CN115830024A - Bridge inhaul cable micro-motion vibration detection method based on image segmentation - Google Patents
Bridge inhaul cable micro-motion vibration detection method based on image segmentation Download PDFInfo
- Publication number
- CN115830024A CN115830024A CN202310120011.2A CN202310120011A CN115830024A CN 115830024 A CN115830024 A CN 115830024A CN 202310120011 A CN202310120011 A CN 202310120011A CN 115830024 A CN115830024 A CN 115830024A
- Authority
- CN
- China
- Prior art keywords
- cable
- frequency
- measuring point
- area
- bridge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention provides a bridge cable micro-motion vibration detection method based on image segmentation, which is a detection method based on machine vision, and compared with the traditional sensor method, the detection method has the advantages of remarkably reducing the cost and higher precision, realizes the motion signal extraction and the motion signal analysis of a cable through an image segmentation and gray mapping model, extracts the cable fundamental frequency through a mathematical method to calculate the cable tension force, can effectively capture the real-time change of the cable, can shoot from a farther place, can shoot more cables from a farther view angle, and simultaneously reduces the erection environment requirement of a camera.
Description
Technical Field
The invention belongs to the technical field of bridge health monitoring, and particularly relates to a bridge inhaul cable micro-motion vibration detection method based on image segmentation.
Background
With the development of society, urban areas and the number of vehicles rapidly increase, urban traffic systems face great pressure which is increasing day by day, and bridges are important components of the traffic systems and are the most dangerous parts of the traffic systems once the structures are unstable. In some cities with developed water systems, municipal bridges are in junction and core positions in municipal transportation, and once problems occur, the municipal bridges not only cause huge direct loss, but also are more likely to cause personal threats and generate huge influences. The cable-stayed bridge and the suspension cable bridge in the bridge with longer span are main bridge types, the stay cable (suspender) serving as a tension member of the bridge type is one of the main members of the bridge, the working state of the stay cable (suspender) can directly reflect the working state of the bridge, the stay cable force of the stay cable is monitored in real time, and the safety risk of the bridge is effectively reduced.
The traditional inhaul cable force measuring method comprises but is not limited to an oil pressure gauge method, a magnetic flux method and a vibration frequency method, wherein when the oil pressure gauge method is generally used for building a bridge, the magnetic flux method is high in cost and high in installation and maintenance difficulty; the vibration frequency method is divided into two methods: contact measurement and non-contact measurement, in which the contact measurement has a time dispersion, cannot effectively capture real-time changes of a cable (boom).
Most of the existing non-contact measurement methods are based on a machine vision technology, a connection network can analyze the real-time state of a bridge in real time by erecting a camera so as to achieve real-time monitoring of the state of the bridge, and the existing measurement mode of the inhaul cable (suspender) inhaul cable force based on the machine vision technology is as follows: and laying the mark points, capturing the mark points to obtain a time-course diagram, and carrying out signal analysis on the time-course diagram to obtain the frequency. The method has the disadvantages that the mark points are greatly influenced by environmental factors, only a single cable can be measured and calculated, the mark points need to be captured during calculation, and the calculation amount is large.
Disclosure of Invention
The invention aims to provide a bridge inhaul cable micro-motion vibration detection method based on image segmentation.
In order to solve the technical problems, the invention adopts the technical scheme that: a bridge inhaul cable micro-motion vibration detection method based on image segmentation comprises the following specific steps: the method comprises the following steps that S1, a vibration video of a bridge inhaul cable is obtained through image acquisition equipment; s2, intercepting a certain frame in the video as a first frame according to requirements, setting a target area on at least one guy cable in the first frame, wherein each target area only comprises one guy cable and a background, framing at least one measuring point area in each target area, and framing the measuring point area at the edge of the corresponding guy cable; s3, performing foreground and background segmentation on each target area, extracting a foreground gray value and a background gray value respectively, substituting the foreground gray value and the background gray value into a gray mapping model, and realizing the positioning of the cable edge in the measuring point area in the corresponding target area; s4, according to the measuring point area and the target area selected by the first frame, executing the step S3 on the video frame by frame, and recording the inhaul cable edge positioning result of the measuring point area of each frame as a motion signal I n （t/Fs）,I n （t/Fs）=[i n (1/Fs),i n (2/Fs),…,i n (Ts/Fs)]T =1,2, \ 8230, wherein n is the number of a measuring point region, t is the frame number, ts is the total frame number of a video, fs is the video acquisition frequency, and Fourier change is respectively carried out on motion signals of all measuring point regions to obtain corresponding frequency-amplitude data; s5, calculating to obtain the fundamental frequency of the corresponding inhaul cable through frequency-amplitude data, and obtaining the fundamental frequency of the corresponding inhaul cable according to the relation T =4 rho L between the inhaul cable force and the natural vibration frequency of the corresponding inhaul cable 2 h 1 2 To find the pulling force, where T is the pulling force, ρ is the linear density of the cable, L is the calculated length of the cable, and h 1 The fundamental frequency of the inhaul cable is a first-order frequency, the proportional relation between the measured inhaul cable force T and the designed inhaul cable force value B of the inhaul cable is set as w, namely w = T/B, and the inhaul cable force is judged according to the w valueThe state of the cable.
Preferably, in step S3, the foreground and background segmentation is to use edge detection to make an edge mask and multiply the edge mask with the image to obtain a gray distribution at the edge, and calculate a threshold segmentation image based on the gray distribution.
Preferably, in step S3, the grayscale mapping model is a mapping relationship between a position of a cable edge in the image and an average grayscale value in the region, the coverage area of the foreground is obtained by combining the average luminance in the target region with the foreground luminance and the background luminance obtained by segmenting the target region, and the positioning of the cable is realized by a geometric relationship between the coverage area and the edge position.
Preferably, in step S3, the average gray-scale value in the target region is I, and the foreground gray-scale value is I 1 Foreground coverage area of S 1 Background gray value of i 0 Background coverage area of S 0 And the total area is S, then S can be obtained according to the proportional relation 1 =S·（I-i 0 ）/(i 1 -i 0 ) Front view coverage area S 1 When the foreground is known to be rectangular, the edge position is located through the edge angle, the included angle between the edge and the area is set to be alpha, when the alpha =0, the shape of the coverage area is rectangular, and the formula of the cable position d at this time is d = S 1 /width, wherein width represents the width of the coverage area; when alpha is not equal to 0, the mapping relation between the coverage area and the position of the inhaul cable is divided into three types of triangle, trapezoid and pentagon according to different coverage shapes, and if Stan alpha/2 = k, S is determined 1 The mapping relation with d is as follows:
preferably, in step S4, fourier transform is performed on the motion signal of the nth measuring point region to obtain the amplitude a of the nth measuring point region n (f p )，A n (f p )=[α n (f 1 ),α n (f 2 ),…,α n (f p )]P =1,2, \ 8230, P, where P denotes the number of points of FFT, f p Frequency component representing a Fourier transform, f p =Fs*p/2*P，α n Representing an and frequencyRate-related variable, α n (f 1 ),α n (f 2 ),…,α n (f p ) Fourier signals A representing motion signals of n number of measuring point areas respectively n Frequency f in 1 、f 2 、…、f p Corresponding amplitudes, frequency-amplitude data is obtained.
Preferably, in step S5, the step of calculating the fundamental frequency of the corresponding cable through the frequency-amplitude data specifically includes: s51, selecting the frequency corresponding to the maximum value by segmenting the frequency amplitude data of each measuring point area, wherein the specific formula is
f n1 =Max{α n (f a1 ),α n (f a1+1 ),…,α n (f b1 )},0.5≤f a1 ∩f b1 ≤0.5+d，
f n2 =Max{α n (f a2 ),α n (f a2+1 ),…,α n (f b2 )},f n1 ＜f a2 ∩f b2 ≤f n1 +d，
…,
f nq =Max{α n (f aq ),α n (f aq+1 ),…,α n (f bq )},f nq-1 ＜f aq ∩f bq ≤f nq-1 + d, where d is the step size, q is the frequency order, f nq Is the q-order natural frequency of the n number measuring point region, and is in the interval [0.5,0.5+ d +]Fundamental frequency h of inhaul cable at n-number measuring point area is searched internally n1 The q-th order frequency is the interval (f) nq-1 ,f nq-1 +d]The frequency corresponding to the peak in (1); step S52, differentiating the q-order natural frequency of the measuring point region n, wherein the difference formula is ^ n (e)={(f n2 -f n1 ),(f n3 -f n2 ),…,(f ne+1 -f ne ) Q-1 is less than or equal to e, wherein e is a positive integer; step S53, approximating the average value of the difference to the fundamental frequency of the inhaul cable,
(ii) a Step S54, matching the guy cable fundamental frequency of another measuring point area of the same guy cable as a comparison group, setting the relative error delta of the guy cable fundamental frequencies of the two measuring point areas,then
Wherein c is a cable number.
Preferably, when the relative error δ ≦ 0.1 in step S54, the fundamental frequency h of the measurement point area 2c of the cable denoted by symbol c 2c Fundamental frequency h as output 1 (ii) a When the < δ > 0.1 < 0.3 >, the average value of two groups of fundamental frequencies of a measuring point area 2c and a measuring point area 2c-1 on a c-number stay is taken as the output fundamental frequency h 1 I.e. h 1 =(h 2c +h 2c-1 ) 2; when delta>At 0.3, calculating a difference value of the fundamental frequencies of the two groups of the measuring point region 2c and the measuring point region 2c-1 on the guy cable No. c 2c (e) And & 2c-1 (e) And calculating the variance of the two difference values, variance σ 2 Is of the formula
Evaluating the accuracy of the two sets of fundamental frequencies, and selecting the variance σ 2 The fundamental frequency corresponding to the small difference value is used as the output fundamental frequency h 1 。
Preferably, in the step S5, the determining the state of the cable according to the w value specifically includes: when w is less than or equal to 0.8, the cable is seriously damaged, or the management and maintenance department is advised to check the reason and maintain in an abnormal working state; when the wire cable is damaged or the working state is abnormal when the wire cable is less than 0.8 and less than 0.9, recommending a management and maintenance department to pay close attention to checking reasons; when the 0.9< w < 1.05, the working state of the inhaul cable is normal; when the range of 1.05< w < 1.15, the load of the guy cable is large, overweight vehicles or the congestion on the bridge is serious, and the management and maintenance department is advised to pay close attention to the situation; when the load of the stay cable is extremely large and the safety of the bridge is threatened in the condition that the load of the stay cable is less than 1.5 w, the management and maintenance department is advised to determine the working state of the bridge, whether a serious overweight vehicle exists or not, and after the recording, the management and maintenance department is in contact with a traffic department for management or in severe weather, and the traffic department is in contact with for evacuation.
Preferably, the video format of the cable vibration video in step S1 is one of avi, mov and mp 4.
Preferably, the target region and the measurement point region in step S2 are manually selected and set.
The scope of the present invention is not limited to the specific combinations of the above-described features, and other embodiments in which the above-described features or their equivalents are arbitrarily combined are also intended to be encompassed. For example, the above features and the technical features (but not limited to) having similar functions disclosed in this application are replaced with each other to form the technical solution.
Due to the application of the technical scheme, compared with the prior art, the invention has the following advantages:
1. compared with the traditional sensor method, the bridge inhaul cable micro-motion vibration detection method based on image segmentation has the advantages that the cost is remarkably reduced, the precision is higher, the real-time change of an inhaul cable (a suspender) can be effectively captured, the image can be shot from a farther place, more inhaul cables can be shot at a farther visual angle, and meanwhile, the requirement on the erection environment of a camera is lowered; different from most methods based on machine vision, the method does not need to lay a large number of lines or mark points, and obtains the vibration data of a plurality of inhaul cables through a small number of cameras;
2. the invention is a non-contact measuring method, which does not damage the measuring target;
3. the invention enables the computer to identify the frequency spectrum and solve the fundamental frequency, can directly obtain the inhaul cable force of each target inhaul cable according to the video data, and can monitor the inhaul cable force state in real time by using the 5G technology to transmit the video on the premise that the position of the camera is relatively stable.
Drawings
FIG. 1 is a schematic flow chart of a bridge cable micro-motion vibration detection method based on image segmentation;
FIG. 2 is a schematic view of a cable positioning method;
FIG. 3 is a schematic flow chart of step S5 of the bridge cable micro-motion vibration detection method based on image segmentation.
Detailed description of the preferred embodiments
The present invention is further described with reference to the accompanying drawings, and the following examples are only for clearly illustrating the technical solutions of the present invention, and should not be taken as limiting the scope of the present invention.
The method for detecting the micro-motion vibration of the bridge cable based on image segmentation as shown in fig. 1 specifically comprises the following steps.
The method comprises the following steps that S1, an image acquisition device is used for shooting a plurality of guys to obtain a guy vibration video with a target guy to be detected, in the embodiment, a civil camera is used for shooting the guy vibration video, the video format of the guy vibration video is common video formats such as avi, mov and mp4, mark points do not need to be laid on the guys when the video is acquired, and the guys are shot at a slightly far position.
According to the Nyquist sampling theorem, the sampling frequency should be more than twice of the interested highest frequency component, the interested highest frequency component of the embodiment is the q-order frequency of the cable, and the cable fundamental frequency h is taken 1 The frame rate of the camera should be greater than 10 xqxh 1 In this embodiment, q =5,h 1 ≈1,10×q×h 1 Approximately equal to 50, the sampling frequency is 60 frames/second.
S2, intercepting a certain frame in a video as a first frame according to requirements, setting a target area on at least one guy cable in the first frame, wherein each target area only comprises one guy cable and a background, framing at least one measuring point area in each target area, and framing the measuring point area at the edge of the corresponding guy cable; and (3) a smaller measuring point area is framed at the edge of the stay cable in the target area, the size of the measuring point area is smaller than that of the target area, and the measuring point area contains the stay cable and can capture the finished stay cable vibration process.
In the embodiment, the selection is performed according to the selection mode, and in order to ensure the validity of data, two target areas ROI1 and ROI2 are taken at the adjacent positions on the inhaul cable, and two measuring point areas are drawn in each target area.
And S3, performing foreground and background segmentation on each target area, extracting a foreground gray value and a background gray value respectively, and substituting the foreground gray value and the background gray value into a gray mapping model to realize the positioning of the cable edge in the measuring point area in the corresponding target area.
In this embodiment, a classical edge detection canny operator is used as an auxiliary to realize the separation of the foreground and the background by a dual-threshold segmentation method to meet the model requirement in S3, and the dual-threshold segmentation method is specifically shown in fig. 2: identifying the edge of the image by using a canny operator, and multiplying the edge of the image by the image point to obtain the gray distribution at the edge; determining a threshold value according to the gray distribution at the edge and the overall gray distribution of the image; segmenting the image according to the threshold value, wherein the foreground is larger than the large threshold value, and the background is the opposite; and multiplying the threshold separation mask and the image to respectively obtain the foreground and background brightness distribution of the image.
Substituting the foreground gray value and the background gray value into a gray mapping model, obtaining foreground brightness and background brightness by combining with a segmentation image, obtaining the coverage area of the foreground by combining with the average brightness of the target area, realizing the positioning of the stay cable of the target to be detected by the geometric relation between the coverage area and the position of the edge, and realizing the positioning of the edge of the stay cable in the area, wherein the gray mapping model is the mapping relation between the position of the edge of the stay cable in the image and the average gray value in the area.
Assuming that the foreground gray value and the background gray value in the image are both relatively uniform, assuming that the average gray value in the target region is I, the foreground gray value is I 1 Foreground coverage area of S 1 Background gray value of i 0 Background coverage area of S 0 And the total area is S, then S can be obtained according to the proportional relation 1 =S·（I-i 0 ）/(i 1 -i 0 ) Front view coverage area S 1 When the foreground is known to be rectangular, the edge position is located through the edge angle, the included angle between the edge and the area is set to be alpha, when the alpha =0, the shape of the coverage area is rectangular, and the formula of the cable position d at this time is d = S 1 /width, wherein width represents the width of the coverage area; when alpha is not equal to 0, the mapping relation between the coverage area and the position of the inhaul cable is divided into three types of triangle, trapezoid and pentagon according to different coverage shapes, and if Stan alpha/2 = k, S is determined 1 The mapping relation with d is as follows:
。
s4, according to the measuring point area and the target area selected by the first frame, executing the step S3 on the video frame by frame, and recording the inhaul cable edge positioning result of the measuring point area of each frame as a motion signal I n （t/Fs）,I n （t/Fs）=[i n (1/Fs),i n (2/Fs),…,i n (Ts/Fs)]T =1,2, \ 8230;, ts, where n is the number of the measurement point region, t is the frame number, ts is the total frame number of the video, fs is the video acquisition frequency, fourier changes are respectively performed on the motion signals of all the measurement point regions to obtain corresponding frequency-amplitude data, I n (t/Fs) represents a set of one-dimensional signal sets, i n (1/Fs),i n (2/Fs),…,i n Each of (Ts/Fs) represents a single value, and respectively represents the detected motion signals of n measuring point areas at the moments of 1/Fs, 2/Fs, \8230, and the moments of Ts/Fs, fourier transformation is carried out on n groups of motion signals to obtain the amplitude data of n groups of motion signals, and the formula is as follows: a. The n (f p )，A n (f p )=[α n (f 1 ),α n (f 2 ),…,α n (f p )]P =1,2, \ 8230, P, where P denotes the number of points of FFT, f p Frequency component representing a Fourier transform, f p =Fs*p/2*P，α n Representing a frequency-dependent variable, alpha n (f 1 ),α n (f 2 ),…,α n (f p ) Fourier signal A representing motion signals of n-numbered point regions n Frequency f in 1 、f 2 、…、f p And obtaining frequency-amplitude data according to the corresponding amplitude.
S5, carrying out fundamental frequency identification on the amplitude data to obtain a fundamental frequency of the inhaul cable, wherein each order of inherent frequency of the inhaul cable is in linear distribution by taking the fundamental frequency as a difference, and a frequency-amplitude spectrum can be analyzed to find a peak value when the inherent frequency of the inhaul cable is identified; the principle is that the amplitude is maximum when the inhaul cable vibrates at the natural frequency of the inhaul cable, the representation on a frequency amplitude diagram is an obvious peak value, and the mathematical representation corresponds to the situation that the amplitude is large in a certain rangeThe invention utilizes the two characteristics of the guy cable to calculate, match and screen fundamental frequency, and the fundamental frequency is obtained by the relation T =4 rho L 2 h 1 2 To find the pulling force, where T is the pulling force, ρ is the linear density of the cable, L is the calculated length of the cable, and h 1 Is the fundamental frequency of the cable, which is the first order frequency.
In step S5, the step of identifying the fundamental frequency of the amplitude data to obtain the fundamental frequency of the cable specifically includes: step S51, the amplitude data is segmented according to the frequency amplitude data of each measuring point area to select the frequency corresponding to the maximum value, and the specific formula is
f n1 =Max{α n (f a1 ),α n (f a1+1 ),…,α n (f b1 )},0.5≤f a1 ∩f b1 ≤0.5+d，
f n2 =Max{α n (f a2 ),α n (f a2+1 ),…,α n (f b2 )},f n1 ＜f a2 ∩f b2 ≤f n1 +d，
…,
f nq =Max{α n (f aq ),α n (f aq+1 ),…,α n (f bq )},f nq-1 ＜f aq ∩f bq ≤f nq-1 + d, where d is the step size, q is the frequency order, f nq Is the q-order natural frequency of the n number measuring point region, and is in the interval [0.5,0.5+ d +]Fundamental frequency h of inhaul cable at n-number measuring point area is searched internally n1 The q-th order frequency is the interval (f) nq-1 ,f nq-1 +d]The frequency corresponding to the peak in (1); and (5) differencing the q-order inherent frequency of the measuring point region n, wherein the difference formula is n (e)={(f n2 -f n1 ),(f n3 -f n2 ),…,(f ne+1 -f ne ) Q-1 is less than or equal to e, wherein e is a positive integer; step S53, approximating the average value of the difference to the fundamental frequency of the inhaul cable,
。
step S54, matching the fundamental frequency of another measuring point area guy cable of the same guy cable as a comparison group,if the relative error delta of the fundamental frequency of the guy cable in the two measuring point areas is set, then
Wherein c is a cable number.
When the relative error delta in the step S54 is less than or equal to 0.1, the fundamental frequency h of the measuring point area 2c of the No. c inhaul cable 2c Fundamental frequency h as output 1 (ii) a When delta is more than 0.1 and less than or equal to 0.3, the average value of two groups of fundamental frequencies of the c number stay cable upper measuring point area 2c and the measuring point area 2c-1 is taken as the output fundamental frequency h 1 I.e. h 1 =(h 2c +h 2c-1 ) 2; when delta is larger than 0.3, the difference value V of the two groups of fundamental frequencies of the measuring point region 2c and the measuring point region 2c-1 on the No. c inhaul cable needs to be calculated 2c (e) And + 2c-1 (e) And calculating the variance of the two difference values, variance σ 2 Is of the formula
Evaluating the accuracy of the two sets of fundamental frequencies, and selecting the variance σ 2 The fundamental frequency corresponding to the small difference value is used as the output fundamental frequency h 1 。
In this embodiment, referring to fig. 3, taking the analysis process of the No. 1 cable, i.e. the ROI1 region as an example, the specific process of this embodiment is as follows: the frequency corresponding to the maximum amplitude value is selected by the amplitude to the frequency amplitude data of each measuring point area in a segmented manner, the step length is 1, the maximum order is 5,
f 1 =Max{α 1 (f a1 ),α 1 (f a1+1 ),…,α 1 (f b1 )},0.5≤f a1 ∩f b1 ≤1.5，
f 2 =Max{α 1 (f a2 ),α 1 (f a2+1 ),…,α 1 (f b2 )},f 1 ＜f a2 ∩f b2 ≤f 1 +1，
…,
f 5 =Max{α 1 (f a5 ),α 1 (f a5+1 ),…,α 1 (f b5 )},f 4 ＜f a5 ∩f b5 ≤f 4 +1, first in the interval [0.5,0.5+1 +]Interval finding temporary fundamental frequency f 1 5 th, 5 thOrder frequency interval (f) 4 ,f 4 +1]The frequency corresponding to the peak in (1); differentiating the frequencies of each group of signals, taking c as a positive integer, wherein the formula is
▽ 1 (c)={(f 2 -f 1 ),(f 3 -f 2 ),(f 4 -f 3 ),(f 5 -f 4 )},
▽ 2 (c)={(f 2 -f 1 ),(f 3 -f 2 ),(f 4 -f 3 ),(f 5 -f 4 ) Approximating the average value of the difference to the fundamental frequency of the stay cable
(ii) a Matching h 11 ，h 21 The fundamental frequency of the No. 1 cable with high reliability is obtained,
when delta is less than or equal to 0.1, the fundamental frequency h of No. 1 stay cable is directly output 21 When delta is more than 0.1 and less than or equal to 0.3, the average value of two groups of fundamental frequencies of the No. 1 stay cable is directly used as an output value h 1 =(h 21 +h 11 )/2;
When delta is larger than 0.3, two groups of difference values of No. 1 cable need to be calculated 1 、▽ 2 Evaluating the accuracy of the two groups of fundamental frequencies by the variance of the two groups of fundamental frequencies, and selecting the variance sigma 2 The small difference value corresponds to the fundamental frequency as an output value,
，
by the relation T =4 ρ L 2 h 1 2 To find the pulling force, where T is the pulling force, ρ is the linear density of the cable, L is the calculated length of the cable, and h 1 Is the fundamental frequency of the cable.
Judging the working state of the stay cable according to the proportional relation w between the measured stay cable force T and a bridge design value obtained from a bridge design scheme, and when w < =0.8, recommending a management department to check the reason and maintain when the stay cable is seriously damaged or has an abnormal working state; when 0.8< -w < =0.9 and the guy cable is damaged or the working state is abnormal, advising management and maintenance departments to pay close attention to checking reasons; when the 0.9 yarn-woven fabrics w < =1.05, the stay rope working state is normal; when 1.05 woven fabrics w < =1.15 are woven fabrics, the load of the guy cable is large, overweight vehicles or bridges are likely to have serious congestion, and management and maintenance departments are advised to pay close attention to the situation; when the 1.5-straw (w) is used, the load of the guy cable is extremely large, the safety of the bridge is threatened, and the management and maintenance department is advised to determine the working state of the bridge, determine whether serious overweight vehicles exist or not and contact the traffic department after recording.
The above embodiments are only for illustrating the technical idea and features of the present invention, and the purpose of the present invention is to enable those skilled in the art to understand the content of the present invention and implement the present invention, and not to limit the protection scope of the present invention by this means. All equivalent changes and modifications made according to the spirit of the present invention should be covered in the protection scope of the present invention.
Claims (10)
1. A bridge inhaul cable micro-motion vibration detection method based on image segmentation is characterized by comprising the following steps: the method comprises the following specific steps: the method comprises the following steps that S1, a vibration video of a bridge inhaul cable is obtained through image acquisition equipment; s2, intercepting a certain frame in the video as a first frame according to requirements, setting a target area on at least one guy cable in the first frame, wherein each target area only comprises one guy cable and a background, and framing at least one measuring point area in each target area, wherein the measuring point area is framed and selected at the edge of the corresponding guy cable; s3, performing foreground and background segmentation on each target area, extracting a foreground gray value and a background gray value respectively, substituting the foreground gray value and the background gray value into a gray mapping model, and realizing the positioning of the cable edge in the measuring point area in the corresponding target area; s4, according to the measuring point area and the target area selected by the first frame, executing the step S3 on the video frame by frame, and recording the inhaul cable edge positioning result of the measuring point area of each frame as a motion signal I n （t/Fs）,I n （t/Fs）=[i n (1/Fs),i n (2/Fs),…,i n (Ts/Fs)]T =1,2, \8230, ts, wherein n is the number of the measuring point region, t is the frame number, ts is the total frame number of the video, fs is the video acquisition frequency, and Fourier changes are respectively carried out on the motion signals of all the measuring point regions to obtain corresponding frequencies-amplitude data; s5, calculating to obtain the fundamental frequency of the corresponding inhaul cable through frequency-amplitude data, and obtaining the fundamental frequency of the corresponding inhaul cable according to the relation T =4 rho L between the inhaul cable force and the natural vibration frequency of the corresponding inhaul cable 2 h 1 2 To find the pulling force, where T is the pulling force, ρ is the linear density of the cable, L is the calculated length of the cable, and h 1 The fundamental frequency of the inhaul cable is a first-order frequency, the proportional relation between the detected inhaul cable force T and the designed inhaul cable force value B of the inhaul cable is set to be w, namely w = T/B, and the state of the inhaul cable is judged according to the w value.
2. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: in step S3, the foreground and background segmentation is to adopt edge detection to make an edge mask and multiply the edge mask with the image to obtain the gray distribution at the edge, and calculate a threshold segmentation image based on the gray distribution.
3. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: in step S3, the gray mapping model refers to a mapping relationship between the position of the edge of the cable in the image and the average gray value in the region, the coverage area of the foreground is obtained by the foreground brightness and the background brightness obtained after the target region is segmented, and the cable is positioned according to the geometric relationship between the coverage area and the edge position.
4. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: in step S3, the average gray value in the target area is I, and the foreground gray value is I 1 Foreground coverage area of S 1 Background gray value of i 0 Background coverage area of S 0 And the total area is S, then S can be obtained according to the proportional relation 1 =S·（I-i 0 ）/(i 1 -i 0 ) Front view coverage area S 1 When the foreground is known to be rectangular, the edge position is located through the edge angle, the included angle between the edge and the region is set to be alpha, and when the alpha =0, the shape of the coverage area is rectangularAt this time, the formula of the cable position d is d = S 1 /width, wherein width represents the width of the coverage area; when alpha is not equal to 0, the mapping relation between the coverage area and the position of the inhaul cable is divided into three types of triangle, trapezoid and pentagon according to different coverage shapes, and if Stan alpha/2 = k, S is determined 1 The mapping relation with d is as follows:
5. the method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: in step S4, fourier transformation is carried out on the motion signal of the nth measuring point area to obtain the amplitude A of the nth measuring point area n (f p )，A n (f p )=[α n (f 1 ),α n (f 2 ),…,α n (f p )]P =1,2, \8230, P, where P represents the number of points of FFT, f p Representing the frequency component of the Fourier transform, f p =Fs*p/2*P，α n Representing a frequency-dependent variable, alpha n (f 1 ),α n (f 2 ),…,α n (f p ) Fourier signals A representing motion signals of n number of measuring point areas respectively n Frequency f in 1 、f 2 、…、f p And obtaining frequency-amplitude data according to the corresponding amplitude.
6. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: in step S5, the step of calculating the fundamental frequency of the corresponding cable through the frequency-amplitude data specifically includes: s51, selecting the frequency corresponding to the maximum value by segmenting the frequency amplitude data of each measuring point area, wherein the specific formula is
f n1 =Max{α n (f a1 ),α n (f a1+1 ),…,α n (f b1 )},0.5≤f a1 ∩f b1 ≤0.5+d，
f n2 =Max{α n (f a2 ),α n (f a2+1 ),…,α n (f b2 )},f n1 ＜f a2 ∩f b2 ≤f n1 +d，
…,
f nq =Max{α n (f aq ),α n (f aq+1 ),…,α n (f bq )},f nq-1 ＜f aq ∩f bq ≤f nq-1 + d, where d is the step size, q is the frequency order, f nq Is the q-order natural frequency of the n number measuring point region, and is in the interval [0.5,0.5+ d +]Fundamental frequency h of inhaul cable at internal search n-number measuring point area n1 The q-th order frequency is the interval (f) nq-1 ,f nq-1 +d]The frequency corresponding to the peak in (1); step S52, differentiating the q-order natural frequency of the measuring point region n, wherein the difference formula is ^ n (e)={(f n2 -f n1 ),(f n3 -f n2 ),…,(f ne+1 -f ne ) Q-1 is less than or equal to e, wherein e is a positive integer; step S53, approximating the average value of the difference to the fundamental frequency of the inhaul cable,
(ii) a Step S54, matching the guy cable fundamental frequency of another measuring point area of the same guy cable as a comparison group, and setting the relative error delta of the guy cable fundamental frequencies of the two measuring point areas, then
Wherein c is a cable number.
7. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 6, wherein the method comprises the following steps: when the relative error delta in the step S54 is less than or equal to 0.1, the fundamental frequency h of the measuring point area 2c of the No. c inhaul cable 2c Fundamental frequency h as output 1 (ii) a When delta is more than 0.1 and less than or equal to 0.3, the average value of two groups of fundamental frequencies of the c number stay cable upper measuring point area 2c and the measuring point area 2c-1 is taken as the output fundamental frequency h 1 I.e. h 1 =(h 2c +h 2c-1 ) 2; when delta is larger than 0.3, a difference value of two groups of fundamental frequencies of the measuring point region 2c on the guy cable c and the measuring point region 2c-1 is required to be calculated 2c (e) And + 2c-1 (e) And calculating the variance of the two difference values, variance σ 2 Is of the formula
Evaluating the accuracy of the two sets of fundamental frequencies, and selecting the variance σ 2 The fundamental frequency corresponding to the small difference value is used as the output fundamental frequency h 1 。
8. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: in step S5, the determining the state of the cable according to the w value specifically includes: when w is less than or equal to 0.8, the stay cable is seriously damaged, or the management and maintenance department is advised to check the reason and maintain in an abnormal working state; when w is more than 0.8 and less than or equal to 0.9, the stay cable is damaged or the working state is abnormal, and the management and maintenance department is advised to pay close attention to the checking reason; when w is more than 0.9 and less than or equal to 1.05, the working state of the inhaul cable is normal; when w is more than 1.05 and less than or equal to 1.15, the load of the guy cable is large, overweight vehicles or bridges have serious congestion, and management and maintenance departments are recommended to pay close attention to the guy cable; when the load of the guy cable is more than 1.5 and w is less than w, the safety of the bridge is threatened, and a management and maintenance department is advised to determine the working state of the bridge, determine whether a serious overweight vehicle exists or not, contact a traffic department to control after recording, or contact the traffic department to evacuate in bad weather.
9. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: the video format of the cable vibration video in the step S1 is one of avi, mov and mp 4.
10. The method for detecting the micro-motion vibration of the bridge guy cable based on the image segmentation as claimed in claim 1, wherein the method comprises the following steps: the target area and the measuring point area in the step S2 are manually selected and set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310120011.2A CN115830024B (en) | 2023-02-16 | 2023-02-16 | Bridge guy cable micro-motion vibration detection method based on image segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310120011.2A CN115830024B (en) | 2023-02-16 | 2023-02-16 | Bridge guy cable micro-motion vibration detection method based on image segmentation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115830024A true CN115830024A (en) | 2023-03-21 |
| CN115830024B CN115830024B (en) | 2023-05-02 |
Family
ID=85521507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310120011.2A Active CN115830024B (en) | 2023-02-16 | 2023-02-16 | Bridge guy cable micro-motion vibration detection method based on image segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115830024B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116295790A (en) * | 2023-05-22 | 2023-06-23 | 合肥工业大学 | Frequency detection method and system based on inter-frame phase difference of bridge inhaul cable characteristic region |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051226A (en) * | 2007-02-27 | 2007-10-10 | 宁波杉工结构监测与控制工程中心有限公司 | Remote online monitor method and device for stayed-cable force |
| CN108106541A (en) * | 2017-12-21 | 2018-06-01 | 浙江大学 | A kind of bridge cable force measuring method based on video image identification |
| CN112764025A (en) * | 2020-12-31 | 2021-05-07 | 江西商思伏沌科技有限公司 | Bridge pier monitoring system and monitoring method based on millimeter wave radar |
| US20220166957A1 (en) * | 2020-11-25 | 2022-05-26 | Zhejiang University | Method for modal analysis of bridge structures based on surveillance videos |
| CN115375924A (en) * | 2022-09-14 | 2022-11-22 | 湖南大学 | Bridge health monitoring method and system based on image recognition |
| CN115683431A (en) * | 2023-01-03 | 2023-02-03 | 湖南大学 | Method, device and equipment for determining cable force of inhaul cable based on linear tracking algorithm |
-
2023
- 2023-02-16 CN CN202310120011.2A patent/CN115830024B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051226A (en) * | 2007-02-27 | 2007-10-10 | 宁波杉工结构监测与控制工程中心有限公司 | Remote online monitor method and device for stayed-cable force |
| CN108106541A (en) * | 2017-12-21 | 2018-06-01 | 浙江大学 | A kind of bridge cable force measuring method based on video image identification |
| US20220166957A1 (en) * | 2020-11-25 | 2022-05-26 | Zhejiang University | Method for modal analysis of bridge structures based on surveillance videos |
| CN112764025A (en) * | 2020-12-31 | 2021-05-07 | 江西商思伏沌科技有限公司 | Bridge pier monitoring system and monitoring method based on millimeter wave radar |
| CN115375924A (en) * | 2022-09-14 | 2022-11-22 | 湖南大学 | Bridge health monitoring method and system based on image recognition |
| CN115683431A (en) * | 2023-01-03 | 2023-02-03 | 湖南大学 | Method, device and equipment for determining cable force of inhaul cable based on linear tracking algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 贾新杰;姜万录;吴红林;: "城市异型钢斜拉桥的车桥耦合振动分析" * |
| 韩州斌;李东升;杨洋;: "等值张拉时基于振动法的平行钢绞线索索力测试分析" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116295790A (en) * | 2023-05-22 | 2023-06-23 | 合肥工业大学 | Frequency detection method and system based on inter-frame phase difference of bridge inhaul cable characteristic region |
| CN116295790B (en) * | 2023-05-22 | 2023-09-05 | 合肥工业大学 | Frequency detection method and system based on inter-frame phase difference of bridge inhaul cable characteristic region |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115830024B (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210381911A1 (en) | Method for identifying spatial-temporal distribution of vehicle loads on bridge based on densely connected convolutional networks | |
| CN111044197B (en) | Non-contact cable force test system based on unmanned aerial vehicle platform | |
| CN101625723B (en) | Rapid image-recognizing method of power line profile | |
| CN108106541A (en) | A kind of bridge cable force measuring method based on video image identification | |
| US11593952B2 (en) | Structural vibration monitoring method based on computer vision and motion compensation | |
| Zhang et al. | Vortex‐induced vibration measurement of a long‐span suspension bridge through noncontact sensing strategies | |
| US20210383093A1 (en) | Method for determining sag of electricity transmission line based on image recognition | |
| CN102768022A (en) | Tunnel surrounding rock deformation detection method adopting digital camera technique | |
| CN111259770B (en) | Unmanned plane platform and deep learning-based cable force rapid test method under complex background | |
| CN115830024A (en) | Bridge inhaul cable micro-motion vibration detection method based on image segmentation | |
| CN109580137B (en) | Bridge structure displacement influence line actual measurement method based on computer vision technology | |
| WO2020093631A1 (en) | Antenna downtilt angle measurement method based on depth instance segmentation network | |
| CN115761487A (en) | Method for quickly identifying vibration characteristics of small and medium-span bridges based on machine vision | |
| JP4025161B2 (en) | Flow velocity distribution measurement system, flow velocity distribution measurement method, flow velocity distribution measurement program, flow rate measurement system, flow rate measurement method, and flow rate measurement program | |
| CN114413854B (en) | Real-scene three-dimensional image checking system based on oblique photography | |
| CN109934151B (en) | Face detection method based on movidius computing chip and Yolo face | |
| CN113569956A (en) | Mountain fire disaster investigation and identification method based on AI algorithm | |
| CN112734732B (en) | Railway tunnel cable leakage clamp detection method based on improved SSD algorithm | |
| CN111553500B (en) | Railway traffic contact net inspection method based on attention mechanism full convolution network | |
| CN117152617A (en) | Urban flood identification method and system | |
| CN112686956A (en) | Method for detecting tilt fault of urban road signal lamp post | |
| CN113959619A (en) | Cable force measuring system, method and storage medium | |
| KR20210057887A (en) | Deep learning and image processing based bolt loosening detection method | |
| CN116256096A (en) | Cable force real-time measurement method based on EDLines linear detection | |
| CN117036611B (en) | Three-dimensional scene construction method and system of non-mine safety monitoring platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |