CN106696825A - Driving assisting method and system - Google Patents
Driving assisting method and system Download PDFInfo
- Publication number
- CN106696825A CN106696825A CN201610997607.0A CN201610997607A CN106696825A CN 106696825 A CN106696825 A CN 106696825A CN 201610997607 A CN201610997607 A CN 201610997607A CN 106696825 A CN106696825 A CN 106696825A
- Authority
- CN
- China
- Prior art keywords
- value
- image information
- input value
- target
- crucial subregion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000002265 prevention Effects 0.000 claims abstract description 13
- 230000006870 function Effects 0.000 claims description 60
- 238000005457 optimization Methods 0.000 claims description 54
- 238000012545 processing Methods 0.000 claims description 27
- 238000012544 monitoring process Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 abstract description 10
- 238000005192 partition Methods 0.000 abstract 2
- 230000008859 change Effects 0.000 description 3
- 230000005484 gravity Effects 0.000 description 3
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000034994 death Effects 0.000 description 1
- 231100000517 death Toxicity 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R1/00—Optical viewing arrangements; Real-time viewing arrangements for drivers or passengers using optical image capturing systems, e.g. cameras or video systems specially adapted for use in or on vehicles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T7/00—Brake-action initiating means
- B60T7/12—Brake-action initiating means for automatic initiation; for initiation not subject to will of driver or passenger
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/30—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of image processing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/80—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement
- B60R2300/8066—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the intended use of the viewing arrangement for monitoring rearward traffic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Transportation (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a driving assisting method and system. The method comprises the steps that when a vehicle is detected to be in a driving state, more than one image information corresponding to the surrounding environment of the vehicle is obtained; aiming at each image information, the executed steps are that each image information is divided into a key partition of the first amount, and an input value of each key partition is determined; optimizing process is conducted on the input values of the first amount through a preset optimized function to generate reference values corresponding to the input values of the first amount; and whether the different values of the reference values and predetermined standard values are not larger than given threshold values or not is judged, and if the different values of the reference values and the predetermined standard values are not larger than the given threshold values, urgent danger prevention process is executed. Because the vehicle driving state can be judged through the surrounding environment of the vehicle, the driving assisting method and system are superior to the method that judgment is conducted by just simply relying on the distance and the speed, and the precision of avoiding accident occurrence can be improved.
Description
Technical field
The present invention relates to technical field of data processing, more particularly to a kind of method and system for aiding in driving.
Background technology
With the development of society, the improvement of people's living standards, automobile turns into product indispensable during people live.By
In in driving procedure, driver can adjust the distance, the judgement of speed etc. there is error, cause to take risk avoidance measures not in time, and
And when urgent danger prevention measure is taken with very big randomness, inappropriate urgent danger prevention measure is taken, causing originally can be with
The generation of the traffic accident for avoiding, so as to cause the injures and deaths for damaging even personnel of vehicle.
At present, some Driving assistant systems, by the distance between detection and front truck and relative velocity, judging distance
And whether relative velocity exceedes the threshold value of setting, after more than the threshold value for setting, warning is issued the user with accordingly or is carried out certainly
Dynamic brake.
But, existing implementation generally can not hazard recognition situation, such as the rapidly close feelings of rear automobile completely
Condition, therefore the accuracy to avoiding accident from occurring is relatively low.
The content of the invention
The invention provides a kind of method and system for aiding in and driving, the accuracy for avoiding accident from occurring can be improved.
In a first aspect, the invention provides a kind of method for aiding in and driving, the method includes：
When vehicle is detected in transport condition, the corresponding at least one image letter of surrounding environment of the vehicle is obtained
Breath；
For each described image information, it is performed both by：Described image information is divided into the crucial subregion of the first quantity, and
Determine the input value of each crucial subregion；
Treatment is optimized to the input value of the first quantity using the majorized function for pre-setting, with generate this first
The corresponding reference value of the input value of quantity；
Judge whether the reference value is not more than given threshold with the difference of predetermined standard value, if so, performing tight
Anxious hedging treatment.
It is preferably, described that treatment is optimized to the input value of the first quantity using the majorized function for pre-setting,
To generate the corresponding reference value of the input value of first quantity, including：
S1：For each the target input value in the input value of the first quantity, it is performed both by：Using what is pre-set
Each first majorized function in first majorized function of the first quantity, optimizes to the target input value respectively, with
Generate the corresponding objective optimization value of the target input value；Determine the corresponding target weight of the objective optimization value, and determine institute
The product that objective optimization value is stated with the target weight is target output value, to complete the suboptimization to the target input value
Treatment；
S2：Judge whether the number of times of the corresponding optimization processing of the target input value reaches given threshold, if so, performing
S3, otherwise, using the target output value as the target input value, and performs S1；
S3：According to the target output value of the first quantity determined, and predetermined second majorized function is utilized, meter
Calculate the corresponding reference value of target output value of first quantity.
Preferably, first majorized function, including：
Wherein, f (xi) for i-th optimal value of crucial subregion in the crucial subregion for characterizing first quantity；xiWith
In the input value for characterizing i-th key subregion；C is used to characterize the parameter of described image information；Radom is described for characterizing
The corresponding empirical value of i-th crucial subregion；
Second majorized function, including：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor table
Levy i-th default weight of crucial subregion in j-th image information；bjFor characterizing j-th image information
Threshold value；N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
Preferably, the input value for determining each crucial subregion, including：
For crucial subregion each described, the crucial subregion is carried out into gray processing treatment, and determine the crucial subregion
Including each pixel gray value, and determine the gray value of each pixel plus and be the crucial subregion
Input value.
Preferably, further include：
When the barycentre offset for monitoring the corresponding driver of the vehicle is more than the first threshold of respective settings, start
Brake hard.
Preferably, further include：
The normally travel direction of the vehicle is determined according to described image information, when the current driving for monitoring the vehicle
When angle between direction and the normally travel direction is more than the Second Threshold of respective settings, start brake hard.
Second aspect, the invention provides a kind of system for aiding in and driving, the system includes：Acquiring unit, determining unit,
Processing unit and judging unit, wherein,
The acquiring unit, for when vehicle is detected in transport condition, obtaining the surrounding environment pair of the vehicle
At least one image information answered；
The determining unit, for for each described image information, being performed both by：Described image information is divided into first
The crucial subregion of quantity, and determine the input value of each crucial subregion；
The processing unit, for being optimized to the input value of the first quantity using the majorized function for pre-setting
Treatment, to generate the corresponding reference value of the input value of first quantity；
The judging unit, for judging the reference value sets with whether the difference of predetermined standard value is not more than
Threshold value, if so, performing urgent danger prevention treatment.
Preferably, the processing unit, including optimization subelement, triggering subelement, computation subunit, wherein,
The optimization subelement, for each the target input value in the input value for the first quantity, holds
OK：Using each first majorized function in the first majorized function of the first quantity for pre-setting, respectively to the target
Input value is optimized, to generate the corresponding objective optimization value of the target input value；Determine that the objective optimization value is corresponding
Target weight, and determine that the objective optimization value is target output value with the product of the target weight, to complete to the mesh
Mark an optimization processing of input value；
Whether the triggering subelement, the number of times for judging the corresponding optimization processing of the target input value reaches setting
Threshold value, if so, triggering the computation subunit, otherwise, using the target output value as the target input value, and triggers institute
State optimization subelement；
The computation subunit, for the target output value according to the first quantity determined, and using predetermined
The second majorized function, calculate the corresponding reference value of target output value of first quantity.
Preferably, first majorized function, including：
Wherein, f (xi) for i-th optimal value of crucial subregion in the crucial subregion for characterizing first quantity；xiWith
In the input value for characterizing i-th key subregion；C is used to characterize the parameter of described image information；Radom is described for characterizing
The corresponding empirical value of i-th crucial subregion；
Second majorized function, including：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor table
Levy i-th default weight of crucial subregion in j-th image information；bjFor characterizing j-th image information
Threshold value；N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
Preferably, the determining unit, specifically for for crucial subregion each described, the crucial subregion being carried out into ash
Degreeization treatment, and determine the gray value of each pixel that the crucial subregion includes, and determine described each pixel
Point gray value plus and be the input value of the crucial subregion.
Preferably, further include：First brake unit, for when the center of gravity for monitoring the corresponding driver of the vehicle
When side-play amount is more than the first threshold of respective settings, start brake hard.
Preferably, the second brake unit, the normally travel direction for determining the vehicle according to described image information, when
Monitor second threshold of the angle between the current driving direction of the vehicle and the normally travel direction more than respective settings
During value, start brake hard.
The invention provides it is a kind of aid in drive method and system, by detect vehicle be in transport condition when,
The corresponding image information of surrounding environment of vehicle is obtained, image information is divided into the crucial subregion of the first quantity, and determine each
The input value of individual crucial subregion, optimizes treatment, with life using the majorized function for pre-setting to the input value of the first quantity
Into the corresponding reference value of input value of first quantity, judge whether reference value is not more than with the difference of predetermined standard value
Given threshold, if so, performing urgent danger prevention treatment.Due to that can sentence to motoring condition from the surrounding environment of vehicle
It is disconnected, better than the method that only simple dependence distance and speed are judged, it is possible to increase the accuracy for avoiding accident from occurring.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are the present invention
Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
These accompanying drawings obtain other accompanying drawings.
Fig. 1 is the flow chart of the method that a kind of auxiliary that one embodiment of the invention is provided drives；
Fig. 2 is a kind of identification model framework map that one embodiment of the invention is provided；
Fig. 3 is the flow chart of the method that another auxiliary that one embodiment of the invention is provided drives；
Fig. 4 is the schematic diagram of the system that a kind of auxiliary that one embodiment of the invention is provided drives；
Fig. 5 is the schematic diagram of the system that another auxiliary that one embodiment of the invention is provided drives.
Specific embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
A part of embodiment of the present invention, rather than whole embodiments, based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained on the premise of creative work is not made, belongs to the scope of protection of the invention.
As shown in figure 1, the embodiment of the invention provides a kind of method for aiding in and driving, the method can include following step
Suddenly：
Step 101：When vehicle is detected in transport condition, the surrounding environment corresponding at least of the vehicle is obtained
Individual image information.
Step 102：For each described image information, it is performed both by：Described image information is divided into the pass of the first quantity
Key subregion, and determine the input value of each crucial subregion.
Step 103：Treatment is optimized to the input value of the first quantity using the majorized function for pre-setting, with life
Into the corresponding reference value of the input value of first quantity.
Step 104：Judge whether the reference value is not more than given threshold with the difference of predetermined standard value, if
It is to perform step 105.
Step 105：Perform urgent danger prevention treatment.
In the embodiment shown in fig. 1, by when vehicle is detected in transport condition, obtaining the surrounding environment of vehicle
Corresponding image information, image information is divided into the crucial subregion of the first quantity, and determines the input value of each crucial subregion,
Treatment is optimized to the input value of the first quantity using the majorized function for pre-setting, to generate the input value of first quantity
Corresponding reference value, judges whether reference value is not more than given threshold with the difference of predetermined standard value, if so, performing tight
Anxious hedging treatment.Due to that can judge motoring condition from the surrounding environment of vehicle, better than only simple dependence away from
Method from being judged with speed, it is possible to increase the accuracy for avoiding accident from occurring.
What deserves to be explained is, high-speed camera can be set on vehicle, believed with the image for shooting vehicle periphery in real time
Breath, therefore corresponding at least one image information of the surrounding environment of the vehicle for getting, can be front, rear, the left side of vehicle
At least one of side, the corresponding image information in right side side and top.
In an embodiment of the invention, it is described using the optimization letter for pre-setting in order to ensure the reliability of reference value
Several input values to the first quantity optimize treatment, to generate the corresponding reference of the input value of first quantity
Value, including：
S1：For each the target input value in the input value of the first quantity, it is performed both by：Using what is pre-set
Each first majorized function in first majorized function of the first quantity, optimizes to the target input value respectively, with
Generate the corresponding objective optimization value of the target input value；Determine the corresponding target weight of the objective optimization value, and determine institute
The product that objective optimization value is stated with the target weight is target output value, to complete the suboptimization to the target input value
Treatment；
S2：Judge whether the number of times of the corresponding optimization processing of the target input value reaches given threshold, if so, performing
S3, otherwise, using the target output value as the target input value, and performs S1；
S3：According to the target output value of the first quantity determined, and predetermined second majorized function is utilized, meter
Calculate the corresponding reference value of target output value of first quantity.
In this embodiment it is possible to build identification model as shown in Figure 2 in advance, input is included in the identification model
Layer, hidden layer and the part of output layer three.Wherein, input layer is used to be input into the input value for determining, hidden layer is used for the defeated of input
Enter value to optimize, output layer is used to export the optimal value after hidden layer optimization.And data can be added in the identification model
Filtering technique so that the data sample for going against accepted conventions is optimizing little number of times later with regard to eliminating, and the data for going against accepted conventions are eliminated in advance, than
Optimization number of times just finds that data are unreasonable more intelligent after reaching threshold value.
In fig. 2, x1、x2、…、xrFor characterizing the input value being input in identification model, o1、o2、…、orFor characterizing
It is input to the optimal value generated after the first majorized function carries out a suboptimization in identification model.
In an embodiment of the invention, in order to further ensure that the reliability of reference value, first majorized function, bag
Include：
Wherein, f (xi) for i-th optimal value of crucial subregion in the crucial subregion for characterizing first quantity；xiWith
In the input value for characterizing i-th key subregion；C is used to characterize the parameter of described image information；Radom is described for characterizing
The corresponding empirical value of i-th crucial subregion；
Second majorized function, including：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor table
Levy i-th default weight of crucial subregion in j-th image information；bjFor characterizing j-th image information
Threshold value；N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
In this embodiment, existing majorized function isIn order to monitor the connection between every suboptimization
System, adds memory function in identification model.Then majorized function is made that on the basis of present majorized function and is changed
Enter, that is, the 3 times optimum results of current optimization as memory value are added in the majorized function of current optimization, therefore
The first majorized function is just obtained.
In an embodiment of the invention, it is described to determine each in order to be accurately analyzed to image information
The input value of the crucial subregion, including：
For crucial subregion each described, the crucial subregion is carried out into gray processing treatment, and determine the crucial subregion
Including each pixel gray value, and determine the gray value of each pixel plus and be the crucial subregion
Input value.
In this embodiment it is possible to using the gray value sum of each pixel included in the crucial subregion as corresponding
Input value, it is also possible to using the average value of the gray value of each pixel included in the crucial subregion as corresponding input
Value.And can be calculated by following formula for the gray value Gray of pixel：
Wherein, R is used to characterize red, and G is used to characterize green, and B is used to characterize blueness, and 0.299 is used to characterize the power of red
Value, 0.587 is used to characterize the weights of green, and 0.144 is used to characterize the weights of red.
In an embodiment of the invention, in order to further ensure that the security of driving, the method that the auxiliary drives can be with
Further include：When the barycentre offset for monitoring the corresponding driver of the vehicle is more than the first threshold of respective settings,
Start brake hard.
In this embodiment, due in case of emergency, what the body of driver can can't help being moved or offseting,
So as to cause the change of position of centre of gravity, thus when monitor driver's barycentre offset more than setting first threshold when, can be with
Start brake hard.Other urgent danger preventions can also be but taken to process, for example, starting alarm sounds etc..
In an embodiment of the invention, in order to further ensure that the security of driving, the method that the auxiliary drives can be with
Further include：The normally travel direction of the vehicle is determined according to described image information, when monitoring the current of the vehicle
When angle between travel direction and the normally travel direction is more than the Second Threshold of respective settings, start brake hard.
In this embodiment, due in case of emergency, when driver drives vehicle may offset setting route, work as prison
Measure angle between the current driving direction of vehicle and normally travel direction more than respective settings Second Threshold when, Ke Yiqi
Dynamic brake hard.Other urgent danger preventions can also be but taken to process, for example, starting alarm sounds etc..
As shown in figure 3, the embodiment of the invention provides a kind of method for aiding in and driving, the method can include following step
Suddenly：
Step 301：When vehicle is detected in transport condition, corresponding at least one figure of surrounding environment of vehicle is obtained
As information.
In this step, can be by setting the image letter that high-speed camera shoots vehicle periphery in real time on vehicle
Breath, can be front, rear, left side side, right side side and the top of vehicle.
Step 302：Image information is divided into the crucial subregion of the first quantity.
In this embodiment, in order to ensure that the accuracy rate that things is recognized can be improved in identification process, can be to image
Information carries out division operation, and the division operation can be that image information is carried out to be divided into some regions, can be believed according to image
The gray value of pixel carrys out subregion in breath, for example, including a toy car in the image information, then can be by the wheel of toy car
Son as a crucial subregion, using the vehicle body of toy car as a crucial subregion.
Step 303：Determine the input value of each crucial subregion.
In this step, corresponding input value can be determined according to the gray value of included pixel in crucial subregion.
In this embodiment it is possible to using the gray value sum of each pixel included in the crucial subregion as corresponding
Input value.What deserves to be explained is, it is also possible to by the average value of the gray value of each pixel included in the crucial subregion
As corresponding input value.
Step 304：It is determined that the first majorized function with input value number the first quantity of identical.
In this step, in order to realize the things to whether including causing traffic accident in image information, it is thus necessary to determine that with
First majorized function of input value number the first quantity of identical.Assuming that in step 302, image information divide into r pass
Key subregion, then be accomplished by determining r the first majorized function.
In this embodiment, due to it needs to be determined that r the first majorized function, you can to obtain r corresponding first optimization
Function, and each majorized function for determining imparts the empirical value of first three suboptimization, so as to obtain r, to differ first excellent
Change function.
Step 305：For each the target input value in the input value of the first quantity, distinguished using the first majorized function
Target input value is optimized, to generate the corresponding objective optimization value of target input value；Determine the corresponding mesh of objective optimization value
Mark weight, and determine that objective optimization value is target output value with the product of target weight, to complete to target input value once
Optimization processing.
In this step, can in advance build identification model as shown in Figure 2, the identification model include input layer,
Hidden layer and the part of output layer three.
In an initial condition, the input value of input layer input is each input value determined in step 303.For example, input
Input value in identification model is respectively x1、x2、x3、…、xr。
In subsequent process, the input value of input layer input can be optimal value and the optimization of the last output of identification model
It is worth the product of corresponding weights.It can also be the optimal value of the last output of identification model.
In hidden layer, each round frame included by hidden layer in Fig. 2 is refer to, the correspondence one the in each round frame
One majorized function, and be handled as follows in each round frame：Using corresponding first majorized function of the round frame respectively to r
Input value is calculated, and so as to get one group of numerical value, r objective optimization value is included in the group.
For example：The first majorized function in the hidden layer corresponding to second round frame is following formula：
Wherein, a, b, c are three empirical values.So, the treatment for being carried out for second round frame in hidden layer includes：Will
Input value x1、x2、x3、…、xrIt is updated in above formula respectively, obtains r objective optimization value f corresponding for second round frame
(x1)、f(x2)、f(x3)、…、f(xr)。
Step 306：Judge whether the number of times of the corresponding optimization processing of target input value reaches given threshold, if so, performing
307, otherwise, using target output value as target input value, and perform 305.
In this step, threshold value sets based on experience value, it is also possible to according to sets itself the need for user.For example,
It it is 62 times, 90 is inferior.
Step 307：According to the target output value of the first quantity determined, and optimize letter using predetermined second
Number, calculates the corresponding reference value of target output value of first quantity.
In this embodiment, target output value can be one group, and the group includes each objective optimization value and corresponding mesh
Mark the product of weights, the target output value can also be one, be each optimization input value and respective objects weights product it
Sum afterwards.The form of the target output value can be determined according to the form of standard value.
And can be calculated using following formula for reference value：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor table
Levy i-th default weight of crucial subregion in j-th image information；bjFor characterizing j-th image information
Threshold value；N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
For example：In this embodiment, target output value is one, then r crucial subregion be to that should have r target output value,
Respectively f (x1)、f(x2)、f(x3)、…、f(xr), then reference value Sj=w1×f(x1)+w2×f(x2)+w3×f(x3)+…+wr
×f(xr)+bj+rand(xi)。
Step 308：Judge whether reference value is not more than given threshold with the difference of predetermined standard value, if so, holding
Row urgent danger prevention is processed, and otherwise, terminates current process.
In this step, urgent danger prevention treatment can be braking deceleration, or give a warning sound, or other
Hedging treatment.
As shown in figure 4, the embodiment of the invention provides a kind of DAS (Driver Assistant System), the DAS (Driver Assistant System) can include：
Acquiring unit 401, determining unit 402, processing unit 403 and judging unit 404, wherein,
The acquiring unit 401, for when vehicle is detected in transport condition, obtaining the surrounding environment of the vehicle
Corresponding at least one image information；
The determining unit 402, for for each described image information, being performed both by：Described image information is divided into
The crucial subregion of the first quantity, and determine the input value of each crucial subregion；
The processing unit 403, for being carried out to the input value of the first quantity using the majorized function for pre-setting
Optimization processing, to generate the corresponding reference value of the input value of first quantity；
The judging unit 404, for judging whether the reference value is not more than with the difference of predetermined standard value
Given threshold, if so, performing urgent danger prevention treatment.
As shown in figure 5, in an embodiment of the invention, in order to ensure the reliability of reference value, the processing unit
403, including optimization subelement 4031, triggering subelement 4032, computation subunit 4033, wherein,
The optimization subelement 4031, for each the target input value in the input value for the first quantity,
It is performed both by：Using each first majorized function in the first majorized function of the first quantity for pre-setting, respectively to described
Target input value is optimized, to generate the corresponding objective optimization value of the target input value；Determine the objective optimization value pair
The target weight answered, and determine that the objective optimization value is target output value with the product of the target weight, to complete to institute
State an optimization processing of target input value；
Whether the triggering subelement 4032, the number of times for judging the corresponding optimization processing of the target input value reaches
Given threshold, if so, the computation subunit 4033 is triggered, otherwise, using the target output value as the target input value,
And trigger the optimization subelement 4031；
The computation subunit 4033, for the target output value according to the first quantity determined, and using advance
The second majorized function for determining, calculates the corresponding reference value of target output value of first quantity.
In this embodiment, by building identification model in advance, many suboptimization are carried out to input value, after reaching optimization number of times
Generation target output value, and reference value is calculated, improve the reliability of reference value.
In an embodiment of the invention, in order to further ensure that the reliability of reference value, first majorized function, bag
Include：
Wherein, f (xi) for i-th optimal value of crucial subregion in the crucial subregion for characterizing first quantity；xiWith
In the input value for characterizing i-th key subregion；C is used to characterize the parameter of described image information；Radom is described for characterizing
The corresponding empirical value of i-th crucial subregion；
Second majorized function, including：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor table
Levy i-th default weight of crucial subregion in j-th image information；bjFor characterizing j-th image information
Threshold value；N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
In an embodiment of the invention, in order to be accurately analyzed to image information, the determining unit, tool
Body is used to be directed to each described crucial subregion, and the crucial subregion is carried out into gray processing treatment, and determines to be wrapped in the crucial subregion
The gray value of each pixel for including, and determine the gray value of each pixel plus and be the crucial subregion
Input value.
In an embodiment of the invention, Fig. 5 is refer to, in order to further ensure that the security of driving, is further included：
First brake unit 501, for when monitoring the barycentre offset of the corresponding driver of the vehicle more than the of respective settings
During one threshold value, start brake hard.
In an embodiment of the invention, Fig. 5 is refer to, in order to further ensure that the security of driving, is further included：
Second brake unit 502, the normally travel direction for determining the vehicle according to described image information, when monitoring the car
Current driving direction and the normally travel direction between angle more than respective settings Second Threshold when, start urgent
Braking.
The contents such as the information exchange between each unit, implementation procedure in said system, due to implementing with the inventive method
Example is based on same design, and particular content can be found in the narration in the inventive method embodiment, and here is omitted.
To sum up, various embodiments of the present invention, at least have the advantages that：
1st, in an embodiment of the present invention, by when vehicle is detected in transport condition, obtaining surrounding's ring of vehicle
The corresponding image information in border, image information is divided into the crucial subregion of the first quantity, and determine the input of each crucial subregion
Value, optimizes treatment, to generate the defeated of first quantity using the majorized function for pre-setting to the input value of the first quantity
Enter and be worth corresponding reference value, judge whether reference value is not more than given threshold with the difference of predetermined standard value, if so, holding
Row urgent danger prevention is processed.Due to that can judge motoring condition from the surrounding environment of vehicle, better than it is only simple according to
By the method that distance and speed are judged, it is possible to increase the accuracy for avoiding accident from occurring.
2nd, in an embodiment of the present invention, by building identification model, the input value to crucial subregion sets according to user
Optimization number of times carry out many suboptimization, the reliability of the target output value for making is improved.And add data mistake in the identification model
Filter technology so that the data sample for going against accepted conventions is optimizing little number of times later with regard to eliminating, and the data for going against accepted conventions is eliminated in advance, than excellent
Change after number of times reaches threshold value and just find that data are illegal more intelligent.
3rd, in an embodiment of the present invention, image information is analyzed by the first majorized function and the second majorized function
Identification, increased the reliability of the reference value obtained by calculating, so as to increased the reliability of judged result.
4th, in an embodiment of the present invention, the skew by using the centre-of gravity shift and vehicle of sensor detection driver is special
Property, to realize urgent early warning, there is the accuracy rate of accident such that it is able to further improve prediction vehicle so that driver is as early as possible
Take measures, or emergency brake of vehicle, reduce and the probability of accident occurs.
It should be noted that herein, such as first and second etc relational terms are used merely to an entity
Or operation makes a distinction with another entity or operation, and not necessarily require or imply these entities or exist between operating
Any this actual relation or order.And, term " including ", "comprising" or its any other variant be intended to it is non-
It is exclusive to include, so that process, method, article or equipment including a series of key elements not only include those key elements,
But also other key elements including being not expressly set out, or also include by this process, method, article or equipment are solid
Some key elements.In the absence of more restrictions, the key element limited by sentence " including a 〃 ", does not arrange
Except also there is other identical factor in the process including the key element, method, article or equipment.
One of ordinary skill in the art will appreciate that：Realizing all or part of step of above method embodiment can pass through
Programmed instruction related hardware is completed, and foregoing program can be stored in the storage medium of embodied on computer readable, the program
Upon execution, the step of including above method embodiment is performed；And foregoing storage medium includes：ROM, RAM, magnetic disc or light
Disk etc. is various can be with the medium of store program codes.
It is last it should be noted that：Presently preferred embodiments of the present invention is the foregoing is only, skill of the invention is merely to illustrate
Art scheme, is not intended to limit the scope of the present invention.All any modifications made within the spirit and principles in the present invention,
Equivalent, improvement etc., are all contained in protection scope of the present invention.
Claims (10)
1. a kind of method that auxiliary drives, it is characterised in that including：
When vehicle is detected in transport condition, corresponding at least one image information of surrounding environment of the vehicle is obtained；
For each described image information, it is performed both by：Described image information is divided into the crucial subregion of the first quantity, and is determined
The input value of each crucial subregion；
Treatment is optimized to the input value of the first quantity using the majorized function for pre-setting, to generate first quantity
The corresponding reference value of the input value；
Judge whether the reference value is not more than given threshold with the difference of predetermined standard value, if so, execution promptly keeps away
Danger treatment.
2. method according to claim 1, it is characterised in that
It is described that treatment is optimized to the input value of the first quantity using the majorized function that pre-sets, with generate this first
The corresponding reference value of the input value of quantity, including：
S1：For each the target input value in the input value of the first quantity, it is performed both by：Using first for pre-setting
Each first majorized function in first majorized function of quantity, optimizes to the target input value respectively, to generate
The corresponding objective optimization value of the target input value；Determine the corresponding target weight of the objective optimization value, and determine the mesh
Mark optimal value is target output value with the product of the target weight, to complete at the suboptimization to the target input value
Reason；
S2：Judge whether the number of times of the corresponding optimization processing of the target input value reaches given threshold, if so, S3 is performed, it is no
Then, using the target output value as the target input value, and S1 is performed；
S3：According to the target output value of the first quantity determined, and predetermined second majorized function is utilized, calculating should
The corresponding reference value of target output value of the first quantity.
3. method according to claim 2, it is characterised in that
First majorized function, including：
Wherein, f (xi) for i-th optimal value of crucial subregion in the crucial subregion for characterizing first quantity；xiFor table
Levy described i-th input value of crucial subregion；C is used to characterize the parameter of described image information；Radom is used to characterize described i-th
The corresponding empirical value of individual crucial subregion；
Second majorized function, including：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor characterizing
State i-th default weight of crucial subregion in j-th image information；bjThreshold value for characterizing j-th image information；
N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
4. method according to claim 1, it is characterised in that
The input value for determining each crucial subregion, including：
For crucial subregion each described, the crucial subregion is carried out into gray processing treatment, and determine that the crucial subregion includes
Each pixel gray value, and determine the gray value of each pixel plus and defeated for the crucial subregion
Enter value.
5. according to any described method in Claims 1-4, it is characterised in that
Further include：
When the barycentre offset for monitoring the corresponding driver of the vehicle is more than the first threshold of respective settings, start urgent
Braking；
And/or,
Further include：
The normally travel direction of the vehicle is determined according to described image information, when the current driving direction for monitoring the vehicle
When being more than the Second Threshold of respective settings with the angle between the normally travel direction, start brake hard.
6. the system that a kind of auxiliary drives, it is characterised in that including：Acquiring unit, determining unit, processing unit and judgement are single
Unit, wherein,
The acquiring unit, for when vehicle is detected in transport condition, the surrounding environment for obtaining the vehicle to be corresponding
At least one image information；
The determining unit, for for each described image information, being performed both by：Described image information is divided into the first quantity
Crucial subregion, and determine the input value of each crucial subregion；
The processing unit, for optimizing place to the input value of the first quantity using the majorized function for pre-setting
Reason, to generate the corresponding reference value of the input value of first quantity；
The judging unit, for judging whether the reference value is not more than setting threshold with the difference of predetermined standard value
Value, if so, performing urgent danger prevention treatment.
7. the system that auxiliary according to claim 6 drives, it is characterised in that
The processing unit, including optimization subelement, triggering subelement, computation subunit, wherein,
The optimization subelement, for each the target input value in the input value for the first quantity, is performed both by：Profit
With each first majorized function in the first majorized function of the first quantity for pre-setting, respectively to the target input value
Optimize, to generate the corresponding objective optimization value of the target input value；Determine the corresponding target power of the objective optimization value
Weight, and determine that the objective optimization value is target output value with the product of the target weight, to complete to be input into the target
Optimization processing of value；
Whether the triggering subelement, the number of times for judging the corresponding optimization processing of the target input value reaches setting threshold
Value, if so, triggering the computation subunit, otherwise, using the target output value as the target input value, and triggers described
Optimization subelement；
The computation subunit, for according to the target output value of the first quantity determined, and using predetermined the
Two majorized functions, calculate the corresponding reference value of target output value of first quantity.
8. the system that auxiliary according to claim 7 drives, it is characterised in that
First majorized function, including：
Wherein, f (xi) for i-th optimal value of crucial subregion in the crucial subregion for characterizing first quantity；xiFor table
Levy described i-th input value of crucial subregion；C is used to characterize the parameter of described image information；Radom is used to characterize described i-th
The corresponding empirical value of individual crucial subregion；
Second majorized function, including：
Wherein, SjReference value for characterizing j-th image information at least one image information；wijFor characterizing
State i-th default weight of crucial subregion in j-th image information；bjThreshold value for characterizing j-th image information；
N is used to characterize first quantity；rand(xi) for characterizing described i-th equilibrium valve of crucial subregion.
9. the system that auxiliary according to claim 6 drives, it is characterised in that
The determining unit, specifically for for crucial subregion each described, the crucial subregion being carried out into gray processing treatment, and
Determine the gray value of each pixel that the crucial subregion includes, and the gray value of each pixel described in determination
Plus and be the input value of the crucial subregion.
10. the system for being driven according to any described auxiliary in claim 6 to 9, it is characterised in that
Further include：First brake unit, for being more than when the barycentre offset for monitoring the corresponding driver of the vehicle
During the first threshold of respective settings, start brake hard；
And/or,
Further include：Second brake unit, the normally travel direction for determining the vehicle according to described image information, when
Monitor second threshold of the angle between the current driving direction of the vehicle and the normally travel direction more than respective settings
During value, start brake hard.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610997607.0A CN106696825B (en) | 2016-11-11 | 2016-11-11 | A kind of method and system that auxiliary drives |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610997607.0A CN106696825B (en) | 2016-11-11 | 2016-11-11 | A kind of method and system that auxiliary drives |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106696825A true CN106696825A (en) | 2017-05-24 |
| CN106696825B CN106696825B (en) | 2018-10-19 |
Family
ID=58939803
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610997607.0A Active CN106696825B (en) | 2016-11-11 | 2016-11-11 | A kind of method and system that auxiliary drives |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106696825B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112835048A (en) * | 2021-01-04 | 2021-05-25 | 海门市帕源路桥建设有限公司 | Automatic positioning control method for construction interval |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074014A (en) * | 2011-02-23 | 2011-05-25 | 山东大学 | Stereo matching method by utilizing graph theory-based image segmentation algorithm |
| CN103745241A (en) * | 2014-01-14 | 2014-04-23 | 浪潮电子信息产业股份有限公司 | Intelligent driving method based on self-learning algorithm |
| US20150210312A1 (en) * | 2014-01-30 | 2015-07-30 | Mobileye Vision Technologies Ltd. | Systems and methods for detecting low-height objects in a roadway |
-
2016
- 2016-11-11 CN CN201610997607.0A patent/CN106696825B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074014A (en) * | 2011-02-23 | 2011-05-25 | 山东大学 | Stereo matching method by utilizing graph theory-based image segmentation algorithm |
| CN103745241A (en) * | 2014-01-14 | 2014-04-23 | 浪潮电子信息产业股份有限公司 | Intelligent driving method based on self-learning algorithm |
| US20150210312A1 (en) * | 2014-01-30 | 2015-07-30 | Mobileye Vision Technologies Ltd. | Systems and methods for detecting low-height objects in a roadway |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112835048A (en) * | 2021-01-04 | 2021-05-25 | 海门市帕源路桥建设有限公司 | Automatic positioning control method for construction interval |
| CN112835048B (en) * | 2021-01-04 | 2023-12-15 | 海门市帕源路桥建设有限公司 | Automatic positioning control method for construction space |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106696825B (en) | 2018-10-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108297877B (en) | Control method for vehicle, system and device | |
| CN105702088B (en) | Alarm device | |
| CN108372785B (en) | Image recognition-based automobile unsafe driving detection device and detection method | |
| CN109242251A (en) | Vehicular behavior safety detecting method, device, equipment and storage medium | |
| CN107031629A (en) | A kind of 3D machine vision fork truck traveling supplementary security system and method | |
| CN104827963B (en) | A kind of method, control system and control device for collision prevention of vehicle intelligent early-warning | |
| CN109910879A (en) | A kind of vehicle safety anticollision control method of combination safe distance and collision time | |
| US11926315B2 (en) | Electronic apparatus for detecting risk factors around vehicle and method for controlling same | |
| CN106573589A (en) | Predicted-collision-time calculation device, and predicted-collision-time calculation method | |
| CN112477854A (en) | Monitoring and early warning device and method based on vehicle blind area | |
| CN110901385B (en) | Active speed limiting method based on fatigue state of driver | |
| US11847562B2 (en) | Obstacle recognition assistance device, obstacle recognition assistance method, and storage medium | |
| KR101857554B1 (en) | External data intrusion detection apparatus and method for vehicles | |
| CN110126734A (en) | A kind of collision early warning system being directed to Vehicular visual blind area, device and method | |
| CN104210488A (en) | Method and device for preventing vehicle from colliding with road edge | |
| CN110262521A (en) | A kind of automatic Pilot control method | |
| CN114148332A (en) | Vehicle driving starting prompting method, vehicle driving starting prompting system and computer readable storage medium | |
| CN110803144B (en) | Automatic braking method, automatic braking device and automatic driving vehicle | |
| CN110843775B (en) | Obstacle identification method based on pressure sensor | |
| CN106314275A (en) | Anti-scratching device and method used for car as well as car | |
| EP3640857B1 (en) | Method, vehicle, system, and storage medium for indicating anomalous vehicle scenario using encoder network and discriminator network intermediate layer activation | |
| CN110992709A (en) | Active speed limiting system based on fatigue state of driver | |
| CN106696825A (en) | Driving assisting method and system | |
| CN109774726A (en) | Detection method, device, equipment and the storage medium of vehicle performance | |
| KR20160089933A (en) | Apparatus and method for preventing vehicle collision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |