Search

CN-122020002-A - Phi-OTDR tube cleaner collision signal identification and positioning device and method

CN122020002ACN 122020002 ACN122020002 ACN 122020002ACN-122020002-A

Abstract

The invention discloses a phi The OTDR tube cleaner collision signal identification, positioning and device comprises a data preprocessing module, a kurtosis preliminary judging module, an edge detecting module, a line segment segmentation module and an inverted V matching module, wherein the data preprocessing module preprocesses original time-space diagram data input into the data and outputs 0-1 binary images, the kurtosis preliminary judging module judges images before binarization after band-pass filtering, the edge detecting module detects the 0-1 binary images if judging is passed, left and right oblique line detection results are output, the line segment segmentation module carries out line segment segmentation operation on the left and right oblique line detection results respectively and outputs left and right oblique line segment segmentation results to respectively obtain the number of left and right oblique line segments, the inverted V matching module adopts an iterative mode and takes a pair of left and right oblique line segments with 'the upper end point closest to each time' to be matched into an inverted V pattern according to a distance threshold value, and the final inverted V pattern detection result is output. The invention has good interpretability and stronger anti-interference performance, and can obtain good application effect in the pipe cleaner positioning task.

Inventors

  • ZHENG ZHEYUAN
  • SHA ZHOU
  • RUI XIAOBO
  • HUANG XINJING
  • FENG HAO
  • ZHANG HAIFENG

Assignees

  • 天津大学

Dates

Publication Date
20260512
Application Date
20260206

Claims (7)

  1. 1. Phi shaped The OTDR pipe cleaner collision signal identification and positioning device is characterized by comprising a data preprocessing module, a kurtosis preliminary judging module, an edge detecting module, a line segment dividing module and an inverted V matching module; The data preprocessing module preprocesses the data input into the original space-time diagram and outputs a 0-1 binary image; the kurtosis preliminary judging module judges the images before binarization after the band-pass filtering, and if the images pass the judgment, the edge detecting module is carried out; the edge detection module detects the 0-1 binary image and outputs left and right oblique line detection results; the line segment segmentation module respectively carries out line segment segmentation operation on the left and right oblique line detection results, outputs the left and right oblique line segment segmentation results, and respectively obtains the number of the left and right oblique line segments; the inverted V matching module adopts an iterative mode, a pair of left and right oblique line segments with the upper end points closest to each other are taken each time, the inverted V pattern is matched according to a distance threshold value, and a final inverted V pattern detection result is output.
  2. 2. A phi according to claim 1 The OTDR pipe cleaner collision signal recognition and positioning device is characterized in that the kurtosis preliminary judgment module judges images before binarization after band-pass filtering as follows: judging all the points of the 0-1 binary sequence kbin =0, and if kbin =1 points exist in the left and right 5 adjacent points, setting kbin of the judged points to be 1; if the binary kurtosis sequence has a '1 region' continuously exceeding a threshold value, the edge detection module is carried out if the binary kurtosis sequence passes the judgment, otherwise, the situation that a target event does not exist in the current time-space diagram is judged, and the algorithm is terminated.
  3. 3. A phi according to claim 1 The OTDR pipe cleaner collision signal recognition and positioning device is characterized in that the edge detection module detects 0-1 binary images and outputs left and right oblique line detection results as follows: When both criteria exceed the corresponding threshold values, they are respectively recorded as And When the bright pixel is judged to belong to a certain oblique line and is reserved, otherwise, the pixel is converted into a dark pixel; Wherein, the An absolute criterion; A relative criterion; The bright pixel density of the on-line neighborhood, Bright pixel density of the background neighborhood.
  4. 4. A phi according to claim 1 The OTDR pipe cleaner collision signal recognition and positioning device is characterized in that the line segment segmentation module respectively performs line segment segmentation operation on left and right oblique line detection results as follows: Light pixel New coordinates in a θ -rotating coordinate system : In a theta-rotation coordinate system, performing line segment segmentation operation according to x theta coordinate difference among all the bright pixels, extracting x theta coordinates of all the bright pixels and sequencing from small to large; and calculating the difference of x and theta coordinates between every two, and marking the difference as a vector form: wherein , Representing the total number of bright pixels in the binary image; the threshold value of the x theta coordinate difference of the bright pixel is recorded as For the ith bright pixel and the (i+1) th bright pixel, if The two are judged to be classified into different oblique lines.
  5. 5. A phi according to claim 4 OTDR pig collision signal discernment and positioner, its characterized in that, the line segment segmentation module still includes: The aspect ratio is calculated in the theta-rotation coordinate system, Is the first A segment in which the bar is temporarily identified; 。
  6. 6. A phi according to claim 1 OTDR pig collision signal discernment and positioner, its characterized in that, the matching module of falling V is: For each left oblique line segment obtained in the line segment segmentation module, obtaining the maximum ordinate in the theta-rotation coordinate system Is used as the endpoint, records the rectangular coordinates For the right oblique line segment, the image is turned left and right, then the same operation is carried out to find out the corresponding vertex pixel, and then the rectangular coordinates of the vertex pixel are recorded after the image is turned back; Calculating Euclidean distance between each left oblique line end point and each right oblique line end point, and recording the distance between the ith left oblique line end point and the jth right oblique line end point as Adopting iteration to match only one inverted V each time, and setting the endpoint distance threshold as Taking out Minimum value of (2) And (3) with Comparing; If it is Matching the i-th left oblique line with the j-th right oblique line to form an inverted V, removing the matched oblique line from the matching process, and repeatedly executing matching and removing operations until a certain judgment Indicating that all inverted V in the space-time diagram are successfully identified, and the remaining diagonal lines are isolated diagonal lines.
  7. 7. Phi shaped An OTDR pig collision signal identification and localization method, characterized in that the method comprises: Based on each column of original time domain signal Respectively performing 20-200Hz band-pass filtering to obtain And to the Fourier transformed signal Taking its high frequency component Performing amplitude correction, and calculating to obtain final filtering result ; For each final filtering result Calculating the fractional order kurtosis to obtain a kurtosis sequence k (i), and binarizing the kurtosis sequence k (i) to obtain a binary sequence kbin (i); judging all kbin =0 points once, and if kbin =1 points exist in the left and right 5 adjacent points, setting kbin of the judged points to be 1; judging whether a continuous 1 region with the length exceeding a threshold value exists in the binary kurtosis sequence, and if so, carrying out Gaussian blur processing on the filtered time-space diagram; scanning bright pixels in a region with the kurtosis of 1 point by point, taking a neighborhood of each bright pixel, calculating two criteria, determining whether to reserve the bright pixels according to a threshold value, and obtaining a preliminary screening result of left-bevel-edge pixels after scanning is finished; Turning the space-time diagram left and right, and turning back left and right after scanning is finished to obtain a preliminary screening result of pixels with right hypotenuse; Performing a coordinate system rotation operation of θ= -30 ° on the preliminary screening result of the left-side pixels, so that the horizontal axis xθ of the new coordinate system is perpendicular to the left-side direction, and calculating each bright pixel New coordinates in new coordinate system ; Arranging all the bright pixels from small to large according to the x theta coordinates, and calculating the difference value of the x theta coordinates two by two to obtain a difference vector ; Comparing each in turn And threshold value If the size of (a) Representing a new line segment from the (i+1) th pixel, all After the comparison is finished, the left oblique line segmentation operation is completed; Turning the primary screening result of the right hypotenuse pixels left and right, and turning the result back left and right to finish the right slash segmentation operation; Calculating the length-width ratio of all left and right oblique line segments, and eliminating the pixel blocks with the length-width ratio smaller than 5; for each oblique line segment, taking the pixel with the maximum ordinate in the theta-rotation coordinate system, and recording the rectangular coordinates to obtain a series of left oblique line vertexes and right oblique line vertexes; the Euclidean distance of each 'left oblique line vertex' and 'right oblique line vertex' is calculated in pairs in turn, and the distance between the ith left oblique line endpoint and the jth right oblique line endpoint is recorded as ; Taking out Minimum value of (2) And (3) with Comparing if Matching the i-th left oblique line with the j-th right oblique line to form an inverted V; removing the two left and right oblique lines after the matching is finished until a certain judgment Terminating the algorithm and outputting an inverted V identification result; And calculating a linear equation according to each pixel coordinate of the left and right oblique sides of each inverted V, and calculating an intersection point coordinate according to the equation to obtain the position information of the collision signal of the pipe cleaner.

Description

Phi-OTDR tube cleaner collision signal identification and positioning device and method Technical Field The invention relates to the fields of distributed optical fiber sensing, object detection and event identification, and provides a phi aiming at the problems of difficult identification, inaccurate positioning and the like when a phase-sensitive optical time domain reflectometer (phase-SENSITIVE OPTICAL TIME-Domain Reflectometer, phi-OTDR) is applied to a pipe cleaner tracking taskAn OTDR tube cleaner collision signal identification and positioning and device method. Background In pipeline monitoring applications, sensing fibers are laid in parallel along a pipeline. The phi-OTDR system repeatedly transmits probe pulses along the sensing fiber and samples the vibration signal by detecting Rayleigh back-scattered (Rayleigh Back Scattered, RBS) light. When a disturbance event occurs, the local fiber refractive index at the disturbance location is modulated, changing the RBS optical field phase downstream of the disturbance location. The event related information may be extracted at this time using a direct intensity detection or phase demodulation method. The space-time diagram is a typical phi-OTDR data structure, consisting of a plurality of RBS traces. The space-time diagram can be seen as a two-dimensional matrix, the horizontal and vertical axes representing space and time, respectively, each row (fast axis) being an RBS trace, each column (slow axis) being a one-dimensional time domain signal, the element values of the matrix representing energy intensity. Unlike natural images, where objects have sharp boundaries, "straight lines" and "edges" in the phi-OTDR space-time diagram result from the propagation of a vibration wave along the pipe, which is essentially a series of vibration peaks arranged in a line, the vibration nature of the signal causing the pattern to be discontinuous along the time axis. Furthermore, the coherent fading phenomenon is difficult to avoid in phi-OTDR systems, resulting in a discontinuity of the pattern in the space-time diagram along the spatial axis. Finally, various vibration waves which exist in the surrounding environment and are irrelevant to the pipeline can be captured by the phi-OTDR system and reflected in a space-time diagram in a noise mode, so that the difficulty of event identification is further increased. Disclosure of Invention The invention provides a phiThe invention relates to an OTDR tube cleaner collision signal recognition and positioning device and method, which is characterized in that an image is primarily judged based on kurtosis, then 'pixel density distribution' of a local area in the image is utilized to extract 'oblique lines' in the image, finally 'inverted V matching' method is adopted to pair all the extracted oblique lines to obtain a final inverted V recognition result, the horizontal coordinates of the vertexes of the inverted V represent the position information of a tube cleaner, the space-time slope of the inverted V oblique sides represent the propagation speed of collision sound signals along the tube wall and reflect the type of an event, the method can accurately recognize inverted V feature patterns in a complex image and position vertexes of the inverted V feature patterns under various interferences, has good interpretability and stronger anti-interference performance, and can obtain good application effects in tube cleaner positioning tasks, as described in detail below: Phi shaped The device comprises a data preprocessing module, a kurtosis preliminary judging module, an edge detecting module, a line segment dividing module and an inverted V matching module; The data preprocessing module preprocesses the data input into the original space-time diagram and outputs a 0-1 binary image; the kurtosis preliminary judging module judges the images before binarization after the band-pass filtering, and if the images pass the judgment, the edge detecting module is carried out; the edge detection module detects the 0-1 binary image and outputs left and right oblique line detection results; the line segment segmentation module respectively carries out line segment segmentation operation on the left and right oblique line detection results, outputs the left and right oblique line segment segmentation results, and respectively obtains the number of the left and right oblique line segments; the inverted V matching module adopts an iterative mode, a pair of left and right oblique line segments with the upper end points closest to each other are taken each time, the inverted V pattern is matched according to a distance threshold value, and a final inverted V pattern detection result is output. The kurtosis preliminary judging module judges images before binarization after band-pass filtering as follows: judging all the points of the 0-1 binary sequence kbin =0, and if kbin =1 points exist in the left and right 5 adjacent points, setting kbin