CN1237327C - System and method for discriminating road gap - Google Patents
System and method for discriminating road gap Download PDFInfo
- Publication number
- CN1237327C CN1237327C CN 200410033964 CN200410033964A CN1237327C CN 1237327 C CN1237327 C CN 1237327C CN 200410033964 CN200410033964 CN 200410033964 CN 200410033964 A CN200410033964 A CN 200410033964A CN 1237327 C CN1237327 C CN 1237327C
- Authority
- CN
- China
- Prior art keywords
- crack
- image
- module
- point
- gray
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The present invention discloses a system and a method for discriminating a road gap. The data of the road gap is obtained and stored by utilizing digital imaging technique, through digital image processing and pattern discriminating techniques, the image of the road gap is segmented, the gap is extracted, the characteristic value of the gap is calculated, the gap is positioned and classified, and the extent of the damage of the gap in the image is evaluated according to the evaluation criterion of the gap. The system and the method for discriminating a road gap can fully automatically detect and discriminate the road gap and can also analyze and process the road gap in real time. Therefore, the system and the method for discriminating a road gap overcome the disadvantages of large labor intensity, low safety, traveling interference, low work efficiency and low discriminating result precision of an artificial discriminating method and simultaneously greatly enhance the work efficiency of detection.
Description
Technical field
The present invention relates to the pavement quality detection technique, particularly relate to a kind of system and method for discerning pavement crack.
Background technology
In recent years, along with the continuous enhancing of China's economic strength, highway construction has obtained develop rapidly in China.Highway communication has become China's pillar of economy, and highway communication also becomes a kind of mode of transportation that people generally rely on.And good highway communication be with good pavement quality be prerequisite and the basis.The quality of pavement quality directly has influence on daily life, even has influence on entire economy development.Therefore, guarantee that pavement quality is a major issue that is related to people even whole country vital interests.
Yet, owing to the destruction of weather extremes and use reasons such as frequent or highway is aging to cause the crack occurring on the road surface of highway.Pavement crack gently then influences road surface sight, and is heavy then influence serviceable life of traffic safety and highway, havoc the quality on road surface, influenced people's normal life.
If can in time find and discern pavement crack, and take corresponding repairing measure, then can eliminate safe hidden trouble, improve pavement quality.
In the prior art, the method for identification pavement crack has:
One, manual detection and crack identification.
The testing staff detects all road surfaces to the scene, road surface, if find pavement crack, then position, length, width and the area in crack is measured and record.Then, the testing staff adds up, sorts out and file the data of record, and according to the pavement crack evaluation criterion crack of record is estimated.
Two, utilize laser to gather the road surface data.
This method is detected by the inspection vehicle road pavement.Inspection vehicle mainly is made up of carrying vehicle and laser scanner.When inspection vehicle travelled that road pavement detects on highway, the laser scanner that is installed in the carrying vehicle both sides sent the laser scanning road surface, and optical receiver receives the light of returning from road reflection at a certain angle then.When the crack occurring on the road surface that is scanned, because the crack can be with laser light scattering or the refraction that arrives, so the catoptrical intensity that optical receiver receives will reduce.Therefore, the testing staff can determine whether the road surface in the laser institute scanning area exists the crack according to the variation of the intensity of reflected light of optical receiver output.Then, the testing staff adds up, sorts out and file the pavement crack of determining, and according to the pavement crack evaluation criterion crack of record is estimated.
Three, utilize line-scan camera to gather the road surface data.
This method is detected by the inspection vehicle road pavement.Inspection vehicle mainly is made up of carrying vehicle and line-scan camera.When inspection vehicle travelled that road pavement detects on highway, the line-scan camera road pavement that is installed on the carrying vehicle was carried out linear sweep and image taking.The testing staff can determine whether the road surface in the captured zone of line-scan camera exists the crack according to picture shot.Then, the testing staff adds up, sorts out and file the pavement crack of determining, and according to the pavement crack evaluation criterion crack of record is estimated.
Four, discern based on the pavement crack of camera technique.
Along with the development of video technique, the pavement crack recognition system based on camera technique has appearred at present.This system is made up of carrying vehicle, video camera, stadimeter, computing machine, flashlamp, GPS (GPS) etc.When carrying vehicle travels that road pavement detects on highway, utilize this system to realize that the detailed process of identification pavement crack may further comprise the steps:
1, stadimeter produces trigger pip, and this trigger pip is sent to video camera.
2, after video camera received trigger pip, road pavement was taken pictures, and the flashlamp of simultaneous camera is finished the light filling effect, and uniform illumination is provided, and was beneficial to take, and video camera sends to computing machine with captured image then.
3, after computing machine receives pavement image, image is carried out pre-service, comprise pavement image gray processing, figure image intensifying and image segmentation, only contained the image in road surface background and crack.Image after will handling then is kept in the computing machine.
4, the image after the testing staff handles each width of cloth of preserving in the computing machine carries out crack classification, location, crack and crack mark one by one and calculates the FRACTURE CHARACTERISTICS value;
5, the testing staff estimates the crack in every width of cloth image according to the pavement crack evaluation criterion;
6, GPS provides the positional information of all road surface testing locations in whole road network according to the data of its record to the testing staff.
As seen, in the prior art,
The shortcoming of method one is: because whole process is artificial treatment, so there is very big subjective factor aspect measuring method and reading of data, the error of the feasible pavement crack data that obtain is bigger.Simultaneously, manual detection wastes time and energy, and efficient is very low.And when the road surface of highway was detected, because vehicle flowrate is bigger on the highway, the travel speed of vehicle was higher, thereby can threaten testing staff's personal safety, has potential safety hazard.
The shortcoming of method two is: detect because the crack is the intensity of reflected light of the laser that receives by optical receiver, therefore, in order not to be subjected to the influence of daylight, can only carry out the detection of pavement crack at night.For the testing staff makes troubles.And pavement crack mark, crack statistic of classification processing and fracture evaluation are artificial treatment, therefore waste time and energy, and efficient is extremely low.Because detection speed is slower, therefore limited the travel speed of inspection vehicle again.
The shortcoming of method three is: in order to obtain image under the normal speed of a motor vehicle, line-scan camera needs the illumination condition of high brightness, but then can make the road surface be subjected to serious destruction from the high-strength light that inspection vehicle sends when the direct projection time is longer on bituminous pavement the same area.
The shortcoming of method four is: though this method can utilize video camera to obtain image, utilize computing machine to finish pre-service to image, thereby realized the robotization processing to a certain extent, but still need manually one by one every width of cloth image to be carried out crack mark, crack statistic of classification processing, and the crack of record is estimated, thereby increased testing staff's workload greatly according to the pavement crack evaluation criterion.And manual detection can make result omission and flase drop occur, and the lower situation of accuracy.Because manual procedure is slower, after obtaining crack information, can not discern and estimate the crack immediately, and can only be after having detected all road surfaces, could be by manually carrying out follow-up processing procedure, so can't realize the real-time processing of road pavement crack full-automation.
Summary of the invention
In view of this, fundamental purpose of the present invention is to provide a kind of system and method for discerning pavement crack, thereby realizes handling with full-automatic in real time of road pavement crack identification.
In order to achieve the above object, technical scheme of the present invention is achieved in that
A kind of system that discerns pavement crack, comprise in the Load System other module and the carrying vehicle of stablizing travel speed is provided, the supply module of other each module output stabilized power source in system, range finder module and image capture module, wherein, range finder module calculates the distance that carrying vehicle travels in real time, and when distance element length that the carrying vehicle operating range that calculates equals to set, export apart from trigger pip to image capture module, image capture module receive that range finder module sends apart from trigger pip after, road pavement is carried out image taking, and export captured image, this system also comprises with lower module: the image pretreatment module, crack identification module and fracture evaluation module, wherein
The image pretreatment module, receive the image of image capture module output, and be gray-scale map with received image transitions, and utilize edge detection operator that gray-scale map is carried out the figure image intensifying, and isolate crack point and road surface background dot the image after strengthening, the binary map that is only contained the crack outputs to binary map the crack identification module then;
The crack identification module, receive the binary map that the image pretreatment module is sent, utilize shadow casting technique that binary map is carried out projection, according to the position of perspective view Primary Location crack in image, and tentatively longitudinal crack and transverse crack are divided into a class crack, will chap and piece splits and is divided into a class crack;
The crack identification module utilizes wavelet technique that the zone that Primary Location contains the crack is strengthened, and reuse shadow casting technique and accurately orient residing position in image, crack, and precise classification is carried out in the crack, determine transverse crack and longitudinal crack, and be full of cracks and piece split;
The crack identification module is calculated the eigenwert in crack, statistics crack quantity, and the crack information after will handling outputs to the fracture evaluation module;
The fracture evaluation module is estimated the crack according to the fracture evaluation standard.
The wheel revolutions of described range finder module recording carrying car, and according to the record wheel revolutions calculate the distance that carrying vehicle travels in real time.
Described range finder module further comprises knotmeter, measures carrying vehicle actual travel speed, and according to the physical location of actual travel velocity correction crack in road network.
Described knotmeter comprises radar meter or fifth wheel instrument.
The distance element length of setting in the described range finder module is the path length that a two field picture that described image capture module is shot is covered.
Described range finder module utilizes global position system GPS to determine the position of crack in road network.
Described fracture evaluation module is further used for the positional information of crack in road network deposited in the road net data storehouse.
Described image pretreatment module, crack identification module and fracture evaluation module are integrated on the functional entity of digital signal processor DSP.
A kind of method of discerning pavement crack may further comprise the steps:
The distance element length whether the carrying vehicle operating range that A, real-time judge calculate equals to set, if, execution in step B then, otherwise, steps A returned;
B, send apart from trigger pip, the road pavement image is taken, and is digital picture with the image transitions of taking;
C, digital picture is converted to gray-scale map, and the gray-scale map after using edge detection operator to conversion carries out the figure image intensifying, use binarization method that image is cut apart, only contained the binary map in crack;
D, using shadow casting technique that binary map is carried out projection, according to the position of perspective view Primary Location crack in image, and is longitudinal crack or transverse crack with the crack preliminary classification, and be full of cracks or piece split two class cracks;
E, use wavelet technique with in the gray-scale map of gained among the step C the Primary Location zone of containing the crack strengthen;
F, use edge detection operator carry out edge extracting to the image after strengthening, and image is carried out binary conversion treatment, and accurately orient the position of crack in image;
G, precise classification go out longitudinal crack and transverse crack, and be full of cracks and piece split;
H, calculate the eigenwert in each crack, the quantity in statistics crack, and the crack is carried out the evaluation of the order of severity according to FRACTURE CHARACTERISTICS value and quantity.
In steps A, the distance element length of described setting is the path length that a two field picture is covered.
The described gray-scale map of step C is to carry out addition after the value by the red R of each picture element in the image that will receive, green G and blue B multiply by weighted value respectively, and obtaining with value of making that the value of R, the G of this point and B equals to calculate.
The step that the described use binarization method of step C is cut apart image comprises: the gray-scale value of 1.2 times of background gray shade scales is set at threshold value, gray-scale value in the image is set at the crack point greater than the picture element of this threshold value, and this crack point is made as black, gray-scale value in the image is set at background dot less than the picture element of this threshold value, and this background dot is made as white.
The described step according to the position of perspective view Primary Location crack in image of step D comprises: with the direction of scanning of image as transverse axis, with the gray-scale value summation of the image that exists on the transverse axis orthogonal directions as the longitudinal axis, determine a coordinates regional,
Crack area in the binary map is projected in respectively on the transverse axis and the longitudinal axis in this coordinates regional, obtain two one dimension curves, from the transverse axis one dimension curve projection figure about two end points make straight line respectively perpendicular to transverse axis, two end points up and down of one dimension curve projection figure are made the straight line perpendicular to the longitudinal axis respectively from the longitudinal axis, with these 4 rectangular areas that straight line defined as the position of crack in former binary map;
Step D is described to be longitudinal crack or transverse crack with the crack preliminary classification, and the step that be full of cracks or piece split two class cracks comprises: judge that drop shadow curve on the axle is banded and the crack of fewer obvious peak value is arranged is transverse crack and longitudinal crack, judgement is similar at the projection distribution shape of the transverse axis and the longitudinal axis, and the crack that does not have obvious peak value, perhaps projection has more than the crack of 3 obvious peak value and splits for be full of cracks and piece;
The zone that the described Primary Location of step e contains the crack is the rectangular area that step D is defined.
Between described step F and step G, further comprise:
G11, image is expanded and corrosion treatment, fill discontinuous zone, crack, obtain complete crack area;
Binary map after G12, the filling of lining by line scan, when scanning first black picture element point, write down the coordinate figure of this black picture element point, it is labeled as frontier point, again according to the 8 neighborhood picture elements of counterclockwise analyzing this black picture element point, write down the coordinate figure of the black picture element point of white portion and black region intersection in this 8 neighborhood picture element, and it is labeled as frontier point, the rest may be inferred, up to marking all frontier points and coordinate figure thereof;
G13, according to fracture strike trend, all frontier points that mark are linked in sequence, obtain complete crack profile, utilize thinning algorithm then, the crack contour thinning to single pixel wide.
The described precise classification of step G goes out longitudinal crack and transverse crack comprises: two end points of crack profile are linked to be straight line, judge that whether the angle that institute connects straight line and road xsect spend less than 45, if then this crack is judged to be transverse crack, otherwise, this crack is judged to be longitudinal crack.
The step of the described calculating FRACTURE CHARACTERISTICS of step H value comprises: get the central point at two edges, vertical direction crack, and connect each central point of being got, form curve, length of a curve is set at the equivalent length in this crack;
From an end of curve, get a point every a point, until the other end of curve, per three points carry out fitting a straight line, obtain each bar straight line, with the 2 times breadth extremes that are set at this crack of straight line apart from the ultimate range of crack coboundary.
The described precise classification of step G goes out be full of cracks in the multidirectional crack and piece and splits and comprise: set product threshold value and rectangular block number threshold value, and two 5 * 5 templates are set, " ten " word bit of first template center is changed to 1, and other position is 0, second template is 1 entirely on diagonal line, other position is 0 entirely, the gray-scale map of gained among described two templates and the step C is done product respectively, and get the gained maximum product, judge that whether this maximal value is greater than the product threshold value of setting, if then think to have approximate rectangle crack, otherwise think and be not the rectangle crack, the rest may be inferred, the number of the rectangular block that exists in counting entire image then, judges that whether the rectangular block number that counts is greater than the rectangular block number threshold value of setting, if, then this crack is judged to be piece and splits, otherwise, this crack is judged to be full of cracks.
The step of the described calculating FRACTURE CHARACTERISTICS of step H value comprises: gets the solstics of profile four direction in crack on the image, these four points is linked to be boundary rectangle, and with the area of this boundary rectangle area as this crack.
This method further comprises: when detecting pavement crack, utilize the position of global position system GPS technological orientation crack in road network, and the positional information of crack in road network of determining deposited in the road net data storehouse.
This method further comprises: use radar meter to measure carrying vehicle actual travel speed, and according to the position of actual travel velocity correction crack in the highway section.
As seen, the system and method for the present invention's proposition has the following advantages:
1, the system and method for the present invention's proposition, realized full-automatic detection and Identification pavement crack, entire identification process does not need artificial participation fully, thereby has overcome that the labour intensity that the artificial cognition method has is big, security is low, driving is disturbed, inefficiency and recognition result degree of accuracy lower shortcoming.
2, the system and method for the present invention's proposition can be handled the pavement image that collects in real time, thereby improve testing efficient greatly.
3, the system and method that proposes of the present invention only needs inspection vehicle to travel in the highway section of required detection, can finish the identification and the evaluation in road pavement crack, and therefore, whole testing process is simple and be easy to realize.
4, the system and method for the present invention's proposition, but road pavement is carried out long term monitoring, in time finds pavement crack, and in time the crack of estimating out is safeguarded, thereby improved the travel safety of highway greatly, prolonged the serviceable life of highway, and saved the maintenance of surface expense.
5, the system and method for the present invention's proposition, the maintenance management that can be the road surface provides effectively information support, improve highway maintenance and management level, simultaneously, for further developing the highway checkout equipment, change the present situation of highway engineering in China checkout equipment overwhelming majority dependence on import, economize on resources, the research and development technical force of cultivating oneself has laid manpower and technical foundation.
6, the system and method for the present invention's proposition can provide certain data basis for the intelligent transportation system of China, helps to improve China's transport information technical merit.
Description of drawings
Fig. 1 is the structural representation of system of the present invention.
Fig. 2 is a process flow diagram of realizing embodiments of the invention.
Fig. 3 utilizes shadow casting technique the crack to be carried out the synoptic diagram of Primary Location in realizing embodiments of the invention.
Fig. 4 is the synoptic diagram that the present invention extracts the crack profile.
Fig. 5 calculates in realizing embodiments of the invention laterally or the breadth extreme of longitudinal crack and the synoptic diagram of equivalent length.
Fig. 6 is the synoptic diagram of two 5 * 5 templates using in realizing embodiments of the invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with drawings and the specific embodiments.
The invention provides a kind of road pavement crack in real time and carry out the fully automatic system and the specific implementation method of Flame Image Process and evaluation.
Fig. 1 is the structural representation of system of the present invention.As shown in Figure 1, in the present invention, design system to the vehicular system, the system that promptly discerns pavement crack comprises carrying vehicle 101, and is installed in supply module 102, range finder module 103, image capture module 104, image pretreatment module 105, crack identification module 106 and fracture evaluation module 107 on the carrying vehicle.
Fig. 2 is a process flow diagram of realizing embodiments of the invention.As depicted in figs. 1 and 2, on the basis of system of the present invention, present embodiment realizes that the detailed process of identification pavement crack may further comprise the steps:
Step 201: detection person opens carrying vehicle 101, and uses supply module 102 other each module for power supply in system.
Here, use carrying vehicle 101,, make system have good stable so that stable travel speed to be provided with good resistance shock stability.Because system is an image data processing under the state of motion, so require system that stability preferably will be arranged, the stability of carrying vehicle 101 is good more, and the data message that system obtains is also just reliable more.Such as, when carrying vehicle 101 travels on highway, even Uneven road is smooth or have barriers such as finger stone, because carrying vehicle 101 has good seismic resistance, then still can provide comparatively stable travel speed, then very little to the shake of image capture module 104 generations so, thus the pavement image quality that system obtains guaranteed.
In the present embodiment, supply module 102 uses diesel-driven generators, by diesel-driven generator to the employed power devices of other module of system, because diesel-driven generator can effectively solve plant capacity consumption, and reduce to disturb, therefore, can the assurance system in the operate as normal of other each module.
Step 202: when the path length that is travelled when the carrying vehicle of measuring in the range finder module 103 101 equals the predefined distance element length of system, send apart from trigger pip to image capture module 104.
Wheel revolutions when here, range finder module 103 travels according to the carrying vehicle 101 of its record is calculated the distance that carrying vehicle 101 travels.Carrying vehicle operating range=wheel revolutions * wheel circumference.The distance element length that system sets in advance then is to determine according to the path length that every width of cloth image that image capture module 104 collects can cover.
Step 203: image capture module 104 receive that range finder module 103 sends apart from trigger pip after, road pavement is carried out image taking, and the coloured image of shooting is sent to image pretreatment module 105.
Here, image capture module 104 comprises high-speed charge coupled device (CCD) video camera and servicing lighting.Wherein, the high-speed CCD video camera has higher resolution, and the shutter of very high speed, whenever receive that range finder module 103 sends apart from trigger pip after, high-speed CCD video camera then road pavement carries out image taking, and 24 coloured images will shooting then output to image pretreatment module 105.Servicing lighting provides sufficient and uniform illumination condition for the high-speed CCD video camera, eliminates the road surface shade, guarantees the picture quality that the high-speed CCD video camera is taken.
Step 204: 105 pairs of coloured images that receive of image pretreatment module carry out the image pre-service, comprise image gray processing, figure image intensifying and image segmentation, and the image after will handling then outputs to crack identification module 106.
Here, image pretreatment module 105 inside can comprise a video frequency collection card, because the image that the high-speed CCD video camera is shot is a vision signal, therefore, for making system of the present invention carry out subsequent treatment to image, image pretreatment module 105 uses inner video frequency collection card that vision signal is converted to digital signal when receiving the vision signal that the high-speed CCD video camera sends, thereby makes system of the present invention carry out the image pre-service to the digital picture after the conversion.
The processing procedure of image gray processing is: calculate the brightness value Y of each picture element in the color digital image respectively, Y=0.299 * R+0.587 * G+0.114 * B.Wherein, R is the red color value of each point in 24 coloured images, and G is the green chromatic value of each point in 24 coloured images, and B is the chroma blue value of each point in 24 coloured images.Then, the value of the R of each point, G and B is set to the Y value of this point in the image, obtains gray-scale map.Here, coloured image being converted to gray-scale map, is in order to remove the color information in the image, only to keep monochrome information, so that system carries out follow-up processing to image.
Because in the pavement image of shooting, there is more noise, and the background complexity, in order to extract pavement crack effectively, must carry out image enhancement processing to the gray-scale map after the conversion.Therefore, use edge detection operator that gray-scale map is carried out image enhancement processing, strengthen the light and shade contrast of image and restrain noise in the image.
In the present embodiment, adopt binarization method that image is cut apart.Use the processing procedure that binarization method is cut apart image to be: in the image after enhancing, crack area is brighter than background, the ratio maximum that background is shared, and the gray shade scale in crack is greater than background.And for background, its gray shade scale basically identical.Therefore, in order obviously to distinguish crack and the road surface background that strengthens in the image of back, can think that background is in same gray shade scale, the part that is higher than this gray shade scale then can be thought the crack.The gray-scale value of 1.2 times of background gray shade scales as threshold value.Then in entire image, gray-scale value is set at the crack point greater than the picture element of threshold value, and this point is made as black; Gray-scale value is set at background dot less than the picture element of threshold value, and it is made as white.So far, the binary map that is only contained the crack.
Step 205: after crack identification module 106 receives the pretreated image that image pretreatment module 105 sends, Primary Location goes out the position of crack in image, and the crack carried out preliminary classification, distinguish two class cracks, the first kind is transverse crack and longitudinal crack, and second class is that be full of cracks and piece split.
Here, because crack shared ratio in whole road surface is very little, if directly whole pavement image is handled, not only workload is very big, and processing speed can be very slow.Therefore, at first in image, Primary Location and preliminary classification are carried out in the crack.
The binary map that in step 204, obtains, though obviously distinguish the crack, but because the crack presents irregular shape in binary map, crack identification module 106 can't position it, therefore, must determine shared rectangular area in image, crack, so that it is carried out Primary Location.In the present embodiment, use shadow casting technique that Primary Location is carried out in the crack in the image.Fig. 3 utilizes shadow casting technique the crack to be carried out the synoptic diagram of Primary Location in realizing embodiments of the invention.As shown in Figure 3, the specific implementation process that Primary Location is carried out in the crack in the image is:
The direction that the direction of scanning of image is promptly vertical with carrying vehicle 101 travel directions is as transverse axis (X-axis), with the gray-scale value summation of the image that exists on the X-axis orthogonal directions as the longitudinal axis (Y-axis), determine a coordinates regional.Crack identification module 106 is carried out projection with the crack area in the binary map (the determined zone of curve A BC) in this coordinates regional, be projected as curve A on X-axis ' the determined zone of B ' C ', on Y-axis, be projected as curve A ' the determined zone of D '.The B ' of projection point and C ' point on the X-axis are made the straight line of vertical and X-axis, can define crack scope in vertical direction.In like manner, the A ' of projection point and D ' point on the Y-axis are made the straight line of vertical and Y-axis, can define crack scope in the horizontal direction.So far, define a rectangular area A " B " D " C ", thereby determined the approximate location of crack in former binary map, also just determined the position in original image.
The detailed process of the crack being carried out preliminary classification is: judge that the projection distribution is banded and the crack of fewer obvious peak value is arranged is first kind crack, be transverse crack or longitudinal crack, transverse crack be meant with the vertical direction of carrying vehicle 101 travel directions on the crack, longitudinal crack be meant with carrying vehicle 101 travel direction equidirectionals on the crack; To be more or less the same at the projection distribution shape of X-axis and Y-axis, and do not have the crack of obvious peak value or the crack that projection has a plurality of (greater than 3) obvious peak value, be judged to be the second class crack, promptly be full of cracks or piece split.
Step 206: accurately locate and precise classification in the crack in 106 pairs of images of crack identification module, and calculate the eigenwert in crack, counts the quantity in various cracks, and the crack information after will handling then outputs to fracture evaluation module 107.
Here, the gray-scale map that obtains in the obtaining step 204 at first, and in gray-scale map, utilize wavelet analysis technology in step 205 the Primary Location zone (the rectangular area A ' B ' D ' C ' shown in Fig. 3) of containing the crack strengthen.Wavelet analysis is that signal decomposition is become low frequency a1 and high frequency d1 two parts, and in decomposition, the information that loses among the low frequency a1 is caught by high frequency d1.In the decomposition of following one deck, again a1 is resolved into low frequency a2 and high frequency d2 two parts, the information that loses among the low frequency a2 is caught by high frequency d2, so analogizes down, can carry out deeper decomposition.Because image is after wavelet decomposition, the profile of image is mainly reflected in low frequency part, and detail section then is embodied in HFS.For the pavement crack image, the road surface background belongs to low frequency component, and the crack belongs to high fdrequency component.Therefore,, the high frequency coefficient of dissociation is carried out enhancement process, reach the effect that the crack strengthens, obtain the image that the crack strengthens by the low frequency coefficient of dissociation is carried out attenuation processing.
And then reuse the process that edge detection operator carries out edge extracting and carries out binary conversion treatment, and accurately orient residing position in image, crack.
For first kind crack, need judge the crack is transverse crack or longitudinal crack.But, can lose some Useful Informations through the image after the rim detection, so the crack may have fracture in the image after processing, therefore before precise classification is carried out in the crack, must in image, extract the crack profile.
Here, at first utilize morphological method, to expanding and corrode in the crack, fill discontinuous zone, crack, obtain complete crack area, its specific implementation process is: being expanded in the crack, is that fundamental point is extended to the outside with the crack, fills the crack.Then the crack is corroded,, remove outmost marginal point, make the crack near original shape promptly at the crack that mends.After repeating to expand several times and corroding, got up continuously in the crack of fracture, obtain a complete crack area.
Next, in the binary map that obtains complete crack area, extract the crack profile.Fig. 4 is the synoptic diagram that the present invention extracts the crack profile.As shown in Figure 4, the detailed process that extracts the crack profile in binary map is: according to line by line scan binary map after filling of direction from left to right, as the black pixel point (P that scans first white portion and black region intersection
0Point) time, writes down this P
0The coordinate figure of point, and be P
0Point is given a sign L who represents frontier point.For frontier point P
08 neighborhood picture elements, a point, b point, c point, d point, e point, f point, g point and h point, according to counterclockwise, a point, h point and g point are the white pixel point, and the f point is the black picture element point, and e point, d point, c point and b point are the white pixel point, therefore, having only the f point is the frontier point of white portion and black region intersection, the record f coordinate figure of order, and give for the f point and to identify L.Then, the 8 neighborhood picture elements that f is ordered are analyzed again, write down the coordinate figure of the frontier point that is found, and give sign L for it.The rest may be inferred, and up to finding all frontier points, all there is sign L in the frontier point that promptly finds.According to fracture strike trend,, obtain complete crack profile with existing all frontier points of sign L to be linked in sequence.And, in order to highlight the crack profile and to be convenient to subsequent treatment, utilize thinning algorithm, wide the crack contour thinning to single pixel.
So far, obtain a complete crack profile diagram with continuous boundary.
The detailed process of transverse crack or longitudinal crack being carried out precise classification is: the crack profile that scanning obtains, and find out two end points in crack, and make the line of two end points.Whether the angle of judging this line and road xsect less than 45 degree, if, then think transverse crack less than 45 degree, otherwise, think longitudinal crack.
Next, calculate the FRACTURE CHARACTERISTICS value, i.e. the breadth extreme in horizontal and vertical crack and equivalent length.Fig. 5 calculates in realizing embodiments of the invention laterally or the breadth extreme of longitudinal crack and the synoptic diagram of equivalent length.As shown in Figure 5, vertical earlier back transversal scanning crack area is found out the central point at two edges, vertical direction crack, and is connected these central points, forms curve C.With the length of curve C equivalent length as this crack.
From an end of curve C, get a point every a point, per three points carry out fitting a straight line, and resulting straight line is designated as Li, and judges the distance D i of Li apart from the crack coboundary.The rest may be inferred, up to the other end of curve C.Find out the Di of length value maximum, and with the breadth extreme of 2Di as this crack.
So far, type, equivalent length and the breadth extreme of horizontal or longitudinal crack have been drawn.
When precise classification is carried out in the crack,, need judge the crack and be piece and split or chap for the second class crack.Owing to typical piece splits is to make the road surface split into polygon by the vertical and horizontal crack is staggered, therefore, roughly becomes the square shape in image; The net then formed of be full of cracks by a series of polygon fritters, so adopt the mode of template matches piece is split or to chap and carry out precise classification.
Fig. 6 is the synoptic diagram of two 5 * 5 templates using in realizing embodiments of the invention.As shown in Figure 6, design two 5 * 5 templates: " ten " word bit at template 1 center is changed to 1, and other position is 0; Template 2 is 1 entirely on diagonal line, and other position is 0 entirely.Rule of thumb value preestablishes a product threshold value, and rectangular block number threshold value.Above-mentioned two templates and former gray-scale map are done product respectively, and get gained maximum product max.Whether judge max greater than predefined product threshold value, if greater than, then think to have approximate rectangle crack, otherwise think and be not the rectangle crack.The rest may be inferred, the number of the rectangular block that exists in counting entire image.Then, judge whether to count the rectangular block number greater than predefined rectangular block number threshold value, if, think that then piece splits, otherwise, think be full of cracks.
When calculating the FRACTURE CHARACTERISTICS value, in order to calculate the area that be full of cracks or piece split, on image, find out the solstics of the four direction that belongs to the crack, make boundary rectangle according to these four points, and with the area of this boundary rectangle approximate area as this crack.
At last, count the quantity that transverse crack in the image, longitudinal crack, be full of cracks and piece split respectively.
Step 207: fracture evaluation module 107 is according to the crack information that receives, the quantity that comprises all types of cracks, laterally or the length and the width of longitudinal crack, the area that be full of cracks or piece split, and estimate according to the evaluation criterion road pavement crack of all kinds of cracks, the road surface order of severity.
Here, the area that splits according to length, be full of cracks and the piece of wall scroll longitudinal crack and transverse crack and the area in total highway section calculate the cracking ratio in this highway section, finish the evaluation in road pavement crack.Such as, for longitudinal crack, according to the evaluation criterion of the pavement crack order of severity, if the breadth extreme of the longitudinal crack that draws in the image, thinks then that the damaged degree of this longitudinal crack is light level less than 5 millimeters.
Preferably, the present invention deposits the positional information of crack in road network in the road net data storehouse in, so that managerial personnel safeguard better to whole road network.
Preferably, the present invention utilizes the position of GPS (GPS) technological orientation crack in whole road network.
Preferably, the range finder module of system of the present invention further comprises knotmeter, is used to measure carrying vehicle actual travel speed, and according to the physical location of actual travel velocity correction crack in road network.Because the influence of tire pressure size and pavement behavior, have bigger deviation between the feasible carrying vehicle travel speed that is provided with and its actual travel speed, and it is big that this kind deviation can become gradually along with the time, therefore can cause when calculating the position of crack, deviation occurring in whole highway section.Such as, draw the 5th meter in highway section, a crack after the calculating, but because to calculate the carrying vehicle travel speed travel speed of using actual with it inequality, therefore, in fact this crack is at the 5.1st meter in highway section.
In order to guarantee the correct position of resulting crack in the highway section, range finder module uses radar meter to measure the actual travel speed of carrying vehicle.Radar meter is a kind of equipment that tests the speed that utilizes Doppler's principle that the speed of moving object is carried out continuous coverage, and its metering system is a non-contact measurement, and measuring process is not subjected to the influence of tire state and pavement behavior, and rate accuracy is higher.When the crack was estimated, the speed of using radar meter to measure calculated the actual travel distance of carrying vehicle when taking every width of cloth image.Thereby proofread and correct the position of crack in whole highway section.
Preferably, the present invention is integrated into the function of image capture module 104, image pretreatment module 105, crack identification module 106 and fracture evaluation module 10 on the entity digital signal processor (DSP), and, in order to reach real-time processing image, adopt parallel all that finish digital picture of at least two DSP to handle, and the evaluation in road pavement crack.
In a word, the above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (20)
1, a kind of system that discerns pavement crack, comprise in the Load System other module and the carrying vehicle of stablizing travel speed is provided, the supply module of other each module output stabilized power source in system, range finder module and image capture module, wherein, range finder module calculates the distance that carrying vehicle travels in real time, and when distance element length that the carrying vehicle operating range that calculates equals to set, export apart from trigger pip to image capture module, image capture module receive that range finder module sends apart from trigger pip after, road pavement is carried out image taking, and export captured image, it is characterized in that, this system also comprises with lower module: the image pretreatment module, crack identification module and fracture evaluation module, wherein
The image pretreatment module, receive the image of image capture module output, and be gray-scale map with received image transitions, and utilize edge detection operator that gray-scale map is carried out the figure image intensifying, and isolate crack point and road surface background dot the image after strengthening, the binary map that is only contained the crack outputs to binary map the crack identification module then;
The crack identification module, receive the binary map that the image pretreatment module is sent, utilize shadow casting technique that binary map is carried out projection, according to the position of perspective view Primary Location crack in image, and tentatively longitudinal crack and transverse crack are divided into a class crack, will chap and piece splits and is divided into a class crack;
The crack identification module utilizes wavelet technique that the zone that Primary Location contains the crack is strengthened, and reuse shadow casting technique and accurately orient residing position in image, crack, and precise classification is carried out in the crack, determine transverse crack and longitudinal crack, and be full of cracks and piece split;
The crack identification module is calculated the eigenwert in crack, statistics crack quantity, and the crack information after will handling outputs to the fracture evaluation module;
The fracture evaluation module is estimated the crack according to the fracture evaluation standard.
2, system according to claim 1 is characterized in that, the wheel revolutions of described range finder module recording carrying car, and according to the record wheel revolutions calculate the distance that carrying vehicle travels in real time.
3, system according to claim 1 is characterized in that, described range finder module further comprises knotmeter, measures carrying vehicle actual travel speed, and according to the physical location of actual travel velocity correction crack in road network.
4, system according to claim 3 is characterized in that, described knotmeter comprises radar meter or fifth wheel instrument.
5, system according to claim 1 is characterized in that, the distance element length of setting in the described range finder module is the path length that a two field picture that described image capture module is shot is covered.
6, system according to claim 1 is characterized in that, described range finder module utilizes global position system GPS to determine the position of crack in road network.
7, system according to claim 1 is characterized in that, described fracture evaluation module is further used for the positional information of crack in road network deposited in the road net data storehouse.
8, system according to claim 1 is characterized in that, described image pretreatment module, crack identification module and fracture evaluation module are integrated on the functional entity of digital signal processor DSP.
9, a kind of method of discerning pavement crack is characterized in that, this method may further comprise the steps:
The distance element length whether the carrying vehicle operating range that A, real-time judge calculate equals to set, if, execution in step B then, otherwise, steps A returned;
B, send apart from trigger pip, the road pavement image is taken, and is digital picture with the image transitions of taking;
C, digital picture is converted to gray-scale map, and the gray-scale map after using edge detection operator to conversion carries out the figure image intensifying, use binarization method that image is cut apart, only contained the binary map in crack;
D, using shadow casting technique that binary map is carried out projection, according to the position of perspective view Primary Location crack in image, and is longitudinal crack or transverse crack with the crack preliminary classification, and be full of cracks or piece split two class cracks;
E, use wavelet technique with in the gray-scale map of gained among the step C the Primary Location zone of containing the crack strengthen;
F, use edge detection operator carry out edge extracting to the image after strengthening, and image is carried out binary conversion treatment, and accurately orient the position of crack in image;
G, precise classification go out longitudinal crack and transverse crack, and be full of cracks and piece split;
H, calculate the eigenwert in each crack, the quantity in statistics crack, and the crack is carried out the evaluation of the order of severity according to FRACTURE CHARACTERISTICS value and quantity.
10, method according to claim 9 is characterized in that, in steps A, the distance element length of described setting is the path length that a two field picture is covered.
11, method according to claim 9, it is characterized in that, the described gray-scale map of step C is to carry out addition after the value by the red R of each picture element in the image that will receive, green G and blue B multiply by weighted value respectively, and obtaining with value of making that the value of R, the G of this point and B equals to calculate.
12, method according to claim 9, it is characterized in that, the step that the described use binarization method of step C is cut apart image comprises: the gray-scale value of 1.2 times of background gray shade scales is set at threshold value, gray-scale value in the image is set at the crack point greater than the picture element of this threshold value, and this crack point is made as black, gray-scale value in the image is set at background dot less than the picture element of this threshold value, and this background dot is made as white.
13, method according to claim 9 is characterized in that,
The described step according to the position of perspective view Primary Location crack in image of step D comprises: with the direction of scanning of image as transverse axis, with the gray-scale value summation of the image that exists on the transverse axis orthogonal directions as the longitudinal axis, determine a coordinates regional,
Crack area in the binary map is projected in respectively on the transverse axis and the longitudinal axis in this coordinates regional, obtain two one dimension curves, from the transverse axis one dimension curve projection figure about two end points make straight line respectively perpendicular to transverse axis, two end points up and down of the perspective view of a dimension curve are made the straight line perpendicular to the longitudinal axis respectively from the longitudinal axis, with these 4 rectangular areas that straight line defined as the position of crack in former binary map;
Step D is described to be longitudinal crack or transverse crack with the crack preliminary classification, and the step that be full of cracks or piece split two class cracks comprises: judge that drop shadow curve on the axle is banded and the crack of fewer obvious peak value is arranged is transverse crack and longitudinal crack, judgement is similar at the projection distribution shape of the transverse axis and the longitudinal axis, and the crack that does not have obvious peak value, perhaps projection has more than the crack of 3 obvious peak value and splits for be full of cracks and piece;
The zone that the described Primary Location of step e contains the crack is the rectangular area that step D is defined.
14, method according to claim 12 is characterized in that, further comprises between described step F and step G:
G11, image is expanded and corrosion treatment, fill discontinuous zone, crack, obtain complete crack area;
Binary map after G12, the filling of lining by line scan, when scanning first black picture element point, write down the coordinate figure of this black picture element point, it is labeled as frontier point, again according to the 8 neighborhood picture elements of counterclockwise analyzing this black picture element point, write down the coordinate figure of the black picture element point of white portion and black region intersection in this 8 neighborhood picture element, and it is labeled as frontier point, the rest may be inferred, up to marking all frontier points and coordinate figure thereof;
G13, according to fracture strike trend, all frontier points that mark are linked in sequence, obtain complete crack profile, utilize thinning algorithm then, the crack contour thinning to single pixel wide.
15, method according to claim 14, it is characterized in that, the described precise classification of step G goes out longitudinal crack and transverse crack comprises: two end points of crack profile are linked to be straight line, judge whether the angle that institute connects straight line and road xsect spends less than 45, if, then this crack is judged to be transverse crack, otherwise, this crack is judged to be longitudinal crack.
16, method according to claim 15, it is characterized in that the step of the described calculating FRACTURE CHARACTERISTICS of step H value comprises: get the central point at two edges, vertical direction crack, and connect each central point of being got, form curve, length of a curve is set at the equivalent length in this crack;
From an end of curve, get a point every a point, until the other end of curve, per three points carry out fitting a straight line, obtain each bar straight line, with the 2 times breadth extremes that are set at this crack of straight line apart from the ultimate range of crack coboundary.
17, method according to claim 9, it is characterized in that, the described precise classification of step G goes out be full of cracks in the multidirectional crack and piece and splits and comprise: set product threshold value and rectangular block number threshold value, and two 5 * 5 templates are set, " ten " word bit of first template center is changed to 1, and other position is 0, second template is 1 entirely on diagonal line, other position is 0 entirely, the gray-scale map of gained among described two templates and the step C is done product respectively, and get the gained maximum product, judge that whether this maximal value is greater than the product threshold value of setting, if then think to have approximate rectangle crack, otherwise think and be not the rectangle crack, the rest may be inferred, the number of the rectangular block that exists in counting entire image then, judges that whether the rectangular block number that counts is greater than the rectangular block number threshold value of setting, if, then this crack is judged to be piece and splits, otherwise, this crack is judged to be full of cracks.
18, method according to claim 17, it is characterized in that, the step of the described calculating FRACTURE CHARACTERISTICS of step H value comprises: gets the solstics of profile four direction in crack on the image, these four points is linked to be boundary rectangle, and with the area of this boundary rectangle area as this crack.
19, method according to claim 9, it is characterized in that, this method further comprises: when detecting pavement crack, utilize the position of global position system GPS technological orientation crack in road network, and the positional information of crack in road network of determining deposited in the road net data storehouse.
20, method according to claim 9 is characterized in that, this method further comprises: use radar meter to measure carrying vehicle actual travel speed, and according to the position of actual travel velocity correction crack in the highway section.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410033964 CN1237327C (en) | 2004-04-20 | 2004-04-20 | System and method for discriminating road gap |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410033964 CN1237327C (en) | 2004-04-20 | 2004-04-20 | System and method for discriminating road gap |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1563891A CN1563891A (en) | 2005-01-12 |
| CN1237327C true CN1237327C (en) | 2006-01-18 |
Family
ID=34481430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410033964 Expired - Fee Related CN1237327C (en) | 2004-04-20 | 2004-04-20 | System and method for discriminating road gap |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1237327C (en) |
Families Citing this family (69)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4566074B2 (en) * | 2005-06-21 | 2010-10-20 | 株式会社パスコ | House change judgment method and house change judgment program |
| CN100389237C (en) * | 2005-12-06 | 2008-05-21 | 东南大学 | Intelligent hot repair process of asphalt road surface by microwave heating |
| CN101593292B (en) * | 2009-05-07 | 2012-01-04 | 长沙融威电子科技有限公司 | Anti-counterfeiting method and anti-counterfeiting device for separating and counting non-touch paper currency or tickets |
| CN101941012B (en) | 2009-07-03 | 2012-04-25 | 泰怡凯电器(苏州)有限公司 | Cleaning robot, dirt identification device thereof and cleaning method of robot |
| CN101701919B (en) * | 2009-11-20 | 2011-05-11 | 长安大学 | Pavement crack detection system based on image and detection method thereof |
| CN101915764B (en) * | 2010-08-10 | 2012-02-01 | 武汉武大卓越科技有限责任公司 | Road surface crack detection method based on dynamic programming |
| CN101923651B (en) * | 2010-09-16 | 2012-10-10 | 中公高科养护科技股份有限公司 | Calibration method of automatic recognition accuracy of damage data based on pavement images |
| CN102663344A (en) * | 2012-03-04 | 2012-09-12 | 南京理工大学常熟研究院有限公司 | Damaged road detection device based on image segmentation |
| CN102706883B (en) * | 2012-05-06 | 2015-06-10 | 山西省交通科学研究院 | System and method for recognizing holes in paved waterproof board of tunnel |
| CN103422417A (en) * | 2012-05-17 | 2013-12-04 | 上海嘉任电子科技有限公司 | Dynamic identification system and method for detecting road surface damages |
| CN102720118B (en) * | 2012-06-01 | 2014-10-22 | 中公高科养护科技股份有限公司 | Method for determining detection width of road surface |
| CN102706880B (en) * | 2012-06-26 | 2014-04-02 | 哈尔滨工业大学 | Road information extraction device based on two-dimensional image and depth information and road crack information detection method based on same |
| CN103590312A (en) * | 2012-08-15 | 2014-02-19 | 圣东营造股份有限公司 | Checking, reporting and restoring method for missing pavement |
| CN102867302B (en) * | 2012-08-30 | 2015-04-08 | 四川大学 | Core fracture identification method based on three-dimensional image information processing |
| CN102842203B (en) * | 2012-09-04 | 2014-07-09 | 广东省公路管理局 | Method for monitoring bridge fault on basis of video image |
| CN103050012B (en) * | 2013-01-11 | 2015-04-15 | 福建工程学院 | Road maintenance information acquisition method based on floating car technology |
| CN103278864B (en) * | 2013-05-10 | 2016-12-28 | 中国石油天然气股份有限公司 | The geologic characteristic parameter α of hole seam type reservoir stratum and the assay method of distribution and device |
| CN103258059A (en) * | 2013-06-05 | 2013-08-21 | 周岚 | Transverse crack information processing method and system for expressway bituminous pavement |
| CN103306186B (en) * | 2013-06-20 | 2015-05-27 | 长安大学 | Algorithm for detecting structure depth of cement concrete pavement |
| CN103440657B (en) * | 2013-08-27 | 2015-11-04 | 武汉大学 | A kind of online crack on road screening method |
| CN103528527A (en) * | 2013-10-15 | 2014-01-22 | 北京交通大学长三角研究院 | Area selection-based automatic crack size measurement method |
| CN103744119B (en) * | 2013-12-24 | 2016-08-17 | 桂林电子科技大学 | The three-dimensional feature detection method in a kind of building body crack and system |
| CN103870833B (en) * | 2014-03-31 | 2016-03-02 | 武汉工程大学 | The pavement crack estimated based on concavity extracts and assessment method |
| CN103903268B (en) * | 2014-04-09 | 2017-02-15 | 南京理工大学 | Road surface crack detection method based on multiple lower-layer marked features |
| CN103955923B (en) * | 2014-04-18 | 2017-03-01 | 南京理工大学 | A kind of quickly pavement disease detection method based on image |
| CN104636750B (en) * | 2014-12-17 | 2018-09-21 | 长安大学 | A kind of pavement crack recognizer and system based on double scale clustering algorithms |
| CN104655639A (en) * | 2014-12-23 | 2015-05-27 | 北京恒达锦程图像技术有限公司 | Pavement crack detection method capable of accurately acquiring crack characteristics and pavement crack detection system capable of accurately acquiring crack characteristics |
| CN105784710B (en) * | 2014-12-23 | 2018-08-28 | 桂林电子科技大学 | A kind of glue into concrete beam cracks detection device based on Digital Image Processing |
| CN104504216B (en) * | 2015-01-06 | 2017-11-17 | 东南大学 | Sheet pavement in-situ heat regeneration engineering sampling paragraph determines method |
| CN104655642B (en) * | 2015-02-06 | 2017-07-18 | 江苏大学 | A kind of automatic measurement of steel cracking type defect, sign sorting technique and its system |
| CN106152962B (en) * | 2015-04-07 | 2018-12-14 | 南京理工大学 | Oil drilling casing deformation detecting method based on pipe crawling device |
| CN105113375B (en) * | 2015-05-15 | 2017-04-19 | 南京航空航天大学 | Pavement cracking detection system and method based on line structured light |
| CN105136114B (en) * | 2015-08-11 | 2017-12-19 | 武汉武大卓越科技有限责任公司 | Big cross section measuring method based on line-structured light under complicated luminous environment |
| CN107923745B (en) * | 2015-08-26 | 2021-07-23 | 住友建机株式会社 | Measuring device for excavator |
| CN105389793B (en) * | 2015-10-14 | 2018-10-19 | 北京智博联科技股份有限公司 | A kind of method of automatic identification body surface fracture strike and width |
| CN105825169B (en) * | 2016-03-10 | 2019-02-15 | 辽宁工程技术大学 | A kind of pavement crack recognition methods based on road image |
| CN105862556A (en) * | 2016-05-30 | 2016-08-17 | 江苏智途科技股份有限公司 | Intelligent vehicle-mounted road information acquisition device and method for acquiring road information |
| CN106087677B (en) * | 2016-06-02 | 2018-07-31 | 上海华城工程建设管理有限公司 | Asphalt pavement crack type automatic identifying method |
| CN106226834B (en) * | 2016-07-14 | 2019-04-23 | 昆山市交通工程试验检测中心有限公司 | A kind of vehicular road surface manhole missing detection method |
| CN106225702A (en) * | 2016-08-03 | 2016-12-14 | 武汉科技大学 | Fracture width detection apparatus and method |
| CN107780324A (en) * | 2016-08-28 | 2018-03-09 | 上海华测导航技术股份有限公司 | A kind of airfield pavement method for inspecting and system |
| CN106226157B (en) * | 2016-08-31 | 2019-04-16 | 北京华横新技术开发公司 | Concrete structure member crevices automatic detection device and method |
| CN106758715B (en) * | 2017-01-23 | 2019-01-11 | 中公高科养护科技股份有限公司 | A kind of measuring method of the pavement current repair project amount based on image recognition result |
| CN106988193B (en) * | 2017-04-12 | 2019-09-03 | 南京南化建设有限公司 | A kind of efficient pavement damage crack detection system |
| CN107316064B (en) * | 2017-06-26 | 2020-07-14 | 长安大学 | Asphalt pavement crack classification and identification method based on convolutional neural network |
| CN107766886A (en) * | 2017-10-20 | 2018-03-06 | 华中师范大学 | High-speed railway rail fastening spring fatigue fracture automatic identifying method |
| CN108038883B (en) * | 2017-12-06 | 2021-01-26 | 陕西土豆数据科技有限公司 | Crack detection and identification method applied to highway pavement video image |
| CN108074230A (en) * | 2017-12-14 | 2018-05-25 | 中国科学院长春光学精密机械与物理研究所 | A kind of pavement crack detection device and method |
| CN108592811A (en) * | 2018-03-26 | 2018-09-28 | 中震（北京）工程检测股份有限公司 | A kind of building structure Fracture Deformation real-time dynamic monitoring system and recognition methods |
| CN108517752B (en) * | 2018-04-28 | 2020-01-21 | 中建路桥集团有限公司 | Multifunctional pavement crack pouring machine and crack pouring method |
| CN108961230B (en) * | 2018-06-28 | 2020-09-11 | 武汉大学 | Identification and extraction method for structural surface crack characteristics |
| CN109035218B (en) * | 2018-07-09 | 2021-08-31 | 武汉光谷卓越科技股份有限公司 | Pavement repair area detection method |
| CN109447944B (en) * | 2018-09-21 | 2020-08-11 | 中国石油天然气股份有限公司 | Lithofacies identification method and system for carbonate rock |
| CN109632007B (en) * | 2019-01-17 | 2020-12-04 | 北京理工大学 | Edge point extraction method and gear high-precision vision measurement system |
| CN110512501B (en) * | 2019-08-12 | 2021-02-19 | 武汉理工大学 | Crack disease evaluation method based on ground penetrating radar and FWD combined detection |
| CN110588272B (en) * | 2019-09-23 | 2020-10-27 | 无锡职业技术学院 | Automobile suspension system based on visual sensing technology and road surface unevenness measuring method |
| CN110926359B (en) * | 2019-11-05 | 2021-10-29 | 长安大学 | Three-dimensional crack curved surface contour detection method for optimizing two-phase scanning pavement |
| CN110794389B (en) * | 2019-11-18 | 2023-02-28 | 淮阴工学院 | Image processing-based non-contact distance measurement method |
| CN111311559B (en) * | 2020-02-07 | 2022-06-14 | 中国石油大学(华东) | Non-deviation crack characterization method |
| CN111462060A (en) * | 2020-03-24 | 2020-07-28 | 湖南大学 | Method and device for detecting standard section image in fetal ultrasonic image |
| CN111516692A (en) * | 2020-04-20 | 2020-08-11 | 重庆长安汽车股份有限公司 | Control system and method for vehicle running on hollow road surface |
| CN111735409A (en) * | 2020-06-04 | 2020-10-02 | 深圳职业技术学院 | Soft robot arm shape measuring method, system and storage medium |
| CN111719400B (en) * | 2020-07-08 | 2022-01-14 | 贵州恒创建设工程有限公司 | Method for repairing damaged road surface |
| CN114333141A (en) * | 2020-09-25 | 2022-04-12 | 深圳富桂精密工业有限公司 | Carton recycling method, electronic device and storage medium |
| CN112815885B (en) * | 2020-12-30 | 2023-03-28 | 广东百年基业水利水电建设有限公司 | Method and system for monitoring house wall cracks |
| CN113538733B (en) * | 2021-07-09 | 2023-03-14 | 江苏金海星导航科技有限公司 | Black box system based on satellite communication and cloud computing technology |
| CN114740010B (en) * | 2022-04-14 | 2023-05-23 | 石云飞 | Intelligent monitoring method for pavement diseases |
| CN116448016A (en) * | 2023-04-26 | 2023-07-18 | 成都智达万应科技有限公司 | Intelligent rapid detection system and detection vehicle with same |
| CN116612470B (en) * | 2023-07-17 | 2023-09-26 | 临沂农业科技职业学院(筹) | Bread detection method and system based on visual characteristics |
-
2004
- 2004-04-20 CN CN 200410033964 patent/CN1237327C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1563891A (en) | 2005-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1237327C (en) | System and method for discriminating road gap | |
| Moaveni et al. | Evaluation of aggregate size and shape by means of segmentation techniques and aggregate image processing algorithms | |
| CN101354241B (en) | Method for evaluating aggregate digital image | |
| CN1945596A (en) | Vehicle lane Robust identifying method for lane deviation warning | |
| CN107064169B (en) | Bridge crack detection device and detection method | |
| CN1223964C (en) | Apparatus and method for measuring vehicle queue length | |
| US20160292518A1 (en) | Method and apparatus for monitoring changes in road surface condition | |
| CN101469985A (en) | Single-frame image detection apparatus for vehicle queue length at road junction and its working method | |
| CN101029824A (en) | Method and apparatus for positioning vehicle based on characteristics | |
| CN107798293A (en) | A kind of crack on road detection means | |
| CN101075376A (en) | Intelligent video traffic monitoring system based on multi-viewpoints and its method | |
| CN101063662A (en) | Method for detecting empty bottle bottom defect and device for detecting empty bottle bottom defect based on DSP | |
| CN109272548B (en) | Method for measuring diameter of bubbles in flotation process | |
| US20110221906A1 (en) | Multiple Camera System for Automated Surface Distress Measurement | |
| CN104021574A (en) | Method for automatically identifying pavement diseases | |
| CN104392224A (en) | Crack detection method for road surface | |
| CN113537016B (en) | Method for automatically detecting and early warning road damage in road patrol | |
| CN106969708A (en) | A kind of detection means and method of aggregate quality morphology | |
| CN110390256B (en) | Asphalt pavement crack extraction method | |
| CN109146859A (en) | A kind of pavement crack detection system based on machine vision | |
| CN1519143A (en) | System and method for visual detecting overhead contact line equipment of electric railways | |
| Wang et al. | Automated pavement distress survey: a review and a new direction | |
| JPH03160349A (en) | Device for detecting crack | |
| Medina et al. | Automated visual inspection of road surface cracks | |
| CN1790378A (en) | Binary method and system for image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20060118 Termination date: 20110420 |