CN109801282A - Pavement behavior detection method, processing method, apparatus and system - Google Patents
Pavement behavior detection method, processing method, apparatus and system Download PDFInfo
- Publication number
- CN109801282A CN109801282A CN201910070867.7A CN201910070867A CN109801282A CN 109801282 A CN109801282 A CN 109801282A CN 201910070867 A CN201910070867 A CN 201910070867A CN 109801282 A CN109801282 A CN 109801282A
- Authority
- CN
- China
- Prior art keywords
- pavement
- image
- corrupted
- road surface
- damage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 64
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 43
- 238000003707 image sharpening Methods 0.000 claims abstract description 12
- 239000000284 extract Substances 0.000 claims abstract description 8
- 230000006399 behavior Effects 0.000 claims description 137
- 238000012360 testing method Methods 0.000 claims description 49
- 238000012549 training Methods 0.000 claims description 38
- 238000012423 maintenance Methods 0.000 claims description 36
- 238000004422 calculation algorithm Methods 0.000 claims description 24
- 238000000605 extraction Methods 0.000 claims description 24
- 238000005457 optimization Methods 0.000 claims description 19
- 239000003550 marker Substances 0.000 claims description 16
- 230000001771 impaired effect Effects 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 238000003709 image segmentation Methods 0.000 claims description 4
- 238000010191 image analysis Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 13
- 230000000694 effects Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000000638 solvent extraction Methods 0.000 description 7
- 230000008901 benefit Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000003862 health status Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000009835 boiling Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005336 cracking Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Landscapes
- Road Repair (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
本发明提供了一种路面状况检测方法、处理方法、装置及系统，涉及图像分析技术领域。其中，该方法包括通过车载摄像头和定位装置获取车辆所经路面的路面图像及路面图像对应的地理位置；将路面图像进行图像去噪处理和图像锐化处理后，输入至预训练的路面状况识别模型进行坏损路面标记，并对带有坏损路面标记的路面图像进行损坏参数的提取；将带有坏损路面标记的路面图像、该路面图像对应的地理位置和损坏参数汇总至路面状况检测结果中，这种利用车载摄像头拍摄路面的方式，减少了人力的使用，通过路面状况识别模型对路面进行坏损路面标记，可以准确实现对多级路段、多坏损路面进行损坏识别和标记，以诊断多种坏损情况。
The invention provides a road condition detection method, processing method, device and system, which relate to the technical field of image analysis. Wherein, the method includes obtaining a road image of the road the vehicle passes through and a geographic location corresponding to the road image through a vehicle-mounted camera and a positioning device; after performing image denoising and image sharpening processing on the road image, inputting it into the pre-trained road surface condition recognition The model marks the damaged pavement, and extracts the damage parameters from the pavement image with the damaged pavement mark; summarizes the pavement image with the damaged pavement mark, the geographic location and damage parameters corresponding to the pavement image to the road surface condition detection In the results, this method of using the vehicle camera to photograph the road surface reduces the use of manpower. The road surface condition recognition model is used to mark the road surface with damage, which can accurately identify and mark the damage of multi-level road sections and multi-damaged road surfaces. to diagnose a variety of damage conditions.
Description
Technical field
The present invention relates to image analysis technology fields, more particularly, to a kind of pavement behavior detection method, processing method, dress
It sets and system.
Background technique
Highroad is generally made of the higher bituminous pavement of receiving dynamics or concrete road surface at present, carload,
Temperature and humidity repeated action, sleet corrode etc. conditions it is long-term under the influence of, military service performance and the service level of highway are gradually reduced, and are deposited
In the different degrees of damage such as surface deformation, depression, cracking, frost boiling, pit-hole, check crack.Currently, in order to which road pavement situation is tieed up
It repairs and conserves, highway administration department needs the pavement behavior of periodic detection highway section, and now common detection mode is artificial inspection
It surveys and carries out pavement detection using automated intelligent detection device, wherein artificial detection is manually to measure record road surface corrupted situation
With location information.This manually mode that detects point by point the problem of there is only inefficiency, higher cost, and on highway road
Easily there is personal safety accident in manual inspection in section.
During carrying out pavement detection using automated intelligent detection device, though used automated intelligent detection device
Detection efficiency can be so improved to a certain extent, but cost of use is too high, technology is complicated, and operation difficulty is big, and mostly single work
Environmental preparation is unfavorable for universal and maintenance.And the problems such as being directed to multiple features corrupted, miniature corrupted, track corrupted, is helpless,
This kind of detection device can only play basic automatic diagnostic function to part road surface corrupted situation simultaneously, can not be to highway health status
Accomplish effectively detection and forecast function with service life.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of pavement behavior detection method, processing method, device and being
System needs a large amount of manpowers to alleviate existing highway corrupted auto-check system and method, and can not diagnose multiple types highway it is bad
The technical issues of damage.
In a first aspect, this method is applied to mobile unit the embodiment of the invention provides a kind of pavement behavior detection method
On, which is configured with vehicle-mounted camera and positioning device, this method comprises: by vehicle-mounted camera and positioning device,
Obtain respectively vehicle pavement image through road surface and the corresponding geographical location of pavement image；Road pavement image carries out image denoising
Processing and image sharpening processing, the pavement image after being optimized；By the road surface shape of the pavement image input pre-training after optimization
Condition identification model carries out corrupted pavement markers, wherein corrupted pavement markers include types of damage label and failure area label；It is right
Pavement image with corrupted pavement markers carries out the extraction of damage parameter；Wherein, damage parameter includes at least following one: damage
Bad area, damage shape, damage color and damage texture；Pavement image, the pavement image pair of corrupted pavement markers will be had
The geographical location and damage parameter answered are summarized into pavement behavior testing result.
With reference to first aspect, the embodiment of the invention provides the first possible embodiments of first aspect, wherein will
The step of pavement behavior identification model of pavement image input pre-training after optimization carries out corrupted pavement markers, comprising: to excellent
Pavement image after change carries out image segmentation, generates multiple road surface block of pixels；Wherein, road surface block of pixels is marked with station location marker,
Station location marker is the position in the pavement image of road surface block of pixels after optimization；By the road surface shape of road surface block of pixels input pre-training
Condition identification model carries out corrupted pavement markers；According to the station location marker of the road surface block of pixels with corrupted pavement markers, by the road
The corrupted pavement markers of face block of pixels load in the corresponding position of station location marker.
The possible embodiment of with reference to first aspect the first, the embodiment of the invention provides second of first aspect
Possible embodiment, wherein the training process of pavement behavior identification model, comprising: obtain road surface sample image；Wherein, road
Face sample image is marked with corrupted pavement markers sample；Wherein, corrupted pavement markers sample includes types of damage sample and damage
Area sample；It is obtained according to the training algorithm of setting using road surface sample image and corrupted pavement markers sample training initial model
To trained pavement behavior identification model.
With reference to first aspect, the embodiment of the invention provides the third possible embodiments of first aspect, wherein right
Pavement image with corrupted pavement markers carries out the step of extraction of damage parameter, comprising: extracts and has corrupted pavement markers
Pavement image in failure area；Calculate the area coverage that failure area accounts for the pavement image with corrupted pavement markers；
According to the camera calibration parameter of vehicle-mounted camera, three-dimensional reconstruction is carried out to the pavement image with corrupted pavement markers, is obtained
The actual size of the pavement image；Using area coverage and actual size, the impaired area of failure area is calculated.
With reference to first aspect, the embodiment of the invention provides the 4th kind of possible embodiments of first aspect, wherein right
Pavement image with corrupted pavement markers carries out the step of extraction of damage parameter, comprising: extracts and has corrupted pavement markers
Pavement image, utilize the damage parameter model of pre-training to extract the damage ginseng of the pavement image with corrupted pavement markers
Number.
Second aspect, the embodiment of the present invention also provide a kind of pavement behavior processing method, and this method is applied to Cloud Server
On；This method comprises: obtaining the pavement behavior testing result that mobile unit generates；The pavement behavior testing result is mobile unit
It is obtained based on first aspect the method；Store pavement behavior testing result；It is bad according to being had in pavement behavior testing result
The corresponding geographical location of pavement image for damaging pavement markers, the corrupted pavement markers of the pavement image are embodied and are located in map
On；The maintenance status information on road surface is determined according to pavement behavior testing result, maintenance status information includes at least one of: being made
With service life, maintenance cost and maintenance frequency.
The third aspect, the embodiment of the present invention also provide a kind of pavement behavior detection device, which answers
For mobile unit, which is configured with vehicle-mounted camera and positioning device, which includes: to obtain
Modulus block, for by vehicle-mounted camera and positioning device, obtain respectively vehicle pavement image and pavement image through road surface
Corresponding geographical location；Image processing module carries out image denoising processing and image sharpening processing for road pavement image, obtains
Pavement image after optimization；Mark module, the pavement behavior identification model for the pavement image input pre-training after optimizing
Carry out corrupted pavement markers, wherein corrupted pavement markers include types of damage label and failure area label；Extraction module is used
In the extraction for carrying out damage parameter to the pavement image with corrupted pavement markers；Wherein, damage parameter include at least it is following it
One: impaired area, damage shape, damage color and damage texture；Summarizing module, for the road surface of corrupted pavement markers will to be had
Image, the corresponding geographical location of the pavement image and damage parameter summarize into pavement behavior testing result.
Fourth aspect, the embodiment of the present invention also provide a kind of pavement behavior processing unit, and device is applied on Cloud Server；
The device includes: acquisition object module, obtains the pavement behavior testing result that mobile unit generates；The pavement behavior testing result
It is obtained for mobile unit based on first aspect the method；Memory module, for storing pavement behavior testing result；Positioning mould
Block, for according in pavement behavior testing result have corrupted pavement markers the corresponding geographical location of pavement image, by the road
The corrupted pavement markers of face image, which are embodied, to be located on map；Maintenance block of state is determined, for detecting according to pavement behavior
As a result determine road surface maintenance status information, maintenance status information include at least one of: service life, maintenance cost and support
Protect frequency.
5th aspect, the embodiment of the invention provides a kind of pavement behavior detection system, the system include mobile unit and
Cloud Server, mobile unit are configured with vehicle-mounted camera and positioning device；Mobile unit is installed on vehicle；Vehicle-mounted camera is used
In shooting vehicle the pavement image through road surface, and pavement image is sent to mobile unit；Positioning device is for positioning road surface
The corresponding geographical location of image, and geographical location is sent to mobile unit；Mobile unit includes road described in the third aspect
Face condition detection apparatus；Cloud Server includes pavement behavior processing unit described in fourth aspect.
6th aspect, the embodiment of the present invention also provides a kind of computer storage medium, for storing computer program instructions,
When computer executes shown computer program instructions, method as described in relation to the first aspect is executed, or execute such as second aspect
The method.
The embodiment of the present invention bring it is following the utility model has the advantages that
Above-mentioned pavement behavior detection method provided in an embodiment of the present invention, processing method, apparatus and system, are taken the photograph by vehicle-mounted
As head and positioning device, obtain respectively vehicle pavement image through road surface and the corresponding geographical location of pavement image；Road pavement
Image carries out image denoising processing and image sharpening processing, the pavement image after being optimized；Pavement image after optimization is defeated
The pavement behavior identification model for entering pre-training carries out corrupted pavement markers, and carries out to the pavement image with corrupted pavement markers
Damage the extraction of parameter；By pavement image, the corresponding geographical location of the pavement image and damage ginseng with corrupted pavement markers
Number summarize into pavement behavior testing result, it is this by vehicle-mounted camera shooting vehicle through road surface in the way of, reduce people
The use of power, it is this that corrupted pavement markers are carried out by pavement behavior identification model road pavement and damage the side of the extraction of parameter
Formula can accurately realize identification and the label of the failure area to multistage section, more corrupted road surfaces, to diagnose a variety of corrupted feelings
Condition.
Other features and advantages of the present invention will illustrate in the following description, also, partly become from specification
It obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention are in specification and attached drawing
Specifically noted structure is achieved and obtained.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
It, below will be to specific in order to illustrate more clearly of the specific embodiment of the invention or technical solution in the prior art
Embodiment or attached drawing needed to be used in the description of the prior art be briefly described, it should be apparent that, it is described below
Attached drawing is some embodiments of the present invention, for those skilled in the art, without creative efforts,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of pavement behavior detection method provided in an embodiment of the present invention；
A kind of Fig. 2 pavement behavior detection effect figure provided in an embodiment of the present invention；
Fig. 3 is a kind of pavement behavior processing method provided in an embodiment of the present invention；
Fig. 4 is a kind of pavement behavior detection device provided in an embodiment of the present invention；
Fig. 5 is a kind of pavement behavior processing unit provided in an embodiment of the present invention；
Fig. 6 is a kind of pavement behavior detection system provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with attached drawing to the present invention
Technical solution be clearly and completely described, it is clear that described embodiments are some of the embodiments of the present invention, rather than
Whole embodiments.Based on the embodiments of the present invention, those skilled in the art institute without making creative work
The every other embodiment obtained, shall fall within the protection scope of the present invention.
Currently, existing road conditions detection mode is mostly artificial detection or is carried out using highway corrupted auto-check system
Detection, wherein the working efficiency of artificial detection is lower, and personal safety accident easily occurs, diagnoses automatically using highway corrupted
When system is detected, cost of use is too high, and technology is complicated, and operation difficulty is big, and mostly single working environment prepares, and is unfavorable for
Universal and maintenance.And the problems such as being directed to multiple features corrupted, miniature corrupted, track corrupted, is helpless, while this kind of detection is set
It is standby to play basic automatic diagnostic function to part road surface corrupted situation, highway health status can not be accomplished with service life
Effective detection and forecast function.Based on this, a kind of pavement behavior detection method provided in an embodiment of the present invention, processing method,
Apparatus and system, less manpower, which can be used, can be realized diagnostic work to the section corrupted of multiple types.
For convenient for understanding the present embodiment, first to a kind of pavement behavior detection side disclosed in the embodiment of the present invention
Method describes in detail.
Embodiment one:
The embodiment of the invention provides a kind of pavement behavior detection method, this method is applied on mobile unit, this is vehicle-mounted
Equipment is server or electronic equipment etc. with data receiver and processing function, such as has GPU (Graphics
Processing Unit, graphics processor) unit computer equipment, mobile unit can be optionally mounted on any vehicle, should
Mobile unit is configured with vehicle-mounted camera and positioning device.Fig. 1 shows a kind of flow chart of pavement detection method, this method packet
Include following steps:
Step S102, by vehicle-mounted camera and positioning device, obtain respectively vehicle pavement image through road surface and road
The corresponding geographical location of face image；
When specific implementation, driver can drive vehicle driving on needing the road surface detected, which can be with
Select binocular camera or other high-definition cameras, the vehicle-mounted camera that can be fixedly mounted on vehicle tail or head, mirror
Upside down, keep horizontal with road surface, for the pavement image of captured in real-time vehicle running surface, which, which can be, passes through
Frame is carried out to the road surface video of shooting and decomposes acquisition, or driver has found to use vehicle manually when corrupted road surface when driving
The pavement image of camera shooting is carried, which can be communicated by data-interface with mobile unit；Further,
Positioning device can select GPS (Global Positioning System, global positioning system) built-in terminal, the positioning
The geographical location that device obtains can be one-to-one with pavement image, can also set in the short time captured road surface view
Frequency and the road surface video frame decompose the public same geographical location of pavement image obtained.
Step S104, road pavement image carry out image denoising processing and image sharpening processing, the road surface figure after being optimized
Picture；
When specific implementation, during carrying out image denoising processing, it can be filtered with road pavement image.Example
Such as, it can use median filtering road pavement image to be filtered, choose the filter window of certain size, filter window is existed
The gray value of the initial position of filtering is substituted by intermediate value in the neighborhood of set filter window, so that around filter window
Pixel value is more nearly true value, to eliminate isolated noise spot.
It, can be by adjusting the contrast of filtered pavement image, to expand during carrying out image sharpening processing
The otherness of big damaged district and road surface characteristic, in order to which subsequent image processing is diagnosed with corrupted.For example, can be by original RGB
(Red Green Blue, RGB) pavement image is transformed into HSV (Hue Saturation Value, tone saturation degree respectively
And lightness) and color space in, using histogram equalization by histogram to the ash more than number in filtered pavement image
Grading line broadening is spent, the gray level few to number of pixels is reduced, to enhance local contrast without influencing whole pair
Than degree, so that pavement image is relatively sharp, the pavement image after finally obtaining optimization；
The pavement behavior identification model of pavement image input pre-training after optimization is carried out corrupted road marking by step S106
Note, wherein corrupted pavement markers include types of damage label and failure area label；
When specific implementation, which can be by deep learning algorithm and/or machine learning algorithm
Training obtains in advance, and corrupted road surface is usually the road surface for having failure area, and the types of damage of failure area is usually crack, hole
The types such as slot, track, loose, depression, frost boiling, types of damage label and failure area label are for respectively by failure area
The location information of types of damage and failure area marks on the corresponding pavement image in corrupted road surface.
Step S108 carries out the extraction of damage parameter to the pavement image with corrupted pavement markers；Wherein, parameter is damaged
Including at least following one: impaired area, damage shape, damage color and damage texture；
When specific implementation, it can use precompile and extract formula and/or image characteristics extraction algorithm, to obtain pavement image
Damage parameter.
Step S110 will have pavement image, the corresponding geographical location of the pavement image and the damage of corrupted pavement markers
Parameter summarizes into pavement behavior testing result.
The embodiment of the invention provides a kind of pavement behavior detection method, this method is applied on mobile unit, this is vehicle-mounted
Device configuration has vehicle-mounted camera and positioning device, and this method obtains vehicle institute by vehicle-mounted camera and positioning device respectively
The corresponding geographical location of pavement image and pavement image through road surface；Road pavement image carries out image denoising processing and image sharpening
Processing, the pavement image after being optimized；The pavement behavior identification model of pavement image input pre-training after optimization is carried out
Corrupted pavement markers, and the extraction of damage parameter is carried out to the pavement image with corrupted pavement markers；Corrupted road surface will be had
The pavement image of label, the corresponding geographical location of the pavement image and damage parameter summarize into pavement behavior testing result, this
Kind using vehicle-mounted camera shoot vehicle through road surface in the way of, reduce the use of manpower, it is this to be identified by pavement behavior
Model road pavement carries out corrupted pavement markers and damages the mode of the extraction of parameter, can accurately realize to multistage section, how bad
Identification and the label of the failure area on road surface are damaged, to diagnose a variety of corrupted situations.
During carrying out corrupted pavement markers to the pavement image after optimization, in order to by each bad of the pavement image
Damage part carry out clear label, before label, can by the pavement image carry out image dividing processing, then to segmentation after
Pavement image carry out corrupted label, be based on this, step S106, by after optimization pavement image input pre-training pavement behavior
Identification model carries out corrupted pavement markers, can be realized by step 11, step 12 and step 13:
Step 11, image segmentation is carried out to the pavement image after optimization, generates multiple road surface block of pixels；Wherein, road surface picture
Plain block is marked with station location marker, and station location marker is the position in the pavement image of road surface block of pixels after optimization；
During carrying out image segmentation, a kind of partitioning scheme can be for will be having a size of W × H pavement image with default big
Small window carries out pixel segmentation, and the size of the pavement image is fixation picture size W × H that vehicle-mounted camera is shot,
It is also possible to the image for shooting vehicle-mounted camera or video modification into specified size W × H, the format of pavement image can be set
To specify picture format；Further, pavement image is carried out pixel with the window of N × N to be divided into, is obtainedA road
Face pixel, when wherein the road surface pixel is non-integer, with the downward value principle being rounded as the road surface pixel；Utilize this segmentation
When the road surface block of pixels that mode obtains carries out neural computing, it is likely to result in the local feature generated by division size and covers
The global feature of pavement image, and then influence the recognition effect of road pavement block of pixels.
Another partitioning scheme can be with are as follows: establishes rectangular coordinate system (including x-axis and y-axis) on the pavement image, adopts
The pavement image having a size of W × H is appointed with image overlap partition algorithm, pixel segmentation is carried out with the window of N × N size, presets N
Shift step of the window of × N size in x-axis is xstep, shift step on the y axis is ystep, calculated to simplify, this reality
It applies example and the window of N × N size is disposed as synchronous shift step l with the shift step in y-axis in x-axisstep), then x-axis direction
Can carry outSecondary segmentation, y-axis direction can carry outSecondary segmentation, is obtainedA overlapping road surface block of pixels, using the partitioning scheme, when neural network is to each overlapping
When road surface block of pixels is calculated, multiple operation can be carried out to lap, to realize the local feature of road pavement image and whole
The assurance of body characteristics is identified though increasing calculation amount to a certain extent convenient for the damaged district of road pavement image.It is worth
Illustrate, the setting of above-mentioned moving step length need to be adapted to initial model network topology structure and mobile unit it is computational
Can, comprehensively consider the accuracy, sensitivity and specificity of calculating, Rational choice shift step, to realize the identification optimized effect
Fruit.
During station location marker is marked in road pavement block of pixels, can on pavement image the multiple volumes of preliminary making
Code, each road surface block of pixels also can use certain in the block of pixels of road surface using the coding of position as corresponding station location marker
The some station location marker of the coordinate of (central point or frame line) as road surface block of pixels.
Step 12, the pavement behavior identification model of road surface block of pixels input pre-training is subjected to corrupted pavement markers；
Pavement behavior identification model can be with whether failure area be judged in road pavement block of pixels, if it is, can be with
The types of damage of failure area is marked, can also include the corresponding position frame of failure area.In addition, on the road of input pre-training
Before planar condition identification model, need road surface block of pixels carrying out image data vector processing, to convert road surface block of pixels
It is input in pavement behavior identification model for tensor.Further, above-mentioned pavement behavior identification model can pass through following steps 01
It is obtained with step 02 training:
Step 01, road surface sample image is obtained；Wherein, road surface sample image is marked with corrupted pavement markers sample；Wherein,
Corrupted pavement markers sample includes types of damage sample and failure area sample；
Specifically, road surface sample image can be obtained by intercepting the video frame of multiple road surface Sample videos, can also be led to
The acquisition of real scene shooting road surface is crossed, the corrupted pavement markers sample marked on each road surface sample image can be obtained by manual measurement
It arrives, can also be obtained by other pavement behavior detection device automatic measurements.
Step 02, initial using road surface sample image and corrupted pavement markers sample training according to the training algorithm of setting
Model obtains trained pavement behavior identification model.
When specific implementation, during training initial model, machine learning and/or deep learning algorithm pair can be used
Initial model is trained and verifies, to obtain trained pavement behavior identification model.
Further, during above-mentioned trained initial model, machine learning and/or deep learning algorithm structure can be used
Established model network topology structure, is iterated training data and verify data, improves accuracy of identification and reduces and lose, specifically
Ground, can advance with a variety of different training algorithms, for example, K nearest neighbor algorithm, SVM (Support Vector Machine,
Support vector machines), logistic regression, decision tree scheduling algorithm, multiple initial models are trained respectively, wherein initial model
Network topology structure includes to be not limited to the structures such as Alexnet, VGG, Googlenet.Further, it is also possible to be calculated using data verification
The verification algorithms such as method, such as data enhancing algorithm, the verifying of K folding, weight regularization, dropout regularization, to the net of initial model
Network topological structure and size are made rational planning for, to solve over-fitting and poor fitting phenomenon；Finally to different training algorithms and not
The multiple pavement behavior identification models obtained with initial model training are verified, that is, road can be accurately identified and mark by selecting
The training algorithm and initial model of face types of damage and damage position, and the road that the training algorithm and initial model training are obtained
Planar condition identification model is as above-mentioned trained pavement behavior identification model.
Step 13, according to the station location marker of the road surface block of pixels with corrupted pavement markers, by the bad of the road surface block of pixels
Pavement markers load is damaged in the corresponding position of station location marker；
When specific implementation, the road surface block of pixels with corrupted pavement markers can be directly overlayed into corresponding pavement image
Corresponding position, can also according to station location marker, by the position of the failure area in the corrupted pavement markers of road surface block of pixels and
Type is re-flagged in the corresponding position of pavement image.
During road pavement failure area is detected, impaired area, the damage in pavement damage region can be known
Shape, damage color and damage texture, are based on this, and step S108 damages the pavement image with corrupted pavement markers
The extraction of parameter can there are two types of modes, one way in which can be realized by step 21, step 22, step 23 and step 24:
Step 21, the failure area in the pavement image with corrupted pavement markers is extracted；
It when specific implementation, in the pavement image with corrupted pavement markers, is marked according to all failure areas, button choosing
Multiple identified areas images out, the identified areas image include failure area image and damage area of failure area label
The surrounding normal pavement image in domain.
Step 22, the area coverage that failure area accounts for the pavement image with corrupted pavement markers is calculated；
When specific implementation, damage contours extract can be carried out to the failure area image in identified areas image, for example, making
Damage contour edge detection is carried out to failure area image with Sobel edge detection algorithm, then calculates what damage profile was surrounded
The pixel number p of failure area imagedis, and then obtain failure area and account for pavement image (road with corrupted pavement markers
The size of face image is W × H) area coverage be
Step 23, according to the camera calibration parameter of vehicle-mounted camera, to the pavement image with corrupted pavement markers into
Row three-dimensional reconstruction obtains the actual size of the pavement image；
When specific implementation, by taking vehicle-mounted camera is binocular camera as an example, the calibrating parameters of the binocular camera include double
Mesh camera inside and outside parameter, wherein inner parameter is the intrinsic parameter of binocular camera, including picture centre coordinate (xc, yc), two
Camera focus f, two camera distance d, x aspect ratio coefficient Nx, y aspect ratio coefficient Ny, distortion coefficients of camera lens k etc., external parameter
Calibration refer to three-dimensional position and direction relations of the coordinate system relative to a certain world coordinate system of binocular camera.
During calculating the actual size of pavement image, road surface figure in the two images of binocular camera shooting is utilized
As Stereo matching one triangle of composition between fixed point and fixed point itself carries out three-dimensional reconstruction to obtain three-dimensional information
To obtain the actual size of pavement image, wherein the binocular camera that the distance of optical center connection is B shoots space in synchronization
In road surface, respectively left video camera shooting image planes and right video camera shooting image planes on obtain road surface pavement image, sit
Mark is respectively (Xleft, Yleft)、(Xright, Yright), it can be obtained by similar triangles relationshipRoad surface figure
The actual size of picture is
Step 24, using area coverage and actual size, the impaired area of failure area is calculated.
When specific implementation, impaired area is
Further, another way can be realized by step 31 and step 32:
Step 31, the pavement image for having corrupted pavement markers is extracted；
Step 32, joined using the damage that the damage parameter model of pre-training extracts the pavement image with corrupted pavement markers
Number.
The damage parameter model can be the model or algorithm pre-established according to the damage parameter sample of damage road surface sample,
During establishing model or algorithm, the model that is pre-established using neural network, deep learning scheduling algorithm, or right
It damages parameter sample and carries out characteristic parameter extraction and statistics, the algorithm established according to the regularity of the characteristic parameter of statistics；Into one
Step, the damage parameter model can carry out impaired area, damage shape, damage face to the pavement image with corrupted pavement markers
The extraction of the damage parameter such as color and damage texture.Extracting the damage parameter of the above-mentioned pavement image with corrupted pavement markers
It in the process, can also be by, according to default lookup principle, searching to the pavement image with corrupted pavement markers and summarizing the road
The image feature value of failure area in the image of face, and then the area pixel point of failure area is combined, calculate the damage of the pavement image
Bad parameter.
Based on above-mentioned pavement behavior detection method, Fig. 2 shows a kind of pavement behavior detection effect figures；
Wherein, (a) figure in Fig. 2 is the effect picture of 1920 × 1080 sized images 64 × 64 not overlapped partitioning mode；Fig. 2
In (b) figure be 1920 × 1080 sized image, 64 × 64 overlapped partitioning mode, the effect picture that shift step is 32；In Fig. 2
(c) figure is 1920 × 1080 sized image, 64 × 64 overlapped partitioning mode, the effect picture that shift step is 48；
It can be seen from (a) figure, (b) figure in Fig. 2 and (c) figure using overlapped partitioning mode road pavement image at
Reason, when the walk step-length set during the treatment is larger, the result of the pavement behavior detection of detection is more accurate.
Using above-mentioned pavement behavior detection method, it can be achieved that the failure area of road pavement is identified, acquisition types of damage,
It is as follows to damage details, the advantages of this method such as parameter:
(1) using efficient image processing techniques road pavement image carry out image sharpening, enhance failure area contrast,
Regional luminance, region acutance, expand characteristics of image difference, significantly improve the accuracy of image recognition；
(2) many-sided training algorithm for testing a variety of machine learning and deep learning, chooses and merges optimal algorithm, designs
Exclusive initial model network topology structure, realizes the discrimination of optimal road surface damage type and region, while considering that model is known
Other speed, advanced optimizes the network structure of initial model, realizes optimal recognition speed.
(3) it can use pavement behavior identification model to accurately identify in the pavement image and road surface video on polymorphic type road surface
The damage number amount and type of pavement damage, and the damage parameter of the extraction failure area of the multiple angles of energy, including morphological feature, face
Color characteristic, textural characteristics etc., adaptation highway and roads at different levels.
(4) method provided in this embodiment is easy to operate, operates without professional, and user only needs normal driving automobile i.e.
Can, mobile unit can execute this method and automatically process and identification pavement image data.
(5) mobile unit, vehicle-mounted camera and the positioning device that the present embodiment uses adapt to any vehicle, staff
The professional test vehicle and equipment high without procurement price, only need to be by the mobile unit of executable this method and the letter of existing vehicle
Single installation.This method is not high to the required precision of vehicle-mounted camera, positioning device and mobile unit simultaneously, also protects without stringent
Close equipment, it is only necessary to which simple line service is carried out to equipment.
Embodiment two:
The embodiment of the present invention also provides a kind of pavement behavior processing method, with reference to a kind of pavement behavior processing shown in Fig. 3
The flow chart of method, this method are applied on Cloud Server；Method includes the following steps:
Step S202 obtains the pavement behavior testing result that mobile unit generates；The pavement behavior testing result is the vehicle
Carry what equipment was obtained based on pavement behavior detection method described in embodiment one, specifically, mobile unit can examine pavement behavior
It surveys result to transmit to Cloud Server, Cloud Server also can according to need, and actively obtain the pavement behavior in mobile unit
Testing result.
Step S204 stores pavement behavior testing result；
Step S206, according to the corresponding geographical position of the pavement image for having corrupted pavement markers in pavement behavior testing result
It sets, the corrupted pavement markers of the pavement image is embodied and are located on map；
When specific implementation, Cloud Server can by corrupted pavement markers types of damage and failure area mark above-mentioned
On the corresponding geographical location of pavement image, for example, in conjunction with GIS (Geographic Information System, geography information
System) information API (Application Programming Interface, application programming interface) and GPS embedded end
The location information of end record failure area carries out precise positioning to lesion area, can be in order to which driver is during progress
By watching map, the damaged condition of running section is inquired.The detail location letter of each failure area can also be positioned in detail
Breath, in order to which the pavement maintenance for road maintenance personnel provides precisely navigation and positioning service with maintenance work.
Step S208 determines that the maintenance status information on road surface, maintenance status information include according to pavement behavior testing result
At least one of: service life, maintenance cost and maintenance frequency.
When specific implementation, according to the detailed data of the damage parameter of section failure area, section health status and pre- is estimated
Section service life is surveyed, pitch required for maintenance road surface can also be estimated according to the detailed data of impaired area is probably used
Amount.
In addition, the Cloud Server can be configured with mobile application software, which is mounted on mobile device
On, with details such as damages for showing road surface.
Pavement behavior processing method provided in this embodiment, this method is applied to Cloud Server, available and store vehicle
Carry the pavement behavior testing result that equipment generates；And the pavement image for corrupted pavement markers being had in pavement behavior testing result
It is embodied and is located on map, the maintenance status information on road surface can also be determined according to pavement behavior testing result；This record
Periodic maintenance can be arranged in order to which Cloud Server is according to multiple pavement behavior testing result in the mode of pavement behavior testing result
Remind, predict section service life etc., this be embodied the pavement image with corrupted pavement markers is located on map
Mode can provide failure area pinpoint navigation service for the maintenance work of highway, and precise positioning avoids omitting.
Embodiment three:
The embodiment of the present invention also provides a kind of pavement behavior detection device, with reference to a kind of pavement behavior detection shown in Fig. 4
The structural schematic diagram of device, the pavement behavior detection device are applied on mobile unit, which is configured with vehicle-mounted pick-up
Head and positioning device, the pavement behavior detection device include:
Obtain module 302, for by vehicle-mounted camera and positioning device, obtain respectively vehicle the road surface figure through road surface
Picture and the corresponding geographical location of pavement image；
Image processing module 304 carries out image denoising processing and image sharpening processing for road pavement image, is optimized
Pavement image afterwards；
Mark module 306, it is bad for carrying out the pavement behavior identification model of the pavement image input pre-training after optimization
Damage pavement markers, wherein corrupted pavement markers include types of damage label and failure area label；
Extraction module 308, for carrying out the extraction of damage parameter to the pavement image with corrupted pavement markers；Wherein,
It damages parameter and includes at least following one: impaired area, damage shape, damage color and damage texture；
Summarizing module 310, for pavement image, the corresponding geographical location of the pavement image of corrupted pavement markers will to be had
Summarize with damage parameter into pavement behavior testing result.
The embodiment of the invention provides a kind of pavement behavior detection device, which passes through vehicle-mounted pick-up
Head and positioning device, obtain respectively vehicle pavement image through road surface and the corresponding geographical location of pavement image；Road pavement figure
As carrying out image denoising processing and image sharpening processing, the pavement image after being optimized；By the pavement image input after optimization
The pavement behavior identification model of pre-training carries out corrupted pavement markers, and damages to the pavement image with corrupted pavement markers
The extraction of bad parameter；By pavement image, the corresponding geographical location of the pavement image and damage parameter with corrupted pavement markers
Summarize into pavement behavior testing result, it is this by vehicle-mounted camera shooting vehicle through road surface in the way of, reduce manpower
Use, it is this pavement behavior identification model road pavement carry out corrupted pavement markers and damage parameter extraction by way of,
It can realize identification and the label of the failure area to multistage section, more corrupted road surfaces, accurately to diagnose a variety of corrupted situations.
Example IV
The embodiment of the present invention also provides a kind of pavement behavior processing unit, with reference to a kind of pavement behavior processing shown in fig. 5
The structural schematic diagram of device, the device are applied on Cloud Server；The device includes:
Object module 402 is obtained, for obtaining the pavement behavior testing result of mobile unit generation；Pavement behavior detection
As a result it is obtained for the mobile unit based on pavement behavior detection method described in embodiment one；
Memory module 404, for storing pavement behavior testing result；
Locating module 406, for corresponding according to the pavement image for having corrupted pavement markers in pavement behavior testing result
Geographical location, the corrupted pavement markers of the pavement image are embodied and are located on map；
It determines maintenance block of state 408, for determining the maintenance status information on road surface according to pavement behavior testing result, ties up
Protecting status information includes at least one of: service life, maintenance cost and maintenance frequency.
Pavement behavior processing unit provided in this embodiment, pavement behavior inspection that is available and storing mobile unit generation
Survey result；And the pavement image for having corrupted pavement markers in pavement behavior testing result is embodied and is located on map, also
The maintenance status information on road surface can be determined according to pavement behavior testing result；The side of this record pavement behavior testing result
Formula can be arranged periodic maintenance and remind, predict section service life in order to which Cloud Server is according to multiple pavement behavior testing result
Deng, it is this that the pavement image with corrupted pavement markers is embodied the mode being located on map, it can be supported for the maintenance of highway
Nurse makees to provide the service of failure area pinpoint navigation, and precise positioning avoids omitting.
Embodiment five
The embodiment of the invention provides a kind of pavement behavior detection systems, with reference to a kind of pavement behavior detection shown in fig. 6
The structural schematic diagram of system, the system include mobile unit 502 and Cloud Server 504, and mobile unit is configured with vehicle-mounted camera
506 and positioning device 508；Mobile unit is installed on vehicle；Specifically, which is with data receiver and processing function
Server or electronic equipment of energy etc., such as with the electricity of GPU (Graphics Processing Unit, graphics processor) unit
Brain equipment, mobile unit can be optionally mounted on any vehicle.
Vehicle-mounted camera for shoot vehicle the pavement image through road surface, and pavement image is sent to mobile unit；
Specifically, which can select binocular camera or other high-definition cameras, which can fix peace
Mounted in vehicle tail or head, camera lens keeps horizontal downward, with road surface；The vehicle-mounted camera can pass through data-interface and vehicle
Equipment is carried to be communicated.
Geographical location is sent to mobile unit for positioning the corresponding geographical location of pavement image by positioning device；Tool
Body, which can select GPS (Global Positioning System, global positioning system) built-in terminal.
Mobile unit includes pavement behavior detection device described in embodiment three；
Cloud Server includes pavement behavior processing unit described in example IV.
Pavement behavior detection method and pavement behavior processing method provided in an embodiment of the present invention are provided with above-described embodiment
Pavement behavior detection device and pavement behavior processing unit technical characteristic having the same, so also can solve identical technology
Problem reaches identical technical effect.
The embodiment of the invention also provides a kind of server, which includes memory and processor, above-mentioned storage
Device is used to store the program of two the method for program or embodiment for supporting processor to execute one the method for above-described embodiment, on
Processor is stated to be configurable for executing the program stored in the memory.
Further, the embodiment of the present invention also provides a kind of computer storage medium, for storing computer program instructions, when
When computer executes shown computer program instructions, the method as described in above-described embodiment one is executed, or execute such as above-mentioned implementation
Method described in example two.
The computer program of pavement behavior detection method, processing method provided by the embodiment of the present invention, apparatus and system
Product, the computer readable storage medium including storing program code, the instruction that said program code includes can be used for executing
Previous methods method as described in the examples, specific implementation can be found in embodiment of the method, and details are not described herein.
For convenience and simplicity of description, the specific work process of the system of foregoing description and device can refer to aforementioned side
Corresponding process in method embodiment, details are not described herein.
Flow chart and structural block diagram in above-mentioned attached drawing show multiple embodiments according to the present invention method, apparatus and
The architecture, function and operation in the cards of computer program product.In this regard, each side in flowchart or block diagram
Frame can represent a part of a module, section or code, and a part of the module, section or code includes one
Or multiple executable instructions for implementing the specified logical function.It should also be noted that in some implementations as replacements, side
The function of being marked in frame can also occur in a different order than that indicated in the drawings.For example, two continuous boxes are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yiyong
The dedicated hardware based system of defined function or movement is executed to realize, or can be referred to specialized hardware and computer
The combination of order is realized.
In several embodiments provided herein, it should be understood that disclosed method and apparatus, it can be by other
Mode realize.The apparatus embodiments described above are merely exemplary, for example, the division of the unit, only one
Kind of logical function partition, there may be another division manner in actual implementation, in another example, multiple units or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some communication interfaces, the INDIRECT COUPLING or logical of device or unit
Letter connection can be electrical property, mechanical or other forms.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-OnlyMemory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
Above embodiments, only a specific embodiment of the invention, to illustrate technical solution of the present invention, rather than to it
Limitation, scope of protection of the present invention is not limited thereto, although the present invention is described in detail referring to the foregoing embodiments,
It should be understood by those skilled in the art that: anyone skilled in the art in the technical scope disclosed by the present invention,
It still can modify to technical solution documented by previous embodiment or can readily occur in variation, or to part
Technical characteristic is equivalently replaced；And these modifications, variation or replacement, it does not separate the essence of the corresponding technical solution this hair
The spirit and scope of bright embodiment technical solution, should be covered by the protection scope of the present invention.Therefore, protection of the invention
Range should be subject to the protection scope in claims.
Claims (10)
1. a kind of pavement behavior detection method, which is characterized in that the method is applied on mobile unit, and the mobile unit is matched
It is equipped with vehicle-mounted camera and positioning device, which comprises
By the vehicle-mounted camera and the positioning device, obtain respectively vehicle pavement image through road surface and the road surface
The corresponding geographical location of image；
Image denoising processing and image sharpening processing are carried out to the pavement image, the pavement image after being optimized；
The pavement behavior identification model of pavement image input pre-training after optimization is subjected to corrupted pavement markers, wherein described
Corrupted pavement markers include types of damage label and failure area label；
The extraction of damage parameter is carried out to the pavement image with the corrupted pavement markers；Wherein, the damage parameter is at least
Including following one: impaired area, damage shape, damage color and damage texture；
By with the pavement images of the corrupted pavement markers, the corresponding geographical location of the pavement image and damage parameter summarize to
In pavement behavior testing result.
2. the method according to claim 1, wherein by the road surface shape of the pavement image input pre-training after optimization
Condition identification model carries out the step of corrupted pavement markers, comprising:
Image segmentation is carried out to the pavement image after the optimization, generates multiple road surface block of pixels；Wherein, the road surface block of pixels
It is marked with station location marker, the station location marker is position of the road surface block of pixels in the pavement image after the optimization；
The pavement behavior identification model of road surface block of pixels input pre-training is subjected to corrupted pavement markers；
According to the station location marker of the road surface block of pixels with the corrupted pavement markers, by the corrupted road of the road surface block of pixels
Face label load is in the corresponding position of the station location marker.
3. according to the method described in claim 2, it is characterized in that, the training process of the pavement behavior identification model, comprising:
Obtain road surface sample image；Wherein, the road surface sample image is marked with corrupted pavement markers sample；Wherein, described bad
Damaging pavement markers sample includes types of damage sample and failure area sample；
According to the training algorithm of setting, using the road surface sample image and the corrupted pavement markers sample training introductory die
Type obtains trained pavement behavior identification model.
4. the method according to claim 1, wherein being carried out to the pavement image with the corrupted pavement markers
The step of damaging the extraction of parameter, comprising:
Extract the failure area in the pavement image with the corrupted pavement markers；
Calculate the area coverage that the failure area accounts for the pavement image with the corrupted pavement markers；
According to the camera calibration parameter of the vehicle-mounted camera, to the pavement image with the corrupted pavement markers into
Row three-dimensional reconstruction obtains the actual size of the pavement image；
Using the area coverage and the actual size, the impaired area of the failure area is calculated.
5. the method according to claim 1, wherein being carried out to the pavement image with the corrupted pavement markers
The step of damaging the extraction of parameter, comprising:
Extract the pavement image for having the corrupted pavement markers；
The damage parameter of the pavement image with the corrupted pavement markers is extracted using the damage parameter model of pre-training.
6. a kind of pavement behavior processing method, which is characterized in that the method is applied on Cloud Server；The described method includes:
Obtain the pavement behavior testing result that mobile unit generates；The pavement behavior testing result is that the mobile unit is based on
Any one of Claims 1 to 5 the method obtains；
Store the pavement behavior testing result；
According to the corresponding geographical location of pavement image for having corrupted pavement markers in the pavement behavior testing result, by the road
The corrupted pavement markers of face image, which are embodied, to be located on map；
Determine the maintenance status information on road surface according to the pavement behavior testing result, the maintenance status information include with down toward
It is one of few: service life, maintenance cost and maintenance frequency.
7. a kind of pavement behavior detection device, which is characterized in that the pavement behavior detection device is applied on mobile unit, institute
Mobile unit is stated configured with vehicle-mounted camera and positioning device, the pavement behavior detection device includes:
Obtain module, for by the vehicle-mounted camera and the positioning device, respectively acquisition vehicle the road surface through road surface
Image and the corresponding geographical location of the pavement image；
Image processing module, for carrying out image denoising processing and image sharpening processing to the pavement image, after obtaining optimization
Pavement image；
Mark module, the pavement behavior identification model for the pavement image input pre-training after optimizing carry out corrupted road marking
Note, wherein the corrupted pavement markers include types of damage label and failure area label；
Extraction module, for carrying out the extraction of damage parameter to the pavement image with the corrupted pavement markers；Wherein, described
It damages parameter and includes at least following one: impaired area, damage shape, damage color and damage texture；
Summarizing module, for will have the pavement images of the corrupted pavement markers, the corresponding geographical location of the pavement image and
Damage parameter summarizes into pavement behavior testing result.
8. a kind of pavement behavior processing unit, which is characterized in that described device is applied on Cloud Server；Described device includes:
Object module is obtained, for obtaining the pavement behavior testing result of mobile unit generation；The pavement behavior testing result
The mobile unit is obtained based on any one of Claims 1 to 5 the method；
Memory module, for storing the pavement behavior testing result；
Locating module, for having the pavement image of corrupted pavement markers correspondingly according in the pavement behavior testing result
Position is managed, the corrupted pavement markers of the pavement image are embodied and are located on map；
Determine maintenance block of state, it is described for determining the maintenance status information on road surface according to the pavement behavior testing result
Safeguard that status information includes at least one of: service life, maintenance cost and maintenance frequency.
9. a kind of pavement behavior detection system, which is characterized in that described vehicle-mounted the system comprises mobile unit and Cloud Server
Device configuration has vehicle-mounted camera and positioning device；The mobile unit is installed on vehicle；
The vehicle-mounted camera for shoot the vehicle the pavement image through road surface, and the pavement image is sent to institute
State mobile unit；
The geographical location is sent to described by the positioning device for positioning the corresponding geographical location of the pavement image
Mobile unit；
The mobile unit includes pavement behavior detection device as claimed in claim 7；
The Cloud Server includes pavement behavior processing unit according to any one of claims 8.
10. a kind of computer storage medium, which is characterized in that be stored with computer program, institute in the computer storage medium
State the step executed when computer program is run by processor such as pavement behavior detection method described in any one of claim 1 to 5
Suddenly, alternatively, the step of executing pavement behavior processing method as claimed in claim 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910070867.7A CN109801282A (en) | 2019-01-24 | 2019-01-24 | Pavement behavior detection method, processing method, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910070867.7A CN109801282A (en) | 2019-01-24 | 2019-01-24 | Pavement behavior detection method, processing method, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109801282A true CN109801282A (en) | 2019-05-24 |
Family
ID=66558816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910070867.7A Pending CN109801282A (en) | 2019-01-24 | 2019-01-24 | Pavement behavior detection method, processing method, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109801282A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110390666A (en) * | 2019-06-14 | 2019-10-29 | 平安科技（深圳）有限公司 | Road damage detecting method, device, computer equipment and storage medium |
| CN110490839A (en) * | 2019-07-11 | 2019-11-22 | 平安科技（深圳）有限公司 | The method, apparatus and computer equipment of failure area in a kind of detection highway |
| CN110660141A (en) * | 2019-09-06 | 2020-01-07 | 杭州博信智联科技有限公司 | Road surface condition detection method and device, electronic equipment and readable storage medium |
| CN111768373A (en) * | 2020-06-18 | 2020-10-13 | 北京交通大学 | Hierarchical Pavement Marking Damage Detection Method Based on Deep Learning |
| CN111783686A (en) * | 2020-07-03 | 2020-10-16 | 中国交通通信信息中心 | Asphalt pavement health state monitoring system and method |
| CN112241808A (en) * | 2020-09-28 | 2021-01-19 | 四川省交通运输发展战略和规划科学研究院 | Road surface technical condition prediction method, device, electronic equipment and storage medium |
| CN112505142A (en) * | 2020-10-27 | 2021-03-16 | 北京建筑大学 | Method for detecting damage of road structure, autonomous mobile device and storage medium |
| CN112711999A (en) * | 2020-12-24 | 2021-04-27 | 西人马帝言(北京)科技有限公司 | Image recognition method, device, equipment and computer storage medium |
| CN112800911A (en) * | 2021-01-20 | 2021-05-14 | 同济大学 | Pavement damage rapid detection and natural data set construction method |
| CN113129449A (en) * | 2021-04-16 | 2021-07-16 | 浙江孔辉汽车科技有限公司 | Vehicle pavement feature recognition and three-dimensional reconstruction method based on binocular vision |
| CN113191271A (en) * | 2021-04-30 | 2021-07-30 | 水利部长江勘测技术研究所 | Method for detecting surface particulate matter of weir dam based on deep learning |
| CN113516127A (en) * | 2021-09-14 | 2021-10-19 | 南通辑兴紧固件科技有限公司 | Artificial intelligence-based smart city highway maintenance method and system |
| CN113724503A (en) * | 2021-08-31 | 2021-11-30 | 山东交通学院 | Automatic highway state inspection system and method based on cloud platform |
| WO2022007120A1 (en) * | 2020-07-08 | 2022-01-13 | 谢超奇 | Road patching quantitative positioning data sending platform and method |
| CN114037963A (en) * | 2021-11-26 | 2022-02-11 | 中关村科学城城市大脑股份有限公司 | Road sign state monitoring method and device, storage medium and equipment |
| CN114240866A (en) * | 2021-12-09 | 2022-03-25 | 广东省农业科学院环境园艺研究所 | Tissue culture seedling grading method and device based on two-dimensional image and three-dimensional growth information |
| CN114529722A (en) * | 2022-02-09 | 2022-05-24 | 应急管理部国家自然灾害防治研究院 | Color segmentation and object-oriented seismic surface fracture extraction method |
| CN114592411A (en) * | 2022-03-29 | 2022-06-07 | 中国矿业大学 | An intelligent inspection method for carrier parasitic road damage |
| CN115184378A (en) * | 2022-09-15 | 2022-10-14 | 北京思莫特科技有限公司 | Concrete structure disease detection system and method based on mobile device |
| CN115512322A (en) * | 2022-09-28 | 2022-12-23 | 同济大学 | A method for processing multi-dimensional pavement damage data |
| CN115908427A (en) * | 2023-02-27 | 2023-04-04 | 长沙城市发展集团有限公司 | Method and system for pavement disease maintenance cost prediction based on semantic segmentation and SVM |
| CN115984221A (en) * | 2023-01-03 | 2023-04-18 | 广州新粤交通技术有限公司 | Road marking repairing and identifying method, device, equipment and storage medium thereof |
| CN116913077A (en) * | 2023-06-06 | 2023-10-20 | 速度科技股份有限公司 | Expressway emergency management system based on data acquisition |
| CN117495829A (en) * | 2023-11-15 | 2024-02-02 | 广东昭明电子集团股份有限公司 | A quality inspection method for smart watch hardware |
| CN117893988A (en) * | 2024-01-19 | 2024-04-16 | 元橡科技（北京）有限公司 | All-terrain scene pavement recognition method and training method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN204825607U (en) * | 2015-08-25 | 2015-12-02 | 张苇 | Pavement condition detection device |
| CN106845558A (en) * | 2017-02-27 | 2017-06-13 | 广东诚泰交通科技发展有限公司 | The method and apparatus of pavement detection |
| WO2018122818A1 (en) * | 2016-12-30 | 2018-07-05 | 同济大学 | Method for detecting degree of development of asphalt pavement fracture based on infrared thermal image analysis |
| CN108417035A (en) * | 2018-03-29 | 2018-08-17 | 成都精灵云科技有限公司 | Intelligent traffic monitoring system based on cloud platform |
-
2019
- 2019-01-24 CN CN201910070867.7A patent/CN109801282A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN204825607U (en) * | 2015-08-25 | 2015-12-02 | 张苇 | Pavement condition detection device |
| WO2018122818A1 (en) * | 2016-12-30 | 2018-07-05 | 同济大学 | Method for detecting degree of development of asphalt pavement fracture based on infrared thermal image analysis |
| CN106845558A (en) * | 2017-02-27 | 2017-06-13 | 广东诚泰交通科技发展有限公司 | The method and apparatus of pavement detection |
| CN108417035A (en) * | 2018-03-29 | 2018-08-17 | 成都精灵云科技有限公司 | Intelligent traffic monitoring system based on cloud platform |
Non-Patent Citations (2)
| Title |
|---|
| 曹佳煜: "基于图像处理的路面裂缝自动检测技术研究", 《中国硕士学位论文全文数据库信息科技辑》 * |
| 赵芬芬: "基于计算机视觉的路面破损检测与识别的研究", 《中国硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110390666A (en) * | 2019-06-14 | 2019-10-29 | 平安科技（深圳）有限公司 | Road damage detecting method, device, computer equipment and storage medium |
| CN110390666B (en) * | 2019-06-14 | 2023-06-27 | 平安科技（深圳）有限公司 | Road damage detection method, device, computer equipment and storage medium |
| WO2020248371A1 (en) * | 2019-06-14 | 2020-12-17 | 平安科技（深圳）有限公司 | Road damage detection method and apparatus, computer device, and storage medium |
| CN110490839A (en) * | 2019-07-11 | 2019-11-22 | 平安科技（深圳）有限公司 | The method, apparatus and computer equipment of failure area in a kind of detection highway |
| CN110490839B (en) * | 2019-07-11 | 2024-02-27 | 平安科技（深圳）有限公司 | Method and device for detecting damaged area in expressway and computer equipment |
| CN110660141A (en) * | 2019-09-06 | 2020-01-07 | 杭州博信智联科技有限公司 | Road surface condition detection method and device, electronic equipment and readable storage medium |
| CN111768373A (en) * | 2020-06-18 | 2020-10-13 | 北京交通大学 | Hierarchical Pavement Marking Damage Detection Method Based on Deep Learning |
| CN111768373B (en) * | 2020-06-18 | 2024-08-23 | 北京交通大学 | Hierarchical pavement marking damage detection method based on deep learning |
| CN111783686A (en) * | 2020-07-03 | 2020-10-16 | 中国交通通信信息中心 | Asphalt pavement health state monitoring system and method |
| WO2022007120A1 (en) * | 2020-07-08 | 2022-01-13 | 谢超奇 | Road patching quantitative positioning data sending platform and method |
| CN112241808A (en) * | 2020-09-28 | 2021-01-19 | 四川省交通运输发展战略和规划科学研究院 | Road surface technical condition prediction method, device, electronic equipment and storage medium |
| CN112505142A (en) * | 2020-10-27 | 2021-03-16 | 北京建筑大学 | Method for detecting damage of road structure, autonomous mobile device and storage medium |
| CN112505142B (en) * | 2020-10-27 | 2023-01-24 | 北京建筑大学 | Method, autonomous mobile device, and storage medium for detecting damage to road structures |
| CN112711999A (en) * | 2020-12-24 | 2021-04-27 | 西人马帝言(北京)科技有限公司 | Image recognition method, device, equipment and computer storage medium |
| CN112800911A (en) * | 2021-01-20 | 2021-05-14 | 同济大学 | Pavement damage rapid detection and natural data set construction method |
| CN112800911B (en) * | 2021-01-20 | 2022-12-16 | 同济大学 | A method for rapid pavement damage detection and natural data set construction |
| CN113129449A (en) * | 2021-04-16 | 2021-07-16 | 浙江孔辉汽车科技有限公司 | Vehicle pavement feature recognition and three-dimensional reconstruction method based on binocular vision |
| CN113191271A (en) * | 2021-04-30 | 2021-07-30 | 水利部长江勘测技术研究所 | Method for detecting surface particulate matter of weir dam based on deep learning |
| CN113724503A (en) * | 2021-08-31 | 2021-11-30 | 山东交通学院 | Automatic highway state inspection system and method based on cloud platform |
| CN113516127B (en) * | 2021-09-14 | 2021-12-07 | 南通辑兴紧固件科技有限公司 | Artificial intelligence-based smart city highway maintenance method and system |
| CN113516127A (en) * | 2021-09-14 | 2021-10-19 | 南通辑兴紧固件科技有限公司 | Artificial intelligence-based smart city highway maintenance method and system |
| CN114037963A (en) * | 2021-11-26 | 2022-02-11 | 中关村科学城城市大脑股份有限公司 | Road sign state monitoring method and device, storage medium and equipment |
| CN114240866A (en) * | 2021-12-09 | 2022-03-25 | 广东省农业科学院环境园艺研究所 | Tissue culture seedling grading method and device based on two-dimensional image and three-dimensional growth information |
| CN114240866B (en) * | 2021-12-09 | 2022-07-08 | 广东省农业科学院环境园艺研究所 | Tissue culture seedling grading method and device based on two-dimensional image and three-dimensional growth information |
| CN114529722A (en) * | 2022-02-09 | 2022-05-24 | 应急管理部国家自然灾害防治研究院 | Color segmentation and object-oriented seismic surface fracture extraction method |
| CN114592411A (en) * | 2022-03-29 | 2022-06-07 | 中国矿业大学 | An intelligent inspection method for carrier parasitic road damage |
| CN115184378A (en) * | 2022-09-15 | 2022-10-14 | 北京思莫特科技有限公司 | Concrete structure disease detection system and method based on mobile device |
| CN115184378B (en) * | 2022-09-15 | 2024-03-29 | 北京思莫特科技有限公司 | Concrete structure disease detection system and method based on mobile equipment |
| CN115512322A (en) * | 2022-09-28 | 2022-12-23 | 同济大学 | A method for processing multi-dimensional pavement damage data |
| CN115984221A (en) * | 2023-01-03 | 2023-04-18 | 广州新粤交通技术有限公司 | Road marking repairing and identifying method, device, equipment and storage medium thereof |
| CN115984221B (en) * | 2023-01-03 | 2023-08-04 | 广州新粤交通技术有限公司 | Road marking restoration and identification method, device, equipment and storage medium thereof |
| CN115908427A (en) * | 2023-02-27 | 2023-04-04 | 长沙城市发展集团有限公司 | Method and system for pavement disease maintenance cost prediction based on semantic segmentation and SVM |
| CN116913077B (en) * | 2023-06-06 | 2024-03-29 | 速度科技股份有限公司 | Expressway emergency management system based on data acquisition |
| CN116913077A (en) * | 2023-06-06 | 2023-10-20 | 速度科技股份有限公司 | Expressway emergency management system based on data acquisition |
| CN117495829A (en) * | 2023-11-15 | 2024-02-02 | 广东昭明电子集团股份有限公司 | A quality inspection method for smart watch hardware |
| CN117495829B (en) * | 2023-11-15 | 2024-04-30 | 广东昭明电子集团股份有限公司 | A method for quality inspection of hardware parts of smart watches |
| CN117893988A (en) * | 2024-01-19 | 2024-04-16 | 元橡科技（北京）有限公司 | All-terrain scene pavement recognition method and training method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109801282A (en) | Pavement behavior detection method, processing method, apparatus and system | |
| CN111179345B (en) | Front vehicle line-crossing violation behavior automatic detection method and system based on vehicle-mounted machine vision | |
| CN107253485B (en) | Foreign matter invades detection method and foreign matter invades detection device | |
| CN114913399B (en) | Vehicle trajectory optimization method and intelligent transportation system | |
| US9070023B2 (en) | System and method of alerting a driver that visual perception of pedestrian may be difficult | |
| CN112329747B (en) | Vehicle parameter detection method based on video identification and deep learning and related device | |
| CN105260699B (en) | A kind of processing method and processing device of lane line data | |
| CN113724259B (en) | Manhole cover abnormality detection method, device and application | |
| US20140028842A1 (en) | Calibration device and method for use in a surveillance system for event detection | |
| CN102867417B (en) | Taxi anti-forgery system and taxi anti-forgery method | |
| CN107527038A (en) | A kind of three-dimensional atural object automatically extracts and scene reconstruction method | |
| CN114972177B (en) | Road hazard identification and management method, device and intelligent terminal | |
| CN106781680B (en) | A kind of curb parking intelligent control method based on the detection of image empty parking space | |
| CN119478858B (en) | A radar and video information fusion coding method for vehicle overload warning | |
| CN120495603B (en) | Visual recognition method and system | |
| WO2022104798A1 (en) | 5g-based unmanned electronic traffic police duty system | |
| CN112017243A (en) | A medium visibility identification method | |
| KR101874968B1 (en) | Visibility measuring system base on image information and method for using the same | |
| CN106203302B (en) | The pedestrian detection and statistical method that view-based access control model and wireless aware combine | |
| CN110969135A (en) | Vehicle logo recognition method in natural scene | |
| WO2021138372A1 (en) | Feature coverage analysis | |
| CN112016558A (en) | A medium visibility recognition method based on image quality | |
| CN106339683A (en) | Road test traffic light identification method and device | |
| CN202887450U (en) | Taxi anti-fake system | |
| CN118366107A (en) | Irregular parking identification method on roads |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190524 |
|
| RJ01 | Rejection of invention patent application after publication |