Gas infrared imaging leakage identification and positioning method and device and electronic equipment

Abstract

The invention relates to the field of gas leakage monitoring, and provides a gas infrared imaging leakage identification and positioning method and device and electronic equipment. The gas infrared imaging leakage identification and positioning method comprises the following steps: identifying a gas cloud cluster according to an infrared view field image, establishing a cloud cluster track point data set, a gas cloud grid data set and a maximum concentration point set in first N minutes, and judging whether the type of a leakage scene is a continuous leakage scene or an intermittent leakage scene; for a continuous leakage scene, identifying a cloud cluster aggregation point as a leakage source position according to the gas cloud grid data set and the maximum concentration point set; and for an intermittent leakage scene, fitting a cloud cluster movement trajectory curve according to the cloud cluster trajectory point data set, and determining a leakage source position according to a trajectory curve starting point. According to the embodiment of the invention, the identification efficiency and accuracy of leakage source identification are improved.

Description

Technical Field

[0001] The present invention relates to the field of gas leakage monitoring, and particularly to a method for identifying and locating gas leakage by gas infrared imaging, a device for identifying and locating gas leakage by gas infrared imaging, an electronic device, and a computer-readable storage medium. Background Art

[0002] Chemical materials have characteristics such as high temperature, high pressure, flammable, explosive, toxic, and harmful. Once leaked, they are extremely likely to cause hazards in multiple aspects such as safety, environment, and health. Therefore, timely detection of leaks has always been the unremitting pursuit of refining enterprises.

[0003] Currently, traditional gas monitoring methods belong to fixed-point monitoring, with low operating safety. To achieve large-scale dynamic monitoring is time-consuming and laborious, the result accuracy is poor, and regular debugging and calibration are required. Due to its advantages such as large range, all-weather, and long distance, the gas leakage infrared imaging monitoring technology has become an effective means for gas leakage monitoring. The infrared leakage monitoring device acquires infrared video images, and through filtering, differencing, binarization processing, morphological operations, enhancement processing, etc., realizes the identification and marking of gas leakage clouds in the infrared field of view.

[0004] Currently, the identification and marking algorithms based on gas leakage infrared imaging mostly take gas leakage clouds as the object. After a gas leakage occurs, the leakage clouds drift and move under the influence of the environment, and there is still a lack of leakage source identification and location technology. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a method, device, and electronic device for identifying and locating gas leakage by gas infrared imaging to solve some problems in the prior art.

[0006] To achieve the above purpose, in the first aspect of the present invention, a method for identifying and locating gas leakage by gas infrared imaging is provided. The method includes: identifying gas clouds according to infrared field-of-view images, and establishing a data set of cloud trajectory points in the previous N minutes, a data set of gas cloud grids, and a set of maximum concentration points; determining whether the type of the leakage scenario is a continuous leakage scenario or an intermittent leakage scenario; for the continuous leakage scenario, identifying the cloud aggregation point as the leakage source location according to the data set of gas cloud grids and the set of maximum concentration points; for the intermittent leakage scenario, fitting a cloud movement trajectory curve according to the data set of cloud trajectory points, and determining the leakage source location according to the starting point of the trajectory curve.

[0007] Preferably, identifying gas clouds according to infrared field-of-view images includes: Step 1: For any frame of infrared field-of-view image, taking the lower left corner of the infrared field of view as the origin and the field-of-view plane as the XY coordinate system, obtaining the position data and gas cloud concentration C of all pixel points, and constructing the XYC three-dimensional coordinate data S of any pixel point S i of S i (xi , y i , c i ), and establish a data set S of all pixel points within the infrared field of view; where S i represents the i-th pixel point, (x i , y i , c i ) represents the X-axis coordinate, Y-axis coordinate, and C-axis coordinate of the pixel point; the gas cloud concentration C is characterized by the differential image gray value or the concentration value obtained by converting according to the differential image gray value; for the pixel point S i (x i , y i , c i ), if it is identified as a gas cloud pixel point in the infrared field of view image, then c i is not 0, otherwise it is identified as a non-gas cloud pixel point, then c i is 0; the cloud recognition method for the infrared field of view image includes one of various machine recognition algorithms; Step 2: For the data set S, traverse the gas cloud pixel points S j (x j , y j , c j ), that is, c j ≠0, and delete the pixel points S of non-gas clouds that meet the following conditions from the data set S k , to form a data set SA: where L is the pixel distance threshold, L takes 10 or 1 / 40 of the number of pixels of the diagonal of the field of view; x k and y k are the X-axis coordinate and Y-axis coordinate of S k respectively; Step 3: On the basis of the data set SA, supplement the three-dimensional coordinate data S of the 4 boundary pixel points of the infrared field of view n (x n , y n , 0), where (x n , y n ) belongs to the pixel points on the 4 boundaries of the infrared field of view, to form a data set SAN; Step 4: According to the data set SAN, take the minimum non-zero value of c i in SAN as D, use the spatial interpolation algorithm to reconstruct the image of the infrared field of view, and set the minimum value of c i after reconstruction to 0, the XY-axis interpolation accuracy is 1 / 4 to 1 / 2 of the existing pixel accuracy, and draw a closed contour line of c i = D / 2; Step 5: The area enclosed by the closed contour line of c i = D / 2 is the gas cloud; the number of closed contour lines of c i = D / 2 is the number of clouds in this frame of infrared field of view image; the c iAll the pixel points enclosed by the = D / 2 closed contour line are the cloud cluster range; if the area within the closed contour line is less than 10×10 pixel points, it can be recognized as interference and not included in the cloud cluster quantity.

[0008] Preferably, the method further includes: c i The pixel point with the maximum value of the gas cloud concentration C within the range enclosed by the = D / 2 closed contour line is the cloud cluster trajectory point G of this frame of infrared field of view image g (x g , y g , c g ); Define the cloud cluster trajectory point with the maximum gas cloud concentration in this frame of infrared field of view image as the maximum concentration point MAX z (x z , y z , c z ).

[0009] Preferably, for any frame of infrared field of view image, the number of cloud cluster trajectory points is the same as the number of cloud clusters, that is, each cloud cluster has and only has one cloud cluster trajectory point.

[0010] Preferably, the infrared field of view grid is square, and the grid size is preset. Define the infrared field of view grid with pixel points coinciding with the cloud cluster range as the gas cloud grid.

[0011] Preferably, the cloud cluster trajectory point data set G includes the cloud cluster trajectory points identified in the previous N minutes; the gas cloud grid data set includes the gas cloud grids identified in the previous N minutes; the maximum concentration point set includes the maximum concentration points identified in the previous N minutes.

[0012] Preferably, the types of leakage scenarios evaluated include: calculating the frequency fy of the appearance of gas cloud clusters and the frequency fw of the appearance of each gas cloud grid in all infrared field of view images in the previous N minutes; if there is any gas cloud grid with an appearance frequency fw t greater than the preset value, it is a continuous leakage scenario, otherwise it is an intermittent leakage scenario; where, fy = FY / M; fw t = FW t / FY; M is the total number of frames of infrared field of view images included in the previous N minutes, M = N×60×video frame rate; FY is the total number of frames of infrared field of view images in which gas cloud clusters appear in the previous N minutes; FW t is the total number of frames in which the t-th gas cloud grid appears in the FY frames of infrared field of view images in which gas cloud clusters appear in the previous N minutes.

[0013] Preferably, if the frequency fy of the appearance of the gas cloud clusters is less than the preset value, that is, the total number of frames of infrared field of view images with gas cloud clusters in the previous N minutes is insufficient, then extend the statistics to the infrared field of view images in the previous 2*N minutes.

[0014] Preferably, based on the gas cloud grid data set and the maximum concentration point set, identifying the cloud cluster point as the leakage source location includes: selecting fw t The gas cloud grids greater than the preset value are the target gas cloud grids. The target gas cloud grid with the most occurrences of the maximum concentration point is the leakage point grid, and the aggregation position of the maximum concentration point in this leakage point grid is the leakage source location.

[0015] Preferably, the calculation steps for the aggregation position of the maximum concentration point are as follows:

[0016]

[0017]

[0018] where NMAX is the number of maximum concentration points, x z , y z , c z are the X-axis, Y-axis, and C-axis coordinates of the z-th maximum concentration point, and the value range of z is from 1 to NMAX; X source is the X-axis coordinate of the leakage source location, where Y source is the Y-axis coordinate of the leakage source location.

[0019] Preferably, fitting the cloud movement trajectory curve according to the cloud trajectory point data set includes: using the minimum X and maximum X of all gas cloud grids as the X-axis interval for fitting the cloud movement trajectory curve; using the minimum Y and maximum Y of all gas cloud grids as the Y-axis interval for fitting the cloud movement trajectory curve; for all cloud trajectory points, based on curve fitting methods such as least squares, performing polynomial fitting of the third order and above to obtain the optimal cloud movement trajectory curve.

[0020] Preferably, determining the leakage source location according to the starting point of the trajectory curve includes: for all cloud trajectory points G(x g , y g , c g ), calculating the correlation coefficient between c g and y g and the correlation coefficient between x g and c g , taking the one with the largest absolute value of the two as the significant correlation coefficient, and the corresponding X data axis or Y data axis as the significant correlation data axis; if the significant correlation coefficient is positive, using the maximum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point; if the significant correlation coefficient is negative, using the minimum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point; determining the leakage point based on the starting point of the optimal cloud movement trajectory curve.

[0021] Preferably, determining the leakage point based on the starting point of the optimal cloud movement trajectory curve includes: taking the gas cloud grid where the starting point of the optimal cloud movement trajectory curve is located or the gas cloud grid closest to the starting point of the cloud movement trajectory curve as the leakage point grid, and taking the maximum concentration point aggregation position that appears in the leakage point grid as the leakage source position. The calculation method of the maximum concentration point aggregation position is as follows:

[0022]

[0023]

[0024] wherein, X source is the X-axis coordinate of the leakage source position, and Y source is the Y-axis coordinate of the leakage source position; the gas cloud grid contains at least 1 or more maximum concentration points.

[0025] The present invention also provides a leakage identification and positioning device for gas infrared imaging. The device includes: a cloud processing module for identifying a gas cloud according to an infrared field-of-view image and establishing a data set of cloud trajectory points, a data set of gas cloud grids, and a set of maximum concentration points in the previous N minutes; a scene judgment module for judging whether the type of the leakage scene is a continuous leakage scene or an intermittent leakage scene; a first positioning module for, in the case of a continuous leakage scene, identifying the cloud aggregation point as the leakage source position according to the data set of gas cloud grids and the set of maximum concentration points; and a second positioning module for, in the case of an intermittent leakage scene, fitting a cloud movement trajectory curve according to the data set of cloud trajectory points and determining the leakage source position according to the starting point of the trajectory curve.

[0026] The present invention also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the foregoing leakage identification and positioning method for gas infrared imaging.

[0027] The present invention also provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to cause the computer to execute the foregoing leakage identification and positioning method for gas infrared imaging.

[0028] In a fourth aspect of the present invention, there is also provided a computer-readable storage medium. Instructions are stored in the storage medium, and when the instructions are run on a computer, the computer is caused to execute the steps of the foregoing leakage identification and positioning method for gas infrared imaging.

[0029] The present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the aforementioned method for leak identification and location of gas infrared imaging.

[0030] The above technical solution has at least the following beneficial effects:

[0031] In the implementation of this method, for the behavioral characteristics of gas clouds in two different scenarios of continuous leakage and intermittent leakage, a method for locating the leakage source in the intermittent leakage scenario based on curve fitting of the cloud movement trajectory and confirmation of the starting point, and a method for locating the leakage source in the continuous leakage scenario based on identification of cloud aggregation points are proposed. At the same time, considering the specific problems of severe cloud scatter and coexistence of multiple clouds in the infrared field of view image of gas leakage, a method for identifying the number of clouds, cloud range, and cloud trajectory points applicable to the characteristics of infrared cloud images is provided, which serves as necessary support data for fitting the cloud movement trajectory curve and analyzing cloud aggregation points. In addition, the present invention uses historical infrared field of view image data for a current period of time to participate in the analysis and calculation, strengthens the capture effect of the leakage gas cloud, and overall improves the accuracy of leakage source location under the infrared field of view.

[0032] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification. They are used together with the following specific implementation to explain the embodiments of the present invention, but do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0034] Figure 1 Schematically shows a schematic diagram of the steps of the method for leak identification and location of gas infrared imaging according to an embodiment of the present invention;

[0035] Figure 2 Schematically shows a schematic diagram of the steps of the method for leak identification and location of gas infrared imaging according to another embodiment of the present invention;

[0036] Figure 3 Schematically shows a schematic diagram of the steps for the intermittent leakage scenario according to an embodiment of the present invention;

[0037] Figure 4 Schematically shows a schematic diagram of dividing an infrared field of view into infrared field of view grids according to an embodiment of the present invention;

[0038] Figure 5 Schematically shows a schematic diagram of the structure of the device for leak identification and location of gas infrared imaging according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] The following is a detailed description of the specific implementation manners of the embodiments of the present invention in conjunction with the accompanying drawings. It should be understood that the specific implementation manners described herein are only used to illustrate and explain the embodiments of the present invention, and are not used to limit the embodiments of the present invention.

[0040] Embodiment 1

[0041] Figure 1 Schematically shows a schematic diagram of the steps of a method for leak identification and location of gas infrared imaging according to an embodiment of the present invention. As Figure 1 shown, a method for leak identification and location of gas infrared imaging includes:

[0042] S01. Identify a gas cloud mass based on an infrared field-of-view image, and establish a data set of cloud mass trajectory points, a gas cloud grid data set, and a maximum concentration point set for the previous N minutes. An infrared imaging device usually has a two-dimensional field of view, which is composed of a large number of pixel points, such as 320*240 pixel points, and the observed object is displayed as several pixel points within the infrared field of view. Currently, the recognition algorithm for gas leakage cloud masses based on infrared imaging, such as infrared image differential calculation, uses each pixel point as an identification object, and forms a cloud mass image by color-labeling multiple pixel points in the two-dimensional field of view. Therefore, there are problems such as cloud mass scatterization, difficulty in identifying the number and range of cloud masses, and further considering the possible coexistence of multiple gas cloud masses. In this embodiment, based on the infrared field-of-view image, an XYC coordinate system of the field-of-view plane and the gas cloud concentration is established, and the infrared field of view is reconstructed by using a spatial interpolation algorithm to identify the closed contour lines of the gas cloud concentration, so as to realize gas cloud mass recognition. Since the diffusion of the gas cloud mass has a certain randomness, there may be a large error in identifying only based on the infrared field-of-view image of the current frame. Therefore, the historical infrared field-of-view image data of the current period of time is selected to participate in the analysis and calculation at the same time. At the same time, considering the irregular shape of the gas cloud mass, it is difficult to directly perform relevant calculations on the cloud mass drifting trajectory and leakage source location based on the gas cloud mass contour data. Therefore, based on the gas cloud mass range of each frame of infrared field-of-view image, the cloud mass trajectory points, the maximum concentration points, and the gas cloud grid in the external field-of-view image of this frame are obtained, and data sets of the above three are established respectively, providing necessary data for the accurate location of the subsequent leakage source.

[0043] S02. Evaluate the leakage scenario type. The leakage scenarios are divided into 2 types from the perspectives of the behavior characteristics of the gas cloud and the location of the leakage source, namely, the continuous leakage scenario where an identifiable gas cloud always exists at the leakage source location, and the intermittent leakage scenario where the gas cloud at the leakage source location is unstable (an identifiable gas cloud does not always exist) and the cloud drifts and moves with the wind after it appears; due to the significant differences in the cloud behavior characteristics of the 2 types of leakage scenarios, different methods are required to identify the leakage source location. For the continuous leakage scenario, the key is to determine the location of the stable and continuous gas cloud aggregation point, and for the intermittent leakage scenario, the key is to determine the starting point of the cloud movement trajectory. To improve the accuracy of the leakage source location, it is necessary to identify the leakage scenario type in advance, and based on the occurrence frequency of the gas cloud grid, judge whether there is an identifiable gas cloud that always exists, so as to realize the discrimination between the continuous leakage scenario and the intermittent leakage scenario.

[0044] S03. For the continuous leakage scenario, based on the gas cloud grid data set and the maximum concentration point set, identify the cloud aggregation point as the leakage source location.

[0045] S03'. For the intermittent leakage scenario, fit the cloud movement trajectory curve according to the cloud trajectory point data set, and determine the leakage source location according to the starting point of the trajectory curve;

[0046] Through the above implementation methods, based on the infrared field of view to locate the leakage source location, the location efficiency and accuracy of the leakage source location can be improved.

[0047] In some alternative embodiments, identifying the gas cloud according to the infrared field of view image includes: Step 1: For any frame of the infrared field of view image, taking the lower left corner of the infrared field of view as the origin and the field of view plane as the XY coordinate system, obtain the position data and gas cloud concentration C of all pixel points, and construct the XYC three-dimensional coordinate data S of any pixel point S i of S i (x i , y i , c i ), and establish a data set S of all pixel points within the infrared field of view; where S i represents the i-th pixel point, (x i , y i , c i ) represents the X-axis coordinate, Y-axis coordinate, and C-axis coordinate of the pixel point; the gas cloud concentration C is characterized by the differential image gray value or the concentration value obtained by converting the differential image gray value; for the pixel point S i (x i , y i , c i ), if it is identified as a gas cloud pixel point in the infrared field of view image, then c i is not 0, otherwise it is identified as a non-gas cloud pixel point, then c iis 0; the method for identifying cloud clusters in the infrared field of view image includes one of various machine recognition algorithms; Step 2: For the data set S, traverse the gas cloud pixel points S j (x j , y j , c j ), that is, c j ≠0, and delete the pixel points S of non-gas clouds that meet the following conditions from the data set S k , to form the data set SA: This step is used to eliminate the values at other positions outside the cloud pixel points, leaving an interpolation space for the subsequent reconstruction of the spatial interpolation algorithm; where L is the pixel distance threshold, and L is taken as 10 or 1 / 40 of the number of pixels in the diagonal of the field of view; x k and y k are the X-axis coordinate and Y-axis coordinate of S k respectively; Step 3: Based on the data set SA, supplement the three-dimensional coordinate data S of the pixel points on the 4 boundaries of the infrared field of view n (x n , y n , 0), where (x n , y n ) belongs to the pixel points on the 4 boundaries of the infrared field of view, to form the data set SAN; Step 4: According to the data set SAN, take the minimum non-zero value of c i in SAN as D, use the spatial interpolation algorithm to reconstruct the image of the infrared field of view, and set the minimum value of c i reconstruction to 0, the XY-axis interpolation accuracy is 1 / 4 to 1 / 2 of the existing pixel accuracy, and draw the closed contour line of c i =D / 2; Step 5: The area enclosed by the c i =D / 2 closed contour line is the gas cloud; the number of c i =D / 2 closed contour lines is the number of cloud clusters in this frame of infrared field of view image; all the pixel points enclosed by the c i =D / 2 closed contour line are the cloud cluster range; if the area inside the closed contour line is less than 10×10 pixel points, it can be recognized as interference and not counted in the number of cloud clusters.

[0048] In some alternative embodiments, the method further includes: the pixel point with the maximum value of the gas cloud concentration C within the range enclosed by the c i =D / 2 closed contour line is the cloud cluster trajectory point G g (x g , y g , c g ) of this frame of infrared field of view image; define the cloud cluster trajectory point with the maximum gas cloud concentration in this frame of infrared field of view image as the maximum concentration point MAX z (x z , yz , c z ).

[0049] In some alternative embodiments, for any frame of infrared field-of-view image, the number of cloud trajectory points is the same as the number of clouds, that is, each cloud has exactly one cloud trajectory point.

[0050] In some alternative embodiments, the infrared field-of-view grid is square, and the grid size is preset. An infrared field-of-view grid with pixel points coinciding with the cloud range is defined as a gas cloud grid.

[0051] In some alternative embodiments, the cloud trajectory point data set G includes the cloud trajectory points identified in the previous N minutes; the gas cloud grid data set includes the gas cloud grids identified in the previous N minutes; the maximum concentration point set includes the maximum concentration points identified in the previous N minutes.

[0052] In some alternative embodiments, the types of leakage scenarios to be judged include: calculating the frequency fy of the appearance of gas clouds and the frequency fw of the appearance of each gas cloud grid in all infrared field-of-view images in the previous N minutes; if there is any gas cloud grid with an appearance frequency fw t greater than the preset value, it is a continuous leakage scenario, otherwise it is an intermittent leakage scenario; where, fy = FY / M; fw t = FW t / FY; M is the total number of frames of infrared field-of-view images included in the previous N minutes, M = N × 60 × video frame rate; FY is the total number of frames of infrared field-of-view images in which gas clouds appear in the previous N minutes; FW t is the total number of frames in which the t-th gas cloud grid appears in the FY frames of infrared field-of-view images in which gas clouds appear in the previous N minutes.

[0053] In some alternative embodiments, if the frequency fy of the appearance of the gas cloud is less than the preset value, that is, the total number of frames of infrared field-of-view images with gas clouds in the previous N minutes is insufficient, then the statistical period is extended to the infrared field-of-view images in the previous 2*N minutes.

[0054] In some alternative embodiments, according to the gas cloud grid data set and the maximum concentration point set, identifying the cloud aggregation point as the leakage source location includes: selecting the gas cloud grid with fw t greater than the preset value as the target gas cloud grid, taking the target gas cloud grid with the most occurrences of the maximum concentration point as the leakage point grid, and taking the aggregation position of the maximum concentration points appearing in this leakage point grid as the leakage source location.

[0055] In some alternative embodiments, the calculation steps for the aggregation position of the maximum concentration points are as follows:

[0056]

[0057]

[0058] Among them, NMAX is the number of maximum concentration points, and x z , y z , c z are the X-axis, Y-axis, and C-axis coordinates of the z-th maximum concentration point, and the value range of z is from 1 to NMAX; X source is the X-axis coordinate of the leakage source position, where Y source is the Y-axis coordinate of the leakage source position.

[0059] In some alternative embodiments, fitting the cloud movement trajectory curve according to the set of cloud trajectory point data includes: using the minimum X and maximum X of all gas cloud grids as the X-axis interval for fitting the cloud movement trajectory curve; using the minimum Y and maximum Y of all gas cloud grids as the Y-axis interval for fitting the cloud movement trajectory curve; for all cloud trajectory points, performing polynomial fitting of the third order and above based on curve fitting methods such as the least squares method to obtain the optimal cloud movement trajectory curve.

[0060] In some alternative embodiments, determining the leakage source position according to the starting point of the trajectory curve includes: for all cloud trajectory points G(x g , y g , c g ), calculating the correlation coefficient between y g and c g and the correlation coefficient between x g and c g , taking the one with the largest absolute value of the two as the significant correlation coefficient, and the corresponding X data axis or Y data axis as the significant correlation data axis; if the significant correlation coefficient is positive, using the maximum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point; if the significant correlation coefficient is negative, using the minimum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point; determining the leakage point based on the starting point of the optimal cloud movement trajectory curve.

[0061] In some alternative embodiments, determining the leakage point based on the starting point of the optimal cloud movement trajectory curve includes: using the gas cloud grid where the starting point of the optimal cloud movement trajectory curve is located or the gas cloud grid closest to the starting point of the cloud movement trajectory curve as the leakage point grid, and using the aggregation position of the maximum concentration points appearing in the leakage point grid as the leakage source position. The calculation method of the aggregation position of the maximum concentration points is as follows:

[0062]

[0063]

[0064] Among them, X source is the X-axis coordinate of the leakage source position, Y sourceis the Y-axis coordinate of the leakage source position; the gas cloud grid includes at least 1 or more maximum concentration points.

[0065] Embodiment 2

[0066] Figure 2 Schematically shows a step schematic diagram of a leakage identification and positioning method for gas infrared imaging according to another embodiment of the present invention. As Figure 2 shown, it includes the following steps:

[0067] Step 1: Obtain infrared field-of-view images in the recent period;

[0068] Step 2: For each frame of infrared field-of-view image, identify the number of gas cloud clusters and the range of the gas cloud clusters;

[0069] Step 3: Calculate the frequency of the appearance of gas cloud clusters in all infrared field-of-view images;

[0070] Step 4: Determine whether the total number of frames of the gas cloud cluster image is sufficient. If it is sufficient, execute Step 5. Otherwise, expand the acquisition time range of the infrared field-of-view image data and return to execute Step 1;

[0071] Step 5: Based on the number of cloud clusters and the range of the cloud clusters, identify cloud cluster trajectory points, gas cloud grids, and maximum concentration points;

[0072] Step 6: Establish a cloud cluster trajectory point data set, a gas cloud grid data set, and a maximum concentration point set;

[0073] Step 7: Calculate the frequency of the appearance of the gas cloud grid in all infrared field-of-view images;

[0074] Step 8: Evaluate the leakage scenario: continuous leakage scenario, intermittent leakage scenario;

[0075] Step 9: Determine whether it is an intermittent leakage scenario? If so, execute Step 10. Otherwise, execute Step 11;

[0076] Step 10: Fit a cloud cluster movement trajectory curve according to the cloud cluster trajectory point data set, determine the leakage source position according to the starting point of the trajectory curve, and end;

[0077] Step 11: Identify the cloud cluster aggregation point as the leakage source position according to the gas cloud grid data set and the maximum concentration point set, and end.

[0078] Embodiment 3

[0079] Figure 3 Schematically shows a step schematic diagram for an intermittent leakage scenario according to an embodiment of the present invention. As Figure 3As shown in the figure, for the intermittent leakage scenario, this embodiment exemplifies a method for fitting a cloud movement trajectory curve from a set of cloud trajectory point data and determining the starting point and the leakage source location.

[0080] For the intermittent leakage scenario, the gas cloud at the leakage source location does not persist continuously, and after the gas cloud appears, it drifts and moves with the wind. That is, it is difficult to directly locate the leakage source using the identification method for continuous leakage clouds. Therefore, a leakage source location method based on the cloud movement trajectory is adopted. This method first constructs a cloud movement trajectory curve. During the physical process of gas diffusion, the leakage source must be near the starting point direction of the cloud movement trajectory. Since it is necessary to determine the trajectory starting point, and considering that the starting point of the cloud movement trajectory is the fitting result and cannot be directly used as the leakage source, based on the trajectory starting point, the grids where possible leakage sources are located are first screened, and then the aggregation location of the maximum concentration points in the grids is further calculated to determine the leakage source location.

[0081] As Figure 3 shown in the figure, it includes the following steps:

[0082] Step 1: Obtain the gas cloud grid data set;

[0083] Step 2: Use the minimum X and maximum X of all gas cloud grids as the X-axis interval for fitting the cloud movement trajectory curve; use the minimum Y and maximum Y of all gas cloud grids as the Y-axis interval for fitting the cloud movement trajectory curve;

[0084] Step 3: Obtain the cloud trajectory point data set, and based on curve fitting methods such as the least squares method, perform polynomial fitting of the third order and above;

[0085] Step 4: Obtain the optimal cloud movement trajectory curve;

[0086] Step 5: Obtain the cloud trajectory point data set, and calculate the correlation coefficient between c g , y g , c g ) and y g as well as the correlation coefficient between x g and c g based on all cloud trajectory points G(x g );

[0087] Step 6: Take the one with the largest absolute value of the two as the significant correlation coefficient;

[0088] Step 7: Determine whether the significant correlation coefficient is positive. If so, use the optimal cloud movement trajectory curve, and take the maximum value position in the direction of the axis where the significant correlation coefficient is located as the starting point; if not, use the optimal cloud movement trajectory curve, and take the minimum value position in the direction of the axis where the significant correlation coefficient is located as the starting point.

[0089] Step 8: Take the gas cloud grid where the starting point is located (or the gas cloud grid closest in distance) as the leakage point grid; it is necessary to ensure that this gas cloud grid contains at least one maximum concentration point.

[0090] Step 9: Screen out the maximum concentration points that appear in this leakage point grid, and take the aggregation position of the maximum concentration points as the leakage source position.

[0091] Embodiment 4

[0092] This embodiment exemplifies a grid division method. In this infrared field of view grid division method, according to Johnson's rule, in order to achieve an identifiable effect of the observed target in the infrared field of view, the image formed by it must occupy more than 12 pixels in the critical dimension direction. Considering the accuracy of the identification effect and the allowable deviation of leakage source positioning, the grid size of the infrared field of view grid division is set to be at least larger than the area of 48 pixel points. The grid is set to be N*N times the size of the infrared field of view, and it is recommended that N be taken as 1 / 30 - 1 / 10 of the short side of the infrared field of view. Determine the origin within the infrared field of view range, and divide the infrared field of view range into several infrared field of view grids according to the grid size and the origin. Here, the origin is preferably the lower left corner boundary point of the infrared field of view range.

[0093] For a certain infrared imaging device, its field of view resolution is 320×240, that is, it contains 320 pixels horizontally and 240 pixels vertically. Set N to 1 / 10, then the size of each grid can be set to 24×24, that is, 14 grids are divided horizontally and 10 grids are divided vertically, and the entire infrared field of view is divided into 140 grids. Figure 4 Schematically shows a schematic diagram of dividing the infrared field of view grid based on the infrared field of view range in the embodiment of the present invention, and the obtained division result is as Figure 4 shown.

[0094] Embodiment 5

[0095] This embodiment schematically shows the calculation method of the air mass movement trajectory. The air mass movement trajectory adopts the discrete point curve fitting method, taking the cloud mass trajectory points as discrete points and the fitted curve as the air mass movement trajectory; since the air mass movement trajectory is a curve with a non-fixed shape, that is, an irregular curve and the curve equation is unknown, it is impossible to define the curve data model. A polynomial fitting model of the third order or higher or a logarithmic fitting model is preferably used. Common curve trajectory drawing methods include least square fitting, Bezier curve fitting, B-spline curve fitting, parabola blending curve fitting, cubic parametric spline curve fitting, etc., and least square fitting is preferably used.

[0096] Embodiment 6

[0097] Based on the same inventive concept, the present invention also provides a leakage identification and positioning device for gas infrared imaging. Figure 5Schematically shows a structural diagram of a leakage identification and positioning device for gas infrared imaging according to an embodiment of the present invention. As Figure 5 shown, the device includes: a cloud processing module for identifying a gas cloud according to an infrared field-of-view image and establishing a data set of cloud trajectory points, a gas cloud grid data set, and a maximum concentration point set for the first N minutes; a scene judgment module for judging whether the type of leakage scene is a continuous leakage scene or an intermittent leakage scene; a first positioning module for, in response to a continuous leakage scene, identifying a cloud aggregation point as the leakage source location according to the gas cloud grid data set and the maximum concentration point set; and a second positioning module for, in response to an intermittent leakage scene, fitting a cloud movement trajectory curve according to the cloud trajectory point data set and determining the leakage source location according to the starting point of the trajectory curve.

[0098] In some alternative embodiments, identifying a gas cloud according to an infrared field-of-view image includes: Step 1: For any frame of infrared field-of-view image, taking the lower left corner of the infrared field of view as the origin and the field-of-view plane as the XY coordinate system, obtaining the position data and gas cloud concentration C of all pixel points, constructing the XYC three-dimensional coordinate data S i of any pixel point S i (x i , y i , c i ), and establishing a data set S of all pixel points within the infrared field of view; where S i represents the i-th pixel point, (x i , y i , c i ) represents the X-axis coordinate, Y-axis coordinate, and C-axis coordinate of the pixel point; the gas cloud concentration C is characterized by the differential image gray value or the concentration value obtained by converting according to the differential image gray value; for the pixel point S i (x i , y i , c i ), if it is identified as a gas cloud pixel point in the infrared field-of-view image, then c i is not 0, otherwise it is identified as a non-gas cloud pixel point, then c i is 0; the cloud identification method for the infrared field-of-view image includes one of multiple machine recognition algorithms; Step 2: For the data set S, traverse the gas cloud pixel points S j (x j , y j , c j ), that is, c j ≠0, and delete the pixel points S k of non-gas clouds that meet the following conditions from the data set S to form a data set SA: where L is a pixel distance threshold, L is taken as 10 or 1 / 40 of the number of pixels of the field-of-view diagonal; x k and yk are S respectively k The X-axis coordinate and Y-axis coordinate of; Step 3: On the basis of the data set SA, supplement the three-dimensional coordinate data S of the four boundary pixel points of the infrared field of view n (x n , y n , 0), where (x n , y n ) belongs to the pixel points on the four boundaries of the infrared field of view, forming a data set SAN; Step 4: According to the data set SAN, take the minimum non-zero value of c i in SAN as D, and use the spatial interpolation algorithm to reconstruct the infrared field of view image, and set the minimum value of c i for reconstruction to be 0, and the XY-axis interpolation accuracy is 1 / 4 to 1 / 2 of the existing pixel accuracy, and draw the closed contour line of c i = D / 2; Step 5: The area enclosed by the closed contour line of c i = D / 2 is the gas cloud; the number of closed contour lines of c i = D / 2 is the number of clouds in this frame of infrared field of view image; all pixel points enclosed by the closed contour line of c i = D / 2 are the cloud ranges; if the area within the closed contour line is less than 10×10 pixel points, it can be identified as interference and not counted in the cloud number.

[0099] In some alternative embodiments, the device further includes: the pixel point with the maximum value of the gas cloud concentration C within the range enclosed by the closed contour line of c i = D / 2 is the cloud trajectory point G of this frame of infrared field of view image g (x g , y g , c g ); Define the cloud trajectory point with the maximum gas cloud concentration in this frame of infrared field of view image as the maximum concentration point MAX z (x z , y z , c z ).

[0100] In some alternative embodiments, for any frame of infrared field of view image, the number of cloud trajectory points is the same as the number of clouds, that is, each cloud has and only has one cloud trajectory point.

[0101] In some alternative embodiments, the infrared field of view grid is square, and the grid size is preset, and the infrared field of view grid that has pixel points coincident with the cloud range is defined as the gas cloud grid.

[0102] In some alternative embodiments, the set G of cloud trajectory point data includes cloud trajectory points identified in the previous N minutes; the set of gas cloud grid data includes gas cloud grids identified in the previous N minutes; and the set of maximum concentration points includes maximum concentration points identified in the previous N minutes.

[0103] In some alternative embodiments, the types of leakage scenarios to be judged include: calculating the frequency fy of the appearance of gas clouds and the frequency fw of the appearance of each gas cloud grid in all infrared field-of-view images in the previous N minutes; if there is any gas cloud grid with an appearance frequency fw t greater than a preset value, it is a continuous leakage scenario, otherwise it is an intermittent leakage scenario; where, fy = FY / M; fw t = FW t / FY; M is the total number of infrared field-of-view images included in the previous N minutes, M = N × 60 × video frame rate; FY is the total number of frames in which gas clouds appear in the infrared field-of-view images in the previous N minutes; FW t is the total number of frames in which the t-th gas cloud grid appears in the FY frames of infrared field-of-view images in which gas clouds appear in the previous N minutes.

[0104] In some alternative embodiments, if the frequency fy of the appearance of the gas cloud is less than the preset value, that is, the total number of frames of infrared field-of-view images with gas clouds in the previous N minutes is insufficient, then the statistical period is extended to the infrared field-of-view images in the previous 2*N minutes.

[0105] In some alternative embodiments, according to the set of gas cloud grid data and the set of maximum concentration points, identifying the cloud cluster point as the leakage source location includes: selecting the gas cloud grid with fw t greater than the preset value as the target gas cloud grid, taking the target gas cloud grid with the most occurrences of the maximum concentration point as the leakage point grid, and taking the aggregation position of the maximum concentration points appearing in this leakage point grid as the leakage source location.

[0106] Preferably, the calculation steps for the aggregation position of the maximum concentration points are as follows:

[0107]

[0108]

[0109] where, NMAX is the number of maximum concentration points, x z 、y z 、c z are the X-axis, Y-axis, and C-axis coordinates of the z-th maximum concentration point, and the value range of z is from 1 to NMAX; X source is the X-axis coordinate of the leakage source location, where Y source is the Y-axis coordinate of the leakage source location.

[0110] In some alternative embodiments, fitting the cloud movement trajectory curve based on the cloud trajectory point data set includes: using the minimum X and maximum X of all gas cloud grids as the X-axis interval for fitting the cloud movement trajectory curve; using the minimum Y and maximum Y of all gas cloud grids as the Y-axis interval for fitting the cloud movement trajectory curve; for all cloud trajectory points, performing polynomial fitting of the third order or higher based on curve fitting methods such as the least squares method to obtain the optimal cloud movement trajectory curve.

[0111] In some alternative embodiments, determining the leakage source location based on the starting point of the trajectory curve includes: for all cloud trajectory points G(x g , y g , c g ), calculating the correlation coefficient between y g and c g , and the correlation coefficient between x g and c g , taking the one with the largest absolute value of the two as the significant correlation coefficient, and the corresponding X data axis or Y data axis as the significant correlation data axis; if the significant correlation coefficient is positive, using the maximum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point; if the significant correlation coefficient is negative, using the minimum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point.

[0112] For the specific definitions of the various functional modules in the above gas infrared imaging leakage identification and positioning device, reference can be made to the definitions of the gas infrared imaging leakage identification and positioning method in the foregoing text, which will not be elaborated here. Each module in the above device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor of the computer device in hardware form or be independent of it, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above respective modules.

[0113] Embodiment 7

[0114] This embodiment provides a non-transitory (non-volatile) computer storage medium, which stores computer-executable instructions that can execute the methods in any of the above method embodiments and achieve the same technical effects.

[0115] This embodiment also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions that, when executed by a computer, cause the computer to execute the methods described in the above aspects and achieve the same technical effects.

[0116] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an all-hardware embodiment, an all-software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0117] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0118] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0119] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the processes or multiple processes and / or blocks Figure 1 one or more of the blocks or multiple blocks.

[0120] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0121] The memory may include non-permanent memory in the computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0122] A computer-readable medium includes both permanent and non-permanent, removable and non-removable media and can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0123] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or apparatus comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or apparatus. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising the element.

[0124] The above are only embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A leakage identification and location method for gas infrared imaging, characterized in that, The method includes: Identifying a gas cloud based on an infrared field-of-view image, and establishing a data set of cloud trajectory points, a data set of gas cloud grids, and a set of maximum concentration points for the previous N minutes. Judging whether the type of the leakage scenario is a continuous leakage scenario or an intermittent leakage scenario. For a continuous leakage scenario, identifying the cloud aggregation point as the leakage source location according to the data set of gas cloud grids and the set of maximum concentration points. For an intermittent leakage scenario, fitting a cloud movement trajectory curve according to the data set of cloud trajectory points, and determining the leakage source location based on the starting point of the trajectory curve.

2. The method according to claim 1, wherein Identifying a gas cloud based on an infrared field-of-view image includes: Step 1: For any frame of infrared field-of-view image, taking the lower left corner of the infrared field of view as the origin and the field-of-view plane as the XY coordinate system, obtain the position data and gas cloud concentration C of all pixel points, and construct the XYC three-dimensional coordinate data S of any pixel point S i ; i (x i ,y i ,c i ), and establish a data set S of all pixel points within the infrared field of view; where S i represents the i-th pixel point, (x i ,y i ,c i ) represents the X-axis coordinate, Y-axis coordinate, and C-axis coordinate of the pixel point; the gas cloud concentration C is characterized by the differential image gray value or the concentration value obtained by converting according to the differential image gray value; For pixel point S i (x i , y i , c i ), if it is recognized as a gas cloud pixel point in the infrared field of view image, then c i is not 0, otherwise it is recognized as a non-gas cloud pixel point, then c i is 0; The cloud recognition method of the infrared field of view image includes one of various machine recognition algorithms; Step 2: For the data set S, traverse the gas cloud pixel points S j (x j , y j , c j ), that is, c j ≠0, and delete the pixel points of non-gas clouds that meet the following conditions from the data set S k , to form the data set SA: Among them, L is the pixel distance threshold, and L takes 10 or 1 / 40 of the number of pixels of the field-of-view diagonal; x k and y k are respectively the X-axis coordinate and Y-axis coordinate of S k ; Step 3: On the basis of the data set SA, supplement the three-dimensional coordinate data S of the pixel points on the four boundaries of the infrared field of view n (x n , y n , 0), where (x n , y n ) belongs to the pixel points on the four boundaries of the infrared field of view, forming the data set SAN; Step 4: According to the data set SAN, take the minimum non-zero value of c in SAN as D, use the spatial interpolation algorithm to reconstruct the infrared field of view, and set the minimum value of c i for reconstruction to be 0, the XY-axis interpolation accuracy is 1 / 4 to 1 / 2 of the existing pixel accuracy, and draw the closed contour line of c i = D / 2; i ​ Step 5: The area enclosed by the c i = D / 2 closed contour lines is the gas cloud; c i = The number of D / 2 closed contour lines is the number of clouds in the infrared field of view image of this frame; the c i = All the pixel points enclosed by the D / 2 closed contour lines are the cloud range; if the area within the closed contour line is less than 10×10 pixel points, it can be recognized as interference and not counted in the number of clouds.

3. The method according to claim 2, wherein The method further includes: c i The pixel point where the maximum value of the gas cloud concentration C within the range enclosed by the = D / 2 closed contour line is located is the cloud cluster trajectory point G of the infrared field of view image of this frame g (x g ,y g ,c g ); Define the cloud cluster trajectory point with the maximum gas cloud concentration in the infrared field of view image of this frame as the maximum concentration point MAX z (x z ,y z ,c z ).

4. The method according to claim 3, wherein For any frame of the infrared field-of-view image, the number of cloud trajectory points is the same as the number of clouds, that is, each cloud has exactly one cloud trajectory point.

5. The method according to claim 3, wherein The infrared field-of-view grid is square, and the grid size is preset. An infrared field-of-view grid with pixel points coinciding with the cloud range is defined as a gas cloud grid.

6. The method according to claim 3, characterized in that The data set G of cloud trajectory points contains the cloud trajectory points identified in the previous N minutes. The data set of gas cloud grids contains the gas cloud grids identified in the previous N minutes. The set of maximum concentration points contains the maximum concentration points identified in the previous N minutes.

7. The method according to claim 2, wherein Judging the type of the leakage scenario includes: Calculate the frequency \(f_y\) of the appearance of gas clouds and the frequency \(f_w\) of the appearance of each gas cloud grid in all infrared field-of-view images in the previous N minutes; if there is any gas cloud grid with an appearance frequency \(f_w\) t greater than the preset value, it is a continuous leakage scenario, otherwise it is an intermittent leakage scenario; where, fy = FY / M; fw t = FW t / FY; M is the total number of frames of the infrared field-of-view images in the previous N minutes, M = N×60×video frame rate. FY is the total number of frames with gas clouds in the infrared field-of-view images in the previous N minutes. FW t It is the total number of frames in which the t-th gas cloud grid appears in the FY-frame infrared field-of-view image of the gas cloud cluster that appears in the previous N minutes.

8. The method according to claim 7, wherein If the frequency fy of the appearance of the gas cloud is less than a preset value, that is, the total number of frames of the infrared field-of-view images with gas clouds in the previous N minutes is insufficient, then extend the statistics to the infrared field-of-view images of the previous 2*N minutes.

9. The method according to claim 7, wherein Identifying the cloud aggregation point as the leakage source location according to the data set of gas cloud grids and the set of maximum concentration points includes: Select fw t The gas cloud grids with a value greater than the preset value are the target gas cloud grids. The target gas cloud grid with the largest number of occurrences of the maximum concentration point is the leakage point grid, and the aggregation position of the maximum concentration point in this leakage point grid is the leakage source position.

10. The method according to claim 9, wherein The calculation steps for the aggregation position of the maximum concentration point are as follows: where NMAX is the number of maximum concentration points, x z , y z , c z are the X-axis, Y-axis, and C-axis coordinates of the z-th maximum concentration point, and the value range of z is from 1 to NMAX; X source is the X-axis coordinate of the leakage source location, where Y source is the Y-axis coordinate of the leakage source location.

11. The method according to claim 7, wherein Fitting a cloud movement trajectory curve according to the data set of cloud trajectory points includes: Taking the minimum X and maximum X of all gas cloud grids as the X-axis interval for fitting the cloud movement trajectory curve. Taking the minimum Y and maximum Y of all gas cloud grids as the Y-axis interval for fitting the cloud movement trajectory curve. For all cloud trajectory points, based on curve fitting methods such as the least squares method, conduct polynomial fitting of the third order and above to obtain the optimal cloud movement trajectory curve.

12. The method according to claim 11, wherein Determining the leakage source location based on the starting point of the trajectory curve includes: For all cloud trajectory points G(x g , y g , c g ), calculate the correlation coefficient between y g and c g , and the correlation coefficient between x g and c g . Take the one with the largest absolute value of the two as the significant correlation coefficient, and the corresponding X data axis or Y data axis is the significant correlation data axis; If the significant correlation coefficient is positive, take the maximum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point. If the significant correlation coefficient is negative, take the minimum value position of the optimal cloud movement trajectory curve in the direction of the axis where the significant correlation coefficient is located as the starting point. Determining the leakage point based on the starting point of the optimal cloud movement trajectory curve.

13. The method according to claim 12, characterized in that, Determining the leakage point based on the starting point of the optimal cloud movement trajectory curve includes: Taking the gas cloud grid where the starting point of the optimal cloud movement trajectory curve is located or the gas cloud grid closest to the starting point of the cloud movement trajectory curve as the leakage point grid, and taking the aggregation position of the maximum concentration point appearing in this leakage point grid as the leakage source location. The calculation method for the aggregation position of the maximum concentration point is as follows: Among them, X source is the X-axis coordinate of the leakage source position, and Y source is the Y-axis coordinate of the leakage source position; Each gas cloud grid contains at least 1 or more maximum concentration points.

14. A leakage identification and positioning device for gas infrared imaging, characterized in that, The device includes: A cloud processing module, configured to identify a gas cloud based on an infrared field-of-view image, and establish a data set of cloud trajectory points, a data set of gas cloud grids, and a set of maximum concentration points for the previous N minutes; A scenario judgment module, configured to judge whether the type of the leakage scenario is a continuous leakage scenario or an intermittent leakage scenario; A first positioning module, configured to, for a continuous leakage scenario, identify the cloud aggregation point as the leakage source location according to the data set of gas cloud grids and the set of maximum concentration points; and A second positioning module, configured to, for an intermittent leakage scenario, fit a cloud movement trajectory curve according to the data set of cloud trajectory points, and determine the leakage source location according to the starting point of the trajectory curve.

15. An electronic device, characterized in that, Comprising: At least one processor; And a memory communicatively connected to the at least one processor; Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the leakage identification and positioning method for gas infrared imaging according to any one of claims 1 to 13.

16. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to cause the computer to execute the leakage identification and positioning method for gas infrared imaging according to any one of claims 1 to 13.

Images