CN106706647B - Road crack size estimation method based on mobile phone photograph - Google Patents

Road crack size estimation method based on mobile phone photograph Download PDF

Info

Publication number
CN106706647B
CN106706647B CN201611230287.2A CN201611230287A CN106706647B CN 106706647 B CN106706647 B CN 106706647B CN 201611230287 A CN201611230287 A CN 201611230287A CN 106706647 B CN106706647 B CN 106706647B
Authority
CN
China
Prior art keywords
crackle
mobile phone
photo
matrix
road
Prior art date
Application number
CN201611230287.2A
Other languages
Chinese (zh)
Other versions
CN106706647A (en
Inventor
於志文
孔莹莹
陈荟慧
郭斌
Original Assignee
西北工业大学
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 西北工业大学 filed Critical 西北工业大学
Priority to CN201611230287.2A priority Critical patent/CN106706647B/en
Publication of CN106706647A publication Critical patent/CN106706647A/en
Application granted granted Critical
Publication of CN106706647B publication Critical patent/CN106706647B/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using infra-red, visible or ultra-violet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8851Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using infra-red, visible or ultra-violet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8851Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854Grading and classifying of flaws
    • G01N2021/8874Taking dimensions of defect into account
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using infra-red, visible or ultra-violet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/8851Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10004Still image; Photographic image

Abstract

The technical issues of road crack size estimation method based on mobile phone photograph that the invention discloses a kind of, the practicability is poor for solving existing road crack size estimation method.Technical solution is to take pictures first with smart phone to road crackle, mobile phone sensor data when acquiring crackle photo and taking pictures.Collected crackle photo is carried out to include median filter smoothness of image, the sharpening of sobel operator and gray proces, then image segmentation is carried out using OTSU algorithm, crackle is partitioned into from the background of road surface, calculate area of the crackle in photo, determine that mobile phone camera to the distance of crackle, calculates the actual physical size of crackle.The present invention does not need dedicated crack detection vehicle and takes pictures to crackle, and the actual physical size of crackle is calculated merely with the convex lens imaging principle of mobile phone camera, and practicability is good.And when calculating crack width, root-mean-square error is 5.2cm, and when counting crack length, root-mean-square error is 17.1cm.

Description

Road crack size estimation method based on mobile phone photograph

Technical field

The present invention relates to a kind of road crack size estimation method, in particular to a kind of road crackle based on mobile phone photograph Size estimation method.

Background technique

In recent years, the development of smart phone provides new means for traditional municipal services.In traditional urban road In crack detecting method, needs the personnel of profession to check using the equipment of profession, waste a large amount of man power and material in this way. With the development of smart phone, common citizen can be used mobile phone and take pictures to crackle, but due to mobile phone photograph when It waits, the direction taken pictures and the distance taken pictures all are arbitrary, we use the sensor information of mobile phone when taking pictures in the present invention, such as The information of accelerometer, magnetometer, barometer etc. and photo itself estimates road flaw size.

" the pavement distress survey technical research Wuhan University of Technology based on image procossing, 2013 " disclose one kind to document The estimation method of road surface crack size.The method fracture first is refined, and the skeleton in crack is obtained, and is made to the skeleton of crackle Element marking, then the length and width in crack can be obtained by way of scanning and counting.But since the method is using dedicated Crack detection vehicle take pictures to crackle, the distance and angle of camera to crackle be it is determining, so the reality of crackle The proportionate relationship of physical size and crackle between the size of the picture in photo is determining.

In conclusion angle and distance when judging crack size using mobile phone photograph, due to camera from crackle Be it is uncertain, pass through the size of photo only available crackle picture.

Summary of the invention

In order to overcome the shortcomings of existing road crack size estimation method, the practicability is poor, and the present invention provides a kind of based on mobile phone The road crack size estimation method taken pictures.This method takes pictures to road crackle first with smart phone, acquires crackle Photo and mobile phone sensor data when taking pictures.Collected crackle photo is carried out to include median filter smoothness of image, sobel operator Then sharpening and gray proces carry out image segmentation using OTSU algorithm, crackle is partitioned into from the background of road surface, calculate crackle and exist Area in photo determines that mobile phone camera to the distance of crackle, calculates the actual physical size of crackle.The present invention does not need Dedicated crack detection vehicle takes pictures to crackle, and the reality of crackle is calculated merely with the convex lens imaging principle of mobile phone camera Physical size, practicability are good.

The technical solution adopted by the present invention to solve the technical problems: a kind of road crack size based on mobile phone photograph is estimated Meter method, its main feature is that the following steps are included:

Step 1: taken pictures using smart phone to road crackle, mobile phone sensing when acquiring crackle photo and taking pictures Device data.Mobile phone baroceptor, acceleration transducer and magnetometric sensor are as calculating mobile phone camera with a distance from crackle Data foundation.By all mobile phone sensor data preparations at triple<crackle photo, sensor name, sensor values>.Its Middle acceleration transducer, magnetometric sensor sensor values format be a three-dimensional vector<x, y, z>, baroceptor is hand The atmospheric pressure value of machine position<baro>.

Step 2: to crackle photo collected in step 1 carry out include median filter smoothness of image, sobel operator sharpen and Then gray proces carry out image segmentation using OTSU algorithm, are partitioned into crackle from the background of road surface.Treated binaryzation square Battle array is denoted as G.

Step 3: carrying out the processing of mathematical morphology make-before-break to binaryzation matrix G, treated, and matrix is denoted as M, then Dilation erosion and Refinement operation are carried out to matrix M, extract crackle skeleton, treated, and matrix is denoted as N.

Step 4: for pattern cracking, the every a line for the matrix M that scanning step three obtains finds the vertical seat of maximum of crackle Punctuate y1With minimum ordinate point y2.Each column of scan matrix M find the maximum abscissa point x of crackle1With minimum abscissa Point x2.Area s of the pattern cracking in photo is calculated using formula (1):

S=(y1-y2)*(x1-x2) (1)

It is split for perpendicular, every a line of scan matrix M calculates the crackle pixel number of every a line, multiplied by each pixel Length, the as crack width of the row.Then the width of the every a line crackle acquired is averaging to obtain the perpendicular average width split Degree.When seeking the perpendicular length split, the matrix N that scanning step three obtains calculates the pixel number of crackle skeleton, multiplied by each picture The length of element, the as perpendicular length split.Each column of scan matrix M, obtain the crack width of the column, and averaging obtains transverse fissure Mean breadth.Scan matrix N obtains the length of transverse fissure.

Step 5: calculate camera to crackle distance three parameters, the vertical range h of mobile phone to ground, mobile phone The left rotation and right rotation angle beta of tilt fore and aft α, mobile phone.Pass through the distance od of formula (2) calculating camera to crackle:

Od=h/ (cos | α | cos | β |) (2)

The vertical range h on mobile phone to ground is calculated by formula (3), pmIt is the atmospheric pressure value of mobile phone position, passes through hand The baroceptor of machine obtains, pgIt is the atmospheric pressure value on ground:

The tilt fore and aft α of mobile phone, angle beta is rotated left and right by acceleration transducer and magnetometric sensor calculating, pass through Android develops API and obtains.

Step 6: obtaining size of the crackle in photo according to step 4, step 5 obtains camera to crackle The focal length of distance and mobile phone derives that formula (6) calculate crackle actual physics ruler by convex lens formula (4) (5) Very little, d is object distance, and i is image distance, and f is focal length, and imagesize is the size of picture, and physicalsize is the actual physics ruler of crackle It is very little.

The beneficial effects of the present invention are: this method takes pictures to road crackle first with smart phone, crackle is acquired Photo and mobile phone sensor data when taking pictures.Collected crackle photo is carried out to include median filter smoothness of image, sobel operator Then sharpening and gray proces carry out image segmentation using OTSU algorithm, crackle is partitioned into from the background of road surface, calculate crackle and exist Area in photo determines that mobile phone camera to the distance of crackle, calculates the actual physical size of crackle.The present invention does not need Dedicated crack detection vehicle takes pictures to crackle, and the reality of crackle is calculated merely with the convex lens imaging principle of mobile phone camera Physical size, practicability are good.And when calculating crack width, root-mean-square error is 5.2cm, when counting crack length, Root-mean-square error is 17.1cm.

It elaborates with reference to the accompanying drawings and detailed description to the present invention.

Detailed description of the invention

Fig. 1 is the flow chart of the road crack size estimation method the present invention is based on mobile phone photograph.

Specific embodiment

Referring to Fig.1.The present invention is based on the road crack size estimation method of mobile phone photograph, specific step is as follows:

Step 1: being taken pictures using smart phone to road crackle, mobile phone when acquiring crackle photo and take pictures is passed Sensor data.The number of baroceptor, acceleration transducer and magnetometric sensor as calculating mobile phone camera with a distance from crackle According to foundation.And further by all data preparations at triple<crackle photo, sensor name, sensor values>.Wherein plus Velocity sensor, magnetometric sensor sensor values format be a three-dimensional vector<x, y, z>, baroceptor is mobile phone institute Atmospheric pressure value in position<baro>.

Step 2: carrying out image procossing to crackle photo collected in step 1.Image is carried out to photo first to locate in advance Reason, including median filter smoothness of image, sobel operator sharpen, gray proces, then image segmentation are carried out using OTSU algorithm, from road Crackle is partitioned into the background of face.Treated, and binaryzation matrix is denoted as G.

Step 3: being carried out at mathematical morphology make-before-break to be precisely calculated flaw size in photo to matrix G Reason, treated, and matrix is denoted as M, then carries out dilation erosion and Refinement operation to matrix M, extracts crackle skeleton, treated Matrix is denoted as N.

Step 4: calculating size of the crackle in photo.For pattern cracking, matrix M that scanning step three obtains Every a line, find the maximum ordinate point y of crackle1With minimum ordinate point y2.The each column for scanning M, find the maximum of crackle Abscissa point x1With minimum abscissa point x2.Area s of the pattern cracking in photo is calculated using formula (1):

S=(y1-y2)*(x1-x2) (1)

It is split for perpendicular, every a line of scan matrix M calculates the crackle pixel number of every a line, multiplied by each pixel Length, the as crack width of the row.Then the width of the every a line crackle acquired is averaging to obtain the perpendicular average width split Degree.When seeking the perpendicular length split, the matrix N that scanning step three obtains calculates the pixel number of crackle skeleton, multiplied by each picture The length of element, the as perpendicular length split.Similarly, each column of scan matrix M, obtain the crack width of the column, and averaging obtains The mean breadth of transverse fissure.Scan matrix N obtains the length of transverse fissure.

Step 5: mobile phone camera is to crackle determination of distance.The distance for calculating camera to crackle needs three parameters, Left rotation and right rotation angle beta of the mobile phone to the vertical range h on ground, the tilt fore and aft α of mobile phone, mobile phone.It is calculated and is imaged by formula (2) Head arrives the distance od of crackle:

Od=h/ (cos | α | cos | β |) (2)

The vertical range h on mobile phone to ground is calculated by formula (3), pmIt is the atmospheric pressure value of mobile phone position, passes through hand The baroceptor of machine is available, pgIt is the atmospheric pressure value on ground:

The tilt fore and aft α of mobile phone, angle beta is rotated left and right by acceleration transducer and magnetometric sensor calculating, pass through Android develops API and obtains.

Step 6: the calculating of crackle actual physical size.The principle of taking pictures of mobile phone camera is convex lens imaging principle, step Rapid four obtained size of the crackle in photo, that is, as size, step 5 obtained camera to crackle away from From, i.e. object distance, the focal length of mobile phone be it is known, by convex lens formula (4) (5), it is real to derive that formula (6) calculate crackle Border physical size, d are object distances, and i is image distance, and f is focal length, and imagesize is the size of picture, and physicalsize is the reality of crackle Border physical size.

Claims (1)

1. a kind of road crack size estimation method based on mobile phone photograph, it is characterised in that the following steps are included:
Step 1: taken pictures using smart phone to road crackle, mobile phone sensor number when acquiring crackle photo and taking pictures According to;The data of mobile phone baroceptor, acceleration transducer and magnetometric sensor as calculating mobile phone camera with a distance from crackle Foundation;By all mobile phone sensor data preparations at triple<crackle photo, sensor name, sensor values>;Wherein plus Velocity sensor, magnetometric sensor sensor values format be a three-dimensional vector<x, y, z>, baroceptor is mobile phone institute Atmospheric pressure value in position<baro>;
Step 2: carrying out including median filter smoothness of image, the sharpening of sobel operator and gray scale to crackle photo collected in step 1 Then processing carries out image segmentation using OTSU algorithm, is partitioned into crackle from the background of road surface;Binaryzation matrix note that treated For G;
Step 3: carrying out the processing of mathematical morphology make-before-break to binaryzation matrix G, treated, and matrix is denoted as M, then to square Battle array M carries out dilation erosion and Refinement operation, extracts crackle skeleton, treated, and matrix is denoted as N;
Step 4: for pattern cracking, the every a line for the matrix M that scanning step three obtains finds the maximum ordinate point y of crackle1 With minimum ordinate point y2;Each column of scan matrix M find the maximum abscissa point x of crackle1With minimum abscissa point x2;Make Area s of the pattern cracking in photo is calculated with formula (1):
S=(y1-y2)*(x1-x2) (1)
It is split for perpendicular, every a line of scan matrix M calculates the crackle pixel number of every a line, multiplied by the length of each pixel Degree, the as crack width of the row;Then the width of the every a line crackle acquired is averaging to obtain the perpendicular mean breadth split;It asks When the perpendicular length split, the matrix N that scanning step three obtains calculates the pixel number of crackle skeleton, multiplied by each pixel Length, the as perpendicular length split;Similarly, each column of scan matrix M, obtain the crack width of the column, and averaging obtains transverse fissure Mean breadth;Scan matrix N obtains the length of transverse fissure;
Step 5: calculate camera to crackle distance three parameters, the front and back of the vertical range h of mobile phone to ground, mobile phone Inclined angle alpha, the left rotation and right rotation angle beta of mobile phone;Pass through the distance od of formula (2) calculating camera to crackle:
Od=h/ (cos | α | cos | β |) (2)
The vertical range h on mobile phone to ground is calculated by formula (3), pmIt is the atmospheric pressure value of mobile phone position, passes through the gas of mobile phone Pressure sensor obtains, pgIt is the atmospheric pressure value on ground:
The tilt fore and aft α of mobile phone, angle beta is rotated left and right by acceleration transducer and magnetometric sensor calculating, pass through android API is developed to obtain;
Step 6: obtaining size of the crackle in photo according to step 4, step 5 obtains camera to the distance of crackle And the focal length of mobile phone derives that formula (6) calculate crackle actual physical size, d is by convex lens formula (4) (5) Object distance, i are image distances, and f is focal length, and imagesize is the size of picture, and physicalsize is the actual physical size of crackle;
CN201611230287.2A 2016-12-28 2016-12-28 Road crack size estimation method based on mobile phone photograph CN106706647B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611230287.2A CN106706647B (en) 2016-12-28 2016-12-28 Road crack size estimation method based on mobile phone photograph

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611230287.2A CN106706647B (en) 2016-12-28 2016-12-28 Road crack size estimation method based on mobile phone photograph

Publications (2)

Publication Number Publication Date
CN106706647A CN106706647A (en) 2017-05-24
CN106706647B true CN106706647B (en) 2019-05-07

Family

ID=58895550

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611230287.2A CN106706647B (en) 2016-12-28 2016-12-28 Road crack size estimation method based on mobile phone photograph

Country Status (1)

Country Link
CN (1) CN106706647B (en)

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5317586B2 (en) * 2008-08-28 2013-10-16 ラピスセミコンダクタ株式会社 Camera module and manufacturing method thereof
US9303890B2 (en) * 2013-03-27 2016-04-05 Russell Haines Intelligent HVAC register airflow control system
CN104089580A (en) * 2014-06-26 2014-10-08 华南理工大学 Concrete surface crack width measuring instrument and method based on smart phone
CN104316601A (en) * 2014-10-29 2015-01-28 上海斐讯数据通信技术有限公司 Mobile phone with floor internal defect detection function and working principle of mobile phone
CN204288301U (en) * 2014-12-19 2015-04-22 中国科学院武汉岩土力学研究所 A kind of fracture width change monitoring device
CN105160695B (en) * 2015-06-30 2019-02-01 Oppo广东移动通信有限公司 A kind of photo processing method and mobile terminal

Also Published As

Publication number Publication date
CN106706647A (en) 2017-05-24

Similar Documents

Publication Publication Date Title
CN105260699B (en) A kind of processing method and processing device of lane line data
Vijayarani et al. Performance analysis of canny and sobel edge detection algorithms in image mining
US8625851B2 (en) Measurement apparatus, measurement method, and feature identification apparatus
Assali et al. Surveying and modeling of rock discontinuities by terrestrial laser scanning and photogrammetry: Semi-automatic approaches for linear outcrop inspection
Novatnack et al. Scale-dependent/invariant local 3D shape descriptors for fully automatic registration of multiple sets of range images
CA2723875C (en) Registration of street-level imagery to 3d building models
CN105346706B (en) Flight instruments, flight control system and method
Javernick et al. Modeling the topography of shallow braided rivers using Structure-from-Motion photogrammetry
CN101699313B (en) Method and system for calibrating external parameters based on camera and three-dimensional laser radar
CN104330022B (en) Method and system is determined using the volume of structure from motion algorithm
US20090154793A1 (en) Digital photogrammetric method and apparatus using intergrated modeling of different types of sensors
CN103913375B (en) A kind of tensile sample distortion measurement method relevant based on digital picture
CN105823416B (en) The method and apparatus of polyphaser measurement object
CN104359403B (en) Planar part dimension measurement method based on sub-pixel edge algorithm
CN103426165A (en) Precise registration method of ground laser-point clouds and unmanned aerial vehicle image reconstruction point clouds
CN105354531B (en) A kind of mask method of face key point
CN104121902B (en) Implementation method of indoor robot visual odometer based on Xtion camera
JP2012503817A (en) Method and composition for blurring an image
Dugdale et al. Aerial photosieving of exposed gravel bars for the rapid calibration of airborne grain size maps
Gallego et al. A variational stereo method for the three-dimensional reconstruction of ocean waves
CN102005047A (en) Image registration system and method thereof
CN104034269B (en) A kind of monocular vision measuring method and device
CN101673397B (en) Digital camera nonlinear calibration method based on LCDs
EP1474782A1 (en) Method and apparatus for determining symmetry in 2d and 3d images
CN103778626B (en) A kind of fast image registration method of view-based access control model marking area

Legal Events

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