of the vehicles are completely visible and processing the pictures into uniform squares; the early warning device comprises a camera, a power supply and a voice prompt module;
b) marking the picture, namely manually marking the picture processed in the step a), wherein the marking content is the region where the license plate is located and the license plate type in the picture, so as to obtain various license plate data sets of the vehicle in various driving scenes;
c) acquiring a license plate detection model, establishing a training model of the license plate detection network based on a deep learning convolution neural network and multi-frame target detection CAFFE-SSD, dividing the data set in the step b) into a training set, a testing set and a verification set, performing feature extraction processing on the training set data, and training according to the established neural network model to acquire a target detection model for license plate detection;
d) establishing a vehicle distance detection model, establishing a vehicle distance detection mathematical model according to a similar triangle principle in the camera imaging process, and adding the algorithm into the trained license plate detection model;
the distance between the camera of the early warning device and the front end of the vehicle is L, the actual distance of the vehicle is d, the imaging focal length of the camera is f, the actual width of the license plate is H, the pixel width of the license plate corresponding to the imaging sensor is H, and the vehicle can be obtained according to the similar triangle principle in the imaging process of the camera:
f/(d+L)＝h/H (1)
from equation (1):
d＝f*H/h-L (2)
e) obtaining a vehicle distance detection model, taking CAFFE-SSD as a training neural network, and adjusting parameters of the neural network in the training process to enable a loss function of the network to tend to converge in an iterative process, and finally forming a neural network target detection model for vehicle distance detection;
f) building a piecewise function algorithm of the real-time relative speed and distance of the vehicle, and adding the piecewise function algorithm into a distance detection model algorithm;
g) and deploying an algorithm model, namely deploying a trained vehicle distance detection model and a segmentation function algorithm of relative vehicle speed and vehicle distance, synchronously detecting in real time, and automatically giving an alarm by a voice prompt module when detecting that a driver neglects the safe vehicle distance due to fatigue driving so as to avoid traffic accidents.
In the step f), the vehicle distance is obtained by using a formula (2) in the step d) according to the imaging size of the license plate in the current picture; the relative speed is obtained according to the distance change corresponding to two adjacent frames of pictures, and the obtaining formula is as follows:
in the formula (3), v is the current relative vehicle speed, d2Distance of vehicles obtained from the current picture, d1The distance between vehicles obtained according to the previous picture; t is t2Time of acquisition, t, for the current picture1The time at which the previous picture was taken.
The invention relates to a safe vehicle distance detection method based on machine vision, wherein a piecewise function algorithm of the real-time relative vehicle speed and the vehicle distance of a vehicle, which is established in the step f), comprises the following steps:
when the relative speed is not more than 20km/h and the vehicle distance is less than 10m, the requirement of safe vehicle distance is not met;
when the relative speed is 20-30 km/h, and the distance is less than 15m, the requirement of safe distance is not met;
when the relative speed is 30-40 km/h, and the distance is less than 25m, the requirement of safe distance is considered not to be met;
when the relative speed is 40-50 km/h, and the distance is less than 35m, the requirement of safe distance is not met;
when the relative speed is 50-60 km/h, and the distance is less than 45m, the requirement of safe distance is not met;
when the relative speed is 60-70 km/h, and the distance is less than 65m, the requirement of safe distance is not met;
when the relative speed is 70-80 km/h, and the distance is less than 75m, the requirement of safe distance is not met;
when the relative speed is 80-90 km/h, and the distance is less than 85m, the requirement of safe distance is considered not to be met;
and when the relative speed is 90-100 km/h, and the distance is less than 95m, the requirement of the safe distance is not met.
The invention discloses a machine vision-based safe vehicle distance detection method, wherein multiple driving scenes in the step a) comprise a clear day, a cloudy day, a rainy day, and vehicle plate inclination, shadow or stain shielding, and a picture data set acquired in the clear day is S ═ S { S }1,s2,s3,.. }, the picture data set acquired on cloudy days is C ═ C1,c2,c3,.. }, the picture data set acquired in rainy days is R ═ R }1,r2,r3,.. }, the picture data set of the inclination, shadow or stain blocking of the license plate is D ═ D }1,d2,d3,.. }; the pictures are processed to a uniform square size of 500 pixels by 500 pixels.
In the method for detecting the safe vehicle distance based on the machine vision, in the step b), the types of the license plates in the marked picture comprise blue bottom white characters, yellow bottom black characters, green bottom black characters and white bottom black character types, the corresponding size of the license plates of the blue bottom white characters and the white bottom black characters is 440mm multiplied by 140mm, the corresponding size of the license plates of the yellow bottom black characters is 440mm multiplied by 220mm, and the corresponding size of the license plates of the green bottom black characters is 480mm multiplied by 220 mm.
The invention has the beneficial effects that: the invention relates to a safe vehicle distance detection method based on machine vision, which comprises the steps of firstly utilizing a vehicle-mounted early warning device to collect images of a front vehicle in scenes such as sunny days, cloudy days, rainy days and the like, forming a data set after marking and processing, then establishing and training a target detection model for detecting a license plate, adding a vehicle distance detection mathematical model established according to a similar triangle principle, then establishing a piecewise function algorithm of the relative vehicle speed and the vehicle distance of the vehicle, finally deploying the algorithm model into the vehicle-mounted early warning device, obtaining the distance of the front vehicle through the images containing the license plate of the front vehicle acquired in real time, obtaining the relative vehicle speed through the change of the vehicle distance in two adjacent frames of images, and sending out a warning signal when the relative vehicle speed is not matched with the vehicle distance so as to avoid traffic accidents caused by long-time fatigue driving and distraction of a driver.
Drawings
FIG. 1 is a schematic diagram of a similar triangle for license plate imaging in the present invention;
FIG. 2 is a schematic view of a picture including a license plate of a preceding vehicle collected in the present invention;
fig. 3 is a flowchart of a safety inter-vehicle distance detection method based on machine vision according to the present invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
As shown in fig. 3, a flowchart of the safety distance detection method based on machine vision according to the present invention is provided, which is implemented by the following steps:
a) establishing a training material library, collecting pictures of front trucks and small-sized vehicles under various driving scenes by utilizing an early warning device arranged on the vehicles as training materials, wherein the collected pictures are used for ensuring that license plates of the vehicles are completely visible and processing the pictures into uniform squares; the early warning device comprises a camera, a power supply and a voice prompt module;
in this step, the multiple driving scenes include a clear day, a cloudy day, a rainy day, and vehicle plate inclination, shadow or stain shielding, and the image data set acquired in the clear day is S ═ S1,s2,s3,.. }, the picture data set acquired on cloudy days is C ═ C1,c2,c3,.. }, the picture data set acquired in rainy days is R ═ R }1,r2,r3,.. }, the picture data set of the inclination, shadow or stain blocking of the license plate is D ═ D }1,d2,d3,.. }; the pictures are processed to a uniform square size of 500 pixels by 500 pixels. As shown in fig. 2, a schematic diagram of a picture containing a license plate of a front vehicle collected in the present invention is shown.
b) Marking the picture, namely manually marking the picture processed in the step a), wherein the marking content is the region where the license plate is located and the license plate type in the picture, so as to obtain various license plate data sets of the vehicle in various driving scenes;
c) acquiring a license plate detection model, establishing a training model of the license plate detection network based on a deep learning convolution neural network and multi-frame target detection CAFFE-SSD, dividing the data set in the step b) into a training set, a testing set and a verification set, performing feature extraction processing on the training set data, and training according to the established neural network model to acquire a target detection model for license plate detection;
d) establishing a vehicle distance detection model, establishing a vehicle distance detection mathematical model according to a similar triangle principle in the camera imaging process, and adding the algorithm into the trained license plate detection model;
as shown in fig. 1, a schematic diagram of a similar triangle for license plate imaging in the present invention is given:
the distance between the camera of the early warning device and the front end of the vehicle is L, the actual distance of the vehicle is d, the imaging focal length of the camera is f, the actual width of the license plate is H, the pixel width of the license plate corresponding to the imaging sensor is H, and the vehicle can be obtained according to the similar triangle principle in the imaging process of the camera:
f/(d+L)＝h/H (1)
from equation (1):
d＝f*H/h-L (2)
e) obtaining a vehicle distance detection model, taking CAFFE-SSD as a training neural network, and adjusting parameters of the neural network in the training process to enable a loss function of the network to tend to converge in an iterative process, and finally forming a neural network target detection model for vehicle distance detection;
in this step, the parameters defined by the neural network model include: class total amount Class total, rectangular frame Size, Learning Rate, Weight Attenuation Rate Weight, training by using images in a test sample, and outputting a training log; and calculating the change of the accuracy by using the test sample, and performing cross validation to finish training.
f) Building a piecewise function algorithm of the real-time relative speed and distance of the vehicle, and adding the piecewise function algorithm into a distance detection model algorithm;
in the step, the distance between vehicles is obtained by using a formula (3) in the step d) according to the imaging size of the license plate in the current picture; the relative speed is obtained according to the distance change corresponding to two adjacent frames of pictures, and the obtaining formula is as follows:
in the formula (3), v is the current relative vehicle speed, d2Distance of vehicles obtained from the current picture, d1The distance between vehicles obtained according to the previous picture; t is t2Time of acquisition, t, for the current picture1The time at which the previous picture was taken.
In the established piecewise function algorithm of the real-time relative speed and the distance of the vehicle:
when the relative speed is not more than 20km/h and the vehicle distance is less than 10m, the requirement of safe vehicle distance is not met; when the relative speed is 20-30 km/h, and the distance is less than 15m, the requirement of safe distance is not met; when the relative speed is 30-40 km/h, and the distance is less than 25m, the requirement of safe distance is considered not to be met; when the relative speed is 40-50 km/h, and the distance is less than 35m, the requirement of safe distance is not met; when the relative speed is 50-60 km/h, and the distance is less than 45m, the requirement of safe distance is not met; when the relative speed is 60-70 km/h, and the distance is less than 65m, the requirement of safe distance is not met; when the relative speed is 70-80 km/h, and the distance is less than 75m, the requirement of safe distance is not met; when the relative speed is 80-90 km/h, and the distance is less than 85m, the requirement of safe distance is considered not to be met; and when the relative speed is 90-100 km/h, and the distance is less than 95m, the requirement of the safe distance is not met.
g) And deploying an algorithm model, namely deploying a trained vehicle distance detection model and a segmentation function algorithm of relative vehicle speed and vehicle distance, synchronously detecting in real time, and automatically giving an alarm by a voice prompt module when detecting that a driver neglects the safe vehicle distance due to fatigue driving so as to avoid traffic accidents.
During the specific training process, a total of 11210 pictures were collected by step a), in a 7: 2: the proportion of 1 is divided into a training set, a testing set and a verification set, the iteration frequency is not lower than 55000 times, each iteration is 8 pictures, and the current loss value and the accuracy rate are calculated through the verification set every 10 times of iteration. And finally, the accuracy rate of the train distance detection model on the test set after training is over 95 percent.
The piecewise function of the real-time safe vehicle speed and the vehicle distance mentioned in the step f) is based on the implementation regulation of the road traffic safety of the people's republic of China, wherein when the visibility is less than 200 meters, the vehicle speed is not more than 60 kilometers per hour, and the distance between the vehicle speed and a vehicle ahead of the same lane is kept more than 100 meters; when the visibility is less than 100 meters, the vehicle speed can not exceed 40 kilometers per hour, and the distance between the vehicle speed and the vehicle in front of the same lane is kept more than 50 meters. The specific relationship is shown in the following table.
When the real-time speed is not matched with the safe distance, the voice alarm is automatically triggered to remind a driver, and the speed is reduced to ensure the safe distance.
Claims (5)
1. A safe vehicle distance detection method based on machine vision is characterized by comprising the following steps:
a) establishing a training material library, collecting pictures of front trucks and small-sized vehicles under various driving scenes by utilizing an early warning device arranged on the vehicles as training materials, wherein the collected pictures are used for ensuring that license plates of the vehicles are completely visible and processing the pictures into uniform squares; the early warning device comprises a camera, a power supply and a voice prompt module;
b) marking the picture, namely manually marking the picture processed in the step a), wherein the marking content is the region where the license plate is located and the license plate type in the picture, so as to obtain various license plate data sets of the vehicle in various driving scenes;
c) acquiring a license plate detection model, establishing a training model of the license plate detection network based on a deep learning convolution neural network and multi-frame target detection CAFFE-SSD, dividing the data set in the step b) into a training set, a testing set and a verification set, performing feature extraction processing on the training set data, and training according to the established neural network model to acquire a target detection model for license plate detection;
d) establishing a vehicle distance detection model, establishing a vehicle distance detection mathematical model according to a similar triangle principle in the camera imaging process, and adding the algorithm into the trained license plate detection model;
the distance between the camera of the early warning device and the front end of the vehicle is L, the actual distance of the vehicle is d, the imaging focal length of the camera is f, the actual width of the license plate is H, the pixel width of the license plate corresponding to the imaging sensor is H, and the vehicle can be obtained according to the similar triangle principle in the imaging process of the camera:
f/(d+L)＝h/H (1)
from equation (1):
d＝f*H/h-L (2)
e) obtaining a vehicle distance detection model, taking CAFFE-SSD as a training neural network, and adjusting parameters of the neural network in the training process to enable a loss function of the network to tend to converge in an iterative process, and finally forming a neural network target detection model for vehicle distance detection;
f) building a piecewise function algorithm of the real-time relative speed and distance of the vehicle, and adding the piecewise function algorithm into a distance detection model algorithm;
g) and deploying an algorithm model, namely deploying a trained vehicle distance detection model and a segmentation function algorithm of relative vehicle speed and vehicle distance, synchronously detecting in real time, and automatically giving an alarm by a voice prompt module when detecting that a driver neglects the safe vehicle distance due to fatigue driving so as to avoid traffic accidents.
2. The machine-vision-based safe vehicle distance detection method according to claim 1, characterized in that: in the step f), the distance between vehicles is obtained by using the formula (2) in the step d) according to the imaging size of the license plate in the current picture; the relative speed is obtained according to the distance change corresponding to two adjacent frames of pictures, and the obtaining formula is as follows:
in the formula (3), v is the current relative vehicle speed, d2Distance of vehicles obtained from the current picture, d1The distance between vehicles obtained according to the previous picture; t is t2Time of acquisition, t, for the current picture1The time at which the previous picture was taken.
3. The machine vision-based safe vehicle distance detection method according to claim 1 or 2, wherein the step f) is implemented by a piecewise function algorithm of real-time relative vehicle speed and vehicle distance of the vehicle, wherein the piecewise function algorithm comprises the following steps:
when the relative speed is not more than 20km/h and the vehicle distance is less than 10m, the requirement of safe vehicle distance is not met;
when the relative speed is 20-30 km/h, and the distance is less than 15m, the requirement of safe distance is not met;
when the relative speed is 30-40 km/h, and the distance is less than 25m, the requirement of safe distance is considered not to be met;
when the relative speed is 40-50 km/h, and the distance is less than 35m, the requirement of safe distance is not met;
when the relative speed is 50-60 km/h, and the distance is less than 45m, the requirement of safe distance is not met;
when the relative speed is 60-70 km/h, and the distance is less than 65m, the requirement of safe distance is not met;
when the relative speed is 70-80 km/h, and the distance is less than 75m, the requirement of safe distance is not met;
when the relative speed is 80-90 km/h, and the distance is less than 85m, the requirement of safe distance is considered not to be met;
and when the relative speed is 90-100 km/h, and the distance is less than 95m, the requirement of the safe distance is not met.
4. A machine vision based safe vehicle distance detection method according to claim 1 or 2, characterized in that: the multiple driving scenes in the step a) comprise sunny days, cloudy days, rainy days, license plate inclination, shadows or dirt blocking, and the image data set acquired in the sunny days is S ═ S1,s2,s3,.. }, the picture data set acquired on cloudy days is C ═ C1,c2,c3,.. }, the picture data set acquired in rainy days is R ═ R }1,r2,r3,.. }, the picture data set of the inclination, shadow or stain blocking of the license plate is D ═ D }1,d2,d3,.. }; the pictures are processed to a uniform square size of 500 pixels by 500 pixels.
5. A machine vision based safe vehicle distance detection method according to claim 1 or 2, characterized in that: in the picture marking in the step b), the marked types of the license plates in the picture comprise blue bottom white characters, yellow bottom black characters, green bottom black characters and white bottom black characters, the corresponding size of the license plates with the blue bottom white characters and the white bottom black characters is 440mm multiplied by 140mm, the corresponding size of the license plates with the yellow bottom black characters is 440mm multiplied by 220mm, and the corresponding size of the license plates with the green bottom black characters is 480mm multiplied by 220 mm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911314890.2A CN111126237A (en) | 2019-12-19 | 2019-12-19 | Safe vehicle distance detection method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911314890.2A CN111126237A (en) | 2019-12-19 | 2019-12-19 | Safe vehicle distance detection method based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111126237A true CN111126237A (en) | 2020-05-08 |
Family
ID=70500082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911314890.2A Pending CN111126237A (en) | 2019-12-19 | 2019-12-19 | Safe vehicle distance detection method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111126237A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111746545A (en) * | 2020-06-29 | 2020-10-09 | 中国联合网络通信集团有限公司 | Vehicle distance detection method and device and vehicle distance reminding method and device |
| CN111967780A (en) * | 2020-08-19 | 2020-11-20 | 江苏经纬智联航空科技有限公司 | Method and system for supervising special vehicle operation process by means of airplane in airport |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201410950Y (en) * | 2009-04-10 | 2010-02-24 | 楚伟峰 | Intelligent brake device |
| CN101750049A (en) * | 2008-12-05 | 2010-06-23 | 南京理工大学 | Monocular vision vehicle distance measuring method based on road and vehicle information |
| CN105574552A (en) * | 2014-10-09 | 2016-05-11 | 东北大学 | Vehicle ranging and collision early warning method based on monocular vision |
| CN109509223A (en) * | 2018-11-08 | 2019-03-22 | 西安电子科技大学 | Front vehicles distance measuring method based on deep learning |
-
2019
- 2019-12-19 CN CN201911314890.2A patent/CN111126237A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101750049A (en) * | 2008-12-05 | 2010-06-23 | 南京理工大学 | Monocular vision vehicle distance measuring method based on road and vehicle information |
| CN201410950Y (en) * | 2009-04-10 | 2010-02-24 | 楚伟峰 | Intelligent brake device |
| CN105574552A (en) * | 2014-10-09 | 2016-05-11 | 东北大学 | Vehicle ranging and collision early warning method based on monocular vision |
| CN109509223A (en) * | 2018-11-08 | 2019-03-22 | 西安电子科技大学 | Front vehicles distance measuring method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 姚春莲;冯胜男;张芳芳;司慧琳: "利用车牌面积进行车距测量的研究", 系统仿真学报, vol. 29, no. 11, pages 2821 - 2825 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111746545A (en) * | 2020-06-29 | 2020-10-09 | 中国联合网络通信集团有限公司 | Vehicle distance detection method and device and vehicle distance reminding method and device |
| CN111967780A (en) * | 2020-08-19 | 2020-11-20 | 江苏经纬智联航空科技有限公司 | Method and system for supervising special vehicle operation process by means of airplane in airport |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3566903B1 (en) | Method and apparatus for vehicle position detection | |
| DE102009012917B4 (en) | Obstacle detection device for vehicles | |
| Tuohy et al. | Distance determination for an automobile environment using inverse perspective mapping in OpenCV | |
| DE102009017431B4 (en) | Driving assistance system for vehicles | |
| CN109190523B (en) | Vehicle detection tracking early warning method based on vision | |
| CN105512623A (en) | Foggy-day driving visual enhancement and visibility early warning system and method based on multiple sensors | |
| CN202071799U (en) | Intelligent automobile safety warning device | |
| CN105047019B (en) | A kind of passenger stock prevent rear car overtake other vehicles after lane change determination methods and device suddenly | |
| CN201825037U (en) | Lane departure alarm device for vehicles on highway | |
| JP2006184276A (en) | All-weather obstacle collision preventing device by visual detection, and method therefor | |
| CN103837139A (en) | Rough road surface driving assisted equipment and method for rough road driving assisting | |
| CN105206109A (en) | Infrared CCD based foggy day identifying early-warning system and method for vehicle | |
| CN107891808A (en) | Driving based reminding method, device and vehicle | |
| DE112020002567T5 (en) | IMAGE RECOGNITION DEVICE AND IMAGE RECOGNITION METHOD | |
| EP2819115B1 (en) | Three-dimensional object detection device | |
| CN111126237A (en) | Safe vehicle distance detection method based on machine vision | |
| CN107832788B (en) | Vehicle distance measuring method based on monocular vision and license plate recognition | |
| CN114419143A (en) | Depth measuring method, depth measuring apparatus, and storage medium | |
| CN114119955A (en) | Method and device for detecting potential dangerous target | |
| CN116142186A (en) | Early warning method, device, medium and equipment for safe running of vehicle in bad environment | |
| EP3815044B1 (en) | Method for sensor and memory-based depiction of an environment, display apparatus and vehicle having the display apparatus | |
| Miclea et al. | Visibility detection in foggy environment | |
| CN114659489A (en) | Front vehicle distance detection method and device based on convex lens imaging principle | |
| CN102314603B (en) | Intellectualized automobile safety attention device | |
| DE102011055441A1 (en) | Method for determining spacing between preceding and forthcoming motor cars by using mono camera in e.g. adaptive cruise control system, involves determining spacing between cars based on information about license plate number |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200508 |