CN108528431A

CN108528431A - Vehicle travels autocontrol method and device

Info

Publication number: CN108528431A
Application number: CN201710120432.XA
Authority: CN
Inventors: 黄忠伟; 姜波
Original assignee: BYD Co Ltd
Current assignee: BYD Co Ltd
Priority date: 2017-03-02
Filing date: 2017-03-02
Publication date: 2018-09-14
Anticipated expiration: 2037-03-02
Also published as: CN108528431B

Abstract

The present invention proposes a kind of vehicle traveling autocontrol method and device, wherein method includes：Front lines on highway is obtained according to the first image, rear lines on highway is obtained according to third image, front lines on highway is mapped in the second image according to the intertexture mapping relations between the first image and the second image and generates multiple front vehicles identification ranges, rear lines on highway is mapped in the 4th image according to the intertexture mapping relations between third image and the 4th image and generates multiple front vehicle identification ranges, objects ahead vehicle is identified according to multiple front vehicles identification ranges are generated, rear area target vehicle is identified according to multiple front vehicle identification ranges are generated；Cruise control is carried out to the kinematic parameter of main body vehicle according to the kinematic parameter of objects ahead vehicle and rear area target vehicle.Thereby, it is possible to correctly be controlled cruise, the driving safety of main body vehicle is improved.

Description

Vehicle travels autocontrol method and device

Technical field

The present invention relates to technical field of automobile control more particularly to a kind of vehicle traveling autocontrol methods and device.

Background technology

Currently, vehicle self-adaption cruise system is used as distance measuring sensor usually using millimetre-wave radar, laser radar etc..By This, main body vehicle can sense multiple targets in front of main body vehicle by installing the distance measuring sensor of above-mentioned any kind type Vehicle and the kinematic parameter for being adaptively adjusted cruise system.

However, the case where travelling on bend for multiple target vehicles, the distance measuring sensors such as millimetre-wave radar, laser radar Lane line can not be identified well.Therefore only installation carries millimetre-wave radar or the main body vehicle of laser radar is likely to this vehicle The target vehicle in road is identified as in this non-track, and may be identified as the target vehicle in this non-track, in this track, may lead Cause braking or brake latency, the driving safety of main body vehicle of the self-adaption cruise system execution mistake of main body vehicle low.

Invention content

The purpose of the present invention is intended to solve at least some of the technical problems in related technologies.

For this purpose, first purpose of the present invention is to propose a kind of vehicle traveling autocontrol method, main body is enabled to The self-adaption cruise system of vehicle executes correctly braking, reduces unnecessary braking adjustment, effectively reduces rear-end impact Risk improves the driving safety of main body vehicle.

Second object of the present invention is to propose a kind of vehicle traveling automatic control device.

In order to achieve the above object, first aspect present invention embodiment proposes a kind of vehicle traveling autocontrol method, including： The first image and the second image that environment in front of main body vehicle is obtained from preposition 3D cameras obtain main body vehicle from postposition 3D cameras The third image and the 4th image of rear environment, wherein the first image and third image are color or luminance picture, the second image It is depth image with the 4th image；Front lines on highway is obtained according to described first image, is obtained according to the third image Rear lines on highway；According to the intertexture mapping relations between described first image and second image by the front highway Lane line, which maps in second image, generates multiple front vehicles identification ranges, according to the third image and the described 4th Intertexture mapping relations between image, which map to the rear lines on highway in the 4th image, generates multiple rear vehicles Identification range；Objects ahead vehicle is identified according to all front vehicles identification ranges, according to all front vehicle identification ranges Identify rear area target vehicle；According to the kinematic parameter of the objects ahead vehicle and the rear area target vehicle to the main body vehicle Kinematic parameter carry out cruise control.

The vehicle of the embodiment of the present invention travels autocontrol method, obtains ring in front of main body vehicle from preposition 3D cameras first First image and the second image in border obtain the third image and the 4th image of main body vehicle rear environment from postposition 3D cameras, And front lines on highway is obtained according to the first image and rear lines on highway is obtained according to third image, then according to first Intertexture mapping relations between image and the second image, which map to front lines on highway in the second image, generates multiple fronts Rear lines on highway is mapped to by vehicle identification range according to the intertexture mapping relations between third image and the 4th image Generate multiple front vehicle identification ranges in four images, and according to all front vehicles identification ranges identify objects ahead vehicle and Rear area target vehicle is identified according to all front vehicle identification ranges, finally according to objects ahead vehicle and rear area target vehicle Kinematic parameter carries out cruise control to the kinematic parameter of main body vehicle.

In order to achieve the above object, second aspect of the present invention embodiment proposes a kind of vehicle traveling automatic control device, including： First acquisition module, the first image and the second image for obtaining environment in front of main body vehicle from preposition 3D cameras, from postposition 3D cameras obtain the third image and the 4th image of main body vehicle rear environment, wherein the first image and third image are color Or luminance picture, the second image and the 4th image are depth image；Second acquisition module, for being obtained according to described first image Front lines on highway；Third acquisition module, for obtaining rear lines on highway according to the third image；First generates mould Block, for being reflected the front lines on highway according to the intertexture mapping relations between described first image and second image It is incident upon in second image and generates multiple front vehicles identification ranges, according between the third image and the 4th image Intertexture mapping relations the rear lines on highway mapped in the 4th image generate multiple front vehicles identification models It encloses；First identification module, for identifying objects ahead vehicle according to all front vehicles identification ranges, according to all front vehicles Identification range identifies rear area target vehicle；Control module, for according to the objects ahead vehicle and the rear area target vehicle Kinematic parameter and steering indicating light cruise control is carried out to the kinematic parameter of the main body vehicle.

The vehicle of the embodiment of the present invention travels automatic control device, obtains ring in front of main body vehicle from preposition 3D cameras first First image and the second image in border obtain the third image and the 4th image of main body vehicle rear environment from postposition 3D cameras, And front lines on highway is obtained according to the first image and rear lines on highway is obtained according to third image, then according to first Intertexture mapping relations between image and the second image, which map to front lines on highway in the second image, generates multiple fronts Rear lines on highway is mapped to by vehicle identification range according to the intertexture mapping relations between third image and the 4th image Generate multiple front vehicle identification ranges in four images, and according to all front vehicles identification ranges identify objects ahead vehicle and Rear area target vehicle is identified according to all front vehicle identification ranges, finally according to objects ahead vehicle and rear area target vehicle Kinematic parameter carries out cruise control to the kinematic parameter of main body vehicle.

The additional aspect of the present invention and advantage will be set forth in part in the description, and will partly become from the following description Obviously, or practice through the invention is recognized.

Description of the drawings

Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments Obviously and it is readily appreciated that, wherein：

Fig. 1 is the flow diagram that vehicle provided by one embodiment of the present invention travels autocontrol method；

Fig. 2 is the flow diagram that the vehicle that another embodiment of the present invention provides travels autocontrol method；

Fig. 3 is the steering indicating light provided by one embodiment of the present invention for accurately identifying this non-track target vehicle of left rear side Schematic diagram；

Fig. 4 accurately identifies the objects ahead vehicle completion into left bend to be provided by one embodiment of the present invention in straight way The schematic diagram of lane change to the right；

Fig. 5 is provided by one embodiment of the present invention objects ahead vehicle to be accurately identified in bend to the right in straight way The schematic diagram of lane change to the left；

Fig. 6 is the flow diagram that the vehicle that another embodiment of the invention provides travels autocontrol method；

Fig. 7 is the structural schematic diagram that vehicle provided by one embodiment of the present invention travels automatic control device；

Fig. 8 is the structural schematic diagram that the vehicle that another embodiment of the present invention provides travels automatic control device.

Specific implementation mode

The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached The embodiment of figure description is exemplary, it is intended to for explaining the present invention, and is not considered as limiting the invention.

Below with reference to the accompanying drawings the vehicle traveling autocontrol method and device of the embodiment of the present invention are described.

In general, being used as by installation millimetre-wave radar, laser radar etc. more in front of distance measuring sensor sensing main body vehicle A target vehicle and the kinematic parameter for being adaptively adjusted cruise system.

However, in the case where multiple target vehicles travel on bend, the distance measuring sensors such as millimetre-wave radar, laser radar Lane line can not be identified well.The braking or braking that may lead to the self-adaption cruise system execution mistake of main body vehicle are prolonged Late, the driving safety of main body vehicle is low.

To solve the above-mentioned problems, the present invention proposes that a kind of vehicle travels autocontrol method, enables to main body vehicle Self-adaption cruise system execute correctly braking, reduce unnecessary braking and adjust, improve the driving safety of main body vehicle Property.It is specific as follows：

Fig. 1 is the flow diagram that vehicle provided by one embodiment of the present invention travels autocontrol method.

As shown in Figure 1, vehicle traveling autocontrol method includes the following steps：

Step 101, the first image and the second image that environment in front of main body vehicle is obtained from preposition 3D cameras, from postposition 3D Camera obtains the third image and the 4th image of main body vehicle rear environment, wherein the first image and third image be color or Luminance picture, the second image and the 4th image are depth image.

It is understood that from preposition 3D cameras obtain main body vehicle in front of environment the first image and the second image and It is obtained from postposition 3D cameras there are many kinds of the third image of main body vehicle rear environment and the modes of the 4th image, it can be according to reality Border application needs to carry out selection setting.It is illustrated below：

The first example obtains the first image of environment in front of main body vehicle from the imaging sensor of preposition 3D cameras, from The time-of-flight sensor of preposition 3D cameras obtains the second image of environment in front of main body vehicle.

Specifically, the color of environment or luminance picture in front of main body vehicle are obtained by imaging sensor and is used as the first image It is used as the second image with the depth image for obtaining environment in front of main body vehicle by time-of-flight sensor.

Second of example obtains the third image of main body vehicle rear environment from the imaging sensor of postposition 3D cameras, from The time-of-flight sensor of postposition 3D cameras obtains the 4th image of main body vehicle rear environment.

Specifically, the color of main body vehicle rear environment is obtained by imaging sensor or luminance picture is used as third image It is used as the 4th image with the depth image for obtaining main body vehicle rear environment by time-of-flight sensor.

Step 102, front lines on highway is obtained according to the first image, rear road driveway is obtained according to third image Line.

It is understood that the first image can be luminance picture or be color image.According to the first different images Front lines on highway difference is obtained according to the first image, is described as follows：

The first example, when the first image is luminance picture, according to front lines on highway and road surface in the first image Luminance difference identification front lines on highway.

Second of example, when the first image be color image, color image is converted into luminance picture, according to the first image Middle front lines on highway and the luminance difference on road surface identification front lines on highway.

It refers to obtaining road driveway all in front of main body vehicle it should be noted that obtaining front lines on highway Line.

It is understood that third image can be luminance picture or be color image.According to different third images Rear lines on highway difference is obtained according to third image, is described as follows：

The first example, when third image is luminance picture, according to rear lines on highway in third image and road surface Luminance difference identifies rear lines on highway.

Second of example, when third image be color image, color image is converted into luminance picture, according to third image Middle rear lines on highway and the luminance difference on road surface identify rear lines on highway.

It refers to obtaining all road driveways in main body vehicle rear it should be noted that obtaining rear lines on highway Line.

Step 103, front lines on highway is mapped according to the intertexture mapping relations between the first image and the second image To multiple front vehicles identification ranges are generated in the second image, according to the intertexture mapping relations between third image and the 4th image Rear lines on highway is mapped in the 4th image and generates multiple front vehicle identification ranges.

It is understood that the intertexture mapping relations between the first image and the second image indicate each picture of the first image Adjustment of the ranks coordinate of element by equal proportion can at least determine the ranks coordinate of a pixel, therefore root in the second image Each edge pixel location of the front lines on highway obtained according to the first image can at least determine one in the second image Location of pixels, to obtain the front lines on highway of equal proportion adjustment in the second image.

As a result, according to the front lines on highway of the equal proportion obtained in the second image, each two adjacent front highway Lane line uniquely creates a front vehicles identification range, to which multiple two adjacent front lines on highway can generate Multiple front vehicles identification ranges.

It is understood that the intertexture mapping relations between third image and the 4th image indicate each picture of third image Adjustment of the ranks coordinate of element by equal proportion can at least determine the ranks coordinate of a pixel, therefore root in the 4th image Each edge pixel location of the rear lines on highway obtained according to third image can at least determine one in the 4th image Location of pixels, to obtain the rear lines on highway of equal proportion adjustment in the 4th image.

As a result, according to the rear lines on highway of the equal proportion obtained in the 4th image, each two adjacent rear highway Lane line uniquely creates a front vehicle identification range, to which multiple two adjacent rear lines on highway can generate Multiple front vehicle identification ranges.

Step 104, it identifies objects ahead vehicle according to all front vehicles identification ranges, is identified according to all front vehicles Range identifies rear area target vehicle.

It is understood that according to all front vehicles identification ranges identify objects ahead vehicle mode there are many kinds of, Selection setting can be carried out according to the actual application.It is illustrated below：

The first example marks all front vehicles identification ranges on the label in front this track and this non-track of front, According to this track target vehicle of the vehicle identification range identification front of this track label of label front, according to label front Fei Benche This non-track target vehicle of vehicle identification range identification front of road label, knows according to the front vehicles identification range of combination of two Other lane change ahead target vehicle.

Second of example detects the object boundary of objects ahead vehicle using the boundary detection method in image processing algorithm It is identified.

Specifically, objects ahead vehicle always changes at any time relative to the distance of time-of-flight sensor and position, And road surface, isolation strip are approximately indeclinable at any time relative to the distance of time-of-flight sensor and position.Therefore can lead to The second image (depth image) the creation time differential depth image obtained using two width different moments is crossed to detect above-mentioned distance With the variation of position, and then mark this track label front vehicles identification range identify front this track target vehicle, label This non-track target vehicle of front vehicles identification range identification front of this non-track label, knows in the front vehicles of combination of two Other range identifies lane change ahead target vehicle.

It is understood that according to all front vehicle identification ranges identify rear area target vehicle mode there are many kinds of, Selection setting can be carried out according to the actual application.It is illustrated below：

The first example marks all front vehicle identification ranges on the label in this track of rear and this non-track of rear, This track target vehicle of rear is identified according to the vehicle identification range of label this track label of rear, according to label rear Fei Benche The vehicle identification range of road label identifies this non-track target vehicle of rear, is known according to the front vehicle identification range of combination of two Other rear lane change target vehicle.

Second of example detects the object boundary of rear area target vehicle using the boundary detection method in image processing algorithm It is identified.

Specifically, rear area target vehicle always changes at any time relative to the distance of time-of-flight sensor and position, And road surface, isolation strip are approximately indeclinable at any time relative to the distance of time-of-flight sensor and position.Therefore can lead to The second image (depth image) the creation time differential depth image obtained using two width different moments is crossed to detect above-mentioned distance With the variation of position, and then mark this track label front vehicle identification range identify this track target vehicle of rear, label The front vehicle identification range of this non-track label identifies this non-track target vehicle of rear, knows in the front vehicle of combination of two Other range identifies rear lane change target vehicle.

Step 105, according to the kinematic parameter of objects ahead vehicle and rear area target vehicle to the kinematic parameter of main body vehicle Carry out cruise control.

It is understood that the movement according to the kinematic parameter of objects ahead vehicle and rear area target vehicle to main body vehicle Parameter there are many kinds of the modes of cruise control, different control is carried out according to different operating modes, is described as follows：

Specifically, in one embodiment of the invention, objects ahead vehicle model can be generated according to objects ahead vehicle It encloses, rear area target vehicle range is generated according to rear area target vehicle.

It is understood that generated according to objects ahead vehicle there are many kinds of the modes of objects ahead vehicle range, it can be with Selection setting is carried out according to the actual application.It is illustrated below：

The first example generates objects ahead vehicle model according to the enclosed region that the object boundary of objects ahead vehicle surrounds It encloses.

Second of example generates front mesh according to the enclosed region of the extension of the object boundary of objects ahead vehicle surrounded Mark vehicle range.

The third example generates front mesh according to the enclosed region that multiple location of pixels lines of objects ahead vehicle surround Mark vehicle range.

Specifically, in one embodiment of the invention, rear area target vehicle model can be generated according to rear area target vehicle There are many kinds of the modes enclosed, and can carry out selection setting according to the actual application.It is illustrated below：

The first example generates rear area target vehicle model according to the enclosed region that the object boundary of rear area target vehicle surrounds It encloses.

Second of example, square mesh after being generated according to the enclosed region of the extension of the object boundary of rear area target vehicle surrounded Mark vehicle range.

The third example, square mesh after being generated according to the enclosed region that multiple location of pixels lines of rear area target vehicle surround Mark vehicle range.

It specifically, in one embodiment of the invention, can also be according to the intertexture between the first image and the second image Objects ahead vehicle range is mapped in the first image and generates front car light identification region by mapping relations, according to third image and Intertexture mapping relations between 4th image, which map to rear area target vehicle range in third image, generates the identification of rear car light Region.

Specifically, the intertexture mapping relations between the first image and the second image indicate objects ahead vehicle in the second image Adjustment of the ranks coordinate of each pixel of range by equal proportion can at least determine the row of a pixel in the first image Row coordinate, and the imaging of the car light of objects ahead vehicle is included in corresponding objects ahead vehicle range, to first Car light identification region is generated in image.

Specifically, the intertexture mapping relations between third image and the 4th image indicate rear area target vehicle in the 4th image Adjustment of the ranks coordinate of each pixel of range by equal proportion can at least determine the row of a pixel in third image Row coordinate, and the imaging of the car light of rear area target vehicle is included in corresponding rear area target vehicle range, in third Car light identification region is generated in image.

Specifically, in one embodiment of the invention, corresponding front can also be identified according to front car light identification region The steering indicating light of target vehicle identifies the steering indicating light of corresponding rear area target vehicle according to rear car light identification region.

Specifically, it can be identified according to the color, flicker frequency or flashing sequence of tail-light in the car light identification region of front The steering indicating light of corresponding objects ahead vehicle.

For example, its length travel of the initial stage of objects ahead vehicle lane change and lateral displacement are all smaller, it is meant that front The car light identification region size variation of target vehicle is also smaller, and the brightness being imaged only at steering indicating light is changed greatly because of flicker.

Specifically, it can be identified according to the color, flicker frequency or flashing sequence of headlight in the car light identification region of rear The steering indicating light of corresponding rear area target vehicle.

For example, its length travel of the initial stage of rear area target vehicle lane change and lateral displacement are all smaller, it is meant that rear The car light identification region size variation of target vehicle is also smaller, and the brightness being imaged only at steering indicating light is changed greatly because of flicker.

The kinematic parameter of main body vehicle is carried out according to the kinematic parameter of objects ahead vehicle and rear area target vehicle as a result, There are many kinds of the modes of cruise control, can also add the accuracy that steering indicating light further increases control.It is specific as follows：

The first example recognizes this non-track target carriage of front according to the kinematic parameter of objects ahead vehicle and steering indicating light Deceleration lane change to this track operating mode so that the kinematic parameter control system of main body vehicle carries out braking adjustment in advance.As a result, The kinematic parameter control system of main body vehicle and security system are adjusted earlier, improves main body vehicle and its multiplies The driving safety of member so that the lamp system of main body vehicle can be adjusted earlier to remind rear area target vehicle, after being Square target vehicle provides more brakings or adjustment time, more effectively reduces rear-end impact risk.

Second of example recognizes this track target vehicle of front according to the kinematic parameter of objects ahead vehicle and steering indicating light Deceleration lane change to this non-track of front operating mode so that the kinematic parameter control system of main body vehicle without braking adjust.By This so that the kinematic parameter control system of main body vehicle can reduce unnecessary braking adjustment, to reduce due to main body Rear-end impact risk caused by the unnecessary braking adjustment of vehicle.

The third example recognizes this non-track target carriage of rear according to the kinematic parameter of rear area target vehicle and steering indicating light Deceleration lane change to this track operating mode so that the kinematic parameter control system of main body vehicle is adjusted in advance, and/or, with The lamp system of main body vehicle is set to remind rear area target vehicle.As a result, main body vehicle can light brake lamp with police in advance Show that rear area target vehicle driver cancels lane change or deceleration lane change, to slow down knocking into the back for main body vehicle and rear area target vehicle Risk of collision.

Fig. 2 is the flow diagram that the vehicle that another embodiment of the present invention provides travels autocontrol method.

As shown in Fig. 2, vehicle traveling autocontrol method includes the following steps：

Step 201, the first image that environment in front of main body vehicle is obtained from the imaging sensor of preposition 3D cameras, from preposition The time-of-flight sensor of 3D cameras obtains the second image of environment in front of main body vehicle, from the imaging sensor of postposition 3D cameras The third image for obtaining main body vehicle rear environment obtains main body vehicle rear ring from the time-of-flight sensor of postposition 3D cameras Wherein, the first image and third image are color or luminance picture to 4th image in border, and the second image and the 4th image are depth Image.

Specifically, imaging sensor refers to the array or set of luminance pixel sensor, for example, red, green, blue (RGB) or Luminance pixels sensor, the distance between the detection objects such as brightness, coloration (YUV) have relationship, commonly used in obtaining environment Luminance picture.

Specifically, time-of-flight sensor refers to the array or set of flight time element sensor, such as flight time picture Plain sensor can be optical sensor, phase detectors etc., can detect from light-pulse generator, modulated light source light in flight Between flight time for being propagated between element sensor and object to be detected, to detection object distance and obtain depth image.

For example, imaging sensor or time-of-flight sensor can use complementary metal oxide semiconductor (CMOS) technique is made, and luminance pixel sensor and flight time element sensor can be produced in proportion it is same On substrate, such as with 8:The 8 luminance pixel sensors and 1 flight time element sensor composition that 1 ratio is made One big intertexture pixel, wherein the photosensitive area of 1 flight time element sensor can be equal to 8 luminance pixel sensors Photosensitive area, wherein 8 luminance pixel sensors can be arranged by the array format that 2 rows and 4 arrange.

Further, for another example can be made on the substrate of 1 inch optical target surface 360 rows and 480 row it is above-mentioned active The array of intertexture pixel, can obtain 720 rows and 1920 row enliven luminance pixel sensor array, 360 rows and 480 row enliven TOF pixel sensor arrays, the same camera that thus imaging sensor and time-of-flight sensor form can obtain coloured silk simultaneously Color or luminance picture and depth image.

Therefore, in the present embodiment, the same preposition 3D cameras can be utilized to obtain about environment in front of main body vehicle First image and the second image, wherein the first image is colored or luminance picture, and the second image is depth image；It can utilize Same postposition 3D cameras obtain the third image and the 4th image about main body vehicle rear environment, wherein third image is Colored or luminance picture, the 4th image are depth image.

Wherein, colored or luminance picture pixel and depth image pixel are characterized in interleaved arrangement in proportion.And by CMOS technology manufacture can be used to realize and obtain the in the same preposition 3D cameras for obtaining the first image and the second image The same postposition 3D cameras of three images and the 4th image can use CMOS technology manufacture to realize, according to rubbing for semi-conductor industry That law, 3D cameras will be provided with sufficiently low production cost in limited period.

Step 202, when the first image be luminance picture, according in the first image front lines on highway and road surface brightness Lines on highway in front of difference identification, when third image is luminance picture, according to rear lines on highway and road in third image The luminance difference in face identifies rear lines on highway.

It should be noted that front (rear) the existing front of lines on highway (rear) solid line lane line also have front (after Side) dotted line lane line.When first image and third image are all luminance picture, the two identification front (rear) lines on highway Detailed process is identical, below using the first image as luminance picture when, according to front lines on highway and road surface in the first image It is described as follows for the lines on highway of luminance difference identification front：

Specifically, front lines on highway is created according to the luminance information of the first image and preset luminance threshold first Bianry image, it is to be appreciated that can search to obtain by using the luminance difference of front lines on highway and road surface Certain luminance thresholds as preset luminance threshold, preset luminance threshold can using " statistics with histogram-is bimodal " algorithm come Lookup obtains, and the bianry image of front lines on highway is created using preset luminance threshold and luminance picture.

It will also be appreciated that can be by the way that luminance picture be divided into multiple brightness subgraphs, to each brightness subgraph It searches to obtain multiple luminance thresholds as executing " statistics with histogram-is bimodal " algorithm, utilizes each luminance threshold and corresponding bright The two-value subgraph of lines on highway in front of creation of sub-pictures is spent, and utilizes the complete front road driveway of two-value creation of sub-pictures The bianry image of line.

Further, whole edges of straight way solid line lane line are detected in bianry image according to preset detection algorithm Location of pixels or the whole edge pixel locations for detecting solid line lane line of going off the curve.

It is understood that the radius of curvature of front lines on highway can not possibly be too small, and due to camera projection theory Cause the imaging pixel of the nearby opposite lines on highway of front at a distance of front lines on highway more so that the front solid line of bend The pixel that lane line is arranged in a straight line in luminance picture also accounts for the major part of front solid line lane line imaging pixel.

Thus, it is possible to by the preset detection algorithm such as Hough transform algorithm front lines on highway bianry image In detect that whole edge pixel locations of straight way solid line lane line or detection are gone off the curve whole edge pixels of solid line lane line Position.

It should be noted that above-mentioned preset detection algorithm can also be straight in bianry image by isolation strip, electric pole Line edge pixel location detects.According to the Aspect Ratio of imaging sensor, the road of 3D camera lens focal length, highway layout specification Slope range of the lane line in bianry image can be arranged in the installation site of main body vehicle in width range and imaging sensor, It is excluded to be filtered the straight line of non-lane line according to slope range.

It should be noted that the edge pixel location of the solid line lane line of bend always consecutive variations, therefore according to looking into The connected pixel position of the edge pixel location at the initial straight both ends of above-mentioned detection is looked for, and the connected pixel position is incorporated to this Initial straight edge pixel set repeats above-mentioned lookup and is incorporated to the connected pixel position, finally by bend solid line lane line Whole edge pixel locations uniquely determine.

It will be appreciated that aforesaid way can also detected front dotted line lane line whole edge pixel location.

As an example, according to be mutually parallel in the priori of solid line lane line, lane line principle, image sensing Whole edge pixel locations of solid line lane line are projected to the initial straight of dotted line lane line by the projective parameter of device and 3D cameras Edge pixel location with connect the dotted line lane line initial straight edge pixel location and belong to the dotted line lane line other The edge pixel location of shorter lane line, to obtain whole edge pixel locations of dotted line lane line.

As another example, the priori of straight way or bend need not be obtained.Since vehicle is in straight way cruise or perseverance During determining the cruise of steering angle bend, the lateral shift of dotted line lane line can almost be ignored within shorter continuous time, But vertical misalignment is larger, therefore dotted line lane line can be in the bianry image of continuous a few width lines on highway of different moments It is superimposed as a solid line lane line, then obtains the whole of the dotted line lane line by the recognition methods of above-mentioned solid line lane line again Edge pixel location.

It is understood that the vertical misalignment amount due to dotted line lane line is influenced by main body vehicle speed, it can with this With according to the bianry image for the continuous lines on highway for dynamically determining different moments from the speed of wheel speed sensors acquisition Minimum width number by dotted line lane line to be superimposed as a solid line lane line, to obtain whole edge pixel positions of dotted line lane line It sets.

It should be noted that in the present embodiment, whole edge pixel positions of " lines on highway " acute pyogenic infection of finger tip lines on highway It sets.

It should be noted that identifying rear highway according to rear lines on highway in third image and the luminance difference on road surface Lane line includes the bianry image that rear lines on highway is created according to the luminance information of third image and preset luminance threshold； Whole edge pixel locations of straight way solid line lane line are detected in bianry image according to preset detection algorithm or are detected Whole edge pixel locations of bend solid line lane line；Straight way dotted line is detected in bianry image according to preset detection algorithm Whole edge pixel locations of lane line or the illustrating for whole edge pixel locations for detecting dotted line lane line of going off the curve can With referring to identifying the specific of front lines on highway according to front lines on highway in the first image and the luminance difference on road surface Description, and will not be described here in detail.

Step 203, front lines on highway is mapped according to the intertexture mapping relations between the first image and the second image To multiple front vehicles identification ranges are generated in the second image, according to the intertexture mapping relations between third image and the 4th image Rear lines on highway is mapped in the 4th image and generates multiple front vehicle identification ranges.

It should be noted that the description of step S203 is corresponding with above-mentioned steps S103, thus to step S203 retouch The description with reference to above-mentioned steps S103 is stated, details are not described herein.

Step 204, the label that all front vehicles identification ranges are marked with front this track and this non-track of front, according to This track target vehicle of the vehicle identification range identification front of this track label of label front, is marked according to this non-track of label front This non-track target vehicle of front vehicles identification range identification front of label, knows according to the front vehicles identification range of combination of two Other lane change ahead target vehicle.

In being embodiment at one of the present invention, according to the front lines on highway of the equal proportion obtained in the second image, The line number shared by the initial linear portion of each front lines on highway is taken to compare to obtain the first of front lines on highway with columns The slope of beginning straight line, to the front lines on highway where the initial straight according to two front lines on highway of maximum slope The front vehicles identification range of establishment marks the label in this track, and Fei Benche is marked to the front vehicles identification range of other establishments The label in road.

It is understood that the second image is depth image, the light reflection at the back side of the same objects ahead vehicle to TOF The depth sub-picture pack that sensor is formed is containing consistent range information, as long as therefore identifying depth that objects ahead vehicle is formed Position of the image in depth image can obtain the range information of objects ahead vehicle.Wherein, subgraph refers to one of image Divide the combination of pixel.

It is understood that the light reflection at the back side of the same objects ahead vehicle forms depth subgraph to TOF sensor It includes consistent range information to seem, and it includes consecutive variations that the light reflection on road surface forms depth subgraph to be to TOF sensor Range information, therefore include depth of depth subgraph and the range information comprising consecutive variations of consistent range information Image necessarily forms mutation differences in the intersection of the two, and the boundary of these mutation differences forms objects ahead vehicle in depth Object boundary in image.

It is possible to further by using Canny, Sobel, the Laplace etc. on the detection boundary in image processing algorithm A variety of boundary detection methods are to detect the object boundary of objects ahead vehicle.

Further, vehicle identification range is determined by whole location of pixels of lane line, therefore in vehicle identification range The object boundary of interior detection objects ahead vehicle is done the boundary that the road equipments such as isolation strip, light pole, fender pile are formed is reduced It disturbs.

Specifically, the object boundary detected within the scope of each vehicle identification is projected to the row reference axis of image respectively, And one-dimensional lookup is carried out in reference axis of being expert at, you can determine longitudinal mesh of all objects ahead vehicles in front vehicles identification range The line number and row coordinate range shared by boundary are marked, and determines the shared columns and row coordinate position of lateral object boundary.Its In, longitudinal object boundary refers to the object boundary for occupying that number of lines of pixels is more and columns is few, and lateral object boundary, which refers to have, occupies pixel Line number is few and columns more than object boundary.

Further, the columns shared by lateral object boundary all in front vehicles identification range, row coordinate bit It sets, row coordinate position (namely the respective transversal object boundary of all longitudinal object boundaries is searched in front vehicles identification range Row coordinate initial position and final position), and different mesh are distinguished according to the principle that object boundary includes consistent range information The object boundary for marking vehicle, so that it is determined that the position of all objects ahead vehicles and range information in front vehicles identification range.

Therefore, the object boundary of detection acquisition objects ahead vehicle can uniquely determine the depth of objects ahead vehicle formation Position of the subgraph in depth image is spent, to uniquely determine the range information of objects ahead vehicle.

Multiple objects ahead vehicles can be detected by the boundary detection method of above-mentioned example simultaneously as a result, and its apart from letter Breath.

Further, according to above-mentioned example, in front of the label within the scope of the vehicle identification of this track label in front of identification this Track target vehicle, this non-track target vehicle in front of identification within the scope of the vehicle identification of this non-track label in front of the label, Lane change objects ahead vehicle is identified in the front vehicles identification range of combination of two.

Step 205, the label that all front vehicle identification ranges are marked with this track of rear and this non-track of rear, according to It marks the vehicle identification range of this track label of rear to identify this track target vehicle of rear, is marked according to this non-track of label rear The vehicle identification range of label identifies this non-track target vehicle of rear, after being identified according to the front vehicle identification range of combination of two Fang Biandao target vehicles.

In being embodiment at one of the present invention, according to the rear lines on highway of the equal proportion obtained in the 4th image, The line number shared by the initial linear portion of each rear lines on highway is taken to compare to obtain the first of rear lines on highway with columns The slope of beginning straight line, to the rear lines on highway where the initial straight according to two rear lines on highway of maximum slope The front vehicle identification range of establishment marks the label in this track, and Fei Benche is marked to the front vehicle identification range of other establishments The label in road.

It is understood that the second image is depth image, the light reflection at the back side of the same rear area target vehicle to TOF The depth sub-picture pack that sensor is formed is containing consistent range information, as long as therefore identifying depth that rear area target vehicle is formed Position of the image in depth image can obtain the range information of rear area target vehicle.Wherein, subgraph refers to one of image Divide the combination of pixel.

It is understood that the light reflection at the back side of the same rear area target vehicle forms depth subgraph to TOF sensor It includes consistent range information to seem, and it includes consecutive variations that the light reflection on road surface forms depth subgraph to be to TOF sensor Range information, therefore include depth of depth subgraph and the range information comprising consecutive variations of consistent range information Image necessarily forms mutation differences in the intersection of the two, and the boundary of these mutation differences forms rear area target vehicle in depth Object boundary in image.

It is possible to further by using Canny, Sobel, the Laplace etc. on the detection boundary in image processing algorithm A variety of boundary detection methods are to detect the object boundary of rear area target vehicle.

Further, front vehicle identification range is determined by whole location of pixels of lane line, therefore in front vehicle The object boundary of detection rear area target vehicle is formed the road equipments such as isolation strip, light pole, fender pile are reduced in identification range Border interference.

Specifically, the object boundary detected in each front vehicle identification range is projected to the row reference axis of image respectively On, and one-dimensional lookup is carried out in reference axis of being expert at, you can determine the vertical of all rear area target vehicles in front vehicle identification range To the line number and row coordinate range shared by object boundary, and determine the shared columns and row coordinate bit of lateral object boundary It sets.Wherein, longitudinal object boundary refers to the object boundary for occupying that number of lines of pixels is more and columns is few, and lateral object boundary, which refers to have, to be occupied Number of lines of pixels is few and columns more than object boundary.

Further, the columns shared by lateral object boundary all in front vehicle identification range, row coordinate bit It sets, row coordinate position (namely the respective transversal object boundary of all longitudinal object boundaries is searched in front vehicle identification range Row coordinate initial position and final position), and different mesh are distinguished according to the principle that object boundary includes consistent range information The object boundary for marking vehicle, so that it is determined that the position of all rear area target vehicles and range information in front vehicle identification range.

Therefore, the object boundary of detection acquisition rear area target vehicle can uniquely determine the depth of rear area target vehicle formation Position of the subgraph in depth image is spent, to uniquely determine the range information of rear area target vehicle.

Multiple rear area target vehicles can be detected by the boundary detection method of above-mentioned example simultaneously as a result, and its apart from letter Breath.

Further, according to above-mentioned example, rear sheet is identified within the scope of the vehicle identification of label this track label of rear Track target vehicle identifies this non-track target vehicle of rear within the scope of the vehicle identification of label this non-track label of rear, Lane change rear area target vehicle is identified in the front vehicle identification range of combination of two.

Step 206, objects ahead vehicle range is generated according to the enclosed region that the object boundary of objects ahead vehicle surrounds, Objects ahead vehicle range is mapped into life in the first image according to the intertexture mapping relations between the first image and the second image At front car light identification region, according to the intertexture mapping relations between third image and the 4th image by rear area target vehicle range Map to generation rear car light identification region in third image.

Step 207, it is identified according to the color, flicker frequency of tail-light in the car light identification region of front or flashing sequence corresponding The steering indicating light of objects ahead vehicle is known according to the color, flicker frequency of headlight in the car light identification region of rear or flashing sequence The steering indicating light of not corresponding rear area target vehicle.

Specifically, by continuously acquiring the colour or luminance picture and to front (rear) target vehicle of several different moments Car light identification region carry out time diffusion processing with create front (rear) target vehicle time diffusion car light identification region Subgraph.

Wherein, time diffusion car light identification region subgraph will protrude the vehicle continuously flickered of front (rear) target vehicle Taillight (headlight) subgraph.Then time diffusion car light identification region subgraph is projected into row reference axis, carries out one-dimensional look into Starting and the terminal row coordinate position for looking for tail-light (headlight) subgraph for obtaining front (rear) target vehicle, these are risen Begin and terminal row coordinate position projects to time diffusion car light identification region subgraph and searches tail-light (headlight) subgraph Starting and end line coordinate position, by the row, column coordinate position of the starting of tail-light (headlight) subgraph and terminal project Extremely to confirm the tail-light (headlight) of front (rear) target vehicle in the colour or luminance picture of several above-mentioned different moments Color, flicker frequency or flashing sequence, so that it is determined that the row, column coordinate bit of tail-light (headlight) subgraph of flicker It sets.

Further, the row, column coordinate position of tail-light (headlight) subgraph of flicker is only in front (rear) target It can determine that front (rear) target vehicle is beating left steering lamp, the tail-light of flicker when on the left of the car light identification region of vehicle The row, column coordinate position of (headlight) subgraph only can be true at the car light identification region right side of front (rear) target vehicle Fixed front (rear) target vehicle exists beating right turn lamp, the row, column coordinate position of tail-light (headlight) subgraph of flicker It can determine that front (rear) target vehicle is playing double sudden strain of a muscle warnings when the car light identification region both sides of front (rear) target vehicle Lamp.

It should be noted that its length travel or lateral displacement are larger during front (rear) target vehicle lane change Cause the car light identification region size variation of front (rear) target vehicle also larger, can to continuously acquire several it is different when The car light identification region of front (rear) target vehicle at quarter carries out length travel or lateral displacement compensates and is scaled to size one The car light identification region of cause, then time diffusion processing is carried out to the car light identification region of front (rear) target vehicle after adjustment To create the time diffusion car light identification region subgraph of front (rear) target vehicle.

Further, time diffusion car light identification region subgraph is projected into row reference axis, carries out one-dimensional lookup and obtains The starting of tail-light (headlight) subgraph of front (rear) target vehicle and terminal row coordinate position, by these startings and eventually Point range coordinate position projects to time diffusion car light identification region subgraph and searches the starting of tail-light (headlight) subgraph With end line coordinate position, the row, column coordinate position of the starting of tail-light (headlight) subgraph and terminal is projected to above-mentioned In the colour or luminance picture of several different moments with confirm front (rear) target vehicle tail-light (headlight) color, Flicker frequency or flashing sequence, so that it is determined that the row, column coordinate position of tail-light (headlight) subgraph of flicker, last complete At left steering lamp, right turn lamp or double identifications for dodging warning lamp.

Step 208, this non-track target vehicle of front is recognized according to the kinematic parameter of objects ahead vehicle and steering indicating light Deceleration lane change to this track operating mode so that the kinematic parameter control system of main body vehicle carries out braking adjustment in advance.

It should be noted that with the left and right lane line in this track of front be reference, no matter objects ahead vehicle lane change when at In straight way or bend, no matter the leftward or rightward lane change of objects ahead vehicle can identify accurately, to based on vehicle from It adapts to cruise system and accurate motion control foundation is provided.

As a result, main body vehicle can light brake lamp in advance with warn rear area target vehicle driver cancel lane change or Deceleration lane change, to slow down the rear-end impact risk of main body vehicle and rear area target vehicle.

As another realization method, this track of front is recognized according to the kinematic parameter of objects ahead vehicle and steering indicating light Target vehicle deceleration lane change to this non-track of front operating mode so that the kinematic parameter control system of main body vehicle without braking Adjustment.

For example, the left side target side of objects ahead vehicle when the right turn lamp of this track target vehicle lights is recognized The pixel distance of boundary to this track left-hand lane line is determined as lateral distance P through the conversion of camera projection relation；By continuously acquiring N The first image of width different moments and the second image (time for obtaining first image or the second image is T), during which identify And record the variation of the distance R of target vehicle；It recognizes objects ahead vehicle and just completes non-on the right side of lane change to this track Track, for the left side object boundary of objects ahead vehicle to this track right-hand lane line coincidence, this lane width is D at this time；Therefore, Kinematic parameter of objects ahead vehicle during continuous lane change is duration N × T, the distance of opposite main body vehicle be R and Lateral displacement is (D-P).

Therefore, according to the distance R during the objects ahead vehicle lane change of above-mentioned identification, main body vehicle adaptive cruise system As long as system determines that R is consistently greater than the safety cruise braking distance set and can remain a constant speed cruise, in addition even if recognizing in front of Target vehicle just completes this non-track on the right side of lane change to this track, and the left side object boundary of objects ahead vehicle is to originally at this time Track right-hand lane line overlaps and R is less than safety cruise braking distance, and main body vehicle self-adaption cruise system may also reduce dynamic Power exports, and slightly waits for and recognizes target vehicle and continue displacement to the right and generate the lateral displacement of safer bigger that power can be restored Output is with the cruise that remains a constant speed.

As a result, the kinematic parameter control system of main body vehicle can reduce unnecessary braking adjustment, to reduce Rear-end impact risk caused by unnecessary braking adjustment due to main body vehicle.

As another realization method, rear Fei Benche is recognized according to the kinematic parameter of rear area target vehicle and steering indicating light Road target vehicle deceleration lane change to this track operating mode so that the kinematic parameter control system of main body vehicle is adjusted in advance, And/or so that the lamp system of main body vehicle reminds rear area target vehicle.

This non-track target vehicle deceleration lane change of rear is recognized as a result, to the operating mode in this track of main body vehicle so that main Body vehicle can light brake lamp to warn rear area target vehicle driver to cancel lane change or deceleration lane change, to slow down in advance The rear-end impact risk of main body vehicle and rear area target vehicle.

In order to which those skilled in the art more understand that above-described embodiment controls process, illustrated such as with reference to Fig. 3 to Fig. 5 Under：

It is understood that the lateral displacement of above-mentioned identification with the left and right lane line in this track be reference, no matter target carriage When lane change in straight way or bend, no matter the leftward or rightward lane change of target vehicle can identify accurately, thus based on Vehicle self-adaption cruise system provides accurately control foundation.

As shown in figure 3, can control main body vehicle in common straight way or bend operating mode follows this track target of front Vehicle at the uniform velocity cruises, and the distance of identification objects ahead vehicle is RA, and recognizes this non-track of main body vehicle left rear side simultaneously The distance of target vehicle is RB and identifies that the lane change that its right turn lamp is flickering is intended to；When RB is apart from too small, rear area target Vehicle lane change to main body vehicle rear is easy to happen rear-end impact, since identification rear area target vehicle plays the initial lane change of steering indicating light It is intended to, the brake lamp that can control main body vehicle is lighted in advance to warn rear area target vehicle driver to cancel lane change or change of slowing down Road, to slow down the rear-end impact risk of main body vehicle and rear area target vehicle.

It is possible to further accurately identify rear area target vehicle continue by force lane change and lane change do not slow down generation relative to The lateral displacement of this lane line, the cruise system that can control main body vehicle automatically improve speed suitably to reduce main body vehicle With the following distance of objects ahead vehicle, increase at a distance from main body vehicle and rear area target vehicle, to slow down main body vehicle Rear-end impact risk with rear area target vehicle.

However, traditional only rely on millimetre-wave radar or the vehicle identification rear lane change target vehicle of laser radar has foot Enough big lane change lateral displacements could judge that the lane change of rear area target vehicle is intended to, and rear-end impact risk will be caused to increase.

Therefore, the steering indicating light of the kinematic parameter of identification rear area target vehicle and corresponding identification rear area target vehicle, Ke Yishi This non-track target vehicle deceleration lane change of rear is clipped to the operating mode in this track of main body vehicle so that main body vehicle can be bright in advance Brake lamp is played to warn rear area target vehicle driver to cancel lane change or deceleration lane change, to slow down main body vehicle and rear square mesh Mark the rear-end impact risk of vehicle.

Further, the mesh of traditional vehicle self-adaption cruise system identification for only relying on millimetre-wave radar or laser radar It is reference that the lateral displacement of mark vehicle, which is with main body vehicle, is the lateral displacement with reference to the target vehicle of identification with main body vehicle Sometimes the accurate motion control foundation of vehicle self-adaption cruise system cannot be supplied to.

As shown in figure 4, when this track objects ahead vehicle from this track complete lane change to the right be just in it is curved curved to the left When road, the millimetre-wave radar or laser radar of conventional truck may still identify that objects ahead vehicle sections are in this on straight way On track, 250 meters of above-mentioned bend radius of curvature, above-mentioned objects ahead vehicle lane change has travelled 25 meters on bend in the process, with This track right-hand lane line that the left side object boundary of objects ahead vehicle overlaps opposite lane line at 25 meters of the bend Straight way extended line is offset by the left

It is understood that if the millimetre-wave radar or laser radar of above-mentioned conventional truck recognize objects ahead vehicle at this time Distance be 50 meters to 80 meters, i.e., the millimetre-wave radar or laser radar of above-mentioned conventional truck are located on straight way and apart from curved Road entrance still has 25 meters to 55 meters of distance, the millimetre-wave radar or laser radar of above-mentioned conventional truck to know in shortage bend priori Objects ahead vehicle will be recognized in the case of knowledge, and still there are about the vehicle bodies of 1.25 meters of width on this track, and with target Vehicle is continued on to left bend Reduced Speed Now, and the millimetre-wave radar or laser radar of above-mentioned conventional truck recognize objects ahead Vehicle has the vehicle body of bigger width on this track, i.e., the millimetre-wave radar or laser radar of above-mentioned conventional truck will produce not It accurately identifies and conventional truck self-adaption cruise system will be caused to execute continuous inaccurate and unnecessary braking, to lead Conventional truck and the rear-end impact risk of its rear area target vehicle is caused to increase.

Similarly, the millimetre-wave radar or laser radar of above-mentioned conventional truck to this above-mentioned track target vehicle in bend to the right On from this track complete lane change to the left identification there is also inaccuracies.

Therefore, it according to the steering indicating light of the kinematic parameter of the target vehicle of identification and corresponding identification target vehicle, can identify To this track target vehicle deceleration lane change to the operating mode in this non-track of main body vehicle so that the kinematic parameter control system of main body vehicle System can reduce unnecessary braking adjustment, knock into the back caused by the unnecessary braking adjustment due to main body vehicle to reduce Risk of collision.

It is understood that this non-track target vehicle can be identified and be monitored from steering indicating light is played to completing lane change to this vehicle The continuous process in road, and the distance, opposite of duration of objects ahead vehicle during continuous lane change, opposite main body vehicle The kinematic parameters such as speed and lateral displacement are also easy to be monitored, to which the kinematic parameter of objects ahead vehicle can be used for controlling master The kinematic parameter of body vehicle with make earlier braking adjust and improve driving safety and earlier control car light warning after square mesh Vehicle is marked to reduce rear-end impact risk.

As shown in figure 5, main body vehicle is travelled in this track straight way with constant speed mode, and still there are 55 meters apart from bend entrance The distance of (or to 25 meters), bend is bending to the right and radius of curvature is 250 meters, in 25 meters of this tracks in front of bend entrance Right side has this non-track objects ahead vehicle beating left steering lamp to this lane, and a left side for objects ahead vehicle Side object boundary is overlapped with the right-hand lane line in this track.

Thus, it is possible to accurately identify objects ahead vehicle to this lane, due to objects ahead vehicle distances master About 80 meters of body vehicle (or to 50 meters), the dynamical system that can control main body vehicle accurately execute power output reduction even system Dynamic acts, lights brake lamp in time, to ensure main body vehicle and front, the safe distance of rear area target vehicle, to improve The driving safety of main body vehicle and reduce rear-end impact risk.

It should be noted that traditional vehicle self-adaption cruise system identification for only relying on millimetre-wave radar or laser radar Objects ahead vehicle lateral displacement be with main body vehicle be reference, in the case where lacking bend priori will identify The extended line of this track right-hand lane line of objects ahead vehicle distances also there are aboutCross To distance, i.e., mistakenly identification objects ahead vehicle needs to continue about 1.25 meters of above-mentioned millimetre-wave radars of lateral displacement to the left or swash Optical radar just can confirm that objects ahead vehicle initially enters this track.

It is understood that if objects ahead lateral direction of car velocity of displacement is 1 metre per second (m/s), above-mentioned traditional only relies in the least The vehicle self-adaption cruise system of metre wave radar or laser radar will actually enter this track about 1.25 seconds in objects ahead vehicle The action that power output reduces even braking could be executed later, undoubtedly reduce main body vehicle and front, rear area target vehicle Safe distance, the driving safety for resulting in main body vehicle declines and increases rear-end impact risk.

Therefore, it according to the steering indicating light of the kinematic parameter of the target vehicle of identification and corresponding identification target vehicle, can identify To this non-track target vehicle deceleration lane change to the operating mode in this track of main body vehicle so that the kinematic parameter control system of main body vehicle System and security system can adjust earlier, improve main body vehicle and its driving safety of occupant so that main body vehicle Lamp system can adjust earlier to remind rear area target vehicle, for rear area target vehicle provide more brakings or Adjustment time more effectively reduces rear-end impact risk.

Further, be based on above-described embodiment, in order to effectively further reduce rear-end impact risk and further Improve the driving safety of main body vehicle.It can be handled by way of calibrating automatically.Detailed process combination Fig. 6 descriptions It is as follows：

Step 601, objects ahead vehicle heading angle is obtained according to objects ahead vehicle range or front car light identification region.

Specifically, there are many kinds of the modes for obtaining objects ahead vehicle heading angle, for example passes through preset algorithm, formula etc.. It is illustrated below：

As an example, the lens parameters and installation site for obtaining the video camera of the first image or the second image first can By camera calibration technical limit spacing known to prior those skilled in the art, therefore the road using video camera as origin can be established The relationship look-up table of scenery coordinate and the pixel coordinate of the first image or the second image.

Further, can include by objects ahead vehicle range or front car light identification region by above-mentioned relation look-up table Pixel coordinate be converted to the objects ahead vehicle coordinate using video camera as origin, to according to converting using video camera into origin Objects ahead vehicle coordinate calculate using video camera as the objects ahead vehicle heading angle of origin.

Step 602, the intermediate-freuqncy signal for obtaining dead load frequency radar obtains the relatively fast of front radar target according to intermediate-freuqncy signal Degree and azimuth.

It should be noted that the transmitter of dead load frequency radar works on nearly constant wave frequency, therefore dead load The electromagnetic wave bandwidth very little that frequency radar is occupied relative to the frequency modulated(FM) radar of ranging, to which dead load frequency radar can reduce component Using and reduce cost.

It is understood that when objects ahead vehicle and main body vehicle are there are when relative velocity, dead load frequency radar receives The reflection signal of objects ahead vehicle can pass through phase shifter and generate orthogonal reflection signal, orthogonal reflection signal and dead load frequency thunder The transmitting signal reached generates orthogonal intermediate-freuqncy signal by frequency mixer, and orthogonal intermediate-freuqncy signal includes about the how general of above-mentioned relative velocity Frequency is strangled, the size of Doppler frequency is directly proportional to the size of relative velocity, the sign of Doppler frequency and and relative velocity Sign it is identical.

Further, the prominent Doppler can be created in the way of analog-digital converter and complex fast Fourier algorithm etc. The frequency spectrum of the orthogonal intermediate-freuqncy signal of frequency, this that can obtain the frequency spectrum of the orthogonal intermediate-freuqncy signal using peak detection algorithm are more The general size and sign for strangling frequency；According to the size of the Doppler frequency of acquisition and sign i.e. using Doppler range rate measurement Formula determines the size and sign of relative velocity.

Further, dead load frequency radar can obtain the orientation of front radar target comprising more than two receivers Angle.The position difference that each receiver of dead load frequency radar is mutual causes the orthogonal intermediate-freuqncy signal that each receiver obtains same There are phase differences for phase at Doppler frequency.

Further, phase of each receiver obtained according to the frequency spectrum of orthogonal intermediate-freuqncy signal at same Doppler frequency Potential difference and the mutual position relationship of each receiver are the azimuth that front radar target is obtained using phase method angle measurement formula.

Step 603, front is obtained according to objects ahead vehicle heading angle, the relative velocity of front radar target and azimuth The relative velocity of target vehicle calibrates dead load frequency radar according to the objects ahead vehicle of identification automatically.

It is understood that when there are multiple objects ahead vehicles, the azimuth of multiple objects ahead vehicles is obtained, is connect Relative velocity and the azimuth of multiple front radar targets will be obtained by, using single objects ahead vehicle azimuth with it is a certain The relative velocity of front radar target is determined as objects ahead vehicle by the approximately equal principle in azimuth of front radar target Relative velocity.

It is understood that when the installation site of video camera and dead load frequency radar differs farther out, above-mentioned azimuth is approximate Equal principle may lead to error, as long as according to the installation site relationship of video camera and dead load frequency radar by the Bu Tong former of the two The azimuthal coordinate that the azimuthal coordinate of point is calibrated to same origin can eliminate above-mentioned error.

It is understood that when the installation site of dead load frequency radar is in other than driver's cabin, azimuthal measurement result Possible vibrated, temperature change, sleet dirt covering influence, needs to be calibrated automatically.

For example, there are multiple and different azimuthal objects ahead vehicles when recognizing main body vehicle, you can comparison identification The azimuth of multiple objects ahead vehicles and the azimuth of front radar target whether have consistent deviation, if having consistent inclined Difference records the deviation into the reservoir of dead load frequency radar, and dead load frequency radar reads this in subsequent azimuth determination partially Difference is calibrated and compensated for automatically；If there is inconsistent deviation, dead load frequency radar can be sent out to main body vehicle driver can not Warning reminds main body vehicle driver to check or clean dead load frequency radar.

As a result, by way of calibrating automatically, effectively further reduces rear-end impact risk and further increase The driving safety of main body vehicle.

In order to realize that above-described embodiment, the present invention also propose a kind of vehicle traveling automatic control device.

Fig. 7 is the structural schematic diagram that vehicle provided by one embodiment of the present invention travels automatic control device.

As shown in fig. 7, vehicle traveling automatic control device includes：First acquisition module 701, the second acquisition module 702, Third acquisition module 703, the first generation module 704, the first identification module 705 and control module 706.

Wherein, the first acquisition module 701 be used for from preposition 3D cameras obtain main body vehicle in front of environment the first image and Second image obtains the third image and the 4th image of main body vehicle rear environment from postposition 3D cameras, wherein the first image and Third image is color or luminance picture, and the second image and the 4th image are depth image.

Second acquisition module 702 is used to obtain front lines on highway according to the first image.

Third acquisition module 703 is used to obtain rear lines on highway according to third image.

First generation module 704 is used for front highway according to the intertexture mapping relations between the first image and the second image Lane line, which maps in the second image, generates multiple front vehicles identification ranges, according to the friendship between third image and the 4th image It knits mapping relations and maps to rear lines on highway in the 4th image and generate multiple front vehicle identification ranges.

First identification module 705 is used to identify objects ahead vehicle according to all front vehicles identification ranges, according to all Front vehicle identification range identifies rear area target vehicle.

Control module 706 is used for kinematic parameter and steering indicating light according to objects ahead vehicle and rear area target vehicle to master The kinematic parameter of body vehicle carries out cruise control.

Further, in a kind of possible realization method of the embodiment of the present invention, the first acquisition module 701 is in the past The imaging sensor for setting 3D cameras obtains the first image of environment in front of main body vehicle；It is sensed from the flight time of preposition 3D cameras Device obtains the second image of environment in front of main body vehicle；Main body vehicle rear environment is obtained from the imaging sensor of postposition 3D cameras Third image；The 4th image of main body vehicle rear environment is obtained from the time-of-flight sensor of postposition 3D cameras.

Further, in a kind of possible realization method of the embodiment of the present invention, the second acquisition module 702 includes：The One recognition unit 7021 and the first converting unit 7022.

Wherein, the first recognition unit 7021 is used to when the first image be luminance picture, according to front highway in the first image Lane line and the luminance difference on road surface identification front lines on highway.Alternatively,

First converting unit 7022 is used to when the first image be color image, and color image is converted to luminance picture, the One recognition unit 7021 is used for according to front lines on highway in the first image and the luminance difference on road surface identification front highway car Diatom.

In one embodiment of the invention, the first recognition unit 7021 is specifically used for the luminance information according to the first image The bianry image of front lines on highway is created with preset luminance threshold；It is examined in bianry image according to preset detection algorithm It measures whole edge pixel locations of straight way solid line lane line or detects the whole edge pixel locations for solid line lane line of going off the curve； Whole edge pixel locations of straight way dotted line lane line are detected in bianry image according to preset detection algorithm or are detected Whole edge pixel locations of bend dotted line lane line.

Further, in a kind of possible realization method of the embodiment of the present invention, third acquisition module 703 includes：The Two recognition units 7031 and the first converting unit 7032.

Wherein, the second recognition unit 7031 is used to when third image be luminance picture, according to rear highway in third image Lane line and the luminance difference on road surface identify rear lines on highway.Alternatively,

Second converting unit 7032 is used to when third image be color image, and color image is converted to luminance picture, the Two recognition units 7031 are used to identify rear highway car according to rear lines on highway in third image and the luminance difference on road surface Diatom.

In one embodiment of the invention, the second recognition unit 7031 is specifically used for the luminance information according to third image The bianry image of rear lines on highway is created with preset luminance threshold；It is examined in bianry image according to preset detection algorithm It measures whole edge pixel locations of straight way solid line lane line or detects the whole edge pixel locations for solid line lane line of going off the curve； Whole edge pixel locations of straight way dotted line lane line are detected in bianry image according to preset detection algorithm or are detected Whole edge pixel locations of bend dotted line lane line.

In one embodiment of the invention, the first identification module 705 is specifically used for all front vehicles identification ranges The label in label front this track and this non-track of front；Before the vehicle identification range identification of this track label of label front This track target vehicle of side；According to this non-track target carriage of vehicle identification range identification front of this non-track label of label front ；Lane change ahead target vehicle is identified according to the front vehicles identification range of combination of two.To all front vehicle identification ranges Mark the label in this track of rear and this non-track of rear；After the vehicle identification range identification of label this track label of rear This track target vehicle of side；This non-track target carriage of rear is identified according to the vehicle identification range of label this non-track label of rear ；Rear lane change target vehicle is identified according to the front vehicle identification range of combination of two.

In one embodiment of the invention, the first identification module 705 is additionally operable to using the boundary in image processing algorithm The object boundary of detection method detection objects ahead vehicle is identified；Using the boundary detection method inspection in image processing algorithm The object boundary for surveying rear area target vehicle is identified.

Further, in a kind of possible realization method of the embodiment of the present invention, as shown in figure 8, on the basis of Fig. 7 On, vehicle traveling automatic control device further includes：Second generation module 707, third generation module 708, the second identification module 709, the 4th acquisition module 710, the 5th acquisition module 711 and calibration module 712.

Wherein, the second generation module 707 is used to generate objects ahead vehicle range according to objects ahead vehicle, according to rear Target vehicle generates rear area target vehicle range.

Third generation module 708 is used for objects ahead according to the intertexture mapping relations between the first image and the second image Vehicle range, which maps in the first image, generates front car light identification region, according to the intertexture between third image and the 4th image Rear area target vehicle range is mapped to generation rear car light identification region in third image by mapping relations.

Second identification module 709 is used to identify the steering indicating light of corresponding objects ahead vehicle according to front car light identification region, The steering indicating light of corresponding rear area target vehicle is identified according to rear car light identification region.

Further, in a kind of possible realization method of the embodiment of the present invention, the second generation module 707 is used for basis The enclosed region that the object boundary of objects ahead vehicle surrounds generates objects ahead vehicle range, according to the mesh of rear area target vehicle The enclosed region that mark boundary surrounds generates rear area target vehicle range.Alternatively,

Objects ahead vehicle range is generated according to the enclosed region of the extension of the object boundary of objects ahead vehicle surrounded, Rear area target vehicle range is generated according to the enclosed region of the extension of the object boundary of rear area target vehicle surrounded.Alternatively,

Objects ahead vehicle range is generated according to the enclosed region that multiple location of pixels lines of objects ahead vehicle surround, Rear area target vehicle range is generated according to the enclosed region that multiple location of pixels lines of rear area target vehicle surround.

Further, in a kind of possible realization method of the embodiment of the present invention, the second identification module 709 is used for basis Color, flicker frequency or the flashing sequence of tail-light identify the steering of corresponding objects ahead vehicle in the car light identification region of front Lamp；Corresponding rear area target vehicle is identified according to the color, flicker frequency of headlight in the car light identification region of rear or flashing sequence Steering indicating light.

Further, in a kind of possible realization method of the embodiment of the present invention, control module 706 is used for according to front The kinematic parameter and steering indicating light of target vehicle recognize non-this track target vehicle deceleration lane change in front to the operating mode in this track, with The kinematic parameter control system of main body vehicle is set to carry out braking adjustment in advance.Alternatively,

This track target vehicle deceleration lane change of front is recognized according to the kinematic parameter of objects ahead vehicle and steering indicating light extremely The operating mode in this non-track of front, so that the kinematic parameter control system of main body vehicle is adjusted without braking.Alternatively,

This non-track target vehicle deceleration lane change of rear is recognized according to the kinematic parameter of rear area target vehicle and steering indicating light To the operating mode in this track, so that the kinematic parameter control system of main body vehicle is adjusted in advance, and/or, so that main body vehicle Lamp system remind rear area target vehicle.

4th acquisition module 710 is used to obtain objects ahead according to objects ahead vehicle range or front car light identification region Vehicle heading angle.

5th acquisition module 711 is used to obtain the intermediate-freuqncy signal of dead load frequency radar, and front radar is obtained according to intermediate-freuqncy signal The relative velocity of target and azimuth.

Calibration module 712 is used for according to objects ahead vehicle heading angle, the relative velocity of front radar target and azimuth The relative velocity for obtaining objects ahead vehicle, calibrates dead load frequency radar according to the objects ahead vehicle of identification automatically.

It should be noted that the aforementioned explanation for travelling autocontrol method embodiment to vehicle is also applied for the present invention The vehicle of embodiment travels automatic control device, and details are not described herein again.

In conclusion the vehicle of the embodiment of the present invention travels automatic control device, main body is obtained from preposition 3D cameras first The first image and the second image of vehicle front environment, from postposition 3D cameras obtain main body vehicle rear environment third image and 4th image, and front lines on highway is obtained according to the first image and rear lines on highway is obtained according to third image, so Front lines on highway is mapped to according to the intertexture mapping relations between the first image and the second image afterwards raw in the second image At multiple front vehicles identification ranges, according to the intertexture mapping relations between third image and the 4th image by rear road driveway Line, which maps in the 4th image, generates multiple front vehicle identification ranges, and identifies front according to all front vehicles identification ranges Target vehicle and according to all front vehicle identification ranges identify rear area target vehicle, finally according to objects ahead vehicle and rear The kinematic parameter of target vehicle carries out cruise control to the kinematic parameter of main body vehicle.

In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office It can be combined in any suitable manner in one or more embodiments or example.In addition, without conflicting with each other, the skill of this field Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples It closes and combines.

In addition, term " first ", " second " are used for description purposes only, it is not understood to indicate or imply relative importance Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or Implicitly include at least one this feature.In the description of the present invention, the meaning of " plurality " is at least two, such as two, three It is a etc., unless otherwise specifically defined.

Any process described otherwise above or method description are construed as in flow chart or herein, and expression includes It is one or more for realizing custom logic function or process the step of executable instruction code module, segment or portion Point, and the range of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discuss suitable Sequence, include according to involved function by it is basic simultaneously in the way of or in the opposite order, to execute function, this should be of the invention Embodiment person of ordinary skill in the field understood.

Those skilled in the art are appreciated that realize all or part of step that above-described embodiment method carries Suddenly it is that relevant hardware can be instructed to complete by program, the program can be stored in a kind of computer-readable storage medium In matter, which includes the steps that one or a combination set of embodiment of the method when being executed.

In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, it can also That each unit physically exists alone, can also two or more units be integrated in a module.Above-mentioned integrated mould The form that hardware had both may be used in block is realized, can also be realized in the form of software function module.The integrated module is such as Fruit is realized in the form of software function module and when sold or used as an independent product, can also be stored in a computer In read/write memory medium.

Storage medium mentioned above can be read-only memory, disk or CD etc..Although having been shown and retouching above The embodiment of the present invention is stated, it is to be understood that above-described embodiment is exemplary, and should not be understood as the limit to the present invention System, those skilled in the art can be changed above-described embodiment, change, replace and become within the scope of the invention Type.

Claims

1. a kind of vehicle travels autocontrol method, which is characterized in that including：

The first image and the second image that environment in front of main body vehicle is obtained from preposition 3D cameras obtain main body from postposition 3D cameras The third image and the 4th image of rear of vehicle environment, wherein the first image and third image are color or luminance picture, second Image and the 4th image are depth image；

Front lines on highway is obtained according to described first image, rear lines on highway is obtained according to the third image；

The front lines on highway is mapped according to the intertexture mapping relations between described first image and second image To multiple front vehicles identification ranges are generated in second image, according between the third image and the 4th image The rear lines on highway is mapped in the 4th image and generates multiple front vehicle identification ranges by intertexture mapping relations；

Objects ahead vehicle is identified according to all front vehicles identification ranges, and rear is identified according to all front vehicle identification ranges Target vehicle；

According to the kinematic parameter of the objects ahead vehicle and the rear area target vehicle to the kinematic parameter of the main body vehicle Carry out cruise control.

2. the method as described in claim 1, which is characterized in that described to obtain environment in front of main body vehicle from preposition 3D cameras First image and the second image obtain the third image and the 4th image of main body vehicle rear environment from postposition 3D cameras, including：

The first image of environment in front of main body vehicle is obtained from the imaging sensor of preposition 3D cameras；

The second image of environment in front of main body vehicle is obtained from the time-of-flight sensor of preposition 3D cameras；

The third image of main body vehicle rear environment is obtained from the imaging sensor of postposition 3D cameras；

The 4th image of main body vehicle rear environment is obtained from the time-of-flight sensor of postposition 3D cameras.

3. the method as described in claim 1, which is characterized in that described to obtain front road driveway according to described first image Line, including：

When described first image be luminance picture, according in described first image front lines on highway and road surface luminance difference Identify the front lines on highway；Alternatively,

When described first image be color image, the color image is converted into luminance picture, according in described first image Front lines on highway and the luminance difference on road surface identify the front lines on highway.

4. method as claimed in claim 3, which is characterized in that it is described according in described first image front lines on highway with The luminance difference on road surface identifies the front lines on highway, including：

The binary map of the front lines on highway is created according to the luminance information of described first image and preset luminance threshold Picture；

Whole edge pixel locations of straight way solid line lane line are detected in the bianry image according to preset detection algorithm Or detect the whole edge pixel locations for solid line lane line of going off the curve；

Whole edge pixel locations of straight way dotted line lane line are detected in the bianry image according to preset detection algorithm Or detect the whole edge pixel locations for dotted line lane line of going off the curve.

5. the method as described in claim 1, which is characterized in that described to obtain rear road driveway according to the third image Line, including：

When the third image be luminance picture, according to the luminance difference of rear lines on highway and road surface in the third image Identify the rear lines on highway；Alternatively,

When the third image be color image, the color image is converted into luminance picture, according in the third image Rear lines on highway and the luminance difference on road surface identify the rear lines on highway.

6. method as claimed in claim 5, which is characterized in that it is described according to rear lines on highway in the third image with The luminance difference on road surface identifies the rear lines on highway, including：

The binary map of the rear lines on highway is created according to the luminance information of the third image and preset luminance threshold Picture；

7. the method as described in claim 1, which is characterized in that described to identify front mesh according to all front vehicles identification ranges Vehicle is marked, objects ahead vehicle range is generated according to the objects ahead vehicle, including：

All front vehicles identification ranges are marked with the label in front this track and this non-track of front；

According to this track target vehicle of the vehicle identification range identification front of this track label of label front；

According to this non-track target vehicle of vehicle identification range identification front of this non-track label of label front；

Lane change ahead target vehicle is identified according to the front vehicles identification range of combination of two；

All front vehicle identification ranges are marked with the label in this track of rear and this non-track of rear；

This track target vehicle of rear is identified according to the vehicle identification range of label this track label of rear；

This non-track target vehicle of rear is identified according to the vehicle identification range of label this non-track label of rear；

Rear lane change target vehicle is identified according to the front vehicle identification range of combination of two.

8. the method as described in claim 1, which is characterized in that described to identify front mesh according to all front vehicles identification ranges Vehicle is marked, rear area target vehicle is identified according to all front vehicle identification ranges, including：

The object boundary that objects ahead vehicle is detected using the boundary detection method in image processing algorithm is identified；

The object boundary that rear area target vehicle is detected using the boundary detection method in image processing algorithm is identified.

9. the method as described in claim 1, which is characterized in that further include：

Objects ahead vehicle range is generated according to the objects ahead vehicle, rear area target is generated according to the rear area target vehicle Vehicle range；

The objects ahead vehicle range is reflected according to the intertexture mapping relations between described first image and second image It is incident upon in described first image and generates front car light identification region, according to the friendship between the third image and the 4th image It knits mapping relations and the rear area target vehicle range is mapped into generation rear car light identification region in the third image；

The steering indicating light that corresponding objects ahead vehicle is identified according to the front car light identification region is identified according to the rear car light The steering indicating light of the corresponding rear area target vehicle of region recognition；

The movement according to the kinematic parameter of the objects ahead vehicle and the rear area target vehicle to the main body vehicle Parameter carries out cruise control, including：

According to the kinematic parameter and steering indicating light of the objects ahead vehicle and the rear area target vehicle to the main body vehicle Kinematic parameter carry out cruise control.

10. method as claimed in claim 9, which is characterized in that described to generate objects ahead according to the objects ahead vehicle Vehicle range generates rear area target vehicle range according to the rear area target vehicle, including：

Objects ahead vehicle range is generated according to the enclosed region that the object boundary of the objects ahead vehicle surrounds, according to described The enclosed region that the object boundary of rear area target vehicle surrounds generates rear area target vehicle range；Alternatively,

Objects ahead vehicle range is generated according to the enclosed region of the extension of the object boundary of the objects ahead vehicle surrounded, Rear area target vehicle range is generated according to the enclosed region of the extension of the object boundary of the rear area target vehicle surrounded；Or Person,

Objects ahead vehicle range is generated according to the enclosed region that multiple location of pixels lines of the objects ahead vehicle surround, Rear area target vehicle range is generated according to the enclosed region that multiple location of pixels lines of the rear area target vehicle surround.

11. method as claimed in claim 9, which is characterized in that described to be identified accordingly according to the front car light identification region The steering indicating light of objects ahead vehicle identifies the steering indicating light of corresponding rear area target vehicle according to the rear car light identification region, packet It includes：

Corresponding objects ahead is identified according to the color, flicker frequency of tail-light in the front car light identification region or flashing sequence The steering indicating light of vehicle；

Corresponding rear area target is identified according to the color, flicker frequency of headlight in the rear car light identification region or flashing sequence The steering indicating light of vehicle.

12. method as claimed in claim 9, which is characterized in that it is described according to the objects ahead vehicle and it is described after square mesh The kinematic parameter and steering indicating light for marking vehicle carry out cruise control to the kinematic parameter of the main body vehicle, including：

This non-track target vehicle deceleration lane change of front is recognized according to the kinematic parameter of the objects ahead vehicle and steering indicating light To the operating mode in this track, so that the kinematic parameter control system of the main body vehicle carries out braking adjustment in advance；Alternatively,

This track target vehicle deceleration lane change of front is recognized according to the kinematic parameter of the objects ahead vehicle and steering indicating light extremely The operating mode in this non-track of front, so that the kinematic parameter control system of the main body vehicle is adjusted without braking；Alternatively,

This non-track target vehicle deceleration lane change of rear is recognized according to the kinematic parameter of the rear area target vehicle and steering indicating light To the operating mode in this track, so that the kinematic parameter control system of the main body vehicle is adjusted in advance, and/or, so that described The lamp system of main body vehicle reminds rear area target vehicle.

13. the method as described in claim 1-12 is any, which is characterized in that the method further includes：

Objects ahead vehicle heading angle is obtained according to the objects ahead vehicle range or the front car light identification region；

The intermediate-freuqncy signal for obtaining dead load frequency radar obtains relative velocity and the azimuth of front radar target according to intermediate-freuqncy signal；

The phase of objects ahead vehicle is obtained according to objects ahead vehicle heading angle, the relative velocity of front radar target and azimuth To speed, dead load frequency radar is calibrated automatically according to the objects ahead vehicle of identification.

14. a kind of vehicle travels automatic control device, which is characterized in that including：

First acquisition module, the first image and the second image for obtaining environment in front of main body vehicle from preposition 3D cameras, from Postposition 3D cameras obtain the third image and the 4th image of main body vehicle rear environment, wherein the first image and third image are Color or luminance picture, the second image and the 4th image are depth image；

Second acquisition module, for obtaining front lines on highway according to described first image；

Third acquisition module, for obtaining rear lines on highway according to the third image；

First generation module, before will be described according to the intertexture mapping relations between described first image and second image Square lines on highway, which maps in second image, generates multiple front vehicles identification ranges, according to the third image and institute State the intertexture mapping relations between the 4th image the rear lines on highway is mapped in the 4th image generate it is multiple Front vehicle identification range；

First identification module, for identifying objects ahead vehicle according to all front vehicles identification ranges, according to all rear vehicles Identification range identifies rear area target vehicle；

Control module is used for according to the kinematic parameter of the objects ahead vehicle and the rear area target vehicle to the main body vehicle Kinematic parameter carry out cruise control.

15. device as claimed in claim 14, which is characterized in that described to be used for from the first acquisition module：

16. device as claimed in claim 13, which is characterized in that second acquisition module includes：

First recognition unit, for being luminance picture when described first image, according to front road driveway in described first image Line and the luminance difference on road surface identify the front lines on highway；Alternatively,

The color image is converted to luminance picture, institute by the first converting unit for being color image when described first image State the first recognition unit for according in described first image front lines on highway and road surface luminance difference identification it is described before Square lines on highway.

17. device as claimed in claim 16, which is characterized in that first recognition unit is specifically used for：

18. device as claimed in claim 14, which is characterized in that the third acquisition module includes：

Second recognition unit, for being luminance picture when the third image, according to rear road driveway in the third image Line and the luminance difference on road surface identify the rear lines on highway；Alternatively,

The color image is converted to luminance picture, institute by the second converting unit for being color image when the third image State the second recognition unit for according to rear lines on highway in the third image and the luminance difference on road surface identification it is described after Square lines on highway.

19. device as claimed in claim 18, which is characterized in that second recognition unit is specifically used for：

20. device as claimed in claim 14, which is characterized in that first identification module is used for：

21. device as claimed in claim 14, which is characterized in that first identification module is additionally operable to：

22. device as claimed in claim 13, which is characterized in that further include：

Second generation module, for generating objects ahead vehicle range according to the objects ahead vehicle, according to the rear square mesh It marks vehicle and generates rear area target vehicle range；

Third generation module, before will be described according to the intertexture mapping relations between described first image and second image Square mesh mark vehicle range, which maps in described first image, generates front car light identification region, according to the third image and described Intertexture mapping relations between 4th image, which map to the rear area target vehicle range in the third image, generates rear Car light identification region；

Second identification module, the steering indicating light for identifying corresponding objects ahead vehicle according to the front car light identification region, root The steering indicating light of corresponding rear area target vehicle is identified according to the rear car light identification region；

The control module is used for：

According to the kinematic parameter of the objects ahead vehicle and the rear area target vehicle to the kinematic parameter of the main body vehicle And steering indicating light carries out cruise control.

23. device as claimed in claim 22, which is characterized in that second generation module is used for：

24. device as claimed in claim 22, which is characterized in that second identification module is used for：

25. device as claimed in claim 22, which is characterized in that the control module is used for：

26. the device as described in claim 14-25 is any, which is characterized in that described device further includes：

4th acquisition module, for obtaining front mesh according to the objects ahead vehicle range or the front car light identification region Mark vehicle heading angle；

5th acquisition module, the intermediate-freuqncy signal for obtaining dead load frequency radar obtain front radar target according to intermediate-freuqncy signal Relative velocity and azimuth；

Calibration module, before being obtained according to objects ahead vehicle heading angle, the relative velocity of front radar target and azimuth The relative velocity of square target vehicle calibrates dead load frequency radar according to the objects ahead vehicle of identification automatically.