WO2021042908A1 - 轨道振动检测方法及装置、振动检测设备 - Google Patents

轨道振动检测方法及装置、振动检测设备 Download PDF

Info

Publication number
WO2021042908A1
WO2021042908A1 PCT/CN2020/104833 CN2020104833W WO2021042908A1 WO 2021042908 A1 WO2021042908 A1 WO 2021042908A1 CN 2020104833 W CN2020104833 W CN 2020104833W WO 2021042908 A1 WO2021042908 A1 WO 2021042908A1
Authority
WO
WIPO (PCT)
Prior art keywords
track
vibration
image
tested
video
Prior art date
Application number
PCT/CN2020/104833
Other languages
English (en)
French (fr)
Inventor
高风波
Original Assignee
深圳市豪视智能科技有限公司
深圳市广宁股份有限公司
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 深圳市豪视智能科技有限公司, 深圳市广宁股份有限公司 filed Critical 深圳市豪视智能科技有限公司
Publication of WO2021042908A1 publication Critical patent/WO2021042908A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
    • G01H9/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves by using radiation-sensitive means, e.g. optical means
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • 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
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • 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
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30164—Workpiece; Machine component

Definitions

  • This application relates to the field of computer technology, in particular to a track vibration detection method and device, and vibration detection equipment.
  • Track refers to a route paved with strip-shaped steel rails for trains, trams, etc. to travel. Among them, maintenance personnel need to maintain and inspect the track regularly to ensure the safety of transportation. In the track maintenance and inspection work, track vibration detection is an important link.
  • This application provides a track vibration detection method and device, and vibration detection equipment, in order to improve the accuracy of vibration information extraction and the wide range of applications, thereby improving the efficiency, accuracy, and reliability of vibration detection.
  • an embodiment of the present application provides a method for detecting track vibration, which is applied to a vibration detecting device, and the method includes:
  • the stability of the track to be tested is determined according to the actual vibration data.
  • an embodiment of the present application provides a track vibration detection device, which is applied to a vibration detection device.
  • the device includes a processing unit and a communication unit, wherein:
  • the processing unit is used to obtain the track vibration video of the track to be tested, and select the first RGB image according to the track vibration video of the track to be tested; and to linearly convert the first RGB image from the RGB color space to YIQ color space to obtain the first YIQ image; and used to process the Y-channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence; and used to extract the waiting image sequence according to the characteristic image sequence Actual vibration data of the track to be tested; and used to determine the stability of the track to be tested according to the actual vibration data.
  • an embodiment of the present application provides a vibration detection device, including a processor, a memory, a communication interface, and one or more programs.
  • the one or more programs are stored in the memory and configured by The processor executes, and the program includes instructions for executing the steps in any method in the first aspect of the embodiments of the present application.
  • an embodiment of the present application provides a computer-readable storage medium, wherein the above-mentioned computer-readable storage medium stores a computer program for electronic data exchange, and the above-mentioned computer program enables the computer to execute the same as the first aspect of the embodiment of the present application. Part or all of the steps described in any method.
  • an embodiment of the present application provides a computer program product, wherein the above-mentioned computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the above-mentioned computer program is operable to cause a computer to execute as implemented in this application For example, part or all of the steps described in any method of the first aspect.
  • the computer program product may be a software installation package.
  • the embodiment of the present application discloses a method and device for detecting track vibration, by acquiring the track vibration video of the track to be tested, and selecting the first RGB image according to the track vibration video of the track to be tested;
  • the first RGB image is linearly converted from the RGB color space to the YIQ color space to obtain the first YIQ image;
  • the Y channel image in the first YIQ image is processed by a video magnification algorithm to obtain a characteristic image sequence; according to the characteristic image Extract the actual vibration data of the track to be tested in sequence; determine the stability of the track to be tested based on the actual vibration data, improve the accuracy of vibration data extraction and the versatility of engineering applications, thereby improving the efficiency of track vibration detection, Accuracy and reliability.
  • FIG. 1 is a schematic diagram of a system for detecting track vibration provided by an embodiment of the present application
  • FIG. 2a is a schematic flowchart of a method for detecting track vibration according to an embodiment of the application
  • Figure 2b is a schematic diagram of a possible processing orbital vibration video provided by an embodiment of the present application.
  • Figure 2c is a schematic plan view of a possible track fastener provided by an embodiment of the present application.
  • FIG. 3 is a schematic flowchart of another track vibration detection method provided by an embodiment of the present application.
  • FIG. 4 is a schematic flowchart of another method for detecting track vibration provided by an embodiment of the present application
  • FIG. 5 is a schematic structural diagram of a vibration detection device provided by an embodiment of the present application.
  • Fig. 6 is a block diagram of functional units of a track vibration detection device provided by an embodiment of the present application.
  • FIG. 1 is a schematic diagram of a system for detecting track vibration according to an embodiment of the present application.
  • FIG. 1 includes a track 100 and a vibration detection device 110.
  • the track 100 of the present application mainly includes a steel rail 101, a sleeper 102, a track joint 103, a track fastener 104, a track bed 105, and so on.
  • the sleeper 102 includes wooden sleepers 1021 and concrete sleepers 1022, which bear the vertical and horizontal forces transmitted from the rail 101 and distribute these forces on the bed 105 while effectively maintaining the gauge, direction and position of the track 100 . Therefore, the sleeper 102 needs to have a certain degree of firmness, durability and elasticity, and at the same time have sufficient resistance to avoid lateral movement under the action of the train.
  • the rail connection structure of the present application is divided into two types: an intermediate connection structure and a joint connection structure.
  • the intermediate connection is the connection between the rail 101 and the sleeper 102
  • the intermediate connection structure is generally called the track fastener 104. Its main function is to prevent the rail 101 from moving in the vertical and horizontal directions relative to the sleeper 102 and maintain the stable position of the rail 101.
  • the track fastener 104 includes a wooden sleeper fastener 1041 and a concrete sleeper fastener 1042.
  • the wooden sleeper fastener 1041 includes a baffle plate, a spike pad and a common road stud
  • the concrete sleeper fastener 1042 includes a baffle plate, a spike pad and a common road stud.
  • the joint connection is the connection between the rail 101 and the rail 101.
  • the joint connection consists of splints, bolts, spring washers, etc., commonly known as track joint 103, whose function is to connect the rail 101 at the joint so that the rail joint has the same integrity as the rail 101 to resist bending and displacement.
  • the function of the track bed 105 of the present application is to evenly spread the force transmitted from the sleeper 102 to the roadbed, fix the position of the sleeper 102, maintain the stability of the track 100, remove moisture from the roadbed, maintain the elasticity of the track 100, and adjust the track 100 plane and vertical section.
  • the vibration detection device 110 of the present application includes an electronic device 111 with video processing capabilities, a camera 113, and a sensor 112.
  • the electronic device 111 may include various handheld devices with video processing functions, vehicle-mounted devices, wearable devices, computing devices or other video processing devices connected to a wireless modem, as well as various forms of intelligent terminal devices (intelligent terminal devices) and many more.
  • the camera 112 may include an infrared camera and a visible light camera, and the visible light camera may also include a normal camera or a wide-angle camera, which is not limited herein.
  • the camera 113 is mainly used to record a piece of video of the track 100 and upload it to the electronic device 111, and the track vibration detection device in the electronic device 111 then performs vibration detection on the uploaded segment.
  • the sensor 112 includes an ultrasonic sensor for detecting the track 100.
  • Fig. 2a is a schematic flow chart of a method for detecting track vibration according to an embodiment of the application, which is applied to a vibration detecting device, and the method includes:
  • S201 Obtain a track vibration video of the track to be tested, and select a first RGB image according to the track vibration video of the track to be tested;
  • the track vibration video of the track to be tested can be obtained by the following operations: shoot the track to be tested according to the vibration detection device, and obtain the first video and the second video of the track to be tested, where the first video and The second video is a different video shot in the same time on the track to be tested; the first image corresponding to the first video and the second image corresponding to the second video are acquired; the first image and the second image are overlapped, and then Clear the pixels that cannot overlap between the first image and the second image, and obtain the orbit vibration video of the track to be tested.
  • the vibration detection device when shooting the track to be tested according to the vibration detection device, there are some external interference factors, such as camera shaking, camera failure, etc., which cause a certain deviation between the track vibration video of the track to be tested and the real situation. Therefore, by using different cameras to shoot different videos of the track to be tested at the same time, the first video and the second video of the track to be tested are obtained, and corresponding to the first image and the second video corresponding to the first video The second image is overlapped. In the absence of the aforementioned interference factors in the camera, the pixels corresponding to the first image and the second image can be completely overlapped. By clearing the pixels that cannot overlap the first image and the second image, that is, The shooting deviation pixels in different cameras can obtain images with less noise. Similarly, more different videos in the same time period can be shot on the track to be tested to further reduce image noise.
  • external interference factors such as camera shaking, camera failure, etc.
  • RGB is a color standard in the industry that is represented by three color components of the red primary color (Red), the green primary color (Green), and the blue primary color (Blue).
  • the RGB color space uses a linear combination of three color components to represent colors
  • YIQ is the television system standard of the National Television Standards Committee (NTSC).
  • Y represents the brightness components that provide black-and-white TV and color TV
  • I In-phase
  • Q Quadadrature-phase
  • the YIQ color space can separate and extract the brightness components in the video image, and the relationship between the YIQ color space and the RGB color space is linearly transformed. It has the advantages of small calculation amount and good clustering characteristics, and can adapt to changing light intensity. occasion.
  • the value range of a1, a2, a3, b1, b2, b3, c1, c2, and c3 is [-1,1].
  • the linear conversion relationship between RGB color space and YIQ color space is:
  • FIG. 2b is a schematic diagram of a possible processing orbital vibration video provided by an embodiment of the present application.
  • the vibration detection device 110 obtains a segment of track vibration video of the track 100, and collects an image of the vibration of the track fastener 104 in the track vibration video.
  • the red primary color component, the green primary color component and the blue primary color component of the first image are respectively extracted through the RGB color space. According to the linear conversion relationship between the RGB color space and the YIQ color space, the Y image component, the Q image component, and the I image component of the first YIQ image are obtained.
  • S203 Process the Y-channel image in the first YIQ image by using a video magnification algorithm to obtain a characteristic image sequence
  • the video amplification algorithm may include at least one of the following: Laplacian motion amplification algorithm, Euler motion amplification algorithm, complex phase motion amplification algorithm, and Rees pyramid motion amplification algorithm.
  • S205 Determine the stability of the track to be tested according to the actual vibration data.
  • the stability of the track to be tested is determined by comparing the actual vibration data with the vibration data of the track joint 103, the track fastener 104 and the track bed 105 in the track to be tested.
  • the track vibration detection method described in the embodiment of the present application is applied to track vibration equipment.
  • the orbital vibration equipment extracts vibration data from the orbital vibration video through a video amplification algorithm, so as to improve the accuracy of vibration data extraction, thereby improving the efficiency, accuracy and reliability of the orbital vibration detection.
  • the processing the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence includes: performing Fourier analysis on the Y channel image in the first YIQ image Leaf transform; and perform image pyramid decomposition on the Y-channel image after Fourier transform; and perform normalization, spatio-temporal filtering and linear magnification processing on the Y-channel image after the image pyramid decomposition;
  • the I-channel image and the Q-channel image in the YIQ image are synthesized to form a characteristic image sequence.
  • the Y channel image represents brightness information
  • the Y channel image is subjected to Fourier transform to convert the brightness information in the time domain into a phase change in the frequency domain.
  • the image pyramid is a method of multi-resolution processing of images in the spatial domain.
  • the size and contrast of the object image are different, and the multi-resolution analysis of the object image can help analyze the object image and extract the characteristic parameters of the object image.
  • the object image is decomposed into high-frequency and low-frequency components, and then the wavelet decomposition coefficients are obtained through two sampling. According to the wavelet decomposition coefficient, the object image is decomposed into multiple scales and directions to obtain sub-band images of multiple object images.
  • image normalization is a process of performing a series of standard transformation processing on an image and transforming it into a standard image.
  • the standard image is called a normalized image.
  • the image is processed to obtain multiple sub-band images, and the multiple sub-band images can obtain standard images in the same form after undergoing image normalization processing with the same parameters.
  • image normalization is to convert an image into a corresponding unique standard form through a series of transformations, and the standard form image has invariant characteristics to affine transformations such as translation, rotation, and scaling.
  • the processing the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence includes: photographing the track to be tested according to the vibration detection device, and obtaining the track to be tested Orbital vibration video; Obtain multiple frames of the orbital vibration video to-be-processed image, and partition the to-be-processed image; use the pixel points in each partition as the initial feature point, and based on the minimum difference square sum SSD Matching to calculate the flow vector of the initial feature point; calculate the offset distance of the initial feature point according to the flow vector corresponding to the initial feature point; use k- for multiple offset distances corresponding to multiple initial feature points means clustering algorithm performs clustering to obtain multiple clusters; determine whether the average offset distance in each of the multiple clusters is within a preset range; if so, determine all The partition corresponding to the initial feature point in the cluster is a motion partition; the motion partition in the multi-frame to-be-processed image of the target video is retained to form a characteristic image
  • the processing the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence includes: performing Fourier transform on the Y channel image in the first YIQ image; Perform down-sampling processing on the Y-channel image after Fourier transform to obtain a first sub-image, and the down-sampling processing is to reduce the resolution of the image; perform up-sampling processing on the first sub-image to obtain a second sub-image , The up-sampling process is to increase the resolution of the image; perform pixel processing on the Y channel image after Fourier transform and the second sub-image to obtain a third sub-image; and perform the time domain on the third sub-image Filter processing to obtain a target frequency band; determine multiple Y-channel image signals according to the target frequency band and the Y-channel image in the first YIQ image; amplify the multiple Y-channel image signals to obtain multiple amplified Y channel image signal; synthesize the amplified multiple Y channel image signals according to the
  • the processing the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence includes: performing Fourier transform on the Y channel image in the first YIQ image; Perform spatial filtering on the Y-channel image after Fourier transform to obtain Y-channel images with different spatial resolutions; perform time-domain filtering on the Y-channel images with different spatial resolutions to obtain the target frequency band;
  • the Y-channel images of different spatial resolutions are used to determine multiple Y-channel image signals; the multiple Y-channel image signals are amplified to obtain the amplified multiple Y-channel image signals; the amplified multiple Y-channel image signals are obtained according to the complex steerable pyramid reconstruction algorithm
  • Multiple Y-channel image signals of the first YIQ image are obtained; the I-channel image and the Q-channel image in the first YIQ image are obtained, and the combined multiple Y-channel images are combined with the I-channel image and the Q-channel image Add together to form a sequence of characteristic images.
  • the processing the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence includes: performing Fourier transform on the Y channel image in the first YIQ image; Decompose the Fourier-transformed Y-channel image according to the Laplacian pyramid; perform the Rees transform on the decomposed Y-channel image; perform orthogonal and phase processing on the Y-channel image after the Rees transform; perform the processed Y
  • the channel image is filtered in the space and time domain; the method and phase shift are performed on the Y channel image after the space and time domain filtering to form a characteristic image sequence.
  • the orbital vibration device decomposes the Y-channel image in the first YIQ image through the image pyramid, and performs normalization, spatio-temporal filtering, and linear amplification processing on the Y-channel image after the decomposition of the image pyramid to enlarge the orbit.
  • Vibration video improves the accuracy of vibration data extraction, thereby improving the efficiency, accuracy and reliability of track vibration detection.
  • the extracting actual vibration data of the track to be measured according to the characteristic image sequence includes: calculating the cross cross power spectrum of the characteristic image sequence according to a phase correlation formula; and according to the cross interaction
  • the power spectrum extracts the vibration information of the pixels in the characteristic image sequence; and the actual vibration data of the track to be measured is determined according to the vibration information.
  • phase correlation algorithm uses the following formula to calculate the cross cross power spectrum of the characteristic image sequence:
  • F A frame image of a Fourier transform
  • I the conjugate signal of the Fourier transform of the b frame image
  • Means Fa Means Fa
  • R represents the cross cross power spectrum (including frequency domain noise).
  • the vibration detection equipment uses the phase correlation algorithm to calculate the cross cross power spectrum on the feature image sequence after the video is amplified, so as to improve the accuracy of the track vibration detection.
  • the extracting the vibration information of the pixels in the characteristic image sequence according to the cross-interaction power spectrum includes: selecting an adaptive filter according to the peak position of the cross-interaction power spectrum; and according to the self The adaptive filter filters the cross cross power spectrum; and performs inverse Fourier transform on the filtered cross cross power spectrum; and extracts pixels in the characteristic image sequence according to the cross cross power spectrum after inverse Fourier transform Vibration information.
  • the processed Y channel image is normalized, spatio-temporal filtering, and linear amplification processing, and the resulting characteristic image sequence contains redundant signals. Performing difference, weighting and reconstruction on redundant signals will bring a high amount of calculation, and it has no effect on the analysis of vibration information.
  • principal component decomposition to reduce the dimensionality of the characteristic image sequence, and then use an adaptive filter to filter the dimensionality reduction of the characteristic image sequence, it can effectively reduce the calculation processing time and the influence of noise signals.
  • the vibration detection device filters the characteristic image sequence through an adaptive filter to reduce the amount of calculation for track vibration detection, and improve the efficiency of track vibration detection, and effectively ensure the stability of the track.
  • the determining the stability of the track to be tested according to the actual vibration data includes: determining a vibration position of the track to be tested according to the actual vibration data, and the vibration position includes a track buckle. Parts or track joints or track bed; and the stability of the track to be tested is determined according to the vibration position of the track to be tested and the actual vibration data.
  • the vibration detection equipment determines the stability of the track to be tested based on the vibration position of the track to be tested and the actual vibration data, so as to ensure the accuracy and reliability of the track vibration detection.
  • the determining the stability of the track to be tested according to the vibration part of the track to be tested and the actual vibration data includes: collecting N segments of vibration videos of the track fasteners or track joints of the track to be tested; and statistics of the average vibration frequency, average amplitude, and average period of the N segments of vibration videos; and comparison of the average vibration frequency and the average amplitude And the average vibration period and the actual vibration data, and determine the stability of the track to be tested according to the comparison result.
  • FIG. 2c is a schematic plan view of a possible rail fastener provided by an embodiment of the present application.
  • the concrete sleeper fastener 1042 includes a baffle 301, a common spike 302, and a spike pad 303.
  • the vibration detection equipment collects the N-segment vibration video of the concrete sleeper fastener 1042, and calculates the average vibration frequency, average amplitude, and average vibration period of the baffle 301 and the ordinary spike 302 in the vertical and horizontal vibration directions through the N-end vibration video.
  • the vibration detection equipment counts the average vibration data of the track fasteners of the track to be tested, and compares the average vibration data with the actual vibration data to improve the accuracy and reliability of the track vibration detection.
  • determining the stability of the track to be tested according to the vibration location of the track to be tested and the actual vibration data includes: obtaining a track bed coefficient of the track bed; And determine the vibration stiffness and vibration elasticity coefficient of the track bed according to the track bed coefficient; and compare the vibration stiffness and the vibration elasticity coefficient with the actual vibration data, and determine the stability of the track to be tested according to the comparison result .
  • the vibration detection equipment obtains the vibration stiffness and the vibration elastic coefficient of the track bed to be tested, and compares the vibration stiffness and the vibration elastic coefficient with actual vibration data to improve the accuracy and reliability of the track vibration detection.
  • FIG. 3 is a schematic flowchart of another track vibration detection method provided by an embodiment of the present application, which is applied to a vibration detection device, and the method includes:
  • S301 Obtain a track vibration video of the track to be tested, and select a first RGB image according to the track vibration video of the track to be tested;
  • the track vibration video of the track to be tested can be obtained by the following operations: shoot the track to be tested according to the vibration detection device, and obtain the first video and the second video of the track to be tested, where the first video and The second video is a different video shot in the same time on the track to be tested; the first image corresponding to the first video and the second image corresponding to the second video are acquired; the first image and the second image are overlapped, and then Clear the pixels that cannot overlap between the first image and the second image, and obtain the orbit vibration video of the track to be tested.
  • S303 Perform Fourier transform on the Y channel image in the first YIQ image, and perform image pyramid decomposition on the Y channel image after Fourier transform;
  • the Y channel image represents brightness information
  • the Y channel image is subjected to Fourier transform to convert the brightness information in the time domain into a phase change in the frequency domain.
  • image pyramids include Gaussian pyramids, Laplace pyramids, complex manipulable pyramids, Ries pyramids, and so on.
  • S304 Perform normalization, spatiotemporal filtering, and linear amplification processing on the Y channel image decomposed by the image pyramid;
  • S305 Synthesize the processed Y channel image with the I channel image and the Q channel image in the YIQ image, and form a characteristic image sequence
  • S306 Calculate the cross cross power spectrum of the characteristic image sequence according to the phase correlation formula, and select an adaptive filter according to the peak position of the cross cross power spectrum;
  • S308 Extract the vibration information of the pixels in the characteristic image sequence according to the cross cross power spectrum after the inverse Fourier transform, and determine the actual vibration data of the track to be measured according to the vibration information;
  • S310 Collect N segments of vibration videos of the track fasteners or track joints of the track to be tested, and count the average vibration frequency, average amplitude, and average vibration period of the N segments of vibration video;
  • the track vibration detection method described in the embodiment of the present application is applied to track vibration equipment.
  • the track vibration equipment amplifies the track vibration video through a video amplification algorithm, and extracts the actual vibration data of the track vibration video according to the phase correlation formula and inverse Fourier transform, and then compares the average vibration frequency, average amplitude and average of the track fasteners
  • the vibration period and the actual vibration data improve the accuracy of vibration data extraction, thereby improving the efficiency, accuracy, and reliability of track vibration detection.
  • FIG. 4 is a schematic flowchart of another track vibration detection method provided by an embodiment of the present application.
  • S401 Obtain a track vibration video of the track to be tested, and select a first RGB image according to the track vibration video of the track to be tested;
  • S402 Process the first RGB image, and obtain multiple subband image sequences corresponding to multiple resolutions;
  • S403 Filter out at least one sub-band image sequence used for magnification processing from the multiple sub-band image sequences according to a preset partition gray value screening strategy;
  • the screening of at least one sub-band image sequence for magnification processing from the multiple sub-band image sequences according to a preset divisional gray value screening strategy includes: determining the foreground images of the multiple sub-band image sequences and A background image, the foreground image includes an image of the area where the detected product is reciprocating, the background image is an image other than the image of the detected product; and it is determined that the foreground image is in a sub-band The area accounted for in the image; and determine the number of sub-regions of the foreground image according to the area accounted for and a preset sub-region calculation formula, and divide the foreground image into multiple foreground sub-regions according to the number of sub-regions ; And for each sub-band image sequence, perform the following operations (1)-(6) to obtain the gray value change frequency of each sub-band image sequence: (1) Determine each sub-band in the currently processed sub-band image sequence The measured pixel of each foreground sub-region of the image; (2) The gray value of each measured pixel is
  • the preset sub-partition calculation formula is:
  • x is the area ratio
  • y is the number of sub-divisions
  • x is greater than 0 and less than or equal to 1.
  • the determination of the measured pixel points of each foreground sub-region of each sub-band image in the currently processed sub-band image sequence includes at least one of the following: edge pixels of each foreground sub-region; Pixels in the middle area of the foreground sub-division; multiple pixels randomly selected from each of the foreground sub-divisions.
  • S405 Perform fusion processing on the at least one sub-band image sequence after the enlargement processing and the sub-band image sequences of the multiple sub-band image sequences except the at least one sub-band image sequence, and form a characteristic image sequence;
  • S406 Calculate the cross cross power spectrum of the characteristic image sequence according to the phase correlation formula, and select an adaptive filter according to the peak position of the cross cross power spectrum;
  • S408 Extract the vibration information of the pixels in the characteristic image sequence according to the cross cross power spectrum after the inverse Fourier transform, and determine the actual vibration data of the track to be measured according to the vibration information;
  • S409 Determine a vibration location of the track to be tested according to the actual vibration data, where the vibration location includes a track fastener or a track joint or a track bed;
  • S410 Collect N segments of vibration videos of the track fastener or track joint of the track to be tested, and count the average vibration frequency, average amplitude, and average vibration period of the N segments of vibration video;
  • S411 Compare the average vibration frequency, the average amplitude and the average vibration period with the actual vibration data, and determine the stability of the track fastener or the track joint according to the comparison result.
  • the track vibration detection method described in the embodiment of the present application is applied to track vibration equipment.
  • the orbital vibration equipment processes the orbital vibration video through a divisional gray value screening strategy, and extracts vibration data from the processed orbital vibration video, so as to improve the accuracy of vibration data extraction, thereby improving the efficiency of orbital vibration detection. Accuracy and reliability.
  • FIG. 5 is a schematic structural diagram of a vibration detection device 500 provided by an embodiment of the present application.
  • the vibration detection device 500 includes an application processor 510, a memory 520, a communication interface 530, and one or more programs 521, wherein the one One or more programs 521 are stored in the above-mentioned memory 520 and configured to be executed by the above-mentioned application processor 510.
  • the one or more programs 521 include instructions for performing the following steps: obtaining the orbital vibration video of the track to be tested, And select the first RGB image according to the orbit vibration video of the track to be tested; and linearly convert the first RGB image from the RGB color space to the YIQ color space to obtain the first YIQ image;
  • the Y channel image in the first YIQ image is processed to obtain a characteristic image sequence; and the actual vibration data of the track to be measured is extracted according to the characteristic image sequence; and the stability of the track to be measured is determined according to the actual vibration data Sex.
  • the vibration detection device obtains the track vibration video of the track to be tested, and selects the first RGB image according to the track vibration video of the track to be tested; linearly converts the first RGB image from the RGB color space to the YIQ color Space, obtain the first YIQ image; process the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence; extract the actual vibration data of the track to be measured according to the characteristic image sequence; The actual vibration data determines the stability of the track to be tested, improves the accuracy of vibration data extraction and the versatility of engineering applications, thereby improving the efficiency, accuracy and reliability of track vibration detection.
  • the instructions in the program are specifically used to perform the following operations: Perform Fourier transform on the Y channel image in the first YIQ image; and perform image pyramid decomposition on the Y channel image after Fourier transform; and perform normalization, spatio-temporal filtering, and linearization on the Y channel image after the image pyramid decomposition. Enlarging processing; and synthesizing the processed Y channel image with the I channel image and the Q channel image in the YIQ image to form a characteristic image sequence.
  • the instructions in the program are specifically used to perform the following operations: calculating the characteristic image sequence according to a phase correlation formula And extract the vibration information of the pixels in the characteristic image sequence according to the cross-interaction power spectrum; and determine the actual vibration data of the track to be measured according to the vibration information.
  • the instructions in the program are specifically used to perform the following operations: Selecting an adaptive filter at the peak position; and filtering the cross cross power spectrum according to the adaptive filter; and performing inverse Fourier transform on the filtered cross cross power spectrum; and after the inverse Fourier transform
  • the cross cross power spectrum extracts the vibration information of the pixels in the characteristic image sequence.
  • the instructions in the program are specifically used to perform the following operations: determining the track to be tested based on the actual vibration data
  • the vibration part of the track, the vibration part includes a track fastener or a track joint or a track bed; and the stability of the track to be tested is determined according to the vibration part of the track to be tested and the actual vibration data.
  • the The instructions in the program are specifically used to perform the following operations: collect N segments of vibration videos of the track fasteners or track joints of the track to be tested; and count the average vibration frequency, average amplitude, and average period of the N segments of vibration video; And compare the average vibration frequency, the average amplitude and the average vibration period with the actual vibration data, and determine the stability of the track to be tested according to the comparison result.
  • the instructions in the program specifically Used to perform the following operations obtain the track bed coefficient of the track bed; and determine the vibration stiffness and vibration elastic coefficient of the track bed according to the track bed coefficient; and compare the vibration stiffness and the vibration elastic coefficient with the actual vibration data, and The stability of the track to be tested is determined according to the comparison result.
  • the vibration detection device includes hardware structures and/or software modules corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • the embodiments of the present application may divide the vibration detection device into functional units according to the foregoing method examples.
  • each functional unit may be divided corresponding to each function, or two or more functions may be integrated into one processing unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • FIG. 6 is a block diagram of functional units of a track vibration detection device 600 provided by an embodiment of the present application.
  • the track vibration detection device 600 is applied to vibration detection equipment, and the device includes a processing unit 601 and a communication unit 602.
  • the processing unit 601 is configured to perform any step in the above method embodiment, and when performing data transmission such as sending, it can optionally call the communication unit 602 to complete the corresponding operation, which will be described in detail below .
  • the processing unit 601 is specifically configured to: obtain the orbit vibration video of the track to be tested, and select the first RGB image according to the orbit vibration video of the track to be tested; and linearly convert the first RGB image from the RGB color space to YIQ color space, obtain the first YIQ image; and process the Y channel image in the first YIQ image by a video magnification algorithm to obtain a characteristic image sequence; and extract the actual track of the track to be measured according to the characteristic image sequence Vibration data; and determining the stability of the track to be tested according to the actual vibration data.
  • the embodiment of the present application discloses a track vibration detection device, which acquires the track vibration video of the track to be tested, and selects the first RGB image according to the track vibration video of the track to be tested;
  • the image is linearly converted from the RGB color space to the YIQ color space to obtain the first YIQ image;
  • the Y-channel image in the first YIQ image is processed by the video magnification algorithm to obtain the characteristic image sequence;
  • the characteristic image sequence is extracted according to the characteristic image sequence
  • the actual vibration data of the track to be tested; the stability of the track to be tested is determined according to the actual vibration data, and the accuracy of vibration data extraction and the versatility of engineering applications are improved, thereby improving the efficiency, accuracy and reliability of track vibration detection Sex.
  • the processing unit 601 is specifically configured to: Perform Fourier transform on the Y channel image in the image; and perform image pyramid decomposition on the Y channel image after Fourier transform; and perform normalization, spatio-temporal filtering and linear amplification on the Y channel image after the image pyramid decomposition; And synthesize the processed Y channel image with the I channel image and the Q channel image in the YIQ image to form a characteristic image sequence.
  • the processing unit 601 is specifically configured to: calculate the cross-intersection of the characteristic image sequence according to a phase correlation formula. And extract the vibration information of the pixels in the characteristic image sequence according to the cross-interaction power spectrum; and determine the actual vibration data of the track to be measured according to the vibration information.
  • the processing unit 601 is specifically configured to: select according to the peak position of the cross-interaction power spectrum Adaptive filter; and filtering the cross cross power spectrum according to the adaptive filter; and performing inverse Fourier transform on the filtered cross cross power spectrum; and according to the cross cross cross power spectrum after the inverse Fourier transform
  • the power spectrum extracts the vibration information of the pixels in the characteristic image sequence.
  • the processing unit 601 is specifically configured to: determine the vibration of the track to be tested according to the actual vibration data.
  • the vibration part includes a track fastener or a track joint or a track bed; and the stability of the track to be tested is determined according to the vibration part of the track to be tested and the actual vibration data.
  • the processing unit 601 is specifically configured to: collect N segments of vibration videos of the track fasteners or track joints of the track to be tested; and count the average vibration frequency, average amplitude, and average vibration period of the N segments of vibration video; and compare the results.
  • the average vibration frequency, the average amplitude and the average vibration period are compared with the actual vibration data, and the stability of the track to be measured is determined according to the comparison result.
  • the processing unit 601 specifically determines the stability of the track to be tested according to the vibration location of the track to be tested and the actual vibration data. Used to: obtain the track bed coefficient of the track bed; and determine the vibration stiffness and the vibration elastic coefficient of the track bed according to the track bed coefficient; and compare the vibration stiffness and the vibration elastic coefficient with the actual vibration data, and according to the comparison result Determine the stability of the track to be tested.
  • An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method as recorded in the above method embodiment ,
  • the above-mentioned computer includes a vibration detection device.
  • the embodiments of the present application also provide a computer program product.
  • the above-mentioned computer program product includes a non-transitory computer-readable storage medium storing a computer program.
  • the above-mentioned computer program is operable to cause a computer to execute any of the methods described in the above-mentioned method embodiments. Part or all of the steps of the method.
  • the computer program product may be a software installation package, and the above-mentioned computer includes a vibration detection device.
  • the disclosed device may be implemented in other ways.
  • the device embodiments described above are only illustrative, for example, the division of the above-mentioned units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or integrated. To another system, or some features can be ignored, or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical or other forms.
  • the units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable memory.
  • the technical solution of the present application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory.
  • a number of instructions are included to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the foregoing methods of the various embodiments of the present application.
  • the aforementioned memory includes: U disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
  • the program can be stored in a computer-readable memory, and the memory can include: a flash disk , Read-only memory (English: Read-Only Memory, abbreviation: ROM), random access device (English: Random Access Memory, abbreviation: RAM), magnetic disk or optical disc, etc.

Landscapes

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

Abstract

一种轨道振动检测方法，包括：获取待测轨道的轨道振动视频，并根据待测轨道的轨道振动视频选取第一RGB图像（S201）；将第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像（S202）；通过视频放大算法对第一YIQ图像中的Y通道图像进行处理，获得特征图像序列（S203）；根据特征图像序列提取待测轨道的实际振动数据（S204）；根据实际振动数据确定待测轨道的稳定性（S205）。该方法可提升振动数据提取的准确性以及工程应用的广泛性，进而提升轨道振动检测的效率、准确和可靠性。还提供一种轨道振动检测装置、振动检测设备。

Description

轨道振动检测方法及装置、振动检测设备
本申请要求于2019年09月04日提交中国专利局、申请号为2019108358154、发明名称为“轨道振动检测方法及装置、振动检测设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，具体涉及一种轨道振动检测方法及装置、振动检测设备。
背景技术
轨道是指用条形的钢轨铺成的供火车、电车等行驶的路线。其中，维护人员需要定期对轨道进行维护与检查，以保证交通运输的安全。在轨道维护与检查工作中，轨道振动检测是一个重要的环节。
在传统振动检测中，维护人员往往通过个人经验或者简单振动检测设备排查轨道接头、轨道扣件和道床等是否存在安全隐患，这也导致轨道振动检测的检测效率低以及检测结果不精准等问题。此外，在相关技术中，通过相关设备拍摄轨道振动时的视频，并对轨道振动视频进行处理和分析，以减小人工成本和检测时间，但是轨道振动视频也存在振动部位的振动幅度细微，肉眼难以观测，导致振动信息的提取难度大。
发明内容
本申请提供了一种轨道振动检测方法及装置、振动检测设备，以期望提升振动信息提取的准确性和应用的广泛性，进而提高振动检测的效率、准确和可靠性。
第一方面，本申请实施例提供一种轨道振动检测方法，应用于振动检测设备，所述方法包括：
获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；
将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；
通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；
根据所述特征图像序列提取所述待测轨道的实际振动数据；
根据所述实际振动数据确定所述待测轨道的稳定性。
第二方面，本申请实施例提供一种轨道振动检测装置，应用于振动检测设备，所述装置包括处理单元和通信单元，其中：
所述处理单元，用于获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；以及用于将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；以及用于通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；以及用于根据所述特征图像序列提取所述待测轨道的实际振动数据；以及用于根据所述实际振动数据确定所述待测轨道的稳定性。
第三方面，本申请实施例提供一种振动检测设备，包括处理器、存储器、通信接口，以及一个或多个程序，所述一个或多个程序被存储在所述存储器中，并且被配置由所述处理器执行，所述程序包括用于执行本申请实施例第一方面任一方法中的步骤的指令。
第四方面，本申请实施例提供一种计算机可读存储介质，其中，上述计算机可读存储介质存储用于电子数据交换的计算机程序，并且上述计算机程序使得计算机执行如本申请实施例第一方面任一方法中所描述的部分或全部步骤。
第五方面，本申请实施例提供一种计算机程序产品，其中，上述计算机程序产品包括存储了计算机程序的非瞬时性计算机可读存储介质，并且上述计算机程序可操作来使计算机执行如本申请实施例第一方面任一方法中所描述的部分或全部步骤。该计算机程序产品可以为一个软件安装包。
可以看出，本申请实施例公开了一种轨道振动检测的方法及装置，通过获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；根据所述特征图像序列提取所述待测轨道的实际振动数据；根据所述实际振动数据确定所述待测轨道的稳定性，提升振动数据提取的准确性以及工程应用的广泛性，进而提升轨道振动检测的效率、准确和可靠性。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根 据这些附图获得其他的附图。
图1是本申请实施例提供的一种轨道振动检测的系统示意图；
图2a为本申请实施例提供的一种轨道振动检测方法的流程示意图；
图2b是本申请实施例提供的一种可能的处理轨道振动视频的示意图；
图2c是本申请实施例提供的一种可能的轨道扣件的平面结构示意图；
图3是本申请实施例提供的另一种轨道振动检测方法的流程示意图；
图4是本申请实施例提供的另一种轨道振动检测方法的流程示意图
图5是本申请实施例提供的一种振动检测设备的结构示意图；
图6是本申请实施例提供的一种轨道振动检测装置的功能单元组成框图。
具体实施方式
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别不同对象，而不是用于描述特定顺序。此外，术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其他步骤或单元。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
下面对本申请实施例进行详细介绍。
请参阅图1，图1是本申请实施例提供的一种轨道振动检测的系统示意图。图1包括轨道100和振动检测设备110。
具体的，本申请的轨道100主要包括钢轨101、轨枕102、轨道接头103、轨道扣件104、道床105等等。其中，轨枕102包括木枕1021和混凝土轨枕1022，其承受来自钢轨101 传来的垂直力和水平力，并将这些力分布于床道105上，同时有效地保持轨道100的轨距、方向和位置。因此，轨枕102需要有一定的坚固性、耐久性和弹性，同时具有足够的阻力，以避免在列车作用下发生横向移动。
具体的，本申请的钢轨连接结构分为中间连接结构和接头连接结构两类。其中，中间连接为钢轨101与轨枕102之间的连接，中间连接结构通称轨道扣件104，其主要作用是阻止钢轨101作相对与轨枕102的纵横向移动，并保持钢轨101的稳固位置。轨道扣件104包括木枕扣件1041和混泥土轨枕扣件1042。木枕扣件1041包括挡板、道钉垫板和普通道钉，混泥土轨枕扣件1042包括挡板、道钉垫板和普通道钉。其次，接头连接为钢轨101与钢轨101的连接。接头连接由夹板、螺栓、弹簧垫圈等组成，通称轨道接头103，其作用是在接头处把钢轨101连接起来，使钢轨接头部分具有与钢轨101一样的整体性，以抵抗弯曲和位移。
具体的，本申请的道床105的作用是把轨枕102传来的力均匀地传布到路基上，固定轨枕102位置，保持轨道100的稳定性，排除路基水分，保持轨道100的弹性，并调整轨道100的平面及纵断面。
具体的，本申请的振动检测设备110包括具备视频处理能力的电子设备111、摄像头113和传感器112。其中，电子设备111可以包括各种具有视频处理功能的手持设备、车载设备、可穿戴设备、计算设备或连接到无线调制解调器的其他视频处理设备，以及各种形式的智能终端设备(intelligent terminal device)等等。摄像头112可以包括红外摄像头和可见光摄像头，可见光摄像头也可以包括普通摄像头或广角摄像头，在此不作限定。摄像头113主要用于录制轨道100的一段视频，并上传给电子设备111，电子设备111中的轨道振动检测装置再对上传的一段进行振动检测。传感器112包括超声波传感器，用于检测轨道100。
下面具体描述一种轨道振动检测的执行步骤，请参阅图2a。图2a为本申请实施例提供的一种轨道振动检测方法的流程示意图，应用于振动检测设备，所述方法包括：
S201、获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；
具体的，所述待测轨道的轨道振动视频可以通过以下操作获得：根据所述振动检测设备拍摄待测轨道，并得到所述待测轨道的第一视频和第二视频，其中第一视频和第二视频为所述待测轨道在相同时间内拍摄的不同视频；获取述第一视频对应的第一图像和第二视 频对应的第二图像；对第一图像和第二图像进行重叠，再清除第一图像和第二图像不能重叠的像素点，并获得所述待测轨道的轨道振动视频。
其中，在根据所述振动检测设备拍摄待测轨道时，存在一些外界干扰因素，例如摄像头晃动、摄像头故障等，导致所述待测轨道的轨道振动视频与真实情况存在一定的偏差。因此，通过采用不同摄像头在同一时间内拍摄所述待测轨道的不同视频，获得所述待测轨道的第一视频和第二视频，并且对第一视频对应的第一图像和第二视频对应的第二图像进行重叠。在摄像头不存在上述干扰因素的情况下，所述第一图像和所述第二图像对应的像素点可以完全重叠，通过清除所述第一图像和所述第二图像不能重叠的像素点，即不同摄像头中的拍摄偏差像素点，可以获得噪声更少的图像。同样，可以对所述待测轨道拍摄更多的同一时间段内的不同视频，以进一步减少图像噪声。
S202、将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；
具体的，RGB是工业界的一种由红基色(Red)、绿基色(Green)和蓝基色(Blue)三个颜色分量表示的颜色标准。RGB颜色空间利用三个颜色分量的线性组合来表示颜色，而YIQ是美国全国电视标准委员会(National Television Standards Committee，NTSC)的电视系统标准。Y表示提供黑白电视和彩色电视的亮度分量，I(In-phase)表示色彩从橙色到青色的色度分量，Q(Quadrature-phase)表示色彩从紫色到黄绿色的色度分量。YIQ颜色空间能够将视频图像中的亮度分量分离提取出来，并且YIQ颜色空间与RGB颜色空间之间是线性变换的关系，具有计算量小和聚类特性好等优点，可以适应光照强度不断变化的场合。
具体的，RGB颜色空间和YIQ颜色空间的线性转换关系为：
Y＝a1*R+a2*G+a3*B；
I＝b1*R+b2*G+b3*B；
Q＝c1*R+c2*G+c3*B；
其中，a1、a2、a3、b1、b2、b3、c1、c2和c3取值范围为[-1,1]。
优选的，RGB颜色空间和YIQ颜色空间的线性转换关系为：
Y＝0.299*R+0.587*G+0.114*B；
I＝0.596*R-0.275*G-0.321*B；
Q＝0.212*R-0.523*G+0.311*B。
举例中，请参阅图2b，图2b是本申请实施例提供的一种可能的处理轨道振动视频的示意图。其中，振动检测设备110拍摄得到一段轨道100的轨道振动视频，并采集到轨道振动视频中轨道扣件104振动的图像。通过RGB颜色空间分别提取所述第一图像的红基色分量、绿基色分量和蓝基色分量。根据RGB颜色空间和YIQ颜色空间的线性转换关系，得到第一YIQ图像的Y图像分量、Q图像分量和I图像分量。
S203、通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；
具体的，所述视频放大算法可以包括以下至少一种：拉普拉斯运动放大算法、欧拉运动放大算法、复数相位运动放大算法、里斯金字塔运动放大算法。
S204、根据所述特征图像序列提取所述待测轨道的实际振动数据；
S205、根据所述实际振动数据确定所述待测轨道的稳定性。
具体的，通过将实际振动数据与待测轨道中的轨道接头103、轨道扣件104和道床105的振动数据进行比较，确定所述待测轨道的稳定性。
可以看出，本申请实施例中所描绘的轨道振动检测方法，应用于轨道振动设备。所述轨道振动设备通过视频放大算法从轨道振动视频中提取振动数据，以提升了振动数据提取的准确性，进而提高轨道振动检测的效率、准确和可靠性。
在一个可能的示例中，所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列，包括：对所述第一YIQ图像中的Y通道图像进行傅里叶变换；以及对傅里叶变换后的Y通道图像进行图像金字塔分解；以及对图像金字塔分解后的Y通道图像进行归一化、时空滤波以及线性放大处理；将处理后的Y通道图像与所述YIQ图像中的I通道图像和Q通道图像进行合成，并形成特征图像序列。
其中，Y通道图像表示亮度信息，通过对Y通道图像进行傅里叶变换，以将时域的亮度信息转化为频域的相位变化。
具体的，图像金字塔是一种空域对图像进行多分辨率处理的方法。物体图像的尺寸和对比度的不同，通过对物体图像进行多分辨率的分析，能有利于分析物体图像和提取物体图像的特征参数。此外，通过时频小波分解，将物体图像分解为高频和低频分量，再经过二抽样取得小波分解系数。根据小波分解系数将物体图像分解为多个尺度和方向，以获得多种物体图像的子带图像。
具体的，图像归一化是一种对图像进行一系列标准变换处理，并使其变换为标准图像 的过程，所述标准图像称作归一化图像。对图像进行处理，并得到多种子带图像，所述多种子带图像在经过相同参数的图像归一化处理后能够得到相同形式的标准图像。进一步的，图像归一化就是通过一系列变换将图像转换成相应的唯一标准形式，所述该标准形式图像对平移、旋转、缩放等仿射变换具有不变特性。
可选的，所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列，包括：根据所述振动检测设备拍摄待测轨道，并得到所述待测轨道的轨道振动视频；获取所述轨道振动视频的多帧的待处理图像，并将所述待处理图像进行分区；将每个分区中的像素点作为初始特征点，并基于最小差值平方和SSD匹配，以计算所述初始特征点的流向量；根据所述初始特征点对应的流向量计算所述初始特征点的偏移距离；对多个初始特征点对应的多个偏移距离采用k-means聚类算法进行聚类，获得多个聚类类簇；确定所述多个聚类类簇中每个聚类类簇中的偏移距离平均值是否处于预设范围；若是，则确定所述聚类类簇中的初始特征点对应的所述分区为运动分区；保留所述目标视频的多帧待处理图像中的运动分区，形成特征图像序列。
可选的，所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列，包括：对所述第一YIQ图像中的Y通道图像进行傅里叶变换；对傅里叶变换后的Y通道图像进行下采样处理，得到第一子图像，所述下采样处理为降低图像的分辨率；对所述第一子图像进行上采样处理，得到第二子图像，所述上采样处理为提升图像的分辨率；将傅里叶变换后的Y通道图像与所述第二子图进行像素处理，得到第三子图；将所述第三子图进行时域滤波处理，得到目标频带；根据所述目标频段与所述第一YIQ图像中的Y通道图像确定多个Y通道图像信号；对所述多个Y通道图像信号进行放大，得到放大后的多个Y通道图像信号；根据拉普拉斯金字塔重建算法对放大后的多个Y通道图像信号进行合成；获取所述第一YIQ图像中的I通道图像与Q通道图像，并将合成后的多个Y通道图像与所述I通道图像和所述Q通道图像进行相加，形成特征图像序列。
可选的，所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列，包括：对所述第一YIQ图像中的Y通道图像进行傅里叶变换；对傅里叶变换后的Y通道图像进行空间滤波，得到不同空间分辨率的Y通道图像；对不同空间分辨率的Y通道图像进行时域滤波处理，获取目标频段；根据所述目标频段与所述不同空间分辨率的Y通道图像确定多个Y通道图像信号；对所述多个Y通道图像信号进行放大，得到放大后的多个Y通道图像信号；根据复数可操纵金字塔重建算法对放大后的多个Y通道图像 信号进行合成；获取所述第一YIQ图像中的I通道图像与Q通道图像，并将合成后的多个Y通道图像与所述I通道图像和所述Q通道图像进行相加，形成特征图像序列。
可选的，所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列，包括：对所述第一YIQ图像中的Y通道图像进行傅里叶变换；根据拉普拉斯金字塔对傅里叶变换后的Y通道图像进行分解；将分解后的Y通道图像进行里斯变换；对里斯变换后的Y通道图像进行正交和相位处理；对处理后的Y通道图像进行空间时域滤波；对空间时域滤波后的Y通道图像进行方法和相移，形成特征图像序列。
可以看出，所述轨道振动设备通过图像金字塔对第一YIQ图像中的Y通道图像进行分解，并对图像金字塔分解后的Y通道图像进行归一化、时空滤波以及线性放大处理，以放大轨道振动视频，提升振动数据提取的准确性，进而提高轨道振动检测的效率、准确和可靠性。
在一个可能的示例中，所述根据所述特征图像序列提取所述待测轨道的实际振动数据，包括：根据相位相关公式计算所述特征图像序列的交叉互功率谱；以及根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息；以及根据所述振动信息确定所述待测轨道的实际振动数据。
具体的，所述相位相关算法采用如下的公式计算所述特征图像序列的交叉互功率谱：
Figure PCTCN2020104833-appb-000001
其中，F a为a帧图像的傅立叶变换，
Figure PCTCN2020104833-appb-000002
为b帧图像的傅里叶变换的共轭信号，
Figure PCTCN2020104833-appb-000003
表示F a和
Figure PCTCN2020104833-appb-000004
的相关积的模，R表示交叉互功率谱(包含频域噪音)。
可以看出，振动检测设备通过对视频放大后的特征图像序列采用相位相关算法计算交叉互功率谱，以提高轨道振动检测的准确性。
在一个可能示例中，所述根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息，包括：根据所述交叉互功率谱的峰值位置选择自适应滤波器；以及根据所述自适应滤波器对所述交叉互功率谱进行滤波；以及对滤波后的交叉互功率谱进行反傅里叶变换；以及根据反傅里叶变换后的交叉互功率谱提取所述特征图像序列中像素的振动信息。
具体的，由于待测轨道的轨道振动是有方向性的，对处理后的Y通道图像进行归一化、时空滤波以及线性放大处理，得到的特征图像序列包含有冗余信号。对冗余信号都进行差值、加权和重构，将带来很高的运算量，而且其对振动信息的分析没有作用。通过采用主 成分分解对特征图像序列进行降维处理，再用自适应滤波器对降维后的特征图像序列进行滤波，能有效减小运算处理时间和噪声信号的影响。
可以看出，振动检测设备通过自适应滤波器对特征图像序列进行滤波，以减小轨道振动检测的运算量，以及提高轨道振动检测的效率，有效保证轨道的稳定性。
在一个可能的示例中，所述根据所述实际振动数据确定所述待测轨道的稳定性，包括：根据所述实际振动数据确定所述待测轨道的振动部位，所述振动部位包括轨道扣件或轨道接头或道床；以及根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性。
可以看出，振动检测设备通过待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性，以保证轨道振动检测的准确和可靠性。
在一个可能的示例中，所述振动部位包括轨道扣件或轨道接头时，所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性，包括：采集所述待测轨道的轨道扣件或轨道接头的N段振动视频；以及统计所述N段振动视频的平均振动频率、平均振幅和平均振动周期；以及比较所述平均振动频率、所述平均振幅和所述平均振动周期与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
具体的，图2c是本申请实施例提供的一种可能的轨道扣件的平面结构示意图。其中，混泥土轨枕扣件1042包括挡板301、普通道钉302和道钉垫板303。振动检测设备采集混泥土轨枕扣件1042的N段振动视频，并通过N端振动视频统计挡板301和普通道钉302在上下振动和左右振动方向的平均振动频率、平均振幅和平均振动周期。
可以看出，振动检测设备通过统计待测轨道的轨道扣件的平均振动数据，并将所述平均振动数据比较实际振动数据，以提高轨道振动检测的准确和可靠性。
在一个可能的示例中，所述振动部位包括道床时，所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性，包括：获取道床的道床系数；以及根据所述道床系数确定所述道床的振动刚度和振动弹性系数；以及比较所述振动刚度和所述振动弹性系数与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
可以看出，振动检测设备通过获取待测轨道的道床的振动刚度和振动弹性系数，并将所述振动刚度和所述振动弹性系数比较实际振动数据，以提高轨道振动检测的准确和可靠性。
与上述图2a所述的实施例一致，请参阅图3。图3是本申请实施例提供的另一种轨道 振动检测方法的流程示意图，应用于振动检测设备，所述方法包括：
S301、获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；
具体的，所述待测轨道的轨道振动视频可以通过以下操作获得：根据所述振动检测设备拍摄待测轨道，并得到所述待测轨道的第一视频和第二视频，其中第一视频和第二视频为所述待测轨道在相同时间内拍摄的不同视频；获取述第一视频对应的第一图像和第二视频对应的第二图像；对第一图像和第二图像进行重叠，再清除第一图像和第二图像不能重叠的像素点，并获得所述待测轨道的轨道振动视频。
S302、将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；
具体的，RGB颜色空间和YIQ颜色空间的线性转换关系为：
Y＝0.299*R+0.587*G+0.114*B；
I＝0.596*R-0.275*G-0.321*B；
Q＝0.212*R-0.523*G+0.311*B。
S303、对所述第一YIQ图像中的Y通道图像进行傅里叶变换，并对傅里叶变换后的Y通道图像进行图像金字塔分解；
其中，Y通道图像表示亮度信息，通过对Y通道图像进行傅里叶变换，以将时域的亮度信息转化为频域的相位变化。
其中，图像金字塔包括高斯金字塔、拉普拉斯金字塔、复数可操纵金字塔、里斯金字塔等等。
S304、对图像金字塔分解后的Y通道图像进行归一化、时空滤波以及线性放大处理；
S305、将处理后的Y通道图像与所述YIQ图像中的I通道图像和Q通道图像进行合成，并形成特征图像序列；
S306、根据相位相关公式计算所述特征图像序列的交叉互功率谱，并根据所述交叉互功率谱的峰值位置选择自适应滤波器；
S307、根据所述自适应滤波器对所述交叉互功率谱进行滤波，并对滤波后的交叉互功率谱进行反傅里叶变换；
S308、根据反傅里叶变换后的交叉互功率谱提取所述特征图像序列中像素的振动信息，并根据所述振动信息确定所述待测轨道的实际振动数据；
S309、根据所述实际振动数据确定所述待测轨道的振动部位，所述振动部位包括轨道扣件或轨道接头或道床；
S310、采集所述待测轨道的轨道扣件或轨道接头的N段振动视频，并统计所述N段振动视频的平均振动频率、平均振幅和平均振动周期；
S311、比较所述平均振动频率、所述平均振幅和所述平均振动周期与所述实际振动数据，并根据比较结果确定所述轨道扣件或轨道接头的稳定性。
可以看出，本申请实施例中所描绘的轨道振动检测方法，应用于轨道振动设备。所述轨道振动设备通过视频放大算法对轨道振动视频进行放大，并根据相位相关公式和反傅里叶变换提取轨道振动视频的实际振动数据，再比较轨道扣件的平均振动频率、平均振幅和平均振动周期与所述实际振动数据，以提升了振动数据提取的准确性，进而提高轨道振动检测的效率、准确和可靠性。
请参阅图4，图4是本申请实施例提供的另一种轨道振动检测方法的流程示意图。
S401、获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；
S402、处理所述第一RGB图像，并得到多个分辨率对应的多个子带图像序列；
S403、按照预设的分区灰度值筛选策略从所述多个子带图像序列筛选出用于放大处理的至少一个子带图像序列；
具体的，所述按照预设的分区灰度值筛选策略从所述多个子带图像序列筛选出用于放大处理的至少一个子带图像序列包括：确定所述多个子带图像序列的前景图像和后景图像，所述前景图像包含所述被检测产品的发生往复运动的区域影像，所述后景图像为除所述被检测产品的影像之外的影像；以及确定所述前景图像在子带图像中的面积占比；以及根据所述面积占比和预设的子分区计算公式确定所述前景图像的子分区数量，并按照所述子分区数量划分所述前景图像为多个前景子分区；以及针对每个子带图像序列，执行以下操作(1)-(6)以得到所述每个子带图像序列的灰度值变化频率：(1)确定当前处理的子带图像序列中每个子带图像的每个前景子分区的被测像素点；(2)根据每个被测像素点在当前子带图像序列包含的多个子带图像的灰度值生成每个被测像素点的灰度值时域变化波形图；(3)针对所述每个前景子分区执行如下(a)(b)(c)操作：(a)根据当前处理的前景子分区包含的多个被测像素点的灰度值时域变化波形图，确定所述当前处理的前景子分区是否包含灰度值周期性变化的被测像素点；(b)若是，则标记所述当前处理的前景子分区为 被选择的前景子分区；(c)若否，则标记所述当前处理的前景子分区为未被选择的前景子分区；(4)针对标记后的被选择的多个前景子分区，按照区域关联性将具有相邻关系的前景子分区拼接为振动参考区域；(5)确定所述振动参考区域中多个像素点中灰度值发生周期性变化的多个参考像素点，以及确定每个参考像素点的灰度值变化频率；(6)加权计算所述振动参考区域中所述多个参考像素点的灰度值变化频率，得到所述当前处理的子带图像序列的灰度值变化频率；以及根据所述每个子带图像序列的灰度值变化频率筛选出符合预设参考振动频率的至少一个子带图像序列。
其中，预设的子分区计算公式为：
Figure PCTCN2020104833-appb-000005
其中，x为面积占比，y为子分区数量，x大于0且小于或等于1。
其中，所述确定当前处理的子带图像序列中每个子带图像的每个前景子分区的被测像素点包括以下至少一种：所述每个前景子分区的边缘像素点；所述每个前景子分区的中间区域的像素点；所述每个前景子分区的随机筛选出来的多个像素点。
S404、放大处理所述至少一个子带图像序列，并得到放大处理后的至少一个子带图像序列；
S405、将所述放大处理后的至少一个子带图像序列和所述多个子带图像序列中除所述至少一个子带图像序列之外的子带图像序列进行融合处理，并形成特征图像序列；
S406、根据相位相关公式计算所述特征图像序列的交叉互功率谱，并根据所述交叉互功率谱的峰值位置选择自适应滤波器；
S407、根据所述自适应滤波器对所述交叉互功率谱进行滤波，并对滤波后的交叉互功率谱进行反傅里叶变换；
S408、根据反傅里叶变换后的交叉互功率谱提取所述特征图像序列中像素的振动信息，并根据所述振动信息确定所述待测轨道的实际振动数据；
S409、根据所述实际振动数据确定所述待测轨道的振动部位，所述振动部位包括轨道扣件或轨道接头或道床；
S410、采集所述待测轨道的轨道扣件或轨道接头的N段振动视频，并统计所述N段振动视频的平均振动频率、平均振幅和平均振动周期；
S411、比较所述平均振动频率、所述平均振幅和所述平均振动周期与所述实际振动数 据，并根据比较结果确定所述轨道扣件或轨道接头的稳定性。
可以看出，本申请实施例中所描绘的轨道振动检测方法，应用于轨道振动设备。所述轨道振动设备通过分区灰度值筛选策略对轨道振动视频进行处理，并从处理后的轨道振动视频中提取振动数据，以提升了振动数据提取的准确性，进而提高轨道振动检测的效率、准确和可靠性。
与上述图2a和图3所述的实施例一致，请参阅图5。图5是本申请实施例提供的一种振动检测设备500的结构示意图，所述振动检测设备500包括应用处理器510、存储器520、通信接口530以及一个或多个程序521，其中，所述一个或多个程序521被存储在上述存储器520中，并且被配置由上述应用处理器510执行，所述一个或多个程序521包括用于执行以下步骤的指令：获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；以及将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；以及通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；以及根据所述特征图像序列提取所述待测轨道的实际振动数据；以及根据所述实际振动数据确定所述待测轨道的稳定性。
可以看出，振动检测设备通过获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；根据所述特征图像序列提取所述待测轨道的实际振动数据；根据所述实际振动数据确定所述待测轨道的稳定性，提升振动数据提取的准确性以及工程应用的广泛性，进而提升轨道振动检测的效率、准确和可靠性。
在一个可能示例中，在所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列方面，所述程序中的指令具体用于执行以下操作：对所述第一YIQ图像中的Y通道图像进行傅里叶变换；以及对傅里叶变换后的Y通道图像进行图像金字塔分解；以及对图像金字塔分解后的Y通道图像进行归一化、时空滤波以及线性放大处理；以及将处理后的Y通道图像与所述YIQ图像中的I通道图像和Q通道图像进行合成，并形成特征图像序列。
在一个可能示例中，在所述根据所述特征图像序列提取所述待测轨道的实际振动数据方面，所述程序中的指令具体用于执行以下操作：根据相位相关公式计算所述特征图像序列的交叉互功率谱；以及根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信 息；以及根据所述振动信息确定所述待测轨道的实际振动数据。
在一个可能示例中，在所述根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息方面，所述程序中的指令具体用于执行以下操作：根据所述交叉互功率谱的峰值位置选择自适应滤波器；以及根据所述自适应滤波器对所述交叉互功率谱进行滤波；以及对滤波后的交叉互功率谱进行反傅里叶变换；以及根据反傅里叶变换后的交叉互功率谱提取所述特征图像序列中像素的振动信息。
在一个可能示例中，在所述根据所述实际振动数据确定所述待测轨道的稳定性方面，所述程序中的指令具体用于执行以下操作：根据所述实际振动数据确定所述待测轨道的振动部位，所述振动部位包括轨道扣件或轨道接头或道床；以及根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性。
在一个可能示例中，所述振动部位包括轨道扣件或轨道接头时，在所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性方面，所述程序中的指令具体用于执行以下操作：采集所述待测轨道的轨道扣件或轨道接头的N段振动视频；以及统计所述N段振动视频的平均振动频率、平均振幅和平均振动周期；以及比较所述平均振动频率、所述平均振幅和所述平均振动周期与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
在一个可能示例中，所述振动部位包括道床时，在所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性方面，所述程序中的指令具体用于执行以下操作：获取道床的道床系数；以及根据所述道床系数确定所述道床的振动刚度和振动弹性系数；以及比较所述振动刚度和所述振动弹性系数与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
上述主要从方法侧执行过程的角度对本申请实施例的方案进行了介绍。可以理解的是，振动检测设备为了实现上述功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所提供的实施例描述的各示例的单元及算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对振动检测设备进行功能单元的划分，例如，可 以对应各个功能划分各个功能单元，也可以将两个或两个以上的功能集成在一个处理单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。需要说明的是，本申请实施例中对单元的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
图6是本申请实施例提供的一种轨道振动检测装置600的功能单元组成框图。所述轨道振动检测装置600应用于振动检测设备，所述装置包括处理单元601和通信单元602。
其中，所述处理单元601，用于执行如上述方法实施例中的任一步骤，且在执行诸如发送等数据传输时，可选择的调用所述通信单元602来完成相应操作，下面进行详细说明。
所述处理单元601具体用于：获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；以及将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；以及通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；以及根据所述特征图像序列提取所述待测轨道的实际振动数据；以及根据所述实际振动数据确定所述待测轨道的稳定性。
可以看出，本申请实施例公开了一种轨道振动检测装置，通过获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；根据所述特征图像序列提取所述待测轨道的实际振动数据；根据所述实际振动数据确定所述待测轨道的稳定性，提升振动数据提取的准确性以及工程应用的广泛性，进而提升轨道振动检测的效率、准确和可靠性。
在一个可能的示例中，在所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列方面，所述处理单元601具体用于：对所述第一YIQ图像中的Y通道图像进行傅里叶变换；以及对傅里叶变换后的Y通道图像进行图像金字塔分解；以及对图像金字塔分解后的Y通道图像进行归一化、时空滤波以及线性放大处理；以及将处理后的Y通道图像与所述YIQ图像中的I通道图像和Q通道图像进行合成，并形成特征图像序列。
在一个可能的示例中，在所述根据所述特征图像序列提取所述待测轨道的实际振动数据方面，所述处理单元601具体用于：根据相位相关公式计算所述特征图像序列的交叉互功率谱；以及根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息；以及根 据所述振动信息确定所述待测轨道的实际振动数据。
在一个可能的示例中，在所述根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息方面，所述处理单元601具体用于：根据所述交叉互功率谱的峰值位置选择自适应滤波器；以及根据所述自适应滤波器对所述交叉互功率谱进行滤波；以及对滤波后的交叉互功率谱进行反傅里叶变换；以及根据反傅里叶变换后的交叉互功率谱提取所述特征图像序列中像素的振动信息。
在一个可能的示例中，在所述根据所述实际振动数据确定所述待测轨道的稳定性方面，所述处理单元601具体用于：根据所述实际振动数据确定所述待测轨道的振动部位，所述振动部位包括轨道扣件或轨道接头或道床；以及根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性。
在一个可能的示例中，所述振动部位包括轨道扣件或轨道接头时，在所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性方面，所述处理单元601具体用于：采集所述待测轨道的轨道扣件或轨道接头的N段振动视频；以及统计所述N段振动视频的平均振动频率、平均振幅和平均振动周期；以及比较所述平均振动频率、所述平均振幅和所述平均振动周期与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
在一个可能的示例中，所述振动部位包括道床时，在所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性方面，所述处理单元601具体用于：获取道床的道床系数；以及根据所述道床系数确定所述道床的振动刚度和振动弹性系数；以及比较所述振动刚度和所述振动弹性系数与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
本申请实施例还提供一种计算机存储介质，其中，该计算机存储介质存储用于电子数据交换的计算机程序，该计算机程序使得计算机执行如上述方法实施例中记载的任一方法的部分或全部步骤，上述计算机包括振动检测设备。
本申请实施例还提供一种计算机程序产品，上述计算机程序产品包括存储了计算机程序的非瞬时性计算机可读存储介质，上述计算机程序可操作来使计算机执行如上述方法实施例中记载的任一方法的部分或全部步骤。该计算机程序产品可以为一个软件安装包，上述计算机包括振动检测设备。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的 动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本申请所必须的。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置，可通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如上述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性或其它的形式。
上述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
上述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储器中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储器中，包括若干指令用以使得一台计算机设备(可为个人计算机、服务器或者网络设备等)执行本申请各个实施例上述方法的全部或部分步骤。而前述的存储器包括：U盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、移动硬盘、磁碟或者光盘等各种可以存储程序代码的介质。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成，该程序可以存储于一计算机可读存储器中，存储器可以包括：闪存盘、只读存储器(英文：Read-Only Memory，简称：ROM)、随机存取器(英文： Random Access Memory，简称：RAM)、磁盘或光盘等。
以上对本申请实施例进行了详细介绍，本文中应用了具体个例对本申请的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本申请的方法及其核心思想；同时，对于本领域的一般技术人员，依据本申请的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本申请的限制。

Claims (10)

  1. 一种轨道振动检测方法，其特征在于，应用于振动检测设备，所述方法包括：
    获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；
    将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；
    通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；
    根据所述特征图像序列提取所述待测轨道的实际振动数据；
    根据所述实际振动数据确定所述待测轨道的稳定性。
  2. 根据权利要求1所述的方法，其特征在于，所述通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列，包括：
    对所述第一YIQ图像中的Y通道图像进行傅里叶变换；
    对傅里叶变换后的Y通道图像进行图像金字塔分解；
    对图像金字塔分解后的Y通道图像进行归一化、时空滤波以及线性放大处理；
    将处理后的Y通道图像与所述YIQ图像中的I通道图像和Q通道图像进行合成，并形成特征图像序列。
  3. 根据权利要求1所述的方法，其特征在于，所述根据所述特征图像序列提取所述待测轨道的实际振动数据，包括：
    根据相位相关公式计算所述特征图像序列的交叉互功率谱；
    根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息；
    根据所述振动信息确定所述待测轨道的实际振动数据。
  4. 根据权利要求3所述的方法，其特征在于，所述根据所述交叉互动功率谱提取所述特征图像序列中像素的振动信息，包括：
    根据所述交叉互功率谱的峰值位置选择自适应滤波器；
    根据所述自适应滤波器对所述交叉互功率谱进行滤波；
    对滤波后的交叉互功率谱进行反傅里叶变换；
    根据反傅里叶变换后的交叉互功率谱提取所述特征图像序列中像素的振动信息。
  5. 根据权利要求1所述的方法，其特征在于，所述根据所述实际振动数据确定所述待测轨道的稳定性，包括：
    根据所述实际振动数据确定所述待测轨道的振动部位，所述振动部位包括轨道扣件或 轨道接头或道床；
    根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性。
  6. 根据权利要求5所述的方法，其特征在于，所述振动部位包括轨道扣件或轨道接头时，所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性，包括：
    采集所述待测轨道的轨道扣件或轨道接头的N段振动视频；
    统计所述N段振动视频的平均振动频率、平均振幅和平均振动周期；
    比较所述平均振动频率、所述平均振幅和所述平均振动周期与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
  7. 根据权利要求5所述的方法，其特征在于，所述振动部位包括道床时，所述根据所述待测轨道的振动部位和所述实际振动数据确定所述待测轨道的稳定性，包括：
    获取道床的道床系数；
    根据所述道床系数确定所述道床的振动刚度和振动弹性系数；
    比较所述振动刚度和所述振动弹性系数与所述实际振动数据，并根据比较结果确定所述待测轨道的稳定性。
  8. 一种轨道振动检测装置，其特征在于，应用于振动检测设备，所述装置包括处理单元和通信单元，其中：
    所述处理单元，用于获取待测轨道的轨道振动视频，并根据所述待测轨道的轨道振动视频选取第一RGB图像；以及用于将所述第一RGB图像由RGB颜色空间线性转换到YIQ颜色空间，获得第一YIQ图像；以及用于通过视频放大算法对所述第一YIQ图像中的Y通道图像进行处理，获得特征图像序列；以及用于根据所述特征图像序列提取所述待测轨道的实际振动数据；以及用于根据所述实际振动数据确定所述待测轨道的稳定性。
  9. 一种振动检测设备，其特征在于，包括处理器、存储器，所述存储器用于存储一个或多个程序，并且被配置由所述处理器执行，所述程序包括用于执行如权利要求1-7任一项所述的方法中的步骤的指令。
  10. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质用于存储计算机程序，所述计算机程序被处理器执行，以实现如权利要求1-7任一项所述的方法。
PCT/CN2020/104833 2019-04-26 2020-07-27 轨道振动检测方法及装置、振动检测设备 WO2021042908A1 (zh)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN201910346543 2019-04-26
CN201910835815.4A CN110631812B (zh) 2019-04-26 2019-09-04 轨道振动检测方法及装置、振动检测设备
CN201910835815.4 2019-09-04

Publications (1)

Publication Number Publication Date
WO2021042908A1 true WO2021042908A1 (zh) 2021-03-11

Family

ID=68970829

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/104833 WO2021042908A1 (zh) 2019-04-26 2020-07-27 轨道振动检测方法及装置、振动检测设备

Country Status (2)

Country Link
CN (1) CN110631812B (zh)
WO (1) WO2021042908A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115452125A (zh) * 2022-09-19 2022-12-09 哈尔滨工程大学 一种基于空间时向除法的结构振动视频测量方法及系统
KR20230049273A (ko) * 2021-10-06 2023-04-13 서울시립대학교 산학협력단 실시간 위상기반 모션확대기법을 이용한 미소진동 측정방법
CN117576091A (zh) * 2024-01-15 2024-02-20 深圳昱拓智能有限公司 一种基于视频检测的冷却塔风机振动检测方法及系统

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110631812B (zh) * 2019-04-26 2022-04-22 深圳市豪视智能科技有限公司 轨道振动检测方法及装置、振动检测设备
CN112113655B (zh) * 2020-09-21 2021-06-01 西南交通大学 一种地铁轨道扣件振动信号检测装置及健康评估方法
CN113464380B (zh) * 2021-08-09 2022-09-27 观为监测技术无锡股份有限公司 一种塔筒安全性能确定方法、装置、设备及存储介质
CN113682342A (zh) * 2021-08-25 2021-11-23 深圳市永安环保实业有限公司 轨道的异常检测方法及相关产品
CN113791140B (zh) * 2021-11-18 2022-02-25 湖南大学 基于局部振动响应的桥梁梁底内部无损检测方法和系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2205465A (en) * 1987-05-13 1988-12-07 Ricoh Kk Image transmission system
JPH08294142A (ja) * 1995-04-24 1996-11-05 Omron Corp 画像情報圧縮方法、及びこの方法を使用した画像表示装置
CN105865735A (zh) * 2016-04-29 2016-08-17 浙江大学 一种基于视频监控的桥梁振动测试与动力特性识别方法
CN109520690A (zh) * 2018-10-30 2019-03-26 西安交通大学 一种基于视频的旋转机械转子模态振型全局测量装置及方法
CN110631812A (zh) * 2019-04-26 2019-12-31 深圳市豪视智能科技有限公司 轨道振动检测方法及装置、振动检测设备

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9927225B2 (en) * 2015-05-28 2018-03-27 The Arizona Board Of Regents On Behalf Of The University Of Arizona Measuring dynamic displacement
CN206255021U (zh) * 2016-11-21 2017-06-16 武汉利德测控技术有限公司 一种货车智能在线监测装置
CN106525865A (zh) * 2016-11-30 2017-03-22 南京理工大学 一种基于图像处理的晶圆图像分析装置及方法
EP3242036B1 (de) * 2016-12-30 2020-10-28 Grundfos Holding A/S Verfahren zum erfassen eines zustandes eines pumpenaggregats
CN106657713B (zh) * 2016-12-30 2019-03-22 华中科技大学 一种视频运动放大方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2205465A (en) * 1987-05-13 1988-12-07 Ricoh Kk Image transmission system
JPH08294142A (ja) * 1995-04-24 1996-11-05 Omron Corp 画像情報圧縮方法、及びこの方法を使用した画像表示装置
CN105865735A (zh) * 2016-04-29 2016-08-17 浙江大学 一种基于视频监控的桥梁振动测试与动力特性识别方法
CN109520690A (zh) * 2018-10-30 2019-03-26 西安交通大学 一种基于视频的旋转机械转子模态振型全局测量装置及方法
CN110631812A (zh) * 2019-04-26 2019-12-31 深圳市豪视智能科技有限公司 轨道振动检测方法及装置、振动检测设备

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20230049273A (ko) * 2021-10-06 2023-04-13 서울시립대학교 산학협력단 실시간 위상기반 모션확대기법을 이용한 미소진동 측정방법
KR102651084B1 (ko) * 2021-10-06 2024-03-22 서울시립대학교 산학협력단 실시간 위상기반 모션확대기법을 이용한 미소진동 측정방법
CN115452125A (zh) * 2022-09-19 2022-12-09 哈尔滨工程大学 一种基于空间时向除法的结构振动视频测量方法及系统
CN117576091A (zh) * 2024-01-15 2024-02-20 深圳昱拓智能有限公司 一种基于视频检测的冷却塔风机振动检测方法及系统
CN117576091B (zh) * 2024-01-15 2024-04-09 深圳昱拓智能有限公司 一种基于视频检测的冷却塔风机振动检测方法及系统

Also Published As

Publication number Publication date
CN110631812B (zh) 2022-04-22
CN110631812A (zh) 2019-12-31

Similar Documents

Publication Publication Date Title
WO2021042908A1 (zh) 轨道振动检测方法及装置、振动检测设备
Liang et al. No-reference perceptual image quality metric using gradient profiles for JPEG2000
CN105678700B (zh) 基于预测梯度的图像插值方法及系统
CN105049851B (zh) 基于色彩感知的通用无参考图像质量评价方法
CN111277833B (zh) 一种基于多通带滤波器的多目标微小振动视频放大方法
US20170330301A1 (en) Information embedding device, information detecting device, information embedding method, and information detecting method
Ma et al. Visual horizontal effect for image quality assessment
CN103354617B (zh) 一种基于dct域的边缘强度压缩的图像质量客观评价方法
CN109447903B (zh) 一种半参考型超分辨率重构图像质量评价模型的建立方法
Gu et al. Structural similarity weighting for image quality assessment
DE112016002252T5 (de) Automatische erkennung von panoramagesten
CN104021523A (zh) 一种基于边缘分类的图像超分辨率放大的新方法
WO2021036661A1 (zh) 滤波方法、装置及相关产品
Kuo et al. Improved visual information fidelity based on sensitivity characteristics of digital images
Georgieva et al. Wavelet transform on multi-GPU for real-time pavement distress detection
Moorthy et al. A two-stage framework for blind image quality assessment
CN107220974A (zh) 一种全参考图像质量评价方法和装置
Yalman Histogram based perceptual quality assessment method for color images
Balanov et al. Reduced-reference image quality assessment based on dct subband similarity
Joy et al. RECENT DEVELOPMENTS IN IMAGE QUALITY ASSESSMENT ALGORITHMS: A REVIEW.
Maalouf et al. Offline quality monitoring for legal evidence images in video-surveillance applications
Li et al. Gradient-weighted structural similarity for image quality assessments
Wang et al. Blind quality assessment for 3D-synthesized images by measuring geometric distortions and image complexity
CN112950592B (zh) 基于高维离散余弦变换的无参考光场图像质量评价方法
Avcibas et al. Statistical analysis of image quality measures

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20859867

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20859867

Country of ref document: EP

Kind code of ref document: A1