CN115311260B - Road surface quality detection method for highway traffic engineering - Google Patents

Road surface quality detection method for highway traffic engineering Download PDF

Info

Publication number
CN115311260B
CN115311260B CN202211195054.9A CN202211195054A CN115311260B CN 115311260 B CN115311260 B CN 115311260B CN 202211195054 A CN202211195054 A CN 202211195054A CN 115311260 B CN115311260 B CN 115311260B
Authority
CN
China
Prior art keywords
image
crack
target image
column
curve
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.)
Active
Application number
CN202211195054.9A
Other languages
Chinese (zh)
Other versions
CN115311260A (en
Inventor
张海兵
张�杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inner Mongolia Highway Engineering Consulting And Supervision Co ltd
Original Assignee
Inner Mongolia Highway Engineering Consulting And Supervision Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inner Mongolia Highway Engineering Consulting And Supervision Co ltd filed Critical Inner Mongolia Highway Engineering Consulting And Supervision Co ltd
Priority to CN202211195054.9A priority Critical patent/CN115311260B/en
Publication of CN115311260A publication Critical patent/CN115311260A/en
Application granted granted Critical
Publication of CN115311260B publication Critical patent/CN115311260B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30248—Vehicle exterior or interior
    • G06T2207/30252—Vehicle exterior; Vicinity of vehicle
    • G06T2207/30256—Lane; Road marking
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A30/00—Adapting or protecting infrastructure or their operation
    • Y02A30/60—Planning or developing urban green infrastructure

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Image Processing (AREA)

Abstract

The application relates to the field of image data processing, in particular to a road surface quality detection method for highway traffic engineering. According to the method, a target image corresponding to a target pavement is obtained according to each frame of shooting image corresponding to the target pavement, and each target image comprises a complete crack; calculating crack type judging indexes corresponding to each target image, and judging the crack type of each target image according to the crack type judging indexes; the crack type judging indexes comprise the ratio of the transverse and longitudinal fluctuation range, the difference degree of the average wave peak and the column and the average wave peak, the fluctuation degree of the row and the curve and the fluctuation degree of the column and the curve; and estimating the crack width according to the number of pixel points occupied by the total area of the crack on each target image and the number of edge pixel points. The application realizes the judgment of the crack type and the estimation of the crack width of the target pavement based on the photographed image of the target pavement, namely the detection of the road quality of the highway traffic engineering.

Description

Road surface quality detection method for highway traffic engineering
Technical Field
The application relates to the field of image data processing, in particular to a road surface quality detection method for highway traffic engineering.
Background
The highway is used as an infrastructure construction of China, not only meets the travel of people, but also has a fundamental effect on the development of economy directly, so that the quality of highway engineering is required to be ensured. However, the quality of the current highway subgrade has some problems, which affect the development of highway safety industry.
The crack is one of the most common, most easy occurrence and earliest produced diseases in various damages of the road surface, can influence the beautiful appearance of the road and the comfort of driving, is easy to expand to cause structural damage of the road surface, and shortens the service life of the road surface. Therefore, cracks appear on the pavement, sealing repair should be performed in time, otherwise rainwater and other impurities can enter the surface layer structure and the roadbed along the cracks, so that the bearing capacity of the pavement is reduced, and local or sheet damage of the pavement is accelerated. The types of the cracks mainly comprise longitudinal cracks, transverse cracks and tortoise-shaped net cracks, the causes of each crack are different from each other, the maintenance requirements of the cracks with different widths are also different, and the method has important significance on the maintenance work of the developed highway by judging the types and the corresponding severity of the pavement cracks.
Disclosure of Invention
In order to judge the type and severity of road cracks, the application aims to provide a road quality detection method for road traffic engineering.
The application provides a road surface quality detection method for highway traffic engineering, which comprises the following steps:
acquiring each frame of shooting image corresponding to a target pavement, and carrying out graying and reversing treatment on each frame of shooting image corresponding to the target pavement to obtain each frame of gray reversed image corresponding to the target pavement;
performing enhancement and binarization processing on each frame of gray level reverse image corresponding to the target pavement, judging whether each frame of image is a suspected crack image according to the row and column data corresponding to each frame of processed image, and if the suspected crack image exists in each frame of processed image corresponding to the target pavement and the suspected crack image of the adjacent frame does not exist, marking each suspected crack image as each target image; if the adjacent frame suspected crack images exist, performing image stitching processing on the adjacent frame suspected crack images to obtain 1 or more target images, wherein each target image comprises a complete crack;
calculating crack type judging indexes corresponding to each target image, and judging the crack type of each target image according to the crack type judging indexes; the crack type judging indexes comprise the ratio of the transverse and longitudinal fluctuation range, the difference degree of the average wave peak and the column and the average wave peak, the fluctuation degree of the row and the curve and the fluctuation degree of the column and the curve; and estimating the crack width according to the number of pixel points occupied by the total area of the crack on each target image and the number of edge pixel points.
Further, the step of judging whether each frame image is a suspected crack image according to the row and column data corresponding to each frame image after processing includes:
and carrying out accumulation analysis on pixel values in each frame of binarized image according to the directions of rows and columns, and calculating a row and column white pixel accumulation curve, wherein the formula is as follows:
wherein,accumulating the +.f in the sequence for the line pixels corresponding to the binarized image of a frame>Line sum, line sum->The +.f. in the corresponding column gray-scale accumulation sequence for the frame binarized image>Column sum, head>The total number of lines and the total number of columns corresponding to the image after the frame binarization are adopted; />Representing the +.>Line->Pixel values of column pixels;
if continuous satisfaction in a certain frame of binarized imageThe number of lines of (2) is greater than->Or/and continuously satisfy->The number of columns of (2) is greater than->And judging the image binarized by the frame as a suspected crack image.
Further, the ratio of the corresponding lateral and longitudinal fluctuation ranges of each target image is calculated by using the following formula:
wherein,for the ratio of the corresponding lateral-longitudinal fluctuation range of a certain target image, +.>For +.>Lower limit value of row coordinates, +.>For +.>Upper limit value of row coordinates of (c); />For +.>Lower limit value of column coordinates of>For +.>Upper limit value of column coordinates of>Is->Sum of row pixel values +.>Is->The sum of the column pixel values.
Further, the degree of difference between the average peak and the column and the average peak corresponding to each target image is calculated by using the following formula:
wherein,for the degree of difference of the row and the average peak from the column and the average peak of a certain target image +.>For the size of the peak on the line and curve of the target image, +.>For the number of peaks on the line and curve of the target image,/->For the size of the peak on the column and curve of the target image, +.>Is the number of peaks on the columns and curves of the target image.
Further, the row and curve fluctuation degree and the column and curve fluctuation degree corresponding to each target image are calculated by using the following formulas:
wherein,for the degree of fluctuation of the line and curve of a certain target image, +.>For the degree of fluctuation of the column and curve of the target image, +.>For the fluctuation ratio of the line and curve of the target image, +.>For the fluctuation ratio of the column and curve of the target image, +.>For the average peak-to-valley amplitude variation on the line and curve of the target image,/for>For the average peak-to-valley amplitude variation on the column and curve of the target image,/for>Summing the absolute values of the differences between all adjacent peaks and valleys on the line and curve of the target image, +.>Summing the absolute values of the differences between the columns of the target image and all adjacent peaks and valleys on the curve, +.>For the number of peaks on the line and curve of the target image,/->For the number of peaks on the columns and curves of the target image,/->For +.>Lower limit value of row coordinates, +.>For +.>Upper limit value of row coordinates of (c); />For +.>Lower limit value of column coordinates of>For +.>Upper limit value of column coordinates of>Is->Sum of row pixel values +.>Is->The sum of the column pixel values.
Further, the fracture types include transverse fractures, longitudinal fractures, and crazing network fractures.
Further, the judging the crack type of each target image according to the crack type judging index includes:
if the ratio of the transverse and longitudinal fluctuation ranges isDegree of difference between average peak and column and average peakAnd the degree of fluctuation of the row and curve->Column and curve degree of fluctuation->Judging that the transverse crack exists at the corresponding position of the target image;
if the ratio of the transverse and longitudinal fluctuation ranges isDegree of difference between average peak and column and average peakAnd the degree of fluctuation of the row and curve->Column and curve degree of fluctuation->Judging that a longitudinal crack exists at the corresponding position of the target image;
if the first two conditions are not satisfied, it is determined that a crazing network crack exists at a position corresponding to the target image.
Further, the estimating the slit width according to the number of pixels occupied by the total area of the slit and the number of edge pixels on each target image includes:
adding the ordinate values on the columns and the curves, and estimating the number of pixels occupied by the total area of the crack as follows:
wherein,for a certain purposeThe number of pixels occupied by the total area of the slits of the target image, is->For the total number of columns of the target image,is->A sum of column pixel values;
extracting edges by Canny edge detection on the target image, and recording the total value of edge pixels asThen the seam width estimation formula is:
wherein,for the slit width of the slit on the target image, +.>Is a proportional unit to the real size of the pixel.
The beneficial effects are that: the application realizes the judgment of the crack type and the estimation of the crack width of the target pavement based on the photographed image of the target pavement, namely the detection of the road traffic engineering pavement quality, and the targeted maintenance of the target pavement can be realized based on the obtained crack type and the crack width.
Drawings
FIG. 1 is a flow chart of a method for detecting road surface quality in highway traffic engineering according to the present application;
FIG. 2 is a gray scale reverse plot of the transverse slits of the present application;
FIG. 3 is a gray scale reverse plot of a longitudinal slit of the present application;
FIG. 4 is a gray scale reverse plot of a cracked network crack of the present application;
FIG. 5 is a schematic view of the rows and columns and curves of the transverse slits of the present application;
FIG. 6 is a schematic view of the rows and columns and curves of the longitudinal split of the present application;
FIG. 7 is a schematic view of the rows and columns and curves of a cracked network crack of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.
In order to realize the judgment of the type and severity of the pavement crack, the embodiment provides a pavement quality detection method for highway traffic engineering, as shown in fig. 1, comprising the following steps:
(1) Acquiring each frame of shooting image corresponding to a target pavement, and carrying out graying and reversing treatment on each frame of shooting image corresponding to the target pavement to obtain each frame of gray reversed image corresponding to the target pavement;
in the use process of roads, various quality problems often occur on the road surface, and road surface cracks are the most common type of road diseases. In order to improve the material proportion of the road to be built later, the type of the road crack and the crack width need to be estimated effectively, and accurate advice is provided for road maintenance.
In order to effectively estimate the type and width of the target pavement crack, the embodiment obtains a shot image of the target pavement, the shot image is guaranteed to be a horizontal image of the pavement, a pavement crack detection system for shooting the pavement has been disclosed in the prior art, the pavement crack detection system comprises a pavement crack measuring vehicle, a shooting device arranged on the pavement crack measuring vehicle and the like, the shooting device shoots the pavement image according to a set sampling frequency along with forward running of the pavement measuring vehicle, and a specific image acquisition process is not described in detail in the embodiment.
The implementation isIn the example, the actual shooting size corresponding to the shot image is square, and the pixel size of the shot image is recorded as follows. The image obtained by shooting in this embodiment is a color RGB image, the color RGB image is a three-layer channel, and in order to reduce the calculation amount, the RGB image is subjected to graying processing, so as to obtain a gray image with only one layer of channels. Specifically, in this embodiment, the weighted average is performed on the three components of RGB according to the psychological formula, and in the gray map obtained by the graying process, the lowest gray value 0 is black, and the highest gray value 255 is white. Then gray scale reverse operation is carried out, and 255 is used for subtracting the gray scale value corresponding to each pixel point to obtain a gray scale reverse image; the gray scale reverse image highlights the crack defect information and is convenient to analyze, and as shown in fig. 2, 3 and 4, the gray scale reverse image of the transverse, longitudinal and cracking netlike cracks is respectively shown.
(2) Performing enhancement and binarization processing on each frame of gray level reverse image corresponding to the target pavement, judging whether each frame of image is a suspected crack image according to the row and column data corresponding to each frame of processed image, and if the suspected crack image exists in each frame of processed image corresponding to the target pavement and the suspected crack image of the adjacent frame does not exist, marking each suspected crack image as each target image; if the adjacent frame suspected crack images exist, performing image stitching processing on the adjacent frame suspected crack images to obtain 1 or more target images, wherein each target image comprises a complete crack;
whether it is a low gray-value portion or a high gray-value portion of the crack, it has a distinct pixel difference characteristic from the background. Because the noise introduction in the acquisition process and the characteristics of the pavement texture determine that a large amount of noise is contained in the pavement image, the gray reverse image needs to be enhanced before further processing, namely, the gray reverse image is subjected to preliminary denoising, and the median filter is adopted to perform preliminary denoising on the gray reverse image, so that the method is effective in filtering salt and pepper noise of the gray reverse image.
Then, the image after denoising is subjected to binarization processing, and the obtained image possibly contains crack defects and also possibly is an intact road surface, so that automatic threshold binarization of the image has larger interference. In view of this, the present embodiment applies OSTU large-law binarization to the denoised standard transverse crack image to extract a threshold value with a good segmentation effect. OSTU is also called a maximum inter-class difference method, and the automatic selection of the global threshold is achieved by counting the histogram characteristics of the whole image, and the specific selection process is the prior art and is not repeated here. The threshold value is applied after enhancement of the grey scale reverse image of each frame>The pixel value in the image is larger than +.>Setting a foreground region (corresponding to a suspected pavement crack region) to be white (a pixel value of 1); less than->Setting the background area (corresponding to the road surface area) to black (pixel value 0).
In the actually acquired images, the generation positions of the cracks are random (appear at different positions in the images), and because the road surface crack measuring vehicle is dynamically driven, the shooting device is also dynamic, the shot images show continuous characteristics, and multiple frames of images containing the cracks possibly correspond to the same crack.
In this embodiment, the running speed of the road surface crack measuring vehicle and the frequency of the shooting by the shooting device (camera) are controlled so as to satisfy the condition that the overlap ratio of two adjacent images is about 50%, and the images corresponding to the target road surface are recorded togetherEvery picture is marked as +.>If the current shooting image is +.>Then the previous picture sequence number is +.>And each image has its corresponding geographical location information, giving coordinates for subsequent maintenance work.
If a suspected crack area exists in a certain frame of image, a more obvious white pixel area exists in the frame of image; and if no crack defect exists, the frame image is completely black. In view of this, in this embodiment, the pixel values in each frame of binarized image are accumulated and analyzed according to the directions of the rows and the columns, and the row and column white pixel accumulation curves are calculated, where the formula is as follows:
wherein,accumulating the +.f in the sequence for the line pixels corresponding to the binarized image of a frame>Sum of row pixel values +.>The +.f. in the corresponding column gray-scale accumulation sequence for the frame binarized image>Sum of column pixel values, < >>The total number of lines and the total number of columns corresponding to the image after the frame is binarized. />Representing the +.>Line->The pixel value (0 or 1) of the column pixel.
If continuous satisfaction in a certain frame of binarized imageThe number of lines of (2) is greater than->Or/and continuously satisfy->The number of columns of (2) is greater than->And judging the image of the frame binarized as a suspected crack image, and extracting and analyzing the suspected crack image. Specifically, a row pixel accumulation sequence and a column pixel accumulation sequence corresponding to the frame binarized image are respectively fitted into curves, so that a row sum accumulation curve (H) and a column sum accumulation curve (L) corresponding to the frame binary image are obtained.
Since the crack region is very likely to exist in a single image, there is a high probability that it exists in several adjacent images (i.e., a single frame image does not capture the full view of the crack). If it is extracted in the above processWhat comes is that adjacent frames of binarized images, the overlapping areas in the adjacent frames of binarized images are partially overlapping in rows and columns. Record the adjacent image set asWherein there is->Tension, ordinal number is the smallest +.>Ordinal number is maximum +.>. By adjacent +.>And->For example, the pixel values and the positions thereof corresponding to all the peaks and valleys on the respective rows and curves (H) are obtained, and a row and pixel value sequence is formed from the row positions from small to large (corresponding images from top to bottom)>The coincidence matching (pixel value is the same) of the sequence on the two images is performed and it is required that the coincident sequence portions on both image lines and curves are continuous (in this process, if there is no coincidence between the two consecutive images, the two images are separated and the continuous image set is re-established). Extracting the maximum peak value and its row coordinate in the overlapping part, at +.>The peak and line positions on the image are marked +.>、/>In->The images are marked as +.>、/>Wherein->. Thus, two adjacent images can be spliced to form a new image, the longitudinal dimension of which is +.>The method can obtain:
similarly, the next adjacent image is carried out again according to the methodSub-splicing, i.e. para->Adjacent images are co-processed->Sub-stitching, the image size of the final complete crack can be obtained as +.>. Similarly, if a separate image containing a crack is present, then +.>Size of +.>(order->) Can still use->（/>Positive integer) are expressed.
And marking the obtained image after each stitching and each individual image containing the crack as target images, wherein one or more target images can be obtained according to the embodiment, and each target image comprises 1 complete crack. In this embodiment, 2 or more are pointed.
(3) Calculating crack type judging indexes corresponding to each target image, and judging the crack type of each target image according to the crack type judging indexes; the crack type judging indexes comprise the ratio of the transverse and longitudinal fluctuation range, the difference degree of the average wave peak and the column and the average wave peak, the fluctuation degree of the row and the curve and the fluctuation degree of the column and the curve; and estimating the crack width according to the number of pixel points occupied by the total area of the crack on each target image and the number of edge pixel points.
Cracks are one of the most predominant forms of breakage of asphalt pavement. The cracks on the asphalt pavement are divided into transverse cracks, longitudinal cracks and crazing net-shaped cracks according to different causes, and the different types of cracks reflect the perfection degree of the road construction in terms of different qualities. Wherein:
transverse crack: the cracks are approximately vertical to the central line of the road and are distributed transversely regularly, and one crack appears at intervals; the crack is bent and curved and sometimes accompanied by a small number of branch cracks; the reason is that the transverse construction joint of the road surface is untreated, the joint is not tightly combined well, the temperature is reduced, and the road surface is contracted to cause transverse cracking.
Longitudinal cracking: the segments are cracked along the direction of the route, and some segments are long and some segments are distributed. The formation causes that when the asphalt surface layer is spread in a framing way, two joints are untreated, the compactness of the roadbed is uneven, or the edge of the roadbed is soaked in water to generate uneven subsidence.
Cracking the network cracks: the pavement is locally subjected to netlike cross cracking, and some small pieces are cracked, and meanwhile pavement settlement phenomenon exists in the cracking range. The reasons for this are poor quality of asphalt and asphalt mixture, or soft and mud layers sandwiched in the pavement structure, loose granular layers and poor water stability.
In order to analyze the type and severity of the crack on each target image, any target image is processed as follows:
the update column and formula are:
wherein,is the total number of lines of a certain target image.
The line and formula are unchanged, in this embodiment, the image of the three cracks is taken as an example, the line and column sum curves of the obtained transverse cracks are shown in fig. 5, the line and column sum curves of the obtained longitudinal cracks are shown in fig. 6, and the line and column sum curves of the obtained cracking netlike cracks are shown in fig. 7. The main differences of the three crack images are reflected in the fluctuation degree and the fluctuation interval. Therefore, the present embodiment analyzes the fluctuation range and the fluctuation degree corresponding to the target image, and the procedure is as follows:
(1) analyzing a fluctuation range;
firstly, extracting the fluctuation intervalThe row coordinate range of (2) is marked +.>，/>Is->Lower limit value of row coordinates, +.>Is->Upper limit value of row coordinates of (c); />The column coordinate range of (2) is marked +.>，/>Is->Lower limit value of column coordinates of>Is->Upper limit value of column coordinates of (c). The ratio of the statistical transverse and longitudinal fluctuation ranges is as follows:
wherein,for the ratio of the corresponding transverse and longitudinal fluctuation ranges of a certain target image, the transverse crack is wholly +.>Very small, longitudinal crack->Very large, but the crazing network cracks are local defects, between them.
(2) And (5) fluctuation degree analysis.
Extracting all wave crest and wave trough values on the row and column and curve, and recording the common wave crest and wave trough values on the row and curveWave crest->、The wave trough->Column and curve are shared +>Wave crest->、/>The wave trough->And unify all the peak-valley coordinates thereof to +.>、/>、/>、/>The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>For the size of the peak on the line and curve, +.>For the size of the valley on the row and curve, < >>For the number of peaks on the line sum curve, +.>For the number of valleys on the rows and curves, +.>For the size of the peak on the column sum curve, +.>For the size of the valley on the column and curve, < >>For the number of peaks on the column sum curve, +.>Is the number of valleys on the columns and curves.
Peak difference: the row and upper average peak values of the transverse cracks are far greater than the column and upper average peak values, the longitudinal cracks are diametrically opposed, and the degree of crack reticulate pattern crack differentiation is not obvious. The degree of difference between the average peak and the column and average peak is noted as follows:
wherein,is the degree of difference between the row and the average peak of a certain target image and the column and the average peak.
Fluctuation difference: the horizontal crack has a much greater fluctuation degree than the column and curve, whereas the longitudinal crack has a similar fluctuation degree. In the embodiment, the fluctuation ratio is calculated by using the number of peaks and the number of troughs, and the formula is as follows:
wherein,、/>curves respectively->、/>Is a fluctuation ratio of (1).
Calculating the average peak-to-valley amplitude variation on the row sum, column sum and curve: the abscissa of the wave crest and the wave trough on the line sum curve、/>From as small asLarge arrangement, calculating pixel value +.>、/>Sum of absolute values of differences of (2) to obtain +.>The average peak and trough amplitude variation on the row and curve is:
wherein,the absolute values of the differences between all adjacent peaks and valleys on the line sum curve are summed.
Similarly, the average peak-to-valley amplitude variation on the column and curve is obtained as:
wherein,the absolute values of the differences between all adjacent peaks and valleys on the column sum curve are summed.
Calculating the fluctuation degree of the row sum, the column sum and the curve according to the fluctuation ratio and the average peak-to-valley amplitude variation:
wherein the method comprises the steps ofFor the degree of fluctuation of the rows and curves, +.>Is the degree of fluctuation of the columns and curves.
The cracks can be divided into micro cracks, small cracks, medium cracks and large cracks according to the sizes, the corresponding size intervals are 0-5mm, 5-15mm, 15-25mm and more than 25mm respectively, the corresponding repairing workers are micro, small and medium cracks, slotting and crack cleaning can be carried out, the sealing glue can be filled for crack filling, and the large cracks of the asphalt pavement are repaired by adopting an asphalt thermal regeneration repairing process. Note that the ratio of the real size to the pixels is in units of(unit: mm/pixel), and then the slit width of the slit was estimated.
Estimating the total area of the slit pixels: adding the ordinate values on the columns and the curves, and estimating the number of pixels occupied by the total area of the crack as follows:
estimating the crack width: extracting edges by Canny edge detection on the target image, so that two edges are reserved in a gap, and the total value of edge pixels is recorded asThen the seam width estimation formula is:
wherein,for the purpose ofThe slit width on the target image.
According to the analysis result, judging the crack type and repairing and suggesting, the concrete contents are as follows:
judging the crack type: if the crack characteristics are satisfied、/>And->Then it is determined that a transverse crack exists at the corresponding position of the target image; if the crack characteristics meet->、/>And->Then it is determined that a longitudinal crack exists at the position corresponding to the target image; if the first two conditions are not satisfied, it is determined that a cracked network crack exists at a position corresponding to the target image.
Repair advice: if it isThen it will be recommended to open the slot and pressure pour the sealant to fill the slot; if it isIt is recommended to repair by using asphalt thermal regeneration repair technology.
According to the above process, the embodiment realizes the judgment of the crack type and the estimation of the crack width in each target image, namely the detection of the road traffic engineering pavement quality, and the targeted maintenance of the target pavement can be realized based on the obtained crack type and the crack width.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the application and are intended to be included within the scope of the application.

Claims (6)

1. The road surface quality detection method for the highway traffic engineering is characterized by comprising the following steps of:
acquiring each frame of shooting image corresponding to a target pavement, and carrying out graying and reversing treatment on each frame of shooting image corresponding to the target pavement to obtain each frame of gray reversed image corresponding to the target pavement;
performing enhancement and binarization processing on each frame of gray level reverse image corresponding to the target pavement, judging whether each frame of image is a suspected crack image according to the row and column data corresponding to each frame of processed image, and if the suspected crack image exists in each frame of processed image corresponding to the target pavement and the suspected crack image of the adjacent frame does not exist, marking each suspected crack image as each target image; if the adjacent frame suspected crack images exist, performing image stitching processing on the adjacent frame suspected crack images to obtain 1 or more target images, wherein each target image comprises a complete crack;
the method for acquiring the target image comprises the following steps:
for a section of adjacent image sequence, recording two adjacent images as a first image and a second image, obtaining pixel values and positions thereof corresponding to all peaks and valleys on each row and curve of the first image and the second image, forming a row and pixel value sequence from small to large according to row positions, performing coincidence matching on the row and pixel value sequences of the first image and the second image, and requiring the coincidence on the rows and curves of the two imagesThe sequence combining parts are all continuous; extracting the maximum peak value and its row coordinate in the overlapped part, and marking the peak value and row position on the first image as、/>Marked +.>、/>Wherein->Splicing the first image and the second image to form a new image, wherein the longitudinal dimension of the new image is +.>The method comprises the steps of carrying out a first treatment on the surface of the Splicing the new image with all the rest adjacent images in a section of adjacent image sequence for multiple times to obtain a target image of the section of adjacent image sequence; obtaining target images of all adjacent image sequences;
calculating crack type judging indexes corresponding to each target image, and judging the crack type of each target image according to the crack type judging indexes; the crack type judging indexes comprise the ratio of the transverse and longitudinal fluctuation range, the difference degree of the average wave peak and the column and the average wave peak, the fluctuation degree of the row and the curve and the fluctuation degree of the column and the curve; estimating the crack width according to the number of pixel points occupied by the total area of the crack on each target image and the number of edge pixel points;
calculating the corresponding row and curve fluctuation degree and column and curve fluctuation degree of each target image by using the following formula:
wherein,for the degree of fluctuation of the line and curve of a certain target image, +.>For the degree of fluctuation of the column and curve of the target image, +.>For the fluctuation ratio of the line and curve of the target image, +.>For the fluctuation ratio of the column and curve of the target image, +.>For the average peak-to-valley amplitude variation on the line and curve of the target image,/for>For the average peak-to-valley amplitude variation on the column and curve of the target image,/for>Summing the absolute values of the differences between all adjacent peaks and valleys on the line and curve of the target image, +.>Summing the absolute values of the differences between the columns of the target image and all adjacent peaks and valleys on the curve, +.>For the number of peaks on the line and curve of the target image,/->For the number of peaks on the columns and curves of the target image,/->For the target imageLower limit value of row coordinates, +.>For +.>Upper limit value of row coordinates of (c); />For the target imageLower limit value of column coordinates of>To the object ofIn the image +.>Upper limit value of column coordinates of>Is->Sum of row pixel values +.>Is->A sum of column pixel values;
and judging whether each frame image is a suspected crack image according to the row and column data corresponding to each frame image after processing, comprising the following steps:
and carrying out accumulation analysis on pixel values in each frame of binarized image according to the directions of rows and columns, and calculating a row and column white pixel accumulation curve, wherein the formula is as follows:
wherein,accumulating the +.f in the sequence for the line pixels corresponding to the binarized image of a frame>Line sum, line sum->For the corresponding column gray scale of the frame binarized imageAccumulation of the>Column sum, head>The total number of lines and the total number of columns corresponding to the image after the frame binarization are adopted; />Representing the +.>Line->Pixel values of column pixels;
if continuous satisfaction in a certain frame of binarized imageThe number of lines of (2) is greater than->Or/and continuously satisfy->The number of columns of (2) is greater than->And judging the image binarized by the frame as a suspected crack image.
2. The method for detecting the quality of road surfaces in highway traffic engineering according to claim 1, wherein the ratio of the lateral and longitudinal fluctuation ranges corresponding to each target image is calculated by using the following formula:
wherein the method comprises the steps of，For the ratio of the corresponding lateral-longitudinal fluctuation range of a certain target image, +.>For +.>Lower limit value of row coordinates, +.>For +.>Upper limit value of row coordinates of (c); />For +.>Lower limit value of column coordinates of>For +.>Upper limit value of column coordinates of>Is->Sum of row pixel values +.>Is->Row pixel valuesAnd, a method for producing the same.
3. The method for detecting the quality of road surfaces in highway traffic engineering according to claim 1, wherein the degree of difference between the average peak and the column and the average peak corresponding to each target image is calculated by using the following formula:
wherein,for the degree of difference of the row and the average peak from the column and the average peak of a certain target image +.>For the size of the peak on the line and curve of the target image, +.>For the number of peaks on the line and curve of the target image,/->For the size of the peak on the column and curve of the target image, +.>Is the number of peaks on the columns and curves of the target image.
4. The method for detecting the quality of road traffic engineering pavement according to claim 1, wherein the crack types include transverse cracks, longitudinal cracks and crazing net-shaped cracks.
5. The method for detecting the quality of road surfaces in highway traffic engineering according to claim 4, wherein said judging the type of the crack of each target image according to the crack type judging index comprises:
if the ratio of the transverse and longitudinal fluctuation ranges isDegree of difference between average peak and column and average peak +.>And the degree of fluctuation of the row and curve->Column and curve degree of fluctuation->Judging that the transverse crack exists at the corresponding position of the target image;
if the ratio of the transverse and longitudinal fluctuation ranges isDegree of difference between average peak and column and average peak +.>And the degree of fluctuation of the row and curve->Column and curve degree of fluctuation->Judging that a longitudinal crack exists at the corresponding position of the target image;
if the first two conditions are not satisfied, it is determined that a crazing network crack exists at a position corresponding to the target image.
6. The method for detecting the quality of road surfaces in highway traffic engineering according to claim 1, wherein estimating the crack width according to the number of pixels occupied by the total area of the crack and the number of edge pixels on each target image comprises:
adding the ordinate values on the columns and the curves, and estimating the number of pixels occupied by the total area of the crack as follows:
wherein,the number of pixels occupied by the total area of the slits of a certain target image, +.>For the total number of columns of the target image, +.>Is the firstA sum of column pixel values;
extracting edges by Canny edge detection on the target image, and recording the total value of edge pixels asThen the seam width estimation formula is:
wherein,for the slit width of the slit on the target image, +.>Is a proportional unit to the real size of the pixel.
CN202211195054.9A 2022-09-29 2022-09-29 Road surface quality detection method for highway traffic engineering Active CN115311260B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211195054.9A CN115311260B (en) 2022-09-29 2022-09-29 Road surface quality detection method for highway traffic engineering

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211195054.9A CN115311260B (en) 2022-09-29 2022-09-29 Road surface quality detection method for highway traffic engineering

Publications (2)

Publication Number Publication Date
CN115311260A CN115311260A (en) 2022-11-08
CN115311260B true CN115311260B (en) 2023-12-08

Family

ID=83866080

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211195054.9A Active CN115311260B (en) 2022-09-29 2022-09-29 Road surface quality detection method for highway traffic engineering

Country Status (1)

Country Link
CN (1) CN115311260B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116823808B (en) * 2023-08-23 2023-11-17 青岛豪迈电缆集团有限公司 Intelligent detection method for cable stranded wire based on machine vision
CN117893543B (en) * 2024-03-18 2024-05-10 辽宁云也智能信息科技有限公司 Visual-assistance-based pavement crack intelligent detection method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108038883A (en) * 2017-12-06 2018-05-15 陕西土豆数据科技有限公司 A kind of Crack Detection and recognition methods applied to highway pavement video image
CN112419250A (en) * 2020-11-13 2021-02-26 湖北工业大学 Pavement crack digital image extraction, crack repair and crack parameter calculation method
CN114596551A (en) * 2022-03-03 2022-06-07 湘潭大学 Vehicle-mounted forward-looking image crack detection method
CN114782428A (en) * 2022-06-17 2022-07-22 南通格冉泊精密模塑有限公司 Injection molding bubble defect detection and cause analysis method and artificial intelligence system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108038883A (en) * 2017-12-06 2018-05-15 陕西土豆数据科技有限公司 A kind of Crack Detection and recognition methods applied to highway pavement video image
CN112419250A (en) * 2020-11-13 2021-02-26 湖北工业大学 Pavement crack digital image extraction, crack repair and crack parameter calculation method
CN114596551A (en) * 2022-03-03 2022-06-07 湘潭大学 Vehicle-mounted forward-looking image crack detection method
CN114782428A (en) * 2022-06-17 2022-07-22 南通格冉泊精密模塑有限公司 Injection molding bubble defect detection and cause analysis method and artificial intelligence system

Also Published As

Publication number Publication date
CN115311260A (en) 2022-11-08

Similar Documents

Publication Publication Date Title
CN115311260B (en) Road surface quality detection method for highway traffic engineering
CN104792792B (en) A kind of road surface crack detection method of Stepwise Refinement
US10489903B2 (en) Stepwise refinement detection method for pavement cracks
CN113362285B (en) Steel rail surface damage fine-grained image classification and detection method
Liu et al. Novel approach to pavement cracking automatic detection based on segment extending
CN103886594B (en) Road surface line laser rut detection and recognition methods and disposal system
CN108038883A (en) A kind of Crack Detection and recognition methods applied to highway pavement video image
CN106529593B (en) Pavement disease detection method and system
CN110503637B (en) Road crack automatic detection method based on convolutional neural network
CN109410205B (en) Crack extraction method under complex pavement background
CN107610092B (en) Pavement crack dynamic detection method based on video stream
CN111507971A (en) Tunnel surface defect detection method
Staniek Detection of cracks in asphalt pavement during road inspection processes
CN106023226A (en) Crack automatic detection method based on three-dimensional virtual pavement
Chua et al. Simple procedure for identifying pavement distresses from video images
CN117094914A (en) Smart city road monitoring system based on computer vision
CN115082377A (en) Building surface crack geometric parameter measuring method and system based on unmanned aerial vehicle
CN117474870B (en) Road pavement crack identification decision-making method and system based on big data screening
Wang et al. Advanced crack detection and quantification strategy based on CLAHE enhanced DeepLabv3+
CN117495759A (en) Sleeper crack detection method
CN110969103B (en) Method for measuring length of highway pavement disease based on PTZ camera
CN114418950A (en) Road disease detection method, device, equipment and storage medium
CN112836090A (en) Visibility detection method based on social video
Huang et al. Massively parallel computing techniques might improve highway maintenance
CN118298293A (en) Dynamic tracking method and system for evolution characteristics of highway tunnel lining crack

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
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20231030

Address after: 010000 No.11 Zhanbei Road, Huimin District, Hohhot City, Inner Mongolia Autonomous Region

Applicant after: Inner Mongolia Highway Engineering Consulting and Supervision Co.,Ltd.

Address before: Room 401, No. 42, Guangzhou Road, Development Zone, Nantong City, Jiangsu Province, 226000

Applicant before: Nantong Yiyun Zhilian Information Technology Co.,Ltd.

GR01 Patent grant
GR01 Patent grant