An intelligent monitoring method for dangerous gas leakage based on gas cloud imaging

Abstract

The application discloses a kind of dangerous gas leak intelligent monitoring methods based on gas cloud imaging, it is related to data analysis field, and the method is: based on infrared gas cloud image data and optical gas cloud image data calculation temperature anomaly index J, gas diffusion velocity V, color offset index RGB and gas diffusion index L, solve the problem that temperature fluctuation exists in natural environment, only by single temperature characteristic to determine whether gas leakage occurs, prone to misjudgment;Gas leakage degree is judged by leakage degree evaluation index K and leakage degree threshold set, solve the problem that gas leakage degree cannot be quantified, cannot intuitively understand gas leakage situation.The scheme effectively improves the accuracy and reliability of dangerous gas leakage detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis and dangerous gas detection, and particularly relates to a dangerous gas leakage intelligent monitoring method based on gas cloud imaging. BACKGROUND

[0002] With the continuous expansion and increasing complexity of the chemical industry, the safety requirements for the production process are becoming higher and higher. Dangerous gas leakage is a major safety hazard in the chemical industry. By monitoring the dangerous gas leakage, the danger can be warned in time, the personnel safety can be effectively ensured, and the property loss can be avoided.

[0003] The working principle of the traditional dangerous gas leakage intelligent monitoring method is as follows: real-time scanning and monitoring are performed through the installation of an infrared detector. When dangerous gas leaks, due to the physical properties of the gas and the temperature difference with the surrounding environment, a specific temperature abnormal area will appear in the infrared image. When the abnormal temperature change area is detected, a warning signal is issued.

[0004] The prior art still has the following disadvantages: there are temperature fluctuations in the natural environment. Only a single temperature feature is used to determine whether gas leakage occurs, which is prone to misjudgment. At the same time, the degree of gas leakage cannot be quantified, and the gas leakage situation cannot be intuitively understood. SUMMARY

[0005] (I) Technical problems solved

[0006] In view of the deficiencies of the prior art, the present application provides a dangerous gas leakage intelligent monitoring method based on gas cloud imaging. The temperature anomaly index J, the gas diffusion speed V, the color shift index RGB, and the gas diffusion index L are calculated based on infrared gas cloud image data and optical gas cloud image data. The problem of misjudgment caused by only using a single temperature feature to determine whether gas leakage occurs in the natural environment with temperature fluctuations is solved. The problem of being unable to quantitatively determine the degree of gas leakage and intuitively understand the gas leakage situation is solved by judging the degree of gas leakage according to the leakage degree evaluation index K and the leakage degree threshold set.

[0007] (II) Technical solutions

[0008] To achieve the above purpose, the present application is implemented by the following technical solutions: a dangerous gas leakage intelligent monitoring method based on gas cloud imaging, comprising the following steps:

[0009] Dangerous gas monitoring is performed on a chemical industry park based on infrared imaging technology and optical imaging technology, and infrared gas cloud image data and optical gas cloud image data are generated;

[0010] The horizontal temperature gradient A is calculated based on the infrared gas cloud image data x(x, y) and vertical temperature gradient B x (x, y), and further calculate the temperature gradient amplitude AB(x, y); preset the temperature gradient amplitude threshold CH; determine the infrared leakage area range S based on the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x, y); calculate the temperature anomaly index J based on the infrared gas cloud image;

[0011] Image processing algorithms are used to identify the image boundary of the infrared leakage area S and to obtain the coordinates of the boundary points (x, y, y). h y h According to the boundary point coordinates (x) h y h Calculate the centroid coordinates (E) m F m And further calculate the gas diffusion rate V;

[0012] Calculate the coordinates of the boundary points in the optical gas cloud image based on infrared and optical gas cloud image data (c h d h ); Statistical optical cloud image boundary point coordinates (c h d h The optical leakage region U is formed by combining the coordinates of all pixels within the boundary point; the red dispersion σ is calculated based on the optical cloud image data. R Green dispersion σ G and blue dispersion σ B And further calculate the color offset index RGB;

[0013] Based on the boundary point coordinates (c) of the optical cloud image h d h Calculate the perimeter b of the leak area, and further calculate the gas diffusion index L;

[0014] The leakage assessment index K is calculated based on the temperature anomaly index J, gas diffusion velocity V, color shift index RGB, and color gas diffusion index L; a set of leakage thresholds is preset; the degree of gas leakage is determined based on the leakage assessment index K and the set of leakage thresholds; and remedial measures are determined based on the degree of gas leakage.

[0015] In the preferred embodiment of the above-mentioned intelligent monitoring method for hazardous gas leaks based on gas cloud imaging: the method for calculating the temperature gradient amplitude AB(x, y) is as follows:

[0016] Infrared cloud image data includes at least infrared pixels (x, y) and temperature T (x, y);

[0017] Calculate the horizontal temperature gradient A based on temperature T(x, y). x (x, y), based on the following formula:

[0018] A x (x,y) = T(x+1,y) - T(x,y)

[0019] wherein, A x (x,y) is the horizontal temperature gradient of pixel point (x,y) in the infrared gas cloud image data; T(x,y) is the temperature value of pixel point (x,y) in the infrared gas cloud image data; T(x+1,y) is the temperature value of pixel point (x+1,y) in the infrared gas cloud image data;

[0020] According to the temperature T(x,y), the vertical temperature gradient B x (x,y) is calculated, and the formula is:

[0021] B x (x,y) = T(x,y+1) - T(x,y)

[0022] wherein, B x (x,y) is the vertical temperature gradient of pixel point (x,y) in the infrared gas cloud image data; T(x,y+1) is the temperature value of pixel point (x,y+1) in the infrared gas cloud image data;

[0023] According to the horizontal temperature gradient A x (x,y) and the vertical temperature gradient B x (x,y), the temperature gradient amplitude AB(x,y) is calculated, and the formula is:

[0024]

[0025] wherein, AB(x,y) is the temperature gradient amplitude of pixel point (x,y) in the infrared gas cloud image data.

[0026] In the preferred scheme of the above-mentioned dangerous gas leakage intelligent monitoring method based on gas cloud imaging, the method for determining the infrared leakage area range S is:

[0027] According to the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x,y), the infrared leakage area range S is determined, and the specific way is:

[0028] S = {(x,y) | L(x,y) = 1}

[0029]

[0030] In the preferred scheme of the above-mentioned dangerous gas leakage intelligent monitoring method based on gas cloud imaging, the method for calculating the temperature anomaly index J is:

[0031] The infrared gas cloud image data further includes the average temperature of the infrared leakage area range S Temperature standard deviation σ of the infrared leakage area range S T and the average temperature of the area not affected by the leakage

[0032] According to the average temperature Temperature standard deviation σ of the infrared leakage area range S T and the average temperature The temperature anomaly index J is calculated in the following specific manner:

[0033]

[0034] In the preferred scheme of the above-mentioned intelligent monitoring method for dangerous gas leakage based on gas cloud imaging: the method for calculating the centroid coordinates (E m , F m ) is as follows:

[0035] The centroid coordinates (E m , F m ) are calculated according to the boundary point coordinates (x h , y h ) in the following specific manner:

[0036]

[0037] wherein (x h , y h ) are the coordinates of the hth point on the boundary of the infrared leakage area; (x h , y h ) ∈ S; h is the serial number of different points on the boundary of the infrared leakage area, and takes a value of [1, g]; g is the total number of boundary points, and g is a positive integer; (E m , F m ) are the centroid coordinates of the infrared leakage area at time m; m is the serial number of different time points, and takes a positive integer value; w h is the area of the position where the boundary point (x h , y h ) is located.

[0038] In the preferred scheme of the above-mentioned intelligent monitoring method for dangerous gas leakage based on gas cloud imaging: the method for calculating the gas diffusion speed V is as follows:

[0039] The gas diffusion speed V is calculated according to the centroid coordinates (E m , F m ) in the following specific manner:

[0040]

[0041] wherein (E m+k , F m+k) is the centroid coordinate of the leakage area at time m+k; Δt is the time interval; k is the frame number of the time interval.

[0042] In the preferred embodiment of the above-mentioned method for intelligent monitoring of dangerous gas leakage based on gas cloud imaging, the method for calculating the optical leakage area range U is:

[0043] The infrared gas cloud image data further includes infrared imaging system coordinates (X IR , Y IR ), infrared imaging system focal length f IR , infrared imaging horizontal resolution M IR , and infrared imaging vertical resolution N IR .

[0044] The optical gas cloud image data at least includes optical imaging system coordinates (X O , Y O ), optical imaging system focal length f O , optical imaging horizontal resolution M O , and optical imaging vertical resolution N O .

[0045] According to the infrared imaging system coordinates (X IR , Y IR ), infrared imaging system focal length f IR , infrared imaging horizontal resolution M IR , infrared imaging vertical resolution N IR , boundary point coordinates (x h , y h ), optical imaging system coordinates (X O , Y O ), optical imaging system focal length f O , optical imaging horizontal resolution M O , and optical imaging vertical resolution N O , the optical gas cloud image boundary point coordinates (c h , d h ) are calculated, and the specific method is as follows:

[0046]

[0047] wherein (c h , d h ) is the optical gas cloud image coordinates corresponding to the hth point on the infrared leakage area boundary; the optical gas cloud image boundary point coordinates (c h , d h ) and all pixel point coordinates within the boundary points are counted to form the optical leakage area range U.

[0048] In the preferred scheme of the above-mentioned dangerous gas leakage intelligent monitoring method based on gas cloud imaging: the method for calculating the color deviation index RGB is:

[0049] The optical gas cloud image data further includes the number p of pixel points in the optical leakage area range U, the red value Red a , the green value Green a , and the blue value Blue a ;

[0050] The red dispersion σ a is calculated according to the red value Red R :

[0051]

[0052] wherein Red a is the red value of the a-th pixel point in the optical leakage area range U; a is the serial number corresponding to different pixel points in the optical leakage area range U, and takes the value [1, p]; p is the total number of pixel points in the optical leakage area range U, and takes a positive integer value;

[0053] The green dispersion σ a is calculated according to the green value Green G :

[0054]

[0055] wherein Green a is the green value of the a-th pixel point in the optical leakage area range U;

[0056] The blue dispersion σ a is calculated according to the blue value Blue B :

[0057]

[0058] wherein Blue a is the blue value of the a-th pixel point in the optical leakage area range U;

[0059] The color deviation index RGB is calculated according to the red dispersion σ R , the green dispersion σ G , and the blue dispersion σ B :

[0060]

[0061] In the preferred scheme of the above-mentioned dangerous gas leakage intelligent monitoring method based on gas cloud imaging: the method for calculating the gas diffusion index L is:

[0062] According to the optical gas cloud image boundary point coordinates (c h , d h ), the perimeter b of the leakage area is calculated in the following way:

[0063] c min = min{c h |c h ∈ (c h , d h )}

[0064] c max = max{c h |c h ∈ (c h , d h )}

[0065] d min = min{d h |d h ∈ (c h , d h )}

[0066] d max = max{d h |d h ∈ (c h , d h )}

[0067] b = [(c max - c min ) + (d max - d min )] x 2

[0068] wherein c min is the minimum horizontal coordinate in the optical gas cloud image boundary point coordinates; c max is the maximum horizontal coordinate in the optical gas cloud image boundary point coordinates; d min is the minimum vertical coordinate in the optical gas cloud image boundary point coordinates; and d max is the maximum vertical coordinate in the optical gas cloud image boundary point coordinates.

[0069] According to the pixel point number p and the perimeter b, the gas diffusion index L is calculated:

[0070]

[0071] In the preferred scheme of the above-mentioned method for intelligent monitoring of dangerous gas leakage based on gas cloud imaging: the method for judging the gas leakage degree according to the leakage degree evaluation index K and the leakage degree threshold set is:

[0072] A leakage degree evaluation index K is calculated according to the temperature anomaly index J, the gas diffusion speed V, the color shift index RGB and the gas diffusion index L, and the standard is:

[0073] K = β1 * J + β2 * V + β3 * RGB + β4 * L

[0074] Wherein, β1 is the weight coefficient of the temperature anomaly index J, and the value is 0.2-0.4; β2 is the weight coefficient of the gas diffusion speed V, and the value is 0.3-0.4; β3 is the weight coefficient of the color shift index RGB, and the value is 0.2-0.3; β4 is the weight coefficient of the gas diffusion index L, and the value is 0.1-0.3; and β1 + β2 + β3 + β4 = 1;

[0075] The leakage degree threshold set includes a severe leakage threshold KZ and a slight leakage threshold KQ;

[0076] The gas leakage degree is determined according to the leakage degree evaluation index K and the leakage degree threshold set, and the formula is:

[0077]

[0078] (Three) beneficial effects

[0079] The present application provides a kind of based on gas cloud imaging dangerous gas leakage intelligent monitoring method, with following beneficial effects:

[0080] (1) based on infrared imaging technology and optical imaging technology, dangerous gas in chemical industry park is monitored and infrared gas cloud image data and optical gas cloud image data are generated, which is conducive to more accurately determining whether gas leakage occurs. The information obtained by the two imaging methods can complement each other, improving monitoring accuracy and reliability.

[0081] (2) according to the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB (x, y), the infrared leakage area range S is determined, which can more accurately locate the temperature anomaly area near the leakage source, and help to discover potential leakage points in time;Temperature anomaly index J is calculated based on infrared gas cloud image, which can quantify the degree of temperature change caused by gas leakage. By calculating the gas diffusion speed V, the diffusion process of gas leakage can be dynamically tracked. By calculating the color shift index RGB, the distribution of the leaked gas can be reflected from the color feature angle. By calculating the gas diffusion index L, the dynamic characteristics of the leakage can be better understood. These quantitative indexes provide intuitive data support for the staff;

[0082] (3) According to the temperature anomaly index J, the gas diffusion speed V, the color offset index RGB and the color gas diffusion index L, the leakage degree evaluation index K is calculated, which can comprehensively measure the severity of gas leakage from multiple key dimensions, avoid the limitations of single factor judgment, and solve the problem that only a single temperature feature is used to judge whether gas leakage occurs, which is easy to cause misjudgment; according to the leakage degree evaluation index K and the leakage degree threshold set, the gas leakage degree is judged; according to the gas leakage degree, the remedial measures are determined, so that the staff can take targeted measures according to the specific leakage degree, and the accuracy of emergency decision-making is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0083] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings required to be used in the specific embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0084] Figure 1 A working step schematic diagram of the present application is shown in the figure.

[0085] Figure 2 A working step schematic diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0086] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0087] Please refer to Figure 1 The present application provides a kind of based on gas cloud imaging dangerous gas leakage intelligent monitoring method, comprising the following steps:

[0088] Step 1: based on infrared imaging technology and optical imaging technology, dangerous gas in chemical industry park is monitored and infrared gas cloud image data and optical gas cloud image data are generated.

[0089] According to the content of step 1:

[0090] Based on infrared imaging technology and optical imaging technology, the dangerous gas in the chemical industry park is monitored, and infrared gas cloud image data and optical gas cloud image data are generated, which is beneficial to more accurately determine whether gas leakage occurs. The information obtained by the two imaging methods can complement each other, improve the monitoring accuracy and reliability.

[0091] Step 2: Calculate the horizontal temperature gradient A based on the infrared gas cloud image data x (x, y) and the vertical temperature gradient B x (x, y), and further calculate the temperature gradient amplitude AB(x, y); preset the temperature gradient amplitude threshold CH; determine the infrared leakage area range S according to the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x, y); calculate the temperature anomaly index J based on the infrared gas cloud image.

[0092] Use image processing algorithm to identify the image boundary of the infrared leakage area range S, and obtain the boundary point coordinates (x h , y h ); calculate the center of mass coordinates (E h , F h ) according to the boundary point coordinates (x m , y m ), and further calculate the gas diffusion speed V.

[0093] Calculate the optical gas cloud image boundary point coordinates (c h , d h ) based on the infrared gas cloud image and the optical gas cloud image data; count the optical gas cloud image boundary point coordinates (c h , d h ) and all pixel point coordinates within the boundary point, to form the optical leakage area range U; calculate the red dispersion σ R , green dispersion σ G and blue dispersion σ B based on the optical gas cloud image data, and further calculate the color shift index RGB.

[0094] Calculate the leakage area perimeter b according to the optical gas cloud image boundary point coordinates (c h , d h ), and further calculate the gas diffusion index L.

[0095] Step 201: The method for calculating the temperature gradient amplitude AB(x, y) is:

[0096] The infrared gas cloud image data at least includes infrared pixel point (x, y) and temperature T(x, y).

[0097] It should be noted that the infrared pixel point (x, y) is the coordinate of each pixel point in the infrared cloud image, and the origin of the infrared cloud image coordinate is at the top left corner, the x-axis positive direction is to the right, and the y-axis positive direction is downward. The temperature T(x, y) is obtained through the temperature measurement function of the infrared imaging system.

[0098] The horizontal temperature gradient A x (x, y) is calculated according to the temperature T(x, y), and the specific manner is:

[0099] A x (x, y) = T(x+1, y) - T(x, y)

[0100] It should be noted that the formula obtains the horizontal temperature gradient A x (x, y) by calculating the temperature difference of adjacent pixel points in the horizontal direction.

[0101] Wherein, A x (x, y) is the horizontal temperature gradient of the pixel point (x, y) in the infrared cloud image data; T(x, y) is the temperature value of the pixel point (x, y) in the infrared cloud image data; T(x+1, y) is the temperature value of the pixel point (x+1, y) in the infrared cloud image data.

[0102] The vertical temperature gradient B x (x, y) is calculated according to the temperature T(x, y), and the specific manner is:

[0103] B x (x, y) = T(x, y+1) - T(x, y)

[0104] It should be noted that the formula obtains the vertical temperature gradient B x (x, y) by calculating the temperature difference of adjacent pixel points in the vertical direction.

[0105] Wherein, B x (x, y) is the vertical temperature gradient of the pixel point (x, y) in the infrared cloud image data; T(x, y+1) is the temperature value of the pixel point (x, y+1) in the infrared cloud image data.

[0106] The temperature gradient amplitude AB(x, y) is calculated according to the horizontal temperature gradient A x (x, y) and the vertical temperature gradient B x (x, y), and the specific manner is:

[0107]

[0108] It should be noted that the formula is based on the principle of spatial difference operation, and the temperature gradient amplitude AB(x, y) is obtained by comprehensively operating the temperature gradients in the horizontal direction and the vertical direction.

[0109] wherein AB(x, y) is the temperature gradient amplitude of pixel point (x, y) in the infrared gas cloud image data.

[0110] Step 202: The method for determining the infrared leakage area range S is:

[0111] The infrared leakage area range S is determined according to the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x, y), and the specific manner is:

[0112] S = {(x, y) | L(x, y) = 1}

[0113]

[0114] It should be noted that the operation principle of the formula is that at the junction of the leakage area and the surrounding environment, the temperature will rapidly transition from the temperature of the leaked gas to the ambient temperature. By detecting the mutation of the temperature gradient, the temperature gradient amplitude AB(x, y) exceeding the temperature gradient amplitude threshold CH is marked as 1 and counted, the boundary of the suspected leakage area is determined, and thus the infrared leakage area range S is obtained.

[0115] The determination method of the temperature gradient amplitude threshold CH is that in a laboratory environment, various gas leakage scenarios that may occur in a chemical industrial park are simulated, and temperature data of an infrared imaging system is collected under different gas leakage rates, types of leaked gas, and environmental conditions. The collected data is analyzed to determine a temperature gradient amplitude threshold CH that can distinguish between normal temperature fluctuations and temperature changes caused by gas leakage.

[0116] Step 203: The method for calculating the temperature anomaly index J is:

[0117] The infrared gas cloud image data further includes the average temperature of the infrared leakage area range S The temperature standard deviation σ of the infrared leakage area range S T and the average temperature of the area not affected by the leakage

[0118] It should be noted that the average temperature of the infrared leakage area range S reflects the average value of the temperatures of all pixel points in the area, representing the overall temperature condition of the leakage area. The average temperature of the infrared leakage area range S is determined by adding the temperature values of all pixel points in the leakage area and then dividing by the total number of pixel points. T The temperature standard deviation σ of the infrared leakage area range S The square of the difference of the temperatures of all the pixels in the affected area is added up, and then divided by the total number of pixels to obtain the average temperature difference square value. The square root of the average temperature difference square value is the temperature standard deviation σ T The average temperature of the area not affected by the gas leak The average temperature of the area not affected by the gas leak is the average value of the temperatures of all the pixels in the area not affected by the gas leak. It serves as a reference value for comparison with the temperature of the infrared leak area to more accurately determine the leak situation and evaluate the impact of the leak on the surrounding environment temperature. The average temperature is obtained by adding up the temperature values of all the pixels in the area not affected by the gas leak and then dividing by the total number of pixels.

[0119] According to the average temperature The temperature standard deviation σ of the infrared leak area range S T and the average temperature The temperature anomaly index J is calculated as follows:

[0120]

[0121] It should be noted that this formula takes into account the difference between the average temperature of the leak area and the ambient temperature as well as the degree of temperature dispersion. When the leaked gas causes the area temperature to be significantly higher than the ambient temperature and the temperature distribution is relatively concentrated, the temperature anomaly index J value will be larger.

[0122] Step 204: The method for calculating the centroid coordinates (E m , F m ) is as follows:

[0123] The centroid coordinates (E h , F h ) are calculated according to the boundary point coordinates (x m , y m ) as follows:

[0124]

[0125] It should be noted that the operating principle of this formula is that the centroid is the center position of the mass distribution of an object. For a system composed of multiple mass points, the position of the centroid is related to the mass of each mass point, and in the gas cloud image, the position of the centroid is measured by the area of each mass point. The horizontal coordinate of the centroid is determined by multiplying the value of the horizontal coordinate of each point on the boundary of the leak area by the area and then dividing by the total area according to the proportion of the area of the small area where each point is located to the whole. The vertical coordinate of the centroid is determined by multiplying the value of the vertical coordinate of each point on the boundary of the leak area by the area and then dividing by the total area according to the proportion of the area of the small area where each point is located to the whole.

[0126] where (xh , y h ) is the coordinate of the hth point on the boundary of the infrared leakage area; (x h , y h ) ∈ S; h is the serial number of different points on the boundary of the infrared leakage area, and takes a value of [1, g]; g is the total number of boundary points, and g is a positive integer; (E m , F m ) is the centroid coordinate of the infrared leakage area at time m; m is the serial number of different time points, and takes a positive integer value; w h is the area of the position of the boundary point (x h , y h ), which is obtained by conversion according to the parameters of the imaging device and the actual size of the scene and the corresponding relationship of the pixels.

[0127] Step 205: The method for calculating the gas diffusion speed V is:

[0128] The gas diffusion speed V is calculated according to the centroid coordinates (E m , F m ), and the specific way is:

[0129]

[0130] It should be noted that the formula is based on the speed formula, that is, the speed is equal to the ratio of displacement to time, and the velocity components of the centroid coordinates (E m , F m ) in the x-axis direction and the y-axis direction are calculated, so as to obtain the gas diffusion speed V.

[0131] Wherein, (E m+k , F m+k ) is the centroid coordinate of the leakage area at time m+k, which is calculated according to the centroid coordinate (E m , F m ) calculation formula. Δt is the time interval; k is the frame number of the time interval, which is determined by the setting of the image acquisition device, and can usually be obtained in the parameter configuration of the device.

[0132] Step 206: The method for calculating the optical leakage area range U is:

[0133] The infrared gas cloud image data further includes infrared imaging system coordinates (X IR , Y IR ), infrared imaging focal length f IR , infrared imaging horizontal resolution M IR , and infrared imaging vertical resolution N IR .

[0134] The optical gas cloud image data at least includes optical imaging system coordinates (X O , YO focal length of optical imaging system f O horizontal resolution of optical imaging M O vertical resolution of optical imaging N O .

[0135] It should be noted that the infrared imaging system coordinates (X IR , Y IR ) and the optical imaging system coordinates (X O , Y O ) are the coordinate values of the origin of the infrared cloud image coordinate system and the origin of the optical cloud image coordinate system in the actual space, respectively, which are determined by measurement and positioning during equipment installation and calibration. The focal length f IR of the infrared imaging system, the horizontal resolution M IR of the infrared imaging, the vertical resolution N IR of the infrared imaging, the focal length f O of the optical imaging system, the horizontal resolution M O of the optical imaging, and the vertical resolution N O of the optical imaging are determined by the hardware parameters of the imaging system and are determined by consulting the technical specification manual of the equipment.

[0136] According to the infrared imaging system coordinates (X IR , Y IR ), the focal length f IR of the infrared imaging system, the horizontal resolution M IR of the infrared imaging, the vertical resolution N IR of the infrared imaging, the boundary point coordinates (x h , y h ), the optical imaging system coordinates (X O , Y O ), the focal length f O of the optical imaging system, the horizontal resolution M O of the optical imaging, and the vertical resolution N O of the optical imaging, the optical cloud image boundary point coordinates (c h , d h ) are calculated, and the specific method is as follows:

[0137]

[0138] It should be noted that this formula is based on the principle of similar triangles, is the scaling factor in the horizontal direction of the optical imaging system, is a horizontal position adjustment value considering the boundary point coordinates under the infrared imaging system, and finally subtracting X O is the coordinate of the position converted from the coordinate system of the infrared imaging system to the coordinate system of the optical imaging system, and the horizontal coordinate c h of the optical imaging system is obtained. is the scaling factor in the vertical direction of the optical imaging system, is a vertical position adjustment value considering the coordinate of the boundary point under the infrared imaging system, and finally subtracting Y O is the conversion of this position from the coordinate system of the infrared imaging system to the coordinate system of the optical imaging system, to obtain the vertical coordinate d h of the optical imaging system.

[0139] where (c h , d h ) is the optical image coordinate corresponding to the hth point on the boundary of the optical leakage area; the boundary point coordinates (c h , d h ) and all pixel point coordinates inside the boundary point are counted to form the optical leakage area range U.

[0140] Step 207: the method for calculating the color offset index RGB is:

[0141] The optical image data also includes the number p of pixel points in the optical leakage area range U, the red value Red a , the green value Green a and the blue value Blue a .

[0142] It should be noted that the red value Red a refers to the intensity of red in the pixel point, the green value Green a refers to the intensity of green in the pixel point, and the blue value Blue a refers to the intensity of blue in the pixel point. The number p of pixel points in the optical leakage area range U, the red value Red a , the green value Green a and the blue value Blue a are obtained by image processing software.

[0143] The red dispersion σ R is calculated according to the red value Red a :

[0144]

[0145] It should be noted that the running principle of the formula is: is the average value of the red value of the pixel points in the optical leakage area range U, and the red dispersion σ R is obtained by calculating the sum of the difference between the red value of each pixel point in the optical leakage area range U and the average value and dividing by the total number of pixel points.

[0146] where Red aReda is the red value of the a-th pixel point in the optical leakage area range U; a is the serial number corresponding to different pixel points in the optical leakage area range U, and takes the value of p1, p2; p is the total number of pixel points in the optical leakage area range U, and takes the value of a positive integer;

[0147] According to the green value Green a Calculate the green dispersion σ G :

[0148]

[0149] It should be noted that the running principle of the formula is: Green is the average value of the green value of the pixel point in the optical leakage area range U, and the green dispersion σ is obtained by calculating the sum of the difference between the green value of each pixel point in the optical leakage area range U and the average value and dividing the total number of pixel points. G .

[0150] Wherein, Green a is the green value of the a-th pixel point in the optical leakage area range U;

[0151] According to the blue value Blue a Calculate the blue dispersion σ B :

[0152]

[0153] It should be noted that the running principle of the formula is: Blue is the average value of the blue value of the pixel point in the optical leakage area range U, and the blue dispersion σ is obtained by calculating the sum of the difference between the blue value of each pixel point in the optical leakage area range U and the average value and dividing the total number of pixel points. B .

[0154] Wherein, Blue a is the blue value of the a-th pixel point in the optical leakage area range U.

[0155] According to the red dispersion σ R , the green dispersion σ G and the blue dispersion σ B Calculate the color offset index RGB, and the specific way is:

[0156]

[0157] It should be noted that the formula considers the red dispersion σ R , the green dispersion σ G and the blue dispersion σ BThis yields the color shift index RGB. The larger the RGB value, the higher the color shift and the more severe the gas leak.

[0158] Step 208: The method for calculating the gas diffusion index L is as follows:

[0159] Based on the boundary point coordinates (c) of the optical cloud image h d h The perimeter b of the leak area is calculated as follows:

[0160] c min =min{c h |c h ∈(c h d h )}

[0161] c max =max{c h |c h ∈(c h d h )}

[0162] d min =min{d h |d h ∈(c h d h )}

[0163] d max =max{d h |d h ∈(c h d h )}

[0164] b = [(c max -c min )+(d max -d min )]×2

[0165] It should be noted that the formula works as follows: The leaking area is represented as an equivalent rectangle. By iterating through the pixel coordinates of the leaking area, the minimum, maximum, minimum, and maximum values ​​of the horizontal and vertical coordinates are found. The difference between the maximum and minimum horizontal coordinates is the width of the equivalent rectangle, and the difference between the maximum and minimum vertical coordinates is the length. Using the formula for the perimeter of a rectangle, the perimeter *b* of the leaking area is obtained.

[0166] Among them, c min c is the minimum x-coordinate of the boundary point coordinates in the optical cloud image; max d represents the maximum x-coordinate of the boundary point coordinates in the optical cloud image; min d is the minimum ordinate of the boundary point coordinates in the optical cloud image;max is the maximum longitudinal coordinate in the coordinate of the boundary point of the optical gas cloud image;

[0167] According to the pixel point number p and the leakage area perimeter b, the gas diffusion index L is calculated, and the specific method is:

[0168]

[0169] It should be noted that the formula is based on the concept of circularity, and the gas diffusion index L is obtained according to the ratio of the pixel point number p and the square of the leakage area perimeter b. The gas diffusion index L is used to measure the gas diffusion range, and the larger the gas diffusion index L, the larger the gas diffusion range.

[0170] In combination with the contents of steps 201 to 208:

[0171] According to the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x, y), the infrared leakage area range S is determined, which can more accurately locate the temperature abnormal area near the leakage source, and help to discover potential leakage points in time; based on the infrared gas cloud image, the temperature anomaly index J is calculated, which can quantify the degree of temperature change caused by gas leakage. By calculating the gas diffusion speed V, the diffusion process of gas leakage can be dynamically tracked, by calculating the color shift index RGB, the distribution of the leaked gas can be reflected from the color feature angle, and by calculating the gas diffusion index L, the dynamic characteristics of the leakage can be better understood. These quantitative indicators provide intuitive data support for the staff

[0172] Step 3: Calculate the leakage degree evaluation index K according to the temperature anomaly index J, the gas diffusion speed V, the color shift index RGB and the gas diffusion index L; a set of preset leakage degree thresholds is set; determine the gas leakage degree according to the leakage degree evaluation index K and the set of leakage degree thresholds; determine the remedial measures according to the gas leakage degree.

[0173] Step 301: The method for determining the gas leakage degree according to the leakage degree evaluation index K and the set of leakage degree thresholds is:

[0174] According to the temperature anomaly index J, the gas diffusion speed V, the color shift index RGB and the gas diffusion index L, the leakage degree evaluation index K is calculated, and the standard is:

[0175] K=β1×J+β2×V+β3×RGB+β4×L

[0176] It should be noted that the formula is based on the concept of circularity, and the gas diffusion index L is obtained according to the ratio of the pixel point number p and the square of the leakage area perimeter b. The gas diffusion index L is used to measure the gas diffusion range, and the larger the gas diffusion index L, the larger the gas diffusion range.

[0177] wherein β1 is a weight coefficient of the temperature anomaly index J, and has a value of 0.2-0.4, determined according to the influence of the temperature anomaly index J on the leakage degree evaluation index K. β2 is a weight coefficient of the gas diffusion speed V, and has a value of 0.3-0.4, determined according to the influence of the gas diffusion speed V on the leakage degree evaluation index K. β3 is a weight coefficient of the color shift index RGB, and has a value of 0.2-0.3, determined according to the influence of the color shift index RGB on the leakage degree evaluation index K. β4 is a weight coefficient of the gas diffusion index L, and has a value of 0.1-0.3, determined according to the influence of the gas diffusion index L on the leakage degree evaluation index K. And β1 + β2 + β3 + β4 = 1;

[0178] The leakage degree threshold set includes a severe leakage threshold KZ and a mild leakage threshold KQ.

[0179] The gas leakage degree is determined according to the leakage degree evaluation index K and the leakage degree threshold set, specifically in the following manner:

[0180]

[0181] It should be noted that the determination method of the severe leakage threshold KZ and the mild leakage threshold KQ is as follows: records of past dangerous gas leakage accidents in chemical industrial parks, factories or other related places are collected, including accident reports, monitoring data, on-site photos and videos, etc. These data are sorted out, the historical leakage degree evaluation index is calculated according to the above method, and the median, mean and standard deviation of the historical leakage degree evaluation index are further calculated. The severe leakage threshold KZ is determined with reference to the smaller value of the median and the mean, and the mild leakage threshold KQ is determined with reference to the value of the mean minus the standard deviation.

[0182] In combination with the content of step 301:

[0183] The leakage degree evaluation index K is calculated according to the temperature anomaly index J, the gas diffusion speed V, the color shift index RGB and the gas diffusion index L, which can comprehensively measure the severity of gas leakage from multiple key dimensions, avoiding the limitations of single factor judgment, and solving the problem of easy misjudgment by only using a single temperature feature to determine whether gas leakage has occurred. The gas leakage degree is determined according to the leakage degree evaluation index K and the leakage degree threshold set. The remedial measures are determined according to the gas leakage degree, so that the staff can take targeted measures according to the specific leakage degree, thereby enhancing the accuracy of emergency decision-making.

[0184] In another embodiment, the present scheme also discloses a dangerous gas leakage intelligent monitoring system based on gas cloud imaging, which can include the following main modules:

[0185] An image data generation module is capable of monitoring dangerous gas in the chemical industrial park based on infrared imaging technology and optical imaging technology and generating infrared plume image data and optical plume image data;

[0186] An index calculation module is capable of calculating horizontal temperature gradient A x (x, y) and vertical temperature gradient B x (x, y) and further calculating temperature gradient amplitude AB(x, y); presetting a temperature gradient amplitude threshold CH; determining an infrared leakage area range S according to the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x, y); calculating a temperature anomaly index J based on the infrared plume image;

[0187] Using an image processing algorithm to identify an image boundary of the infrared leakage area range S and obtaining boundary point coordinates (x h , y h ); calculating centroid coordinates (E m , F m ) according to the boundary point coordinates (x h , y h ) and further calculating a gas diffusion speed V;

[0188] Calculating optical plume image boundary point coordinates (c h , d h ) based on the infrared plume image data and the optical plume image data; counting the optical plume image boundary point coordinates (c h , d h ) and all pixel point coordinates within the boundary point to form an optical leakage area range U; calculating red dispersion σ R , green dispersion σ G and blue dispersion σ B based on the optical plume image data and further calculating a color shift index RGB;

[0189] Calculating a leakage area perimeter b according to the optical plume image boundary point coordinates (c h , d h ) and further calculating a gas diffusion index L;

[0190] A judgment module is capable of calculating a leakage degree evaluation index K according to the temperature anomaly index J, the gas diffusion speed V, the color shift index RGB and the gas diffusion index L; presetting a leakage degree threshold set; judging a gas leakage degree according to the leakage degree evaluation index K and the leakage degree threshold set; determining a remedial measure according to the gas leakage degree.

[0191] Furthermore, in the system, the various functional modules can work in coordination with each other to execute the method of intelligent monitoring of dangerous gas leakage as set forth in the above embodiments.

[0192] Further, the system can further comprise an alarm module to interact with a remote terminal (such as a personal digital terminal, a mobile phone, a remote control center, etc.) in a wired or wireless manner, to timely send or upload monitoring data, and to send an alarm signal after discovering a leakage, to timely remind monitoring personnel to perform emergency treatment.

[0193] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the technical solutions.

[0194] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiments according to actual needs.

[0195] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered by the protection scope of the present application.

Claims

1. A method for intelligent monitoring of hazardous gas leakage based on gas cloud imaging, characterized in that: The method comprises the following steps: monitoring a dangerous gas in a chemical industrial park based on infrared imaging technology and optical imaging technology to generate infrared gas cloud image data and optical gas cloud image data; Calculating horizontal temperature gradient A based on infrared cloud image data x (x, y) and vertical temperature gradient B x (x, y), and further calculating temperature gradient amplitude AB(x, y); presetting temperature gradient amplitude threshold CH; determining infrared leakage area range S according to temperature gradient amplitude threshold CH and temperature gradient amplitude AB(x, y); calculating a temperature anomaly index J based on the infrared gas cloud image data; An image processing algorithm is used to identify the image boundary of the infrared leakage area range S, and obtain the boundary point coordinates (x h , y h ); the centroid coordinates (E h , F h ) are calculated according to the boundary point coordinates (x m , y m ), and the gas diffusion velocity V is further calculated; calculating optical cloud image boundary point coordinates (c h , d h ) based on infrared cloud image and optical cloud image data h , d h ) and all pixel point coordinates within the boundary points, to form an optical leakage area range U; calculating red dispersion σ R , green dispersion σ G and blue dispersion σ B based on optical cloud image data, and further calculating a color shift index RGB; According to the coordinates of the boundary points (c h , d h ) of the optical cloud image, the perimeter b of the leakage area is calculated, and a gas diffusion index L is further calculated. calculating a leakage degree evaluation index K based on the temperature anomaly index J, a gas diffusion speed V, a color deviation index RGB and a gas diffusion index L; presetting a leakage degree threshold set; judging the gas leakage degree based on the leakage degree evaluation index K and the leakage degree threshold set; determining a remedial measure based on the gas leakage degree; and the method for calculating the temperature gradient amplitude AB(x, y) is: the infrared gas cloud image data at least comprises an infrared pixel point (x, y) and a temperature T(x, y); The horizontal temperature gradient A is calculated from the temperature T(x, y) according to the formula: x (x, y), according to the formula: A x (x, y) = T(x + 1, y) - T(x, y) wherein A x (x, y) is the horizontal temperature gradient of pixel (x, y) in the infrared cloud image data; T(x, y) is the temperature value of pixel (x, y) in the infrared cloud image data; T(x+1, y) is the temperature value of pixel (x+1, y) in the infrared cloud image data; The vertical temperature gradient B is calculated from the temperature T(x, y) according to the formula: x (x, y), according to the formula: B x (x, y) = T(x, y + 1) - T(x, y) wherein B x (x, y) is the vertical temperature gradient of the pixel point (x, y) in the infrared cloud image data; T(x, y+1) is the temperature value of the pixel point (x, y+1) in the infrared cloud image data; According to the horizontal temperature gradient A x (x, y) and the vertical temperature gradient B x (x, y) the temperature gradient amplitude AB(x, y) is calculated according to the formula wherein, AB(x, y) is the temperature gradient amplitude of the pixel point (x, y) in the infrared gas cloud image data. 2.The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: the method for determining the infrared leakage area range S is: determining the infrared leakage area range S based on the temperature gradient amplitude threshold CH and the temperature gradient amplitude AB(x, y), and the specific method is: S={(x,y)|L(x,y)=1} 3.The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: the method for calculating the temperature anomaly index J is: The infrared cloud image data further comprises a mean temperature of the infrared spill region range S The infrared cloud image data further comprises a mean temperature of the infrared spill region range S T and the mean temperature of the region not affected by the spill According to the average temperature Temperature standard deviation σ of the infrared leakage region range S T and the average temperature The temperature anomaly index J is calculated in the following manner: 4.The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: A method for calculating the centroid coordinates (E m , F m ) is: The centroid coordinates (E m , F m ) are calculated from the boundary point coordinates (x h , y h ) as follows: wherein (x h , y h ) is the coordinate of the hth point on the boundary of the infrared leakage area; (x h , y h ) ∈ S; h is the serial number of different points on the boundary of the infrared leakage area, and takes a value of [1, g]; g is the total number of boundary points, and g is a positive integer; (E m , F m ) is the centroid coordinate of the infrared leakage area at time m; m is the serial number corresponding to different time points, and takes a positive integer value; w h is the area of the position where the boundary point (x h , y h ) is located.

5. The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: the method for calculating the gas diffusion speed V is: Based on the centroid coordinates (E) m F m The gas diffusion velocity V is calculated as follows: wherein (E m+k , F m+k ) is the centroid coordinate of the leakage region at time m+k; Δt is the time interval; and k is the number of frames of the time interval. 6.The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: the method for calculating the optical leakage area range U is: The infrared cloud image data further includes infrared imaging system coordinates (X IR , Y IR ), an infrared imaging system focal length f IR , an infrared imaging horizontal resolution M IR , and an infrared imaging vertical resolution N IR ; The optical cloud image data comprises at least optical imaging system coordinates (X O , Y O ), an optical imaging system focal length f O , an optical imaging horizontal resolution M O , and an optical imaging vertical resolution N O ; According to the infrared imaging system coordinates (X IR , Y IR ), the infrared imaging system focal length f IR , the infrared imaging horizontal resolution M IR , the infrared imaging vertical resolution N IR , the boundary point coordinates (x h , y h ), the optical imaging system coordinates (X O , Y O ), the optical imaging system focal length f O , the optical imaging horizontal resolution M O , and the optical imaging vertical resolution N O , the optical cloud image boundary point coordinates (c h , d h ) are calculated in the following manner: wherein (c h , d h ) is the optical cloud image coordinate corresponding to the hth point on the boundary of the infrared leakage area; the boundary point coordinate (c h , d h ) and all pixel point coordinates within the boundary are counted to form the optical leakage area range U.

7. The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 6, characterized in that: the method for calculating the color deviation index RGB is: The optical cloud image data further comprises a number p of pixels in the optical leakage area range U, a red value Red a , a green value Green a , and a blue value Blue a ; According to the red value Red a The red dispersion σ is calculated R : wherein, Red a Reda is a red value of the a-th pixel point in the optical leakage area range U; a is a serial number corresponding to different pixel points in the optical leakage area range U, and takes a value of [1, p]; p is a total number of pixel points in the optical leakage area range U, and takes a positive integer value; According to the green value Green a Computing the green dispersion σ G : wherein Green a is the green value of the a-th pixel point in the optical leakage region range U; According to the blue value Blue a The blue dispersion σ is calculated B : wherein Blue a is the blue value of the a-th pixel point in the range U of the optical leakage region. The color shift index RGB is calculated from the red dispersion σ R , the green dispersion σ G and the blue dispersion σ B : 8.The intelligent monitoring method for dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: the method for calculating the gas diffusion index L is: The perimeter b of the leak area is calculated from the coordinates of the boundary points (c h , d h ) of the optical cloud image, in particular as follows: c min = min{c h | c h ∈ (c h , d h )} c max = max{c h | c h ∈ (c h , d h )} d min = min{d h | d h ∈ (c h , d h )} d max = max{d h | d h ∈ (c h , d h )} b = [(c max - c min ) + (d max - d min )] x 2 wherein c min is the minimum abscissa of the optical cloud image boundary point coordinates; c max is the maximum abscissa of the optical cloud image boundary point coordinates; d min is the minimum ordinate of the optical cloud image boundary point coordinates; d max is the maximum ordinate of the optical cloud image boundary point coordinates; calculating the gas diffusion index L based on the pixel point number p and the perimeter b: 9.The intelligent monitoring method of dangerous gas leakage based on gas cloud imaging according to claim 1, characterized in that: the method for judging the gas leakage degree based on the leakage degree evaluation index K and the leakage degree threshold set is: calculating the leakage degree evaluation index K based on the temperature anomaly index J, the gas diffusion speed V, the color deviation index RGB and the gas diffusion index L, and the standard is: K=β1×J+β2×V+β3×RGB+β4×L wherein, β1 is the weight coefficient of the temperature anomaly index J, and the value is 0.2-0.4; β2 is the weight coefficient of the gas diffusion speed V, and the value is 0.3-0.4; β3 is the weight coefficient of the color deviation index RGB, and the value is 0.2-0.3; β4 is the weight coefficient of the gas diffusion index L, and the value is 0.1-0.3; and β1+β2+β3+β4=1; the leakage degree threshold set comprises a severe leakage threshold KZ and a slight leakage threshold KQ; the formula for judging the gas leakage degree based on the leakage degree evaluation index K and the leakage degree threshold set is:

Images