Method and device for inspecting a pipeline and storage medium

Abstract

A method for inspecting a pipeline, comprising: S1 (S10): Placing a preset inspection device in a pipeline to be inspected, detecting the movement information of the inspection device, and capturing the pipeline images by the inspection device at preset intervals, wherein the movement information includes an angular velocity and an acceleration; S2 (S20): Calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times; characterized in that it further comprises: S3 (S30): Calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times; S4 (S40): Determining the position of the inspection device at the respective detection times in accordance with the first position and the second position corresponding to the respective detection times; wherein calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times comprises in particular the following: S31: Obtaining the initial pipeline images taken when placing the inspection device with respect to the first acquisition time; S32: Calculating the degree of overlap between the first piping images acquired at the first acquisition time and the initial piping images, and determining the second position of the first acquisition time in accordance with the degree of overlap; S33: Using a first piping image of the first acquisition time point as the initial piping image of a second acquisition time point and repeating steps S31-S33 until all acquisition times are calculated to obtain the second position of the inspection device at the respective acquisition times.

Description

TECHNICAL FIELD

[0001] The present invention relates to the technical field of pipeline inspection, in particular to a method and a device for inspecting a pipeline and a storage medium. The features of the preamble of claim 1 are known from DE 203 13 252 U1. STATE OF THE ART

[0002] Fluid transmission pipelines (such as water supply, drainage, and oil pipelines) are widely used as infrastructure in modern society. They are closely linked to people's daily lives and have a positive impact on improving residents' living standards and promoting regional economic development. With the increase in pipeline service life, aging, cracking, corrosion, and other damages inevitably occur under the long-term impact of the operating environment and the transportation of raw materials, which may lead to potential accidents such as fluid leakage and pipeline rupture. Therefore, regular inspection and maintenance of pipelines are required, and risk assessment is realized through the inspection of the pipeline's performance status.

[0003] Currently, there are four methods for inspecting pipelines: pipeline periscope, closed-circuit TV surveillance, pipeline sonar inspection, and human entry inspection. In pipeline periscope inspection, a high-resolution camera and a light source are inserted into an inspection shaft using a handle with an adjustable length. Video is recorded through ground control, and the inspection distance is tens of meters. The obtained video or image detects internal pipeline conditions such as cracks, blockages, and water leaks. Closed-circuit TV surveillance uses a video surveillance system. A crawler equipped with a camera lens records the internal situation of the pipeline through a wired video inspection to identify defects inside the pipeline.In pipeline sonar inspection, a fluid-filled pipeline is detected using an acoustic method. The acoustic detection unit is driven inside the pipeline by a crawler or manually pulled, and the internal condition is inspected by determining the contour of the pipeline. Personnel entry inspection involves personnel entering an inspection chamber and conducting an immediate inspection.

[0004] However, the existing pipeline inspection methods described above still have several shortcomings. For example, the pipeline periscope can only be used for single-point inspection, which cannot capture the entire pipeline; closed-circuit TV monitoring is a wired inspection method, which is inconvenient to operate and lacks accurate position information in the acquired video data; pipeline sonar is expensive and complicated to operate; and the human-entry inspection method requires a large number of personnel to participate, is labor-intensive, and has low efficiency. It poses a certain safety risk to construction personnel.

[0005] For this reason, the current technology still needs to be improved and further developed. CONTENT OF THE PRESENT INVENTION

[0006] In view of the above-described deficiencies of the prior art, it is an object of the present invention to provide a method for inspecting a pipeline with ease of use and the ability to accurately detect the position of the pipeline. This object is achieved by a method having the features of claim 1.

[0007] Specifically, the method according to the invention for inspecting the pipeline comprises the following: S1: Placing a preset inspection device in a pipeline to be inspected, detecting the movement information of the inspection device, and capturing the pipeline images by the inspection device at preset intervals, wherein the movement information includes an angular velocity and an acceleration; S2: Calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times; S3: Calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times; S4: Determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times, in order to position the pipeline in accordance with the positions of the respective detection times;

[0008] In the method for inspecting a pipeline, the interior of the inspection device is equipped with a gyroscope, an acceleration sensor and a recording device.

[0009] In the method for inspecting a pipeline, the acquisition of the movement information of the inspection device and the recording of the pipeline images by the inspection device at preset intervals are particularly as follows: Detecting the angular velocity of the inspection device using a gyroscope at preset intervals; detecting the speed of the inspection device using an acceleration sensor at preset intervals; and recording the pipeline images using a recording device at preset intervals.

[0010] In the method for inspecting a pipeline, calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times comprises, in particular, the following: S21: Obtaining the initial position information of the inspection device at a first detection time, wherein the initial position information includes the initial speed of the initial position, the initial displacement, and the initial angle; S22: Calculating the position information of the first detection time in accordance with the first movement information detected at the first detection time and the initial position information; S23: Using the position information of the first detection time as the starting position of a second detection time and repeating steps S21-S23 until all detection times are calculated to obtain the first position of the inspection device at the respective detection times.

[0011] In the method for inspecting a pipeline, calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times comprises, in particular, the following: S31: Obtaining the initial pipeline images taken when placing the inspection device with respect to the first acquisition time; S32: Calculating the degree of overlap between the first piping images acquired at the first acquisition time and the initial piping images, and determining the second position of the first acquisition time in accordance with the degree of overlap; S33: Using a first piping image of the first acquisition time point as the initial piping image of a second acquisition time point and repeating steps S31-S33 until all acquisition times are calculated to obtain the second position of the inspection device at the respective acquisition times.

[0012] In the method for inspecting a pipeline, the step of determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times in order to position the pipeline in accordance with the positions of the respective detection times comprises, in particular, the following: S41: Judge whether the first position and second position corresponding to each detection time are empty; S42: When the first position and the second position are absolutely not empty, the mean square deviation of the first position and the second position is calculated in accordance with a third position corresponding to a preset detection time, respectively; S43: Performing weighting for the first position and the second position with the mean square deviation as a weighting coefficient to obtain a position corresponding to the detection time, thereby positioning the pipeline in accordance with the positions of the respective detection times.

[0013] In the method for inspecting a pipeline, the step of determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times in order to position the pipeline in accordance with the positions of the respective detection times further comprises the following: S44: If the first position / second position is empty, the second position / first position is used as the position corresponding to the detection time to position the pipeline in accordance with the positions of the respective detection times.

[0014] In the method for inspecting a pipeline, the method further comprises, after the step of determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times in order to position the pipeline in accordance with the positions of the respective detection times: S5: Perform an inspection for all acquired pipeline images and obtain the acquisition time corresponding to a pipeline image with a crack; S6: Obtain the position corresponding to the acquisition time to position the crack.

[0015] A computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs are executable by one or more processors to implement one of the above-described steps in the method for inspecting a pipeline.

[0016] An inspection device in which a gyroscope, an acceleration sensor, a recording device, a processor, and a memory are arranged; wherein the gyroscope, the acceleration sensor, and the recording device are each connected to the memory; and wherein the memory stores a computer-readable program executable by the processor and the data acquired by the gyroscope, the acceleration sensor, and the recording device at the respective acquisition times;

[0017] When the processor executes the computer-readable program, the steps of the method for inspecting a pipeline according to one of the above points are implemented.

[0018] Advantages: Compared to the prior art, the present invention provides a method and apparatus for inspecting a pipeline, as well as a storage medium, the method comprising: placing a preset inspection device in a pipeline to be inspected, acquiring the movement information of the inspection device, and capturing pipeline images at preset intervals; calculating a first position and a second position of the inspection device at the respective acquisition times in accordance with the movement information acquired at the respective acquisition times and the pipeline relationship;Determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times, to position the pipeline in accordance with the positions of the respective detection times. Using the inertial navigation and video combination positioning algorithm, the present invention extracts the position information to achieve accurate positioning of the pipeline to accurately locate cracks in the pipeline in accordance with the captured images and the detected pipeline position. At the same time, the method for inspecting a pipeline can be widely applied to the conventional inspection of fluid pipelines such as urban drainage pipes and mains water pipes. SHORT DESCRIPTION OF THE DRAWING Fig. 1 shows a flowchart of a preferred embodiment of a method for inspecting a pipeline according to the present invention. Fig. 2 shows a schematic structural diagram of a preferred embodiment of an apparatus for inspecting a pipeline according to the present invention. DETAILED DESCRIPTION

[0019] The present invention provides a method and apparatus for inspecting a pipeline, as well as a storage medium. The present invention will be explained in more detail below in conjunction with figures and embodiments so that the object, technical solutions, and advantages of the present invention become clearer and more apparent. It is understood that the detailed embodiments described here serve only to illustrate the present invention and are not intended to limit the present invention.

[0020] Those skilled in the art will understand that the singular forms "a," "an," "the," and "this" as used herein may include plural forms unless expressly stated otherwise. It is further understood that the phrase "comprising" as used in the description of the present invention refers to the existence of the described features, integers, steps, operations, elements, and / or components, while not excluding the existence of one or more other features, integers, steps, operations, elements, components, and / or their combination. It is understood that when an element is referred to as being "connected" or "coupled" to another element, the element may be directly "connected" or "coupled" to the other element, or an intervening intermediate element may also exist.Furthermore, the "connection" or "coupling" as used herein may include a wireless connection or wireless coupling. The term "and / or" as used herein includes all or a unit of one or more associated listed elements and all combinations.

[0021] Those skilled in the art will understand that all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. It is also understood that terms as defined in the general dictionary have meanings consistent with those associated with the prior art, and unless specifically defined as such, they are not explained with an idealized or overly formal meaning.

[0022] The content of the present invention is explained in more detail below in connection with figures and embodiments.

[0023] See Fig. 1, Fig. Figure 1 shows a flowchart of a preferred embodiment of a method for inspecting a pipeline according to the present invention. The method further comprises the following: S10: Placing a preset inspection device in a pipeline to be inspected, detecting the movement information of the inspection device, and capturing the pipeline images by the inspection device at preset intervals, wherein the movement information includes an angular velocity and an acceleration; S20: Calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times; S30: Calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times; S40: Determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times, in order to position the pipeline in accordance with the positions of the respective detection times.

[0024] The present embodiment provides a method for inspecting a pipeline by placing an inspection device in a pipeline to be inspected, and by the inspection device acquiring its own movement information and the pipeline images of the pipeline to be inspected at preset intervals, a first position and a second position of a corresponding inspection time point are calculated in accordance with the movement information and the pipeline images, wherein a final position of the corresponding acquisition time point is determined in accordance with the first position and the second position, and wherein a calculation of the first position, the second position, and the final position is performed sequentially for each acquisition time point to obtain a final position corresponding to the respective acquisition times.This allows for accurate positioning of the pipeline. In the present embodiment, the positions obtained using the inertial navigation positioning algorithm and the video combination positioning algorithm are integrated to achieve accurate positioning of the pipeline, precisely locating cracks in the pipeline in accordance with the captured images and the detected pipeline position. At the same time, the method for inspecting a pipeline can be widely applied to the conventional inspection of fluid pipelines such as urban drainage pipes and mains water pipes.

[0025] Specifically, in step S10, a gyroscope, an acceleration sensor, and a recording device are mounted in the inspection device. The gyroscope detects the angular velocity of the inspection device, and the acceleration sensor detects the speed of the inspection device. The recording device also records the pipeline images. Furthermore, the recording device includes at least one camera and an additional infrared light, allowing the recording device to record the pipeline images.Accordingly, placing a preset inspection device in the pipeline to be inspected specifically involves mounting a gyroscope, an acceleration sensor, and a recording device in a capsule-shaped / spherical inspection device, performing an inspection of the placed pipeline for the inspection device, and acquiring the operating information of the inspection device and pipeline images at preset intervals. In practical applications, after the inspection is completed, the pipeline can be positioned according to the acquired data. Likewise, each time a group of inspection data is acquired, the corresponding inspection data from the acquisition time can be processed to perform positioning for the acquisition time.

[0026] Furthermore, the preset time period can be preset, such as 1 second. Acquisition is performed at preset intervals to obtain a plurality of acquisition times, respectively, and the time when the inspection device is placed in a pipeline to be inspected is recorded as the initial acquisition time. Here, the acquisition times generated by acquisition at preset intervals are respectively recorded as the initial acquisition time, the first acquisition time, the second acquisition time, ......, and the Nth acquisition time.In practical applications, the intervals between the N detection times can be different, namely, the N detection times can be set in advance, at each inspection time, the inspection device reads its own movements and the captured pipeline images, which can set different detection times according to specific situations of the pipeline, so as to further improve the accuracy of pipeline positioning.

[0027] In step S20, the gyroscope and the acceleration sensor of the inspection device preferably inspect the angular velocity and acceleration of the device in real time, which refers to reading the angular velocity and acceleration detected by the gyroscope and the acceleration sensor at each detection time and calculating the first position corresponding to each detection time using the inertial navigation algorithm.

[0028] By way of example, step S20 for calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times comprises in particular the following: S201: Obtaining the initial position information of the inspection device at a first detection time, wherein the initial position information includes the initial speed of the initial position, the initial displacement, and the initial angle; S202: Calculating the position information of the first detection time in accordance with the first movement information detected at the first detection time and the initial position information; S203: Using the position information of the first detection time point as the starting position of a second detection time point and repeating steps S201-S203 until all detection times are calculated to obtain the first position of the inspection device at the respective detection times.

[0029] Specifically, for the first detection time, the initial detection time is when the inspection device is placed in a pipeline to be inspected, namely, the position information when the inspection device is placed in the pipeline to be inspected is the initial position information, where at the initial placement, the initial velocity V0 is 0 and the initial displacement S0 is 0, and for the angular velocity, the initial angular velocity θ0 is detected by the gyroscope, based on the classical Newton's law of motion and the principle of acceleration integration, if the initial target velocity V0, the initial displacement S0 and the initial angle θ0 as well as the acceleration a corresponding to the first detection time and the interval t between the first detection time and the initial time are known, the acceleration can be integrated once,to obtain a first velocity V1 of the first detection time, wherein the first velocity is integrated once to obtain a first displacement S1 of the first detection time, and wherein the first angular velocity signal w measured by the gyroscope is integrated once with the movement time to obtain the first angle θ1. Accordingly, the calculation formula for the first velocity V1, the first displacement S1, and the first angle θ1 can be as follows: {V1=∫1a+V0S1=∫tV1+S0=∫t(∫ta+V0)+S0θ1=∫tw+θ0

[0030] Here, t represents the interval between the first recording time and the initial recording time.

[0031] Preferably, the first displacement S1 and the first angle θ1 corresponding to the first detection time are determined, then the coordinates of the inspection device are calculated using the active positioning algorithm. In the active positioning algorithm, detection and calculation are performed by moving an object by the displacement and angle of its own movement to determine the position of the moving object.In the present embodiment, the initial coordinates (x0, y0) of the inspection device at the initial time are assumed, the first displacement S1 and the first angle θ1 of the inspection device at the first detection time are calculated, and the following calculation formula can be used to calculate the coordinates (x1, y1) of the first position of the inspection device at the first detection time, and the calculation formula can be expressed as follows:. {x1=x0+S1*cosθ1y1=y0+S1*cosθ1

[0032] For the second detection time, the first detection time may be used as the initial detection time, wherein the above process is repeated to obtain the coordinates of the corresponding first position, and wherein the above steps are repeated sequentially to calculate the coordinates (x2, y2) of the first position corresponding to each detection time, and wherein the coordinates (x n , y n ) of the first position corresponding to the Nth acquisition time can be expressed as follows: {xn=xn−1+Sn*cos θnyn=yn−1+Sn*cos θn

[0033] Furthermore, the pipeline images in step S30 are the pipeline images of a pipeline to be inspected recorded by the video recording device carried in the inspection device, wherein the pipeline images are used, on the one hand, to calculate the position of the inspection device and, on the other hand, can be used to inspect the cracks of the pipeline.First, the second position of the inspection device is the second position of the inspection device in the pipeline to be inspected, the second position is calculated by the overlap of the images, namely, the interval between the adjacent acquisition times and the overlap degree of the pipeline images is used to calculate the displacement of the inspection device in the pipeline to be inspected within the interval, and the displacements before the respective acquisition times are accumulated, whereby the second position corresponding to the respective acquisition times can be obtained.

[0034] The step S30 for calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times comprises in particular the following: S301: Obtaining the initial pipeline images taken when placing the inspection device with respect to the first acquisition time; S302: Calculating the degree of overlap between the first piping images acquired at the first acquisition time and the initial piping images, and determining the second position of the first acquisition time in accordance with the degree of overlap; S303: Using a first piping image of the first acquisition time point as the initial piping image of a second acquisition time point and repeating steps S301-S303 until all acquisition times are calculated to obtain the second position of the inspection device at the respective acquisition times.

[0035] Specifically, for the first acquisition time, the pipeline images acquired when the inspection device was placed are used as the initial pipeline images, and the first pipeline images acquired at the first acquisition time are read. First, the feature points of the overlapped image of the initial pipeline image and the first pipeline image are extracted based on SIFT (the SIFT-based extraction of the feature point of the overlapped image of two images belongs to the current computation in the field of image processing and will not be explained in detail here);then, the extracted feature points of the overlapped image are filtered using the RANSAC algorithm to obtain the overlapped feature points (the SIFT-based extraction of the feature point of the overlapped image of two images belongs to the current computation in the field of image processing and is not explained in detail here);Finally, an affine transformation matrix of the initial piping image and the first piping image is calculated according to the feature points obtained by filtering. The image center of the initial piping image is transformed by the affine transformation matrix to determine the displacement of the center of the first piping image with respect to the center of the initial piping image. In this embodiment, the distance calculation process of the affine transformation matrix and the displacement will not be explained in detail; only an example is given. For example, the affine transformation matrix is ​​as follows: [0.9344486593890110.04985775409285160−0.001139111530236740.917963049581227012.424564031206722.56541232561081]

[0036] The affine transformation matrix transforms the image center of the initial pipeline image, resulting in a horizontal displacement Δx of the image center of the initial pipeline image of 16.8 pixels and a vertical displacement Δy of 3.5 pixels. Therefore, the relative displacement pixel Δs of the overlap image is 17.2 pixels.

[0037] After determining the displacement ΔS1 of the first detection time, the displacement can preferably be corrected using the following formula to obtain the first displacement S1: S1=f×Δs1

[0038] Where f is the correction coefficient, ΔS k represents the shift to the K-th acquisition time.

[0039] The correction coefficient is calculated using the total length L of the image of the pipeline to be inspected and the number of acquisition times N. Moreover, the N acquisition times only send N-1 displacements, accordingly, the calculation formula of the correction coefficient f can be as follows: f=L∑k=1N−1Δsk

[0040] For the second acquisition time, the first acquisition time can be used as the initial acquisition time, namely, the first piping image is used as the initial piping image, by repeating the above steps, a second displacement S2 of the second acquisition time is obtained, wherein the calculation formula of the second displacement S2 can be as follows: S2=S1+f×Δs2=f×∑k=12Δsk

[0041] By repeating the above steps, the n-th shift S nof the image center at the nth acquisition time can be expressed as follows: Sn=f×∑k=1nΔsk

[0042] Since the image overlapping algorithm obtains the displacement at each acquisition time, after determining the displacement at each acquisition time, a second position of the inspection device in the pipeline to be inspected is calculated at each acquisition time in accordance with the position of the image center point of the initial pipeline image and the actual pipeline circuit diagram.

[0043] Preferably, after detecting the first position and the second position at the respective detection times in step S40, the first position and the second position can be integrated to obtain the final position of the inspection device at the respective times. Furthermore, the integration of the first position and the second position is carried out sequentially according to the order of the detection times, wherein the calculation of the first position and the second position corresponding to each detection time is carried out after the final calculation of the position of the last time of this detection time.For each acquisition time point, the first position of the acquisition time point can be calculated first, then the second position of the acquisition time point is calculated, and finally the final position is calculated according to the first position and the second position; likewise, the second position of the acquisition time point can be calculated first, then the first position of the acquisition time point is calculated, and finally the final position is calculated according to the first position and the second position. For all acquisition times, the first position, second position, and final position of the current acquisition time point are calculated in the chronological order of the acquisition times, then the first position, second position, and final position of a next acquisition time point are calculated until all acquisition times are completely calculated.In this way, the final position is used as the starting position of the next acquisition time, which can improve the accuracy of the position calculation.

[0044] By way of example, step S40 for determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times in order to position the pipeline in accordance with the positions of the respective detection times comprises, in particular, the following: S401: Judging whether the first position and the second position corresponding to each detection time point are empty; if the first position and the second position are absolutely not empty, step S402 is performed; if the first position / the second position is empty, step S404 is performed.

[0045] Specifically, if the first position is empty, it means that the angular velocity or acceleration of the inspection device is being read at the acquisition time, whereby the first position corresponding to the acquisition time cannot be determined. If the second position is empty, it means that the piping images acquired at the acquisition time and the piping images acquired at the last acquisition time have no overlapping feature points or few overlapping feature points, whereby the affine transformation matrix of the two cannot be calculated, so the displacement of the current acquisition time with respect to the last acquisition time cannot be determined, thus the second position of the current acquisition time cannot be determined.

[0046] S402: The mean square deviation of the first position and the second position is calculated in accordance with a third position corresponding to a preset detection time, respectively.

[0047] In particular, the third position is a position of the inspection device in the pipeline to be inspected at the detection time obtained in advance by a simulation experiment, wherein a mean square deviation of the first position and the second position with respect to the third position is calculated, respectively.

[0048] S403: Weighting is performed on the first position and the second position using the mean square deviation as a weighting coefficient to obtain a position corresponding to the detection time, thereby positioning the pipeline in accordance with the positions of the respective detection times. S404: The second position / first position is used as the position corresponding to the detection time, thereby positioning the pipeline in accordance with the positions of the respective detection times.

[0049] In one embodiment of the present invention, the method for inspecting a pipeline may further comprise: S50: Perform an inspection for all acquired piping images and obtain the acquisition time corresponding to a piping image with a crack.

[0050] Specifically, performing an inspection for all acquired pipeline images involves inspecting each of the acquired pipeline images one after another. However, in practical applications, the imaging device has higher light requirements when obtaining the pipeline images. Uneven light will cause streaks on the acquired images, while excessive longitudinal noise will render the inspection results meaningless. Therefore, grayscale correction of the image should be performed before cracks are detected in the pipeline images. The grayscale correction process can be as follows:

[0051] First, the pipeline image I(p) with cracks is mainly represented by a pipeline background signal I b (p), a pipeline crack signal I c (p), a random noise signal I n(p) and a noise signal I caused by the unevenness of the light a (p), accordingly the piping pattern I(p) can be expressed as follows: I(p)=Ib(p)+Ic(p)+In(p)+Ia(p)

[0052] Second, the difference between each pixel of the image line and the average value of the image line is determined to obtain a difference image line, then a least squares fit is performed for the difference image line using a sine function to determine the sine parameters of the I a (p) component, whereby the I a (p) gray level difference value of each pixel is calculated, which is considered the I a (p) compensation component of the unit gray level is used, where the image after gray level correction is as follows: I'(p)=I(p)−Ia(p)=Ib(p)+Ic(p)+In(p)

[0053] Preferably, after grayscale correction of the image, crack damage detection is performed for the post-corrected image. In the present embodiment, crack detection is performed for a post-corrected pipeline image using a multi-scale strategy crack damage detection model. In the detection process, an image may first be divided into three levels: a pixel-level image, a unit-level image, and a block-level image. Then, the image is segmented using two calculation processes to perform crack detection. The detection process is specifically as follows, by way of example:

[0054] First, an image grayscale calculation is performed from the pixel level to the unit level. This calculation can eliminate the influence of random noise while reducing the calculation time. For example, a 4 × 4 pixel window is selected as a unit, and the unit grayscale value is calculated: Grayunit=λMinunit+(1−λ)Meanunit

[0055] Gray unit for the calculated grayscale value of the unit, Min unit for the minimum grayscale value of the unit, Mean unit for the mean of the unit gray levels and λ for the weighting of the minimum gray value of the unit.

[0056] The formula of λ can be as follows: λ=f(Meanunit,Meanwhole,Devunit,Devwhole)=DevunitDevwholeDevunitDevwhole+MeanunitMeanwhole

[0057] Mean whole for the overall mean of the image, Dev unit for the unit variance and Dev whole for the total variance of the image. The larger the unit mean, the smaller λ, so the grayscale value of the unit is closer to the image mean, meaning that the possibility that the unit contains cracks is always smaller; conversely, the possibility is always greater.

[0058] Second, difference histogram segmentation is performed for the unit-level based image block, and the statistical value Dh g of the pixel change amount of a certain gray level g in the eight neighborhoods (the surrounding eight pixels) is calculated, the calculation formula can be as follows: Dhg=∑ii=n−1∑jj=m−118∑h=−1h=1∑k=−1k=1(Gray((i+h)*m+(j+k))−g) g∈[0.255]

[0059] Where Gray stands for the grayscale value, while n, m stand for the number of image lines. Based on this, the threshold value for grayscale segmentation T Dh determined, where the formula is as follows: TDh=g(Maxg=0g=255(Dhg)) g∈[0.255]

[0060] Based on the threshold, a grayscale image is converted into a binary image Gc(i,j)={255Gray(i,j)≤TDhi∈[0,n−1] 0Gray(i,j)>TDhj∈[0,m−1] converted.

[0061] Where m, n represent the number of horizontal and vertical units of the block diagram, respectively.

[0062] Finally, by edge tracking for a detected crack region, an edge vector of the crack can be obtained, then the geometric features (including point features, line features and surface features) of the crack are extracted, and based on the geometric features, the type of the crack is determined.

[0063] S60: Obtain the position corresponding to the detection time to position the crack.

[0064] Specifically, the capture time of a pipeline image with cracks is obtained, then the position of the crack is determined in accordance with the calculated positions corresponding to the respective capture times to position the crack.

[0065] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement one of the above-described steps in the method for inspecting a pipeline.

[0066] The present invention further provides an inspection device as described in Fig. 2 illustrates an inspection device in which a gyroscope 100, an acceleration sensor 200, a recording device 300, a processor 400, and a memory 500 are arranged; wherein the gyroscope 100, the acceleration sensor 200, and the recording device 300 are each connected to the memory 500; and wherein the memory 500 stores a computer-readable program executable by the processor and the data acquired by the gyroscope, the acceleration sensor, and the recording device at the respective acquisition times; and wherein, upon execution of the computer-readable program by the processor 400, the steps of the method for inspecting a pipeline according to any of the above points are implemented.

[0067] The present invention can further provide a pipeline inspection system comprising an inspection device and a terminal, wherein the inspection device is equipped with a gyroscope, an acceleration sensor, and a recording device, and wherein a terminal comprises a processor and a memory; and wherein the inspection device is placed in a pipeline to be inspected and acquires the pipeline images and inherent movement information of the pipeline to be inspected, and wherein the terminal receives the pipeline images and inherent movement information acquired by the inspection device and calculates the position of the inspection device in the pipeline to be inspected at the respective acquisition times in accordance with the pipeline images and the movement information.

[0068] Preferably, the memory stores a computer-readable program that can be executed by the processor and stores the data acquired by the gyroscope, the acceleration sensor, and the recording device at the respective acquisition times; and wherein, upon execution of the computer-readable program by the processor, the steps of the method for inspecting a pipeline according to any one of the above points are implemented.

[0069] In the above method, the process that the plurality of instructions in the storage medium, the inspection device and the terminal are loaded and executed by the processor is already explained in detail, and will not be explained in detail here.

Claims

[1] A method of inspecting a pipeline comprising: S1 (S10): Placing a preset inspection device in a pipeline to be inspected, detecting the movement information of the inspection device, and capturing the pipeline images by the inspection device at preset intervals, wherein the movement information includes an angular velocity and an acceleration; S2 (S20): Calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times; characterized by that it also includes: S3 (S30): Calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times; S4 (S40): Determining the position of the inspection device at the respective detection times in accordance with the first position and the second position corresponding to the respective detection times; wherein calculating a second position of the inspection device at the respective acquisition times in accordance with the pipeline images taken at the respective acquisition times comprises in particular the following: S31: Obtaining the initial pipeline images taken when placing the inspection device with respect to the first acquisition time; S32: Calculating the degree of overlap between the first piping images acquired at the first acquisition time and the initial piping images, and determining the second position of the first acquisition time in accordance with the degree of overlap; S33: Using a first piping image of the first acquisition time point as the initial piping image of a second acquisition time point and repeating steps S31-S33 until all acquisition times are calculated to obtain the second position of the inspection device at the respective acquisition times. [2] Method for inspecting a pipeline according to claim 1, characterized by that the interior of the inspection device is equipped with a gyroscope, an acceleration sensor and a recording device. [3] Method for inspecting a pipeline according to claim 2, characterized by that the acquisition of the movement information of the inspection device and the recording of the pipeline images by the inspection device at preset intervals is in particular as follows: Detecting the angular velocity of the inspection device using a gyroscope at preset intervals; detecting the acceleration of the inspection device using an acceleration sensor at preset intervals; and recording the pipeline images using a recording device at preset intervals. [4] Method for inspecting a pipeline according to claim 1, characterized by that calculating a first position of the inspection device at the respective detection times in accordance with the movement information detected at the respective detection times comprises in particular the following: S21: Obtaining the initial position information of the inspection device at a first detection time, wherein the initial position information includes the initial speed, the initial displacement, and the initial angle; S22: Calculating the position information of the first detection time in accordance with the first movement information detected at the first detection time and the initial position information; S23: Using the position information of the first detection time as the starting position of a second detection time and repeating steps S21-S23 until all detection times are calculated to obtain the first position of the inspection device at the respective detection times. [5] Method for inspecting a pipeline according to claim 1, characterized by that the step of determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times comprises in particular the following: S41: Judge whether the first position and second position corresponding to each detection time are empty; S42: When the first position and the second position are absolutely not empty, the mean square deviation of the first position and the second position is calculated in accordance with a third position corresponding to a preset detection time, respectively; S43: Performing weighting for the first position and the second position with the mean square deviation as a weighting coefficient to obtain a position corresponding to the detection time. [6] Method for inspecting a pipeline according to claim 5, characterized by that the step of determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times further comprises: S44: If the first position / second position is empty, the second position / first position is used as the position corresponding to the detection time. [7] Method for inspecting a pipeline according to claim 1, characterized by that the method, after the step of determining the position of the inspection device at the respective detection times in accordance with the first position and second position corresponding to the respective detection times, further comprises: S5: Perform an inspection for all acquired pipeline images and obtain the acquisition time corresponding to a pipeline image with a crack; S6: Obtain the position corresponding to the acquisition time to position the crack. [8] Computer-readable storage medium, characterized bythat the computer-readable storage medium stores one or more programs and the one or more programs can be executed by one or more processors to implement the steps of the method for inspecting a pipeline according to one of claims 1 to 7. [9] Inspection device in which a gyroscope (100), an acceleration sensor (200) and a recording device (300) are arranged, characterized by that it comprises a processor (400), a memory (500) and a communication bus; wherein the gyroscope (100), the acceleration sensor (200) and the recording device (300) are each connected to the memory (500); and wherein the memory (500) stores a computer-readable program that can be executed by the processor (400) and the data acquired by the gyroscope (100), the acceleration sensor (200) and the recording device (300) at the respective acquisition times; and wherein, upon execution of the computer-readable program by the processor (400), the steps of the method for inspecting a pipeline according to one of claims 1 to 7 are implemented.

Images