CN107958222A - A kind of pavement detection method, apparatus and terminal - Google Patents

A kind of pavement detection method, apparatus and terminal Download PDF

Info

Publication number
CN107958222A
CN107958222A CN201711315782.8A CN201711315782A CN107958222A CN 107958222 A CN107958222 A CN 107958222A CN 201711315782 A CN201711315782 A CN 201711315782A CN 107958222 A CN107958222 A CN 107958222A
Authority
CN
China
Prior art keywords
line
oblique line
target
demarcation
oblique
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201711315782.8A
Other languages
Chinese (zh)
Other versions
CN107958222B (en
Inventor
高伟杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hisense Group Co Ltd
Original Assignee
Hisense Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hisense Group Co Ltd filed Critical Hisense Group Co Ltd
Priority to CN201711315782.8A priority Critical patent/CN107958222B/en
Publication of CN107958222A publication Critical patent/CN107958222A/en
Application granted granted Critical
Publication of CN107958222B publication Critical patent/CN107958222B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a kind of pavement detection method, apparatus and terminal, belong to automobile assistant driving technical field.This method includes：When getting the corresponding V disparity maps of road conditions image, V disparity maps are subjected to binary conversion treatment, obtain target V disparity maps；Target V disparity maps are divided according to the first line of demarcation and obtain two subregions, and oblique line is detected in two subregions, if the oblique line detected in two subregions is unsatisfactory for preset condition, the position based on the first line of demarcation of oblique line amendment detected obtains target line of demarcation；Based on the oblique line detected in two subregions that target line of demarcation division target V disparity maps obtain, the road surface in target V disparity maps is determined.In embodiments of the present invention target line of demarcation is obtained by correcting the position in the first line of demarcation, therefore, target V disparity maps are divided by the target line of demarcation and obtain two subregions, road surface more accurate can be determined with stablizing according to the oblique line detected in the two subregions.

Description

A kind of pavement detection method, apparatus and terminal
Technical field
The present invention relates to automobile assistant driving technical field, more particularly to a kind of pavement detection method, apparatus and terminal.
Background technology
With intelligentized continuous development, the auxiliary driving technology of vehicle has become the technology heat that each manufacturer competitively chases Point., can be by the road conditions image that is collected to binocular camera in current existing advanced drive assist system Manage to be detected to the road surface of vehicle front, so as to provide data basis for the safe driving of vehicle.
In correlation technique, after road conditions image is got by binocular camera, can be to the road conditions image at Reason, obtains corresponding anaglyph.Afterwards, corresponding V disparity maps are calculated according to the anaglyph.By V disparity maps according to Preset dividing line is divided into two fixed partitions, and detects oblique line respectively in two fixed partitions, and the oblique line that this is detected is i.e. It can be used for representing road surface.
It is to divide V disparity maps by preset dividing line to obtain two fixed partitions when detecting road surface by the above method, In this way, when the position of preset dividing line chooses unreasonable, standard cannot be obtained when dividing V disparity maps by the preset dividing line True two fixed partitions for being used to detect the oblique line for representing uneven road surface, that is to say, that divided by the preset dividing line The accuracy for two fixed partitions arrived is relatively low, on this basis, oblique line is detected in the two fixed partitions, and will detect Oblique line as road surface, larger detection error will be produced.
The content of the invention
In order to solve in correlation technique to choose in the position of preset dividing line it is unreasonable in the case of, pass through the default boundary The accuracy for two fixed partitions that line divides is relatively low, on this basis, by what is detected in the two fixed partitions There are the problem of larger detection error when oblique line is as road surface, an embodiment of the present invention provides a kind of pavement detection method, dress Put and terminal.The technical solution is as follows：
First aspect, there is provided a kind of pavement detection method, the described method includes：
When getting the corresponding V disparity maps of road conditions image, the V disparity maps are subjected to binary conversion treatment, obtain target V Disparity map；
The target V disparity maps are divided according to the first line of demarcation and obtain two subregions, and are detected in described two subregions Oblique line, if the oblique line detected in described two subregions is unsatisfactory for preset condition, based on the oblique line amendment detected The position in first line of demarcation obtains target line of demarcation；
Based on dividing the oblique line that is detected in two subregions obtained after the target V disparity maps in the target line of demarcation, Determine the road surface in the target V disparity maps.
Alternatively, it is described to detect oblique line in described two subregions, if the oblique line detected in described two subregions is not Meet preset condition, then the position for correcting first line of demarcation based on the oblique line detected obtains target line of demarcation, wraps Include：
Oblique line is detected respectively in described two subregions, obtains the first oblique line and the second oblique line；
Judge whether meet preset condition between first oblique line and second oblique line；
When being unsatisfactory for preset condition between first oblique line and second oblique line, based on first oblique line and institute The position that the second oblique line corrects first line of demarcation is stated, dividing the target V based on the first line of demarcation after correction position regards Poor figure obtains two subregions, and detects oblique line respectively in two subregions obtained after division, until the first oblique line detected And second when meeting preset condition between oblique line, the first line of demarcation after last time correction position is determined as the target point Boundary line.
Alternatively, it is described to judge whether meet preset condition between first oblique line and second oblique line, including：
Determine first oblique line and the intersection point in first line of demarcation, obtain the first intersection point, determine second oblique line With the intersection point in first line of demarcation, the second intersection point is obtained；
When the distance between first intersection point and second intersection point are more than pre-determined distance, first oblique line is determined The preset condition is unsatisfactory between second oblique line；
When the distance between first intersection point and second intersection point are less than or equal to the pre-determined distance, institute is determined State and meet the preset condition between the first oblique line and second oblique line.
Alternatively, the position that first line of demarcation is corrected based on first oblique line and second oblique line, bag Include：
Determine the intersection point of first oblique line and second oblique line, obtain the 3rd intersection point；
By after the 3rd intersection point and the straight line parallel with the longitudinal axis of the target V disparity maps are determined as correction position The first line of demarcation.
Alternatively, it is described based in two subregions obtained after the target line of demarcation division target V disparity maps The oblique line of detection, determines the road surface in the target V disparity maps, including：
Determine to divide two detected in two subregions obtained after the target V disparity maps in the target line of demarcation The slope of oblique line, obtains first slope and the second slope；
When the first slope and second slope differ, by an oblique line in two oblique lines by prolonging The broken line that long line obtains after intersecting with another oblique line in two oblique lines is determined as the road in the target V disparity maps Face.
Alternatively, it is described to determine to divide in the target line of demarcation in two subregions obtained after the target V disparity maps The slope of two oblique lines detected, after obtaining first slope and the second slope, further includes：
When the first slope is identical with second slope, two oblique lines and the target line of demarcation are connected Two intersection points；
The road surface broken line that connection obtains being determined as in the target V disparity maps.
Second aspect, there is provided a kind of road surface checking device, described device include：
Processing module, for when getting the corresponding V disparity maps of road conditions image, the V disparity maps to be carried out binaryzation Processing, obtains target V disparity maps；
First determining module, two subregions are obtained for dividing the target V disparity maps according to the first line of demarcation, and in institute State in two subregions and detect oblique line, if the oblique line detected in described two subregions is unsatisfactory for preset condition, based on the inspection The position that the oblique line measured corrects first line of demarcation obtains target line of demarcation；
Second determining module, for based on two obtained after the target line of demarcation division target V disparity maps The oblique line detected in subregion, determines the road surface in the target V disparity maps.
Alternatively, first determining module includes：
Detection sub-module, for detecting oblique line respectively in described two subregions, obtains the first oblique line and the second oblique line；
Judging submodule, for judging whether first oblique line and second oblique line meet preset condition；
Submodule is corrected, for when being unsatisfactory for preset condition between first oblique line and second oblique line, being based on First oblique line and second oblique line correct the position in first line of demarcation, based on the first line of demarcation after correction position Divide the target V disparity maps and obtain two subregions, and oblique line is detected respectively in two subregions obtained after division, until inspection When meeting preset condition between first oblique line and second oblique line that measure, by first after last time correction position Line of demarcation is determined as the target line of demarcation.
Alternatively, the judging submodule is specifically used for：
Determine first oblique line and the intersection point in first line of demarcation, obtain the first intersection point, determine second oblique line With the intersection point in first line of demarcation, the second intersection point is obtained；
When the distance between first intersection point and second intersection point are more than pre-determined distance, first oblique line is determined The preset condition is unsatisfactory between second oblique line；
When the distance between first intersection point and second intersection point are less than or equal to the pre-determined distance, institute is determined State and meet the preset condition between the first oblique line and second oblique line.
Alternatively, the amendment submodule is specifically used for：
Determine the intersection point of first oblique line and second oblique line, obtain the 3rd intersection point；
By after the 3rd intersection point and the straight line parallel with the longitudinal axis of the target V disparity maps are determined as correction position The first line of demarcation.
Alternatively, second determining module includes：
First determination sub-module, for determining to divide obtain after the target V disparity maps two in the target line of demarcation The slope of two oblique lines detected in a subregion, obtains first slope and the second slope；
Second determination sub-module, for when the first slope and second slope differ, by described two tiltedly The broken line that an oblique line in line obtains after being intersected by extended line with another oblique line in two oblique lines is determined as institute State the road surface in target V disparity maps.
Alternatively, second determining module further includes：
Connect submodule, for when the first slope is identical with second slope, connect two oblique lines with Two intersection points in the target line of demarcation；
3rd determination sub-module, the broken line for connection to be obtained are determined as the road surface in the target V disparity maps.
The third aspect, there is provided a kind of pavement detection terminal, the terminal include：
Processor；
CCD camera assembly, for gathering road conditions image, and the road conditions image is sent to the processor and is handled；
For storing the memory of processor-executable instruction；
Wherein, the processor is configured as the step of any one method described in above-mentioned first aspect.
Fourth aspect, there is provided a kind of computer-readable recording medium, is stored with finger on the computer-readable recording medium The step of making, any one method described in above-mentioned first aspect realized when described instruction is executed by processor.
5th aspect, there is provided a kind of computer program product for including instruction, when run on a computer so that meter Calculation machine performs the pavement detection method described in above-mentioned first aspect.
The beneficial effect that technical solution provided in an embodiment of the present invention is brought is：Regarded when getting the corresponding V of road conditions image During difference figure, V disparity maps can be subjected to binary conversion treatment, obtain target V disparity maps, afterwards, mesh be divided according to the first line of demarcation Mark V disparity maps obtain two subregions, and detect oblique line in two subregions, if the oblique line detected in described two subregions is not Meeting preset condition, then the position based on the first line of demarcation of oblique line amendment detected obtains target line of demarcation, and based in mesh The oblique line detected in two subregions that division target V disparity maps in mark line of demarcation obtain, determines the road surface in target V disparity maps. Be, in embodiments of the present invention, when the oblique line detected in two subregions divided in the first line of demarcation be unsatisfactory for it is pre- If, can be by being modified to obtain target line of demarcation to the position in the first line of demarcation during condition.In this way, demarcated by the target Line carries out division to target V disparity maps can obtain more accurately two subregions, in this way, being detected according in the two subregions The road surface that determines of oblique line it is also just more accurate, improve the accuracy of pavement detection.
Brief description of the drawings
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for For those of ordinary skill in the art, without creative efforts, other can also be obtained according to these attached drawings Attached drawing.
Figure 1A is a kind of principle schematic of calculating anaglyph shown in the embodiment of the present invention；
Figure 1B is a kind of principle schematic that V disparity maps are calculated according to anaglyph shown in the embodiment of the present invention；
Fig. 1 C are a kind of system architecture diagrams of pavement detection method provided in an embodiment of the present invention；
Fig. 2A is a kind of flow chart of pavement detection method provided in an embodiment of the present invention；
Fig. 2 B are the flow charts that a kind of V disparity maps after treatment provided in an embodiment of the present invention determine target line of demarcation；
Fig. 2 C are one kind shown in the embodiment of the present invention judges whether to meet between the first oblique line and the second oblique line continuous The schematic diagram of condition；
Fig. 2 D are in the V disparity maps after a kind of two oblique lines based on acquisition shown in the embodiment of the present invention determine processing The schematic diagram on road surface；
Fig. 2 E are that another two oblique lines based on acquisition shown in the embodiment of the present invention are determined in the V disparity maps after processing Road surface schematic diagram；
Fig. 2 F are that one kind shown in the embodiment of the present invention is implemented using correlation technique detection uneven road surface and using the present invention The detection result contrast schematic diagram for the method detection uneven road surface that example provides；
Fig. 3 A are a kind of structure diagrams of road surface checking device provided in an embodiment of the present invention；
Fig. 3 B are a kind of structure diagrams of first determining module provided in an embodiment of the present invention；
Fig. 3 C are a kind of structure diagrams of second determining module provided in an embodiment of the present invention；
Fig. 3 D are the structure diagrams of another second determining module provided in an embodiment of the present invention；
Fig. 4 is a kind of structure diagram of terminal provided in an embodiment of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention Formula is described in further detail.
In order to make it easy to understand, before detailed explanation is carried out to the embodiment of the present invention, first to the embodiment of the present invention The noun being related to explains.
Anaglyph：It is calculated by the left and right two images photographed to binocular camera synchronization.Its In, in the two images of left and right, piece image is as benchmark image, and another piece image is as movement images, and left and right two images Size it is identical.Benchmark image and movement images are positioned in the same coordinate system, by the pixel and benchmark in movement images The one-row pixels point of identical Y-coordinate is matched in image, to be determined from benchmark image and the pixel in movement images The pixel matched somebody with somebody, and the difference of the abscissa between the pixel that each two is mutually matched in movement images and benchmark image is calculated, The difference of the abscissa is the parallax value between two pixels.The parallax value is corresponding as the pixel in benchmark image Pixel value, so as to obtain the anaglyph with benchmark image formed objects.
Figure 1A is a kind of principle schematic of calculating anaglyph shown in the embodiment of the present invention.Where it is assumed that in Figure 1A Left figure on the basis of image, right figure is movement images.For convenience of description, each lattice in Figure 1A can be regarded as a picture Vegetarian refreshments.For the pixel A in movement images, when finding the matched pixel point of pixel A in benchmark image, first, Can centered on the pixel A pixel, the picture element matrix A of one 9 × 9 is formed, such as the dotted line frame institute in right figure in Figure 1A Show.Afterwards, the one-row pixels point with Y-coordinate with the central pixel point can be determined in benchmark image, in Figure 1A left figures Shown in solid box.When the pixel B in pixel A in movement images and benchmark image is carried out matching primitives, can be formed Onesize 9 × 9 picture element matrix B of pixel centered on pixel B, as shown in left figure dotted line frame in Figure 1A.Afterwards, count The pixel difference between the pixel of corresponding position in each pixel and picture element matrix B in picture element matrix A is calculated, and will be multiple Pixel difference adduction, obtain pixel difference and.For other pixels in benchmark image with pixel A with Y-coordinate, can pass through The above method and pixel A carry out matching primitives, it is final it is corresponding obtain multiple pixel differences and.By the plurality of pixel difference and middle minimum Pixel difference and corresponding pixel be determined as the match point of pixel A.Assuming that match points of the pixel A in benchmark image For pixel B, then can using the difference of the abscissa between pixel A and pixel B as the parallax value between two pixels, And using the parallax value as the pixel value with pixel B in the anaglyph of benchmark image formed objects.
V disparity maps：V disparity maps are calculated by anaglyph.From the above-mentioned introduction to anaglyph, disparity map As in, the pixel value of each pixel is corresponding parallax value.And V disparity maps are to keep the ordinate of anaglyph not Become, abscissa is changed into parallax value, and the pixel value at every bit (x, y) place is the pixel that ordinate is y in anaglyph in V disparity maps Parallax value is the total number of the pixel of x in point.
Figure 1B is a kind of schematic diagram that V disparity maps are calculated according to anaglyph shown in the embodiment of the present invention.Wherein, Figure 1B Left figure be anaglyph, from the foregoing introduction to anaglyph, its abscissa and the transverse and longitudinal of ordinate and benchmark image Coordinate is identical, and the pixel value of each pixel is the corresponding parallax value of the pixel.The right figure of Figure 1B is V disparity maps, its vertical seat Mark is corresponding with the ordinate of anaglyph, and abscissa is then parallax value.For convenience of description, can be by each small side in Figure 1B Lattice regard a pixel as, and the value in lattice is the pixel value of the pixel.Such as a line picture in Figure 1B left figures in solid box Vegetarian refreshments, the ordinate of the row pixel is 7, is counted for the pixel value of the row pixel, can obtain pixel value as 4 The number of pixel have 5, pixel value is that the number of 6 pixel has 8, and remaining is the pixel that pixel value is 3, Number is 7, at this time, the pixel of this line in corresponding V disparity maps can be then generated according to the statistical result.Due to disparity map The pixel value of picture is exactly parallax value in fact, so, the number for the pixel that pixel value is 4 is 5, is actually that parallax value is The number of 4 pixel is 5, at this time, the picture for the pixel at position that abscissa is 4 in V disparity maps and ordinate is 7 Element value is 5.Correspondingly, the pixel value for the pixel at position that abscissa is 6 and ordinate is 7 is 8, abscissa 3 And the pixel value of the pixel at the position that ordinate is 7 is 7.Due in anaglyph ordinate be 7 pixel in not , therefore, will for the pixel value for the pixel that remaining ordinate in V disparity maps is 7 there are the pixel of remaining parallax value It is 0.
For in anaglyph per a line pixel, can be counted by the above method, and according to statistics after Result obtain corresponding to the pixel difference of one-row pixels point in V disparity maps, so that V disparity maps finally be calculated.
Next to the present embodiments relate to application scenarios introduced.
In current advanced drive assist system, thunder can be handled by image procossing and computer vision technique Reach, the sensor road conditions image that either camera collects, made prediction according to road conditions image to front pedestrian, barrier, And early warning is carried out to driver in the case of there are potential danger and either controls emergency brake of vehicle.Wherein, to front The accurate detection of barrier is the key for carrying out effective early warning, and it is then detection that road surface region is accurately extracted from road conditions image The steps necessary of front obstacle.
Currently, with the development of camera technology and computer vision technique, road conditions image is gathered using binocular camera, And image is handled by computer vision technique and then achieves preferable detection result to detect the method for barrier.It is and of the invention The pavement detection method that embodiment provides can be applied in the road conditions image detection barrier using binocular camera shooting During road pavement be detected.
Next to the present embodiments relate to system architecture illustrate.
Fig. 1 C are a kind of system architecture diagrams of pavement detection method provided in an embodiment of the present invention.As shown in Figure 1 C, this is System includes automobile 101, binocular camera 102 and terminal 103.Wherein, binocular camera 102 is mounted on automobile 101, and double Mesh camera 102 can communicate with terminal 103.
Wherein, binocular camera 102 can be mounted in the front of automobile 101 and on the longitudinal axis of automobile 101.Such as figure Arrow pointed location in 1C.Also, can be to binocular camera 102 after binocular camera 102 is mounted on automobile 101 Demarcated.During automobile 101 travels, binocular camera 102 can gather road conditions image.
After binocular camera 102 collects road conditions image, which can be sent to terminal 103, it is whole End 103 can be handled the road conditions image, obtained anaglyph, afterwards, V parallaxes are calculated according to the anaglyph Figure, and road surface is detected from the V disparity maps by the hereinafter pavement detection method shown in Fig. 2A.
It should be noted that in embodiments of the present invention, terminal 103 can be the car-mounted terminal installed on automobile 101, In this case, binocular camera 102 can be led to by bluetooth, wireless network either cable network and the terminal 103 Letter.
Alternatively, terminal 103 and binocular camera 102 can be integrated in equipment together, that is to say, terminal 103 can To be the terminal for being integrated with binocular camera and image processing function chip.
Alternatively, the terminal that terminal 103 can also be while communicate with the binocular camera that multiple automobiles carry. It is that the terminal 103 can be the road conditions image returned dedicated for the binocular camera of the multiple automobiles of reception, and to road conditions figure As the terminal analyzed.
, next will be with reference to attached drawing to this after the application scenarios of the embodiment of the present invention and system architecture are introduced The pavement detection method that inventive embodiments provide carries out detailed explanation.
Fig. 2A is a kind of flow chart of pavement detection method provided in an embodiment of the present invention, and this method can be applied to Fig. 1 C Terminal in shown system architecture.Referring to Fig. 2A, this method comprises the following steps：
Step 201：When getting the V disparity maps corresponding to road conditions image, V disparity maps are subjected to binary conversion treatment, are obtained To target V disparity maps.
Introduction based on aforementioned system framework understands that the binocular camera carried on automobile can gather road conditions figure in real time Picture.Wherein, binocular camera includes left camera and right camera, and two width can be collected by left camera and right camera Road conditions image.Afterwards, the road conditions image collected can be sent to terminal, terminal and can adopt left camera by binocular camera The road conditions image collected is as benchmark image, and the image that right camera is collected is as movement images, and according to foregoing terms The method for the calculating anaglyph mentioned in introduction, anaglyph is calculated according to the benchmark image and movement images.Work as meter After calculation obtains anaglyph, terminal can calculate corresponding V according to the anaglyph according to the method for foregoing calculating V disparity maps Disparity map.
Due to the detection of straight lines in subsequent step, it is necessary in V disparity maps, and the method for the detection of straight lines being typically used for In, it is binary image that the image of input, which often requires that, therefore, terminal get the V disparity maps corresponding to road conditions image it Afterwards, the V disparity maps can be subjected to binary conversion treatment, obtains target V disparity maps.
Specifically, for each pixel in V disparity maps, if the pixel value of the pixel is more than presetted pixel value, So, the pixel value of the pixel can be arranged to 255 by terminal, if the pixel value of the pixel is less than or equal to default picture Element value, then, the pixel value of the pixel can be arranged to 0 by terminal, so as to fulfill the binary conversion treatment of V disparity maps.Wherein, Presetted pixel value can be based on the default numerical value of empirical value, and the presetted pixel value can ensure to retain in the V disparity maps to the greatest extent The more pixel of amount.
For example, the presetted pixel value can be 4, at this time, 4 pixel, terminal are more than for pixel value in the V disparity maps The pixel value of these pixels can be disposed as to 255, and be less than or equal to 4 pixel for pixel value in the V disparity maps The pixel value of these pixels can be then disposed as 0 by point, terminal.
It should be noted that above-mentioned is only to provide a kind of feasible value example of presetted pixel value, in practical applications, The presetted pixel value can also be other numerical value, for example, it is also possible to be 5,8 or 10 etc..
Step 202：Target V disparity maps are divided according to the first line of demarcation and obtain two subregions, are detected in two subregions oblique Line, if the oblique line detected in two subregions is unsatisfactory for preset condition, the first line of demarcation is corrected based on the oblique line detected Position obtains target line of demarcation, which is the straight line parallel to the longitudinal axis of target V disparity maps.
When V disparity maps are carried out binary conversion treatment by terminal, after obtaining target V disparity maps, terminal can be by first The position in line of demarcation is modified to obtain target line of demarcation, so that in two target partitions that target line of demarcation is divided Two oblique lines detected are as the oblique line for determining road surface.
Specifically, as shown in Figure 2 B, this step can be realized by following steps：
Step 2021：Terminal determines predetermined cutoff according to distance to a declared goal, and initializes first according to the predetermined cutoff Line of demarcation.
Wherein, which is in vehicle front and the road surface in the range of automobile distance to a declared goal value for expression Plane.Specifically, point at automobile distance to a declared goal value can be calculated in anaglyph in terminal according to the distance to a declared goal In corresponding parallax value, afterwards, by the parallax value in target V disparity maps, corresponding point is determined as default boundary on axis of abscissas Point, and can then be determined as the first boundary by the predetermined cutoff and the straight line parallel with the axis of ordinates of target V disparity maps Line.
For example, it is assumed that the distance to a declared goal is h meters, it that is to say, it is believed that the road surface in the range of h meters of vehicle front It is plane, according to distance to a declared goal h, it is corresponding in anaglyph that the point at h meters of vehicle front can be calculated Parallax value m, at this time, then can be by the axis of abscissas of target V disparity maps since the axis of abscissas of target V disparity maps is parallax value Upper coordinate is determined as predetermined cutoff for the point of (m, 0), by the preset dividing line point and parallel to the vertical seat of target V disparity maps The straight line of parameter namely the first line of demarcation.
Step 2022：Two subregions obtained according to the first line of demarcation division target V disparity maps, in two subregions respectively Oblique line is detected, obtains the first oblique line and the second oblique line, and judge whether the first oblique line and the second oblique line meet preset condition.
Wherein, since first line of demarcation is parallel to the straight line of the axis of ordinates of target V disparity maps, by this Target V disparity maps can be divided into two subregions in left and right by the first line of demarcation.Terminal can detect respectively in the two subregions Oblique line, obtains the first oblique line and the second oblique line.
Wherein, terminal can carry out fitting a straight line to the pixel in two subregions respectively, and the first oblique line is obtained with detection With the second oblique line.Specifically, terminal can detect oblique line respectively by Hough transformation in two subregions, minimum can also be passed through The method of square law isoline fitting detects oblique line respectively in two subregions.
It should be noted that when carrying out fitting a straight line respectively in two subregions, can be special according to possessed by road surface Property, fitting a straight line carries out the pixel in two subregions by default fitting parameter respectively, so that oblique after fitting Line can be accurate as far as possible.
After detection obtains the first oblique line and the second oblique line, in order to avoid the first oblique line and the second oblique line and real road There are the problem of relatively large deviation between oblique line corresponding to face, it is continuous principle that terminal, which can be based on road surface, judges that first is oblique Whether meet preset condition between line and the second oblique line.Wherein, which is to be used to constrain the first oblique line and the second oblique line, To ensure that the first oblique line and the second oblique line can meet that road surface is the condition of this continuous principle.
Specifically, terminal can determine the first oblique line and the intersection point in the first line of demarcation, the first intersection point is obtained, determines that second is oblique Line and the intersection point in the first line of demarcation, obtain the second intersection point.When the distance between the first intersection point and the second intersection point are more than pre-determined distance When, it is determined that preset condition is unsatisfactory between the first oblique line and the second oblique line, when the distance between the first intersection point and the second intersection point During less than or equal to pre-determined distance, it is determined that meet preset condition between the first oblique line and the second oblique line.
In embodiments of the present invention, it is contemplated that road surface is continuous under actual conditions, even road surface is in uneven situation Under, the continuous oblique line being made of the oblique line section of Different Slope is also represented into road surface in target V disparity maps.It that is to say, such as It is reasonable that the position in the first line of demarcation that fruit is used for target V disparity maps being divided into two subregions is chosen, then, in ideal situation Under, the first oblique line and the second oblique line that the pixel in two subregions detects respectively will be continuous.That is, Ideally, first intersection point and the second oblique line in the first oblique line and the first line of demarcation and second intersection point in the first line of demarcation should It is same point.In view of the error in actually detected, even if the first intersection point and the second intersection point have correspondingly deviation, still, In the case of first boundary line position is chosen reasonably, which also should be in certain scope.
Based on above-mentioned consideration, terminal can set pre-determined distance according to the deviation of permission, and actually obtain first is handed over The distance between point and the second intersection point are compared with the pre-determined distance, when the distance between the first intersection point and the second intersection point are more than During the pre-determined distance, then it can determine that the first current boundary line position selection is unreasonable, and according to first line of demarcation The continuous condition in road surface is unsatisfactory between the first oblique line and the second oblique line that are detected respectively in two subregions that division obtains, no It can be used for representing road surface.And if the distance between the first intersection point and the second intersection point are less than or equal to the pre-determined distance, can be with Determine that the first current boundary line position is chosen reasonable, also, in the deviation of permission, first oblique line and the second oblique line can be with It is continuous condition to meet road surface, be that is to say, first oblique line and the second oblique line can be used for representing road surface.
For example, Fig. 2 C be shown in the embodiment of the present invention one kind judge whether to meet between the first oblique line and the second oblique line it is pre- If the schematic diagram of condition.As shown in Figure 2 C, straight line L is the first line of demarcation, for convenience of description, two subregions is referred to as One subregion and the second subregion.Wherein, on the left of straight line L for the first subregion A, on the right side of straight line L for the second subregion B, When detecting oblique line in the first subregion A, the first obtained oblique line is L1, when oblique line is detected in the second subregion B, obtain the Two oblique lines are L2.Wherein, the first oblique line L1With the intersection point O of the first line of demarcation L1For the first intersection point, the second oblique line L2With the first boundary The intersection point O of line L2For the second intersection point.Assuming that pre-determined distance is dt, calculate O1With O2The distance between d, as d ＞ dtWhen, then can be true It is fixed to be unsatisfactory for preset condition between first oblique line and the second oblique line.
When determining to meet preset condition between the first oblique line and the second oblique line by the above method, terminal can perform Step 203, when being unsatisfactory for preset condition between the first oblique line and the second oblique line, terminal then performs following step 2023.
Step 2023：When being unsatisfactory for preset condition between the first oblique line and the second oblique line, based on the first oblique line and second Oblique line corrects the position in the first line of demarcation, and two points are obtained based on the first line of demarcation division target V disparity maps after correction position Area, and oblique line is detected respectively in two subregions that division obtains again.
Understood based on the description in step 2022, when being unsatisfactory for preset condition between the first oblique line and the second oblique line, then It can determine that the position selection for being currently used in the first line of demarcation of division target V disparity maps is unreasonable, and in the thus first boundary The oblique line detected in two subregions that line divides can not accurately represent road surface.Based on this, terminal can be true When fixed first oblique line and the second oblique line are unsatisfactory for preset condition, the first line of demarcation is corrected based on first oblique line and the second oblique line Position.
Specifically, terminal can determine the intersection point of the first oblique line and the second oblique line, the 3rd intersection point is obtained, and the 3rd will be passed through The intersection point and straight line parallel with the longitudinal axis of target V disparity maps is determined as revised first line of demarcation.
Still exemplified by shown in Fig. 2 C, when terminal determines to be unsatisfactory for continuous preset condition between the first oblique line and the second oblique line When, the second oblique line can be intersected by extended line with the first oblique line, obtain the intersection point O of the first oblique line and the second oblique line3, this When, intersection point O3As the 3rd intersection point, that is, revised separation, and pass through the 3rd intersection point O3And with target V disparity maps The parallel straight line L of the longitudinal axis3As revised first line of demarcation.
After terminal determines revised first line of demarcation, it can be repartitioned by revised first line of demarcation Target V disparity maps, obtain two subregions, and afterwards, terminal can detect the first oblique line and again in the two subregions again Two oblique lines, and judge whether meet preset condition between the first oblique line and the second oblique line.If the first oblique line and the second oblique line it Between be still unsatisfactory for preset condition, then continue correct the first line of demarcation position, repeat the above process.
Step 2024：When meeting preset condition between the first oblique line and the second oblique line, after last time correction position The first line of demarcation be determined as target line of demarcation.
If divided by the first line of demarcation determined in step 2021 to target V disparity maps, and obtained in division Two subregions in meet preset condition between the first oblique line and the second oblique line that detect, then can determine first line of demarcation It is reasonable that position is chosen, and first oblique line and the second oblique line can be used for representing road surface, at this time, then can first line of demarcation determine For target line of demarcation.Certainly, if the position selection in the first line of demarcation definite for the first time in step 2021 is unreasonable, step is passed through After method in rapid 2022-2023 is constantly modified the position in the first line of demarcation, the first oblique line and second are detected Meet preset condition between oblique line, at this time, the first line of demarcation after last time correction position can be then determined as mesh by terminal Mark line of demarcation.
In embodiments of the present invention, when being unsatisfactory for preset condition between the first oblique line and the second oblique line, terminal can lead to The position that step 2023 constantly corrects the first line of demarcation is crossed, until in two subregions divided by modified first line of demarcation When meeting preset condition between the first oblique line and the second oblique line that detect, the of division target V disparity maps will be currently used in One line of demarcation is determined as target line of demarcation.It that is to say, in embodiments of the present invention, terminal can be by constantly adjusting the first boundary Line chooses rational boundary line position to find, so that target V disparity maps are carried out rational region division, is ensured with this The oblique line detected in two target partitions is continuous in the case where deviation allows, in this way, being based on continuous two oblique lines Determine road surface continuous in practice, both ensure that the accuracy of pavement detection, and reduced detection error, while again can be to avoid inspection Survey the problem of calculation amount caused by when more oblique lines determine road surface is excessive.
Alternatively, target V disparity maps can also be divided into multiple subregions by terminal by multiple first lines of demarcation, wherein, The conjunction of the two subregions of division can be determined using the above method for two subregions of arbitrary neighborhood in the plurality of subregion The boundary line position of reason, namely target line of demarcation, afterwards, terminal can be based on dividing target V parallaxes in multiple target lines of demarcation The oblique line detected in the multiple subregions obtained after figure determines road surface.The embodiment of the present invention is only to pass through first line of demarcation Division obtains being explained exemplified by two subregions, and still, this, which is not formed, limits the specific of the embodiment of the present invention.
Step 203：Obtain the oblique line for dividing in target line of demarcation and being detected in two subregions obtained after target V disparity maps.
After determining to obtain target line of demarcation by the method in step 202, terminal can be obtained in target line of demarcation Two oblique lines detected in two obtained target partitions are divided, this two oblique lines are that the satisfaction described in step 202 is pre- If the first oblique line and the second oblique line of condition, subsequent terminal can according to this two oblique lines by the method in step 204 more Accurately determine road surface.
Step 204：Two oblique lines based on acquisition determine the road surface in target V disparity maps.
Wherein, two oblique lines of acquisition are the first oblique line and the second oblique line for meeting preset condition, that is to say, that first Oblique line and the second oblique line are continuous within the scope of deviation permission.At this time, first oblique line and second can also be passed through Oblique line represents road surface.In this case, terminal can determine the V disparity maps after processing based on the first oblique line and the second oblique line In road surface.
Specifically, terminal can determine the slope of two oblique lines, obtain first slope and with the second slope, work as first slope When being differed with the second slope, terminal can by an oblique line in two oblique lines by extended line with it is another in two oblique lines The broken line that bar oblique line obtains after intersecting is determined as the road surface in target V disparity maps.When first slope is identical with the second slope, eventually End can connect two oblique lines and two intersection points in target line of demarcation, and at this time, the broken line which obtains can represent processing The road surface in V disparity maps afterwards.
It should be noted that first slope and the second slope may be identical, it is also possible to different.When first slope and second oblique When rate is different, terminal can directly be intersected by two oblique lines of extended line, and the broken line obtained after intersecting is determined as mesh Mark the road surface in V disparity maps.Certainly, at some in special circumstances, two oblique lines are probably parallel, at this time, due to two tiltedly Line and the distance between two intersection points in target line of demarcation are near enough, completely within the scope of deviation permission, therefore, eventually End can be directly connected to two intersection points and obtain a continuous broken line, which is the road surface in the V disparity maps after handling.
Determined for example, Fig. 2 D are a kind of two oblique lines based on acquisition shown in the embodiment of the present invention in target V disparity maps Road surface schematic diagram.As shown in Figure 2 D, two oblique line L1And L2Slope it is different, at this point it is possible to extend oblique line L2Until with it is oblique Line L1It is intersecting, by oblique line L1With oblique line L2The broken line formed afterwards that is connected is the road surface in target V disparity maps, in Fig. 2 D Shown in bold portion.
Fig. 2 E are illustrated that the identical situation of the slope of two oblique lines, in this case, oblique line L1With target line of demarcation L Intersection point be O1, oblique line L2Intersection point with target line of demarcation L is O2, by O1And O2Connection, obtains a continuous broken line, such as Fig. 2 E Shown in middle bold portion, at this time, which is the road surface in the V disparity maps after handling.
In embodiments of the present invention, when getting the corresponding V disparity maps of road conditions image, terminal can by V disparity maps into Row binary conversion treatment, obtains target V disparity maps.Afterwards, terminal can divide target V disparity maps according to the first line of demarcation and obtain two A subregion, and oblique line is detected in two subregions, if the oblique line detected in two the first subregions is unsatisfactory for preset condition, Position based on the first line of demarcation of oblique line amendment detected obtains target line of demarcation, and based in target line of demarcation division target The oblique line detected in two subregions obtained after V disparity maps, determines the road surface in target V disparity maps.It that is to say, of the invention real Apply in example, when the oblique line detected in two subregions divided in the first line of demarcation is unsatisfactory for preset condition, terminal can With the position by constantly adjusting the first line of demarcation, to repartition target V disparity maps, finally determine to obtain target point Boundary line, reality is disclosure satisfy that with this to divide in target line of demarcation to obtain two oblique lines detected in two target partitions Middle road surface is continuous condition, in this way, determining road surface, both to ensure that based on two oblique lines detected in two target partitions The accuracy of pavement detection, reduces detection error, while meter caused by when can determine road surface to avoid more oblique lines are detected again The problem of calculation amount is excessive.
Fig. 2 F show pavement detection method detection uneven road surface based on the embodiment of the present invention and based on correlations Pavement detection method in technology detects detection result comparison diagram during uneven road surface.The thick dotted line of white in figure is uneven road surface The corresponding accurate oblique line in target V disparity maps, wherein, it is actually the place that is uneven in corresponding road surface at flex point Turnover face, in theory, the target partition divided when the flex point is passed through in target line of demarcation be most accurately, at this time, position It can accurately reflect uneven road surface in the oblique line of two Different Slopes of flex point both sides.
Left figure in Fig. 2 F is to use the detection result figure on the pavement detection method detection road surface in correlation technique, wherein, Straight line L1For preset dividing line, preset dividing line L1By the midpoint of target V disparity map width, according to the preset dividing line Target V disparity maps are divided into two subregions, wherein, the oblique line detected in left partition is L2, on the right Subarea detecting arrive Oblique line be L3.And due to preset dividing line L in correlation technique1Position is fixed, as preset dividing line L1Position choose When unreasonable, as shown in left figure, the excessive deviation flex point of the preset dividing line, at this time, what is detected in two subregions is oblique Line can not accurately represent road surface.
Right figure in Fig. 2 F is the detection effect using pavement detection method provided in an embodiment of the present invention detection uneven road surface Fruit is schemed.By pavement detection method provided in an embodiment of the present invention, terminal can constantly be corrected according to the continuous condition in road surface The position in the first line of demarcation, so as to obtain final target line of demarcation L1, at this time, target line of demarcation L1Flex point will be passed through, or Even if deviateing flex point, due to the presence of constraints, departure degree also can be minimum, at this time, L is drawn in target line of demarcation1Get Two subregions in two oblique line L detecting2And L3Obviously road surface can more accurately be represented.
It can be seen from above-mentioned detection result comparison diagram in embodiments of the present invention, terminal is continuously constraint bar with road surface Part, is constantly adjusted the position in the first line of demarcation, and the target line of demarcation for meeting condition finally obtained is actually to pass through Cross road surface in practice and be uneven and be in the line of demarcation of corresponding point in target V disparity maps, in this way, by the target line of demarcation just Can will due to Uneven road and two oblique lines present in target V disparity maps are accurately divided into two subregions, herein On the basis of, oblique line is detected in the two subregions, and the oblique line detected is determined as road surface can effectively to improve road surface inspection The accuracy of survey.
Next, road surface checking device provided in an embodiment of the present invention is introduced.
Referring to Fig. 3 A, an embodiment of the present invention provides a kind of road surface checking device 300, which includes：
Processing module 301, for when getting the corresponding V disparity maps of road conditions image, V disparity maps to be carried out at binaryzation Reason, obtains target V disparity maps；
First determining module 302, two subregions are obtained for dividing the target V disparity maps according to the first line of demarcation, and Oblique line is detected in two subregions, if the oblique line detected in two the first subregions is unsatisfactory for preset condition, based on detecting The position in the first line of demarcation of oblique line amendment obtain target line of demarcation；
Second determining module 303, for based in two subregions obtained after target line of demarcation division target V disparity maps The oblique line of detection, determines the road surface in target V disparity maps.
Alternatively, include referring to Fig. 3 B, the first determining module 302：
Detection sub-module 3021, for detecting oblique line respectively in two subregions, obtains the first oblique line and the second oblique line；
Judging submodule 3022, for judging whether the first oblique line and the second oblique line meet preset condition；
Submodule 3023 is corrected, for when being unsatisfactory for preset condition between the first oblique line and the second oblique line, based on first Oblique line and the second oblique line correct the position in the first line of demarcation, and target V disparity maps are divided based on the first line of demarcation after correction position Two subregions are obtained, and oblique line is detected respectively in two subregions obtained after division, until detecting the first oblique line and second When meeting preset condition between oblique line, revised first line of demarcation of last time is determined as target line of demarcation.
Alternatively, judging submodule 3022 is specifically used for：
Determine the first oblique line and the intersection point in the first line of demarcation, obtain the first intersection point, determine the second oblique line and the first line of demarcation Intersection point, obtain the second intersection point；
When the distance between the first intersection point and the second intersection point are more than pre-determined distance, determine the first oblique line and the second oblique line it Between be unsatisfactory for preset condition；
When the distance between the first intersection point and the second intersection point are less than or equal to pre-determined distance, the first oblique line and second are determined Meet preset condition between oblique line.
Alternatively, submodule 3023 is corrected to be specifically used for：
Determine the intersection point of the first oblique line and the second oblique line, obtain the 3rd intersection point；
It will be determined as revised first boundary by the 3rd intersection point and the straight line parallel with the longitudinal axis of target V disparity maps Line.
Alternatively, include referring to Fig. 3 C, the second determining module 303：
First determination sub-module 3031, for determine to obtain after target line of demarcation division target V disparity maps two points The slope of two oblique lines detected in area, obtains first slope and the second slope；
Second determination sub-module 3032, for when first slope and the second slope differ, by one in two oblique lines The broken line that bar oblique line obtains after being intersected by extended line with another oblique line in two oblique lines is determined as in target V disparity maps Road surface.
Alternatively, further included referring to Fig. 3 D, the second determining module 303：
Submodule 3033 is connected, is demarcated for when first slope is identical with the second slope, connecting two oblique lines with target Two intersection points of line；
3rd determination sub-module 3034, the broken line for connection to be obtained are determined as the road surface in target V disparity maps.
In conclusion in embodiments of the present invention, when getting the corresponding V disparity maps of road conditions image, terminal can be by V Disparity map carries out binary conversion treatment, obtains target V disparity maps.Afterwards, terminal can divide target V parallaxes according to the first line of demarcation Figure obtains two subregions, and detects oblique line in two subregions, if the oblique line detected in two the first subregions be unsatisfactory for it is pre- If condition, then the position based on the first line of demarcation of oblique line amendment detected obtains target line of demarcation, and based in target boundary The oblique line detected in two subregions obtained after line division target V disparity maps, determines the road surface in target V disparity maps.It that is to say, In embodiments of the present invention, when the oblique line detected in two subregions divided in the first line of demarcation is unsatisfactory for preset condition When, terminal can be final to determine by constantly adjusting the position in the first line of demarcation, to be repartitioned to target V disparity maps Target line of demarcation is obtained, divides to obtain two oblique lines detected in two target partitions in target line of demarcation to enable with this It is enough to meet that road surface is continuous condition in practice, in this way, determining road based on two oblique lines detected in two target partitions Face, both ensure that the accuracy of pavement detection, reduce detection error, while can determine road surface to avoid more oblique lines are detected again When caused by calculation amount it is excessive the problem of.
It should be noted that：The road surface checking device that above-described embodiment provides is when carrying out pavement detection, only with above-mentioned each The division progress of function module, can be as needed and by above-mentioned function distribution by different work(for example, in practical application Energy module is completed, i.e., the internal structure of equipment is divided into different function modules, to complete whole described above or portion Divide function.In addition, the road surface checking device that above-described embodiment provides belongs to same design with pavement detection embodiment of the method, it has Body realizes that process refers to embodiment of the method, and which is not described herein again.
Fig. 4 is a kind of structure diagram of terminal 400 provided in an embodiment of the present invention, in the system architecture shown in Fig. 1 C Terminal can be with for the terminal shown in Fig. 4.Specifically, the terminal 400 can be car-mounted terminal, or such as intelligent hand The mobile terminals such as machine, tablet computer, can also be the terminals such as desktop computer, portable computer.
Referring to Fig. 4, which includes：Processor 401 and memory 402.
Processor 401 is the control centre of the terminal 400, utilizes each of the whole terminal of various interfaces and connection Part, by running or performing the software program and/or module that are stored in memory 402, and calls and is stored in memory Data in 402, perform the various functions and processing data of terminal 400, so as to carry out integral monitoring to the terminal 400.It is optional , processor 401 may include one or more processing cores；Preferably, processor 401 can integrate application processor and modulatedemodulate Processor is adjusted, wherein, application processor mainly handles operating system, user interface and application program etc., modem processor Main processing wireless communication.It is understood that above-mentioned modem processor can not also be integrated into processor 401.
Memory 402 can be used for storage software program and module, and processor 401 is stored in memory 402 by operation Software program and module, so as to perform various functions application and data processing.Memory 402 can mainly include storage journey Sequence area and storage data field, wherein, storing program area can storage program area, the application program (ratio needed at least one function Such as sound-playing function, image player function) etc.；Storage data field can be stored uses created number according to terminal 400 According to (such as the image collected, the anaglyph that is calculated or V disparity maps etc.) etc..In addition, memory 402 can include High-speed random access memory, can also include nonvolatile memory, for example, at least disk memory, a flash memories Part or other volatile solid-state parts.Correspondingly, memory 402 can also include Memory Controller, to provide processing Access of the device 401 to memory 402.
In certain embodiments, terminal 400 is also optional includes：Peripheral interface 403 and at least one ancillary equipment. It can be connected between processor 401, memory 402 and peripheral interface 403 by bus or signal wire.Each ancillary equipment It can be connected by bus, signal wire or circuit board with peripheral interface 403.Specifically, ancillary equipment includes：Radio circuit 404th, at least one of touch display screen 405, camera 406, voicefrequency circuit 407, positioning component 408 and power supply 409.
Wherein, CCD camera assembly 406 is used to gather image or video.Alternatively, CCD camera assembly 406 can be included at least Two cameras.In certain embodiments, at least two cameras can be respectively that the left camera of binocular camera and the right side are taken the photograph As head.In certain embodiments, CCD camera assembly 406 can also include flash lamp.Flash lamp can be monochromatic warm flash lamp, It can be double-colored temperature flash lamp.Double-colored temperature flash lamp refers to the combination of warm light flash lamp and cold light flash lamp, can be used for difference Light compensation under colour temperature.
Although being not shown, which can also include various sensors and display screen etc., and details are not described herein.Specifically In the present embodiment, the display unit of terminal 400 is touch-screen display, which has further included memory, and one Either more than one program one of them or more than one program storage is in memory, and be configured to by one or More than one processor of person performs.The one or more programs, which include, to be used to perform the road provided in above-described embodiment Face detection method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided Such as include the memory 402 of instruction, above-metioned instruction can be performed to complete the above method by the processor 401 of the terminal 400.Example Such as, the non-transitorycomputer readable storage medium can be ROM, it is random access memory (RAM), CD-ROM, tape, soft Disk and optical data storage devices etc..
A kind of non-transitorycomputer readable storage medium, when the instruction in the storage medium is by the terminal 400 When managing device execution so that the terminal is able to carry out the pavement detection method provided in above-described embodiment.
It should be noted that the terminal 400 that above-described embodiment provides can include the road surface checking device of previous embodiment, The terminal 400 and the road surface checking device and pavement detection embodiment of the method for previous embodiment that above-described embodiment provides belong to same One design, its specific implementation process refer to embodiment of the method, and which is not described herein again.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment To complete, relevant hardware can also be instructed to complete by program, the program can be stored in a kind of computer-readable In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all the present invention spirit and Within principle, any modification, equivalent replacement, improvement and so on, should all be included in the protection scope of the present invention.

Claims (10)

  1. A kind of 1. pavement detection method, it is characterised in that the described method includes：
    When getting the corresponding V disparity maps of road conditions image, the V disparity maps are subjected to binary conversion treatment, obtain target V parallaxes Figure；
    The target V disparity maps are divided according to the first line of demarcation and obtain two subregions, and oblique line is detected in described two subregions, If the oblique line detected in described two subregions is unsatisfactory for preset condition, described the is corrected based on the oblique line that detects The position in one line of demarcation obtains target line of demarcation；
    Based on the oblique line detected in two subregions obtained after the target line of demarcation division target V disparity maps, determine Road surface in the target V disparity maps.
  2. 2. according to the method described in claim 1, it is characterized in that, described detect oblique line in described two subregions, if in institute State the oblique line detected in two subregions and be unsatisfactory for preset condition, then first boundary is corrected based on the oblique line detected The position of line obtains target line of demarcation, including：
    Oblique line is detected respectively in described two subregions, obtains the first oblique line and the second oblique line；
    Judge whether meet preset condition between first oblique line and second oblique line；
    When being unsatisfactory for preset condition between first oblique line and second oblique line, based on first oblique line and described Two oblique lines correct the position in first line of demarcation, and the target V disparity maps are divided based on the first line of demarcation after correction position Two subregions are obtained, and oblique line is detected respectively in two subregions obtained after division, until the first oblique line for detecting and the When meeting preset condition between two oblique lines, the first line of demarcation after last time correction position is determined as the target and is demarcated Line.
  3. 3. according to the method described in claim 2, it is characterized in that, it is described judgement first oblique line and second oblique line it Between whether meet preset condition, including：
    Determine first oblique line and the intersection point in first line of demarcation, obtain the first intersection point, determine second oblique line and institute The intersection point in the first line of demarcation is stated, obtains the second intersection point；
    When the distance between first intersection point and second intersection point are more than pre-determined distance, first oblique line and institute are determined State and be unsatisfactory for the preset condition between the second oblique line；
    When the distance between first intersection point and second intersection point are less than or equal to the pre-determined distance, described is determined Meet the preset condition between one oblique line and second oblique line.
  4. 4. according to the method described in claim 2, it is characterized in that, described repaiied based on first oblique line and second oblique line The position in just described first line of demarcation, including：
    Determine the intersection point of first oblique line and second oblique line, obtain the 3rd intersection point；
    By after the 3rd intersection point and the straight line parallel with the longitudinal axis of the target V disparity maps are determined as correction position One line of demarcation.
  5. 5. according to any methods of claim 1-4, it is characterised in that described to be based on dividing institute in the target line of demarcation The oblique line detected in two subregions obtained after target V disparity maps is stated, determines the road surface in the target V disparity maps, including：
    Determine to divide two oblique lines detected in two subregions obtained after the target V disparity maps in the target line of demarcation Slope, obtain first slope and the second slope；
    When the first slope and second slope differ, an oblique line in two oblique lines is passed through into extended line The broken line obtained after intersecting with another oblique line in two oblique lines is determined as the road surface in the target V disparity maps.
  6. 6. according to the method described in claim 5, it is characterized in that, described determine to divide the target in the target line of demarcation The slope of two oblique lines detected in two subregions obtained after V disparity maps, after obtaining first slope and the second slope, also Including：
    When the first slope is identical with second slope, two of two oblique lines and the target line of demarcation are connected Intersection point；
    The road surface broken line that connection obtains being determined as in the target V disparity maps.
  7. 7. a kind of road surface checking device, it is characterised in that described device includes：
    Processing module, for when getting the corresponding V disparity maps of road conditions image, the V disparity maps to be carried out binary conversion treatment, Obtain target V disparity maps；
    First determining module, two subregions are obtained for dividing the target V disparity maps according to the first line of demarcation, and described two Oblique line is detected in a subregion, if the oblique line detected in described two subregions is unsatisfactory for preset condition, based on the detection To oblique line correct the position in first line of demarcation and obtain target line of demarcation；
    Second determining module, for based on two subregions obtained after the target line of demarcation division target V disparity maps The oblique line of middle detection, determines the road surface in the target V disparity maps.
  8. 8. device according to claim 7, it is characterised in that first determining module includes：
    Detection sub-module, for detecting oblique line respectively in described two subregions, obtains the first oblique line and the second oblique line；
    Judging submodule, for judging whether first oblique line and second oblique line meet preset condition；
    Submodule is corrected, for when being unsatisfactory for preset condition between first oblique line and second oblique line, based on described First oblique line and second oblique line correct the position in first line of demarcation, based on the first line of demarcation division after correction position The target V disparity maps obtain two subregions, and detect oblique line respectively in two subregions obtained after division, until detecting The first oblique line and the second oblique line between when meeting preset condition, the first line of demarcation after last time correction position is determined as The target line of demarcation.
  9. 9. device according to claim 8, it is characterised in that the judging submodule is specifically used for：
    Determine first oblique line and the intersection point in first line of demarcation, obtain the first intersection point, determine second oblique line and institute The intersection point in the first line of demarcation is stated, obtains the second intersection point；
    When the distance between first intersection point and second intersection point are more than pre-determined distance, first oblique line and institute are determined State and be unsatisfactory for the preset condition between the second oblique line；
    When the distance between first intersection point and second intersection point are less than or equal to the pre-determined distance, described is determined Meet the preset condition between one oblique line and second oblique line.
  10. 10. a kind of pavement detection terminal, it is characterised in that the terminal includes：
    Processor；
    CCD camera assembly, for gathering road conditions image, and the road conditions image is sent to the processor and is handled；
    For storing the memory of processor-executable instruction；
    Wherein, the processor is configured as the step of any one method described in claim 1-6.
CN201711315782.8A 2017-12-12 2017-12-12 Pavement detection method and device and terminal Active CN107958222B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711315782.8A CN107958222B (en) 2017-12-12 2017-12-12 Pavement detection method and device and terminal

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711315782.8A CN107958222B (en) 2017-12-12 2017-12-12 Pavement detection method and device and terminal

Publications (2)

Publication Number Publication Date
CN107958222A true CN107958222A (en) 2018-04-24
CN107958222B CN107958222B (en) 2020-10-16

Family

ID=61957691

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711315782.8A Active CN107958222B (en) 2017-12-12 2017-12-12 Pavement detection method and device and terminal

Country Status (1)

Country Link
CN (1) CN107958222B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110132288A (en) * 2019-05-08 2019-08-16 南京信息工程大学 A kind of minicar vision navigation method on wide road surface
CN116912779A (en) * 2023-09-12 2023-10-20 四川弘和数智集团有限公司 Queuing detection method and device, electronic equipment and readable storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103679127A (en) * 2012-09-24 2014-03-26 株式会社理光 Method and device for detecting drivable area of road pavement
CN104050654A (en) * 2013-03-11 2014-09-17 株式会社理光 Road edge detection method and device
CN103177236B (en) * 2011-12-22 2016-06-01 株式会社理光 Road area detection method and device, lane line detection method and apparatus
CN107392103A (en) * 2017-06-21 2017-11-24 海信集团有限公司 The detection method and device of road surface lane line, electronic equipment

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103177236B (en) * 2011-12-22 2016-06-01 株式会社理光 Road area detection method and device, lane line detection method and apparatus
CN103679127A (en) * 2012-09-24 2014-03-26 株式会社理光 Method and device for detecting drivable area of road pavement
CN104050654A (en) * 2013-03-11 2014-09-17 株式会社理光 Road edge detection method and device
EP2779025A2 (en) * 2013-03-11 2014-09-17 Ricoh Company, Ltd. Method and system for detecting road edge
CN107392103A (en) * 2017-06-21 2017-11-24 海信集团有限公司 The detection method and device of road surface lane line, electronic equipment

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
DAI YIRUO等: "Complex Ground Plane Detection Based on V-disparity Map in Off-road Environment", 《 2013 IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV)》 *
IMAD BENACER AND AICHA HAMISSI 等: "Hardware Design and FPGA Implementation for Road Plane Extraction Based on V-disparity Approach", 《 2015 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS)》 *
上官珺: "基于U-V视差算法的障碍物识别技术研究", 《兰州工业高等专科学校学报》 *
缪小冬等: "路面车辆的视觉检测方法", 《传感器与微系统》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110132288A (en) * 2019-05-08 2019-08-16 南京信息工程大学 A kind of minicar vision navigation method on wide road surface
CN116912779A (en) * 2023-09-12 2023-10-20 四川弘和数智集团有限公司 Queuing detection method and device, electronic equipment and readable storage medium
CN116912779B (en) * 2023-09-12 2023-12-22 四川弘和数智集团有限公司 Queuing detection method and device, electronic equipment and readable storage medium

Also Published As

Publication number Publication date
CN107958222B (en) 2020-10-16

Similar Documents

Publication Publication Date Title
CN107729856B (en) A kind of obstacle detection method and device
CN109284674A (en) A kind of method and device of determining lane line
CN110956069B (en) Method and device for detecting 3D position of pedestrian, and vehicle-mounted terminal
CN109918977B (en) Method, device and equipment for determining idle parking space
US7764284B2 (en) Method and system for detecting and evaluating 3D changes from images and a 3D reference model
JP6544257B2 (en) INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND INFORMATION PROCESSING PROGRAM
CN111862228A (en) Occlusion detection method, system, computer device and readable storage medium
CN113744348A (en) Parameter calibration method and device and radar vision fusion detection equipment
CN108229406A (en) A kind of method for detecting lane lines, device and terminal
CN108319931A (en) A kind of image processing method, device and terminal
CN107958222A (en) A kind of pavement detection method, apparatus and terminal
CN111724444B (en) Method, device and system for determining grabbing point of target object
CN108460333A (en) Ground detection method and device based on depth map
CN107766847A (en) A kind of method for detecting lane lines and device
CN114821497A (en) Method, device and equipment for determining position of target object and storage medium
CN111476062A (en) Lane line detection method and device, electronic equipment and driving system
CN112146666A (en) Vehicle driving route marking method and device, computer equipment and storage medium
CN109151336B (en) Camera synchronization method, device and system, storage medium and computer equipment
JP6988860B2 (en) Information processing system, information processing method and information processing program
KR20130077933A (en) Video data information analysis system using cctv and operating method thereof
US11481996B2 (en) Calculation device, information processing method, and storage medium
JP5964093B2 (en) Vehicle size measuring device, vehicle size measuring method, and program
CN110378168A (en) The method, apparatus and terminal of polymorphic type barrier fusion
CN113506283B (en) Image processing method and device, storage medium and electronic device
CN107911629A (en) A kind of brightness adjusting method and device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant