A method for processing flow field smoke transmittance distribution test data

Abstract

The present invention discloses a method for processing flow field smoke transmittance distribution test data, converting original image data into relative light intensity data to obtain a relative light intensity image sequence of a surface light source; synthesizing a reference light intensity image based on the relative light intensity image data when there is no smoke; dividing the reference light intensity image into a rectangular area array, wherein the number and size of the rectangular areas are the same as the number and size of the chessboard grids; calculating the average relative light intensity of each rectangular area in the reference light intensity image; partitioning each frame of the relative light intensity image sequence according to the area partitioning method of the reference light intensity image; calculating the average relative light intensity of each rectangular area after partitioning; and calculating the average light transmittance of each rectangular area. The present invention uses the transmittance expected values ​​of adjacent areas to calculate the smoothing coefficient of the area in different directions, and finally performs weighted synthesis of the average light transmittance of the area, thereby improving the calculation accuracy of the average transmittance of the area.

Description

Technical Field

[0001] The invention belongs to the technical field of smoke field image processing, and in particular relates to a method for calculating the transmittance of a gunpowder smoke field based on a chessboard surface light source, which is suitable for processing test data of a solid engine plume smoke field transmittance distribution test or a ballistic gun / gun muzzle smoke field transmittance distribution test. Background Art

[0002] Smoke light transmittance is one of the important parameters of the combustion characteristic parameters of propellants and propellant charges. During the operation of solid rocket engines and barreled weapons, a large amount of smoke will be generated at the nozzle or muzzle. On the one hand, this smoke will expose the position of the weapon system, and on the other hand, it will interfere with the secondary aiming of the video guidance system. The stealth and guidance requirements of the weapon system require that the smoke generated during the combustion process of the propellant be minimized during the formulation design and process.

[0003] With the development of high-tech weapons, there is an increasing demand for low-signal propellants, clean gas generators, and clean propellants, and specific smoke light transmittance index requirements are also put forward for large charges. The smoke field generated by solid engines or ballistic guns and running guns occupies a large space, and the distribution and movement of the smoke field are highly random, which increases the difficulty of characterizing the light transmittance of its smoke field.

[0004] In recent years, China has developed a plume smoke field test device and data processing method based on a surface light source and an image sensor. In the solid propellant plume smoke video image processing method (ZL201510245317.6), the grayscale calculation in the early stage of the data processing method did not perform grayscale-light intensity calibration, which will affect the accuracy of the data processing results to a certain extent. In the propellant smoke shielding ability field test device and method (ZL 201510245809.5), the second step of the test method is to use a set of neutral attenuation plates to measure the transmittance of the test device at the test site, and use the measured values ​​and the nominal values ​​to draw a piecewise linear calibration curve, and use the calibration curve to calibrate the data in the fourth step of data processing. In the processing step (8) of a flow field smoke transmittance distribution test device and data processing method (201910729271.3), the transmittance is calibrated using the calibration characteristic curve of the equipment. In fact, the transmittance response characteristics of the system are not only related to the grayscale-light intensity response characteristics of the image sensor and the light intensity distribution of the light source, but are also significantly affected by the light noise in the smoke reflection and refraction environment during the test. On the one hand, calibrating the transmittance characteristic curve of the test device at the test site cannot eliminate the light noise introduced in the test; on the other hand, measuring the transmittance calibration characteristics of the test device at the test site will inevitably consume a certain amount of time, affect the efficiency of the test, and easily introduce environmental noise interference and human operation errors.

[0005] Secondly, in steps (6) and (7) of a flow field smoke transmittance distribution test device and data processing method (201910729271.3), pixels are extracted from the corresponding positions of each grid area to form a submatrix, and the transmittance of the pixel point is calculated by determining the smooth relationship of the transmittance data between the upper and lower or left and right in the submatrix. This data processing method takes the corresponding points of adjacent areas for data processing at the pixel level. There is much room for improvement in the probability of meeting the assumed conditions and the accuracy of the data results.

[0006] Finally, in step (6) of a flow field smoke transmittance distribution test device and data processing method (201910729271.3), the proposed weight coefficient synthesis of the transmittance test results in the horizontal and vertical directions is performed, and a weight coefficient assignment method is given. The assignment method uses the compressed image grayscale information for calculation and judgment, and the noise component contained in the compressed image grayscale information will affect the accuracy of the smoothing coefficient, which has room for further improvement. Summary of the invention

[0007] In view of the above technical problems, the present invention provides a flow field smoke transmittance distribution test data processing method, which converts the grayscale image into a relative light intensity image, redesigns the calculation method of the smoothing coefficient in the data processing method, and limits the scale of regional division to improve the accuracy of the data processing results. This solves the problem that the previous data processing method requires on-site calibration of the transmittance response characteristics of the test device, thereby improving the accuracy of the data processing results.

[0008] The present invention is specifically implemented by the following technical solutions:

[0009] A method for processing flow field smoke transmittance distribution test data, wherein the original data is an image sequence of smoke-free images and smoke-containing images based on a chessboard-shaped background surface light source, and the data processing includes the following steps:

[0010] Step 1, converting the original image data into relative light intensity data to obtain a relative light intensity image sequence of the surface light source;

[0011] Step 2, synthesizing a reference light intensity image based on the relative light intensity image data when there is no smoke;

[0012] Step 3, dividing the reference light intensity image into a rectangular area array, wherein the number and size of each rectangular area are the same as the number and size of the chessboard grid; calculating the average relative light intensity of each rectangular area in the reference light intensity image;

[0013] Step 4, partitioning each frame of the relative light intensity image sequence according to the region partitioning method of the reference light intensity image in step 3; calculating the average relative light intensity of each rectangular area after partitioning;

[0014] Step 5, calculate the average light transmittance of each rectangular area obtained by partitioning in step 4, specifically:

[0015] 1) For any rectangular area A, its left and right adjacent areas are set as B and C, and its upper and lower adjacent rectangular areas are set as D and F. The average light transmittance τ of the horizontal solution area of ​​rectangular area A is h_a Obtained through the following methods:

[0016] If area A is not at the left or right edge of the image, τ is calculated using formula (1): h_a ,

[0017]

[0018] In formula (1), I a Represents the average relative light intensity in area A, I a ′ represents the average relative light intensity of area A in the reference light intensity image; I b represents the average relative light intensity of area B, I b ′ represents the average relative light intensity of area B in the reference light intensity image; I c represents the average relative light intensity in area C, I c ′ represents the average relative light intensity of area C in the reference light intensity image;

[0019] If area A is at the left or right edge of the image, τ is calculated using formula (2): h_a ;

[0020]

[0021] 2) Using the method in step 1), calculate the horizontal average light transmittance τ of the rectangular areas B, C, D, and F h_b , τ h_c , τ h_d , τ h_f ;

[0022] If rectangular area B or C does not exist, let τ h_b =τ h_c or τ h_c =τ h_b Assigning a value to a non-existent area;

[0023] If the rectangular area D or F does not exist, let τ h_d =τ h_f or τ h_f =τ h_d Assigning a value to a non-existent area;

[0024] 3) The average light transmittance τ of the vertical solution area of ​​rectangular area A v_a Obtained through the following methods:

[0025] If area A is not at the upper or lower edge of the image, τ is calculated using formula (3): v_a ;

[0026]

[0027] I d represents the average relative light intensity in area D, I d ′ represents the average relative light intensity of area D in the reference light intensity image; I f represents the average relative light intensity in area F, I f ′ represents the average relative light intensity of area F in the reference light intensity image;

[0028] If area A is at the upper or lower edge of the image, τ is calculated using formula (4): v_a ,

[0029]

[0030] 4) Using the method in step 3), calculate the average light transmittance τ of the vertical solution area of ​​rectangular areas B, C, D, and F v_b , τ v_c , τ v_d , τ v-f ;

[0031] If rectangular area B or C does not exist, let τ v_b =τ v_c or τ v_c =τ v_b Assigning a value to a non-existent area;

[0032] If the rectangular area D or F does not exist, let τ v_d =τ v_f or τ v_f =τ v_d Assigning a value to a non-existent area;

[0033] 5) Calculate the horizontal smoothing coefficient k of area A by using formula (5) and formula (6): h_a and the vertical smoothing coefficient k v_a ,

[0034]

[0035]

[0036] 6) Calculate the average light transmittance τ of area A by formula (7): a ,

[0037] τ a =λ 1 ·τ h_a +λ 2 ·τv_a (7)

[0038] When region A is the region in the first or last row of the image and is not the region at the two ends: 1 =1,λ 2 =0;

[0039] When region A is the region in the first or last column of the image and is not the region at the two ends: 1 =0,λ 2 =1;

[0040] In other cases, if k h_a >k v_a ,λ 1 =0,λ 2 =1; if k h_a <k v_a ,λ 1 =1,λ 2 =0; if k h_a =k v_a ,λ 1 =0.5,λ 2 =0.5;

[0041] In formula (7), λ 1 and λ 2 Represents the weighting coefficient.

[0042] According to step 1) to step 6), the average light transmittance of each rectangular area is finally obtained.

[0043] Preferably, in step 2, the grayscale data is converted into relative light intensity data using the grayscale-relative light intensity response function of the image sensor. When the grayscale-relative light intensity response function of the image sensor is a linear function, the relative light intensity data is directly set equal to the grayscale data, which is equivalent to omitting step 1 and directly performing step 2 calculation.

[0044] Preferably, the step 3 specifically synthesizes the reference light intensity image by averaging the relative light intensity data of multiple frames of smoke-free images, and the smoke-free relative light intensity images are not less than 10 frames.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] (1) The method of the present invention uses the grayscale-relative light intensity response function of the image sensor to calibrate the image sequence, eliminating the systematic error introduced in the test results by the characteristics of the image sensor itself; the data processing process eliminates the systematic error introduced by the light source reference difference. Since the grayscale-relative light intensity characteristics of the image sensor are relatively stable, there is no need for on-site calibration. The efficiency of the test process and the calibration accuracy can be improved.

[0047] (2) The present invention improves the method for calculating the smoothing coefficient of a region, and uses the expected transmittance values ​​of adjacent regions to calculate the smoothing coefficients of the region in different directions. Finally, a weighted synthesis of the average light transmittance of the region is performed, which can improve the calculation accuracy of the average transmittance of the region.

[0048] (3) Taking the area as the basic unit of average transmittance, regional solution can effectively suppress random noise and avoid the reduction of accuracy of data processing results due to the reduction of pixels in the area.

[0049] Other advantages of the present invention are described in detail in the detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flow chart of the method of the present invention.

[0051] Figure 2 These are the grayscale images of the surface light source when there is no smoke. (a) is the grayscale image of the chessboard surface light source, and (b) is the grayscale image of the stripe surface light source.

[0052] Figure 3 is a pseudo-color image of the smoke field transmittance distribution, (a) is the smoke field S 1 (b) is the pseudo-color image of the transmittance distribution of the smoke field S 2 Pseudo-color image of transmittance distribution.

[0053] Figure 4 It is the area light source passing through the smoke field S 1 , S 2 Grayscale image of the image, (a) is the smoke field S of the chessboard surface light source 1 The transmittance distribution grayscale diagram, (b) is the chessboard surface light source smoke field S 2 The transmittance distribution grayscale diagram, (c) is the stripe surface light source smoke field S 1 The transmittance distribution grayscale diagram, (d) is the smoke field S of the stripe surface light source 2 Grayscale diagram of transmittance distribution.

[0054] Figure 5 yes Figure 4 The regional average smoke transmittance distribution diagram of the data processing results corresponding to each image in (a) is the smoke field S of the chessboard surface light source 1 (b) is the transmittance distribution diagram of the chessboard surface light source smoke field S 2 The transmittance distribution diagram of (c) is the smoke field S of the stripe surface light source 1 The transmittance distribution diagram of (d) is the smoke field S of the stripe surface light source 2 Transmittance distribution diagram.

[0055] Figure 6are the calculation results and true values ​​of the light transmittance under the chessboard surface light source and stripe surface light source obtained by the method of the present invention, (a) is the smoke field S 1 The average light transmittance of the area, (b) is the smoke field S 2 The average light transmittance of the area.

[0056] Figure 7 is the maximum error curve between the calculated value and the true value of the regional average light transmittance, (a) is the smoke field S 1 The maximum error curve of (b) is the smoke field S 2 The maximum error curve. DETAILED DESCRIPTION

[0057] The method of the present invention is also applicable to the processing of a stripe surface light source image sequence, wherein the stripe surface light source can be regarded as a chessboard surface light source with only one row or one column.

[0058] The following are specific implementation cases of the present invention. It should be noted that the present invention is not limited to the following specific embodiments. The various specific technical features described in the following specific embodiments can be combined in any suitable manner without contradiction. As long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

[0059] Example 1

[0060] This embodiment uses smoke field simulation data for calculation. This embodiment uses a chessboard-shaped background surface light source. It is assumed that the image of the chessboard surface light source when there is no smoke is a gray and white grid, the grayscale of the gray area is 100, and the grayscale of the white area is 200. Each area is 50×50 pixels, the number of areas is 20×20, and the surface light source imaging resolution is 1000×1000. The grayscale distribution of the chessboard surface light source grayscale image is recorded as G l (x, y), x and y represent the horizontal and vertical coordinates respectively. When there is no smoke, the number of images is 30 frames, and the grayscale image of the surface light source is Figure 2 (a) shown.

[0061] Set up two smoke fields S 1 and S 2 , assuming that the smoke field S 1 The transmittance changes uniformly in the horizontal direction and changes sinusoidally in the vertical direction. The transmittance distribution function is τ 1 (x,y)=0.5(1-x / 1000)+0.5sin(π·y / 1000), x and y represent the ordinate and abscissa of the image pixel respectively. Smoke field S 2 The transmittance decreases at a uniform rate in the vertical direction and changes in a sinusoidal function in the horizontal direction. The transmittance distribution function is

[0062] τ 2 (x,y)=0.5(1-y / 1000)+0.5sin(π·x / 1000).

[0063] Assume the noise distribution is G n (x,y)=randn+50·(rand-0.5), where randn is a normalized Gaussian random noise function, and rand is a random noise uniformly distributed in the interval [0,1].

[0064] The grayscale of the surface light source image when there is smoke is: G(x,y)=G l (x,y)·τ(x,y)+c·G n (x,y), c is the noise amplitude.

[0065] Let c = 10, 20, 30, 40, 50 to generate 5 sets of data in sequence, each set of data contains 6 frames of images. Then the image sequence with smoke and without smoke has a total of 60 frames of grayscale images.

[0066] Smoke Field 1 , S 2 The pseudo-color image of transmittance distribution is as follows Figure 3 As shown in (a) and (b), the grayscale images of the chessboard-shaped surface light source of this embodiment passing through two smoke fields are as follows Figure 4 As shown in (a) and (b) (noise amplitude c=30).

[0067] The specific steps of data processing are:

[0068] Step 1, using the grayscale-relative light intensity response function I(x, y)=G(x, y) of the image sensor to convert the original image data into relative light intensity data, and obtain a relative light intensity image sequence of the surface light source;

[0069] Step 2, taking the first 10 frames of the relative light intensity image sequence and synthesizing a reference light intensity image;

[0070] Step 3, dividing the reference light intensity image into a 20×20 rectangular area array, where the number of pixels in each rectangular area is 50×50; the average relative light intensity of the area is obtained by averaging the relative light intensity of all pixels in the area;

[0071] Step 4, partition each frame of the relative light intensity image sequence according to the regional division method of the reference light intensity image, and calculate the average relative light intensity of each area in the image; wherein the average relative light intensity refers to the sum of the relative light intensities of all pixels in the area divided by the number of pixels in the area.

[0072] Step 5, calculate the average light transmittance of each rectangular area in the image:

[0073] For any rectangular area A, its left and right adjacent areas are set as B and C, and its upper and lower adjacent rectangular areas are set as D and F. The horizontal solution area average light transmittance τ of rectangular area A is h_a And the average light transmittance of the vertical solution area is:

[0074] 1) Horizontal solution of the average light transmittance τ of the rectangular area A h_a Obtained through the following methods:

[0075] If area A is not at the left or right edge of the image, τ is calculated using formula (1): h_a ,

[0076]

[0077] In formula (1), I a Represents the average relative light intensity in area A, I a ′ represents the average relative light intensity of area A in the reference light intensity image; I b represents the average relative light intensity of area B, I b ′ represents the average relative light intensity of area B in the reference light intensity image; I c represents the average relative light intensity in area C, I c ′ represents the average relative light intensity of area C in the reference light intensity image.

[0078] I a ,I b ,I c Calculated in step 4, I a ′、I b ′、I c ′ is calculated in step 3.

[0079] If area A is at the left or right edge of the image, τ is calculated using formula (2): h_a ;

[0080]

[0081] 2) Using the method in step 1), calculate the horizontal average light transmittance τ of the rectangular areas B, C, D, and F h_b , τ h_c , τ h_d , τ h_f ;

[0082] If rectangular area B or C does not exist, let τ h_b =τ h_c or τ h_c =τ h_b Assigning a value to a non-existent area;

[0083] If the rectangular area D or F does not exist, let τ h_d=τ h_f or τ h_f =τ h_d Assigning a value to a non-existent area;

[0084] 3) The average light transmittance τ of the vertical solution area of ​​rectangular area A v_a Obtained through the following methods:

[0085] If area A is not at the upper or lower edge of the image, τ is calculated using formula (3): v_a ;

[0086]

[0087] I d represents the average relative light intensity in area D, I d ′ represents the average relative light intensity of area D in the reference light intensity image; I f represents the average relative light intensity in area F, I f ′ represents the average relative light intensity of area F in the reference light intensity image;

[0088] If area A is at the upper or lower edge of the image, τ is calculated using formula (4): v_a ,

[0089]

[0090] 4) Using the method in step 3), calculate the average light transmittance τ of the vertical solution area of ​​rectangular areas B, C, D, and F v_b , τ v_c , τ v_d , τ v-f ;

[0091] If rectangular area B or C does not exist, let τ v_b =τ v_c or τ v_c =τ v_b Assigning a value to a non-existent area;

[0092] If the rectangular area D or F does not exist, let τ v_d =τ v_f or τ v_f =τ v_d Assigning a value to a non-existent area;

[0093] 5) Calculate the horizontal smoothing coefficient k of area A by using formula (5) and formula (6): h_a and the vertical smoothing coefficient k v_a ,

[0094]

[0095]

[0096] 6) Calculate the average light transmittance τ of area A by formula (7): a ,

[0097] τ a =λ 1 ·τ h_a +λ 2 ·τ v_a (7)

[0098] λ 1 and λ 2 represents the weighting coefficient;

[0099] When region A is the region in the first or last row of the image and is not the region at the two ends: 1 =1,λ 2 =0;

[0100] When region A is the region in the first or last column of the image and is not the region at the two ends: 1 =0,λ 2 =1;

[0101] In other cases, if k h_a >k v_a ,λ 1 =0,λ 2 =1; if k h_a <k v_a ,λ 1 =1,λ 2 =0; if k h_a =k v_a ,λ 1 =0.5,λ 2 =0.5.

[0102] Through calculation, the average light transmittance of each rectangular area is finally obtained.

[0103] Figure 5 (a) and (b) show Figure 4 The regional average light transmittance distribution diagram obtained from the two frames of smoke images.

[0104] Figure 6 The mean curve of the 20 stripe regions obtained by averaging the vertical values ​​is included in (a), where (a) is the smoke field S 1 The average transmittance curve of (b) is the smoke field S 2 The average transmittance curve.

[0105] Example 2

[0106] This embodiment uses a stripe-shaped background surface light source. It is assumed that the image of the stripe surface light source is gray-white stripes when there is no smoke. The grayscale of the gray area is 100, and the grayscale of the white area is 200. The width of the stripe is 50 pixels, the number of stripes is 20, and the imaging resolution of the surface light source is 1000×1000. The number of images when there is no smoke is 30 frames, and the grayscale image of the surface light source is Figure 2 The rest of the settings and data processing are the same as those in Example 1.

[0107] The grayscale image of the chessboard-shaped surface light source of this embodiment through two smoke fields is as follows: Figure 4 As shown in (c) and (d).

[0108] The data results of this embodiment show that the average transmittance distribution of 20 regions is as follows: Figure 5 As shown in (c) and (d). Figure 6 The average transmittance curve of the area containing two frames of smoke images, where (a) is the smoke field S 1 The average transmittance curve of (b) is the smoke field S 2 The average transmittance curve.

[0109] Table 1 is a comparison of the regional average transmittance and the true value of the processing results of Example 1 and Example 2. Table 2 is the absolute value of the error of the processing results of Example 1 and Example 2.

[0110] Table 1 Regional average transmittance

[0111]

[0112] Table 2 Data processing error

[0113]

[0114]

[0115] Figure 7 The error curve is shown, that is, the absolute value of the error between the calculated value and the true value of Example 1 and Example 2. It can be seen that the error of the light transmittance calculation based on the chessboard surface light source is significantly smaller than the error of the light transmittance calculation based on the stripe surface light source. Among them, the error of the chessboard surface light source is the largest in the edge area, which is 0.88%, and the errors in other areas do not exceed 0.24%; while the error of the stripe surface light source is larger at the edge, but the error in the middle area is smaller, and the maximum does not exceed 1.3%.

[0116] In summary, the data processing result of the method of the present invention has high accuracy.

Claims

1. A method for processing flow field smoke transmittance distribution test data, characterized in that: The original data is an image sequence of smoke-free images and smoke-containing images based on a chessboard-shaped background surface light source. The data processing includes the following steps: Step 1, converting the original image data into relative light intensity data to obtain a relative light intensity image sequence of the surface light source; Step 2, synthesizing a reference light intensity image based on the relative light intensity image data when there is no smoke; Step 3, dividing the reference light intensity image into a rectangular area array, wherein the number and size of the rectangular areas are the same as the number and size of the chessboard grids; calculating the average relative light intensity of each rectangular area in the reference light intensity image; Step 4, partitioning each frame of the relative light intensity image sequence according to the region partitioning method of the reference light intensity image in step 3; calculating the average relative light intensity of each rectangular area after partitioning; Step 5, calculate the average light transmittance of each rectangular area obtained by partitioning in step 4, specifically: 1) For any rectangular area A, its left and right adjacent areas are set as B and C, and its upper and lower adjacent rectangular areas are set as D and F. The average light transmittance τ of the horizontal solution area of ​​rectangular area A is h_a Obtained through the following methods: If area A is not at the left or right edge of the image, τ is calculated using formula (1): h_a , In formula (1), I a Represents the average relative light intensity in area A, I a ′ represents the average relative light intensity of area A in the reference light intensity image; I b represents the average relative light intensity in area B, I b ′ represents the average relative light intensity of area B in the reference light intensity image; I c represents the average relative light intensity in area C, I c ′ represents the average relative light intensity of area C in the reference light intensity image; If area A is at the left or right edge of the image, τ is calculated using formula (2): h_a ; 2) Using the method in step 1), calculate the horizontal average light transmittance τ of the rectangular areas B, C, D, and F h_b , τ h_c , τ h_d , τ h_f ; If the rectangular area B or C does not exist, let τ h_b =τ h_c or τ h_c =τ h_b Assigning a value to a non-existent area; If the rectangular area D or F does not exist, let τ h_d =τ h_f or τ h_f =τ h_d Assigning a value to a non-existent area; 3) The average light transmittance τ of the vertical solution area of ​​rectangular area A v_a Obtained through the following methods: If area A is not at the upper or lower edge of the image, τ is calculated using formula (3): v_a ; I d represents the average relative light intensity in area D, I d ′ represents the average relative light intensity of area D in the reference light intensity image; I f represents the average relative light intensity in area F, I f ′ represents the average relative light intensity of area F in the reference light intensity image; If area A is at the upper or lower edge of the image, τ is calculated using formula (4): v_a , 4) Using the method in step 3), calculate the average light transmittance τ of the vertical solution area of ​​rectangular areas B, C, D, and F v_b , τ v_c , τ v_d , τ v-f ; If the rectangular area B or C does not exist, let τ v_b =τ v_c or τ v_c =τ v_b Assigning a value to a non-existent area; If the rectangular area D or F does not exist, let τ v_d =τ v_f or τ v_f =τ v_d Assigning a value to a non-existent area; 5) Calculate the horizontal smoothing coefficient k of area A by using formula (5) and formula (6): h_a and the vertical smoothing coefficient k v_a , 6) Calculate the average light transmittance τ of area A by formula (7): a , t a =λ1·τ h_a +λ2·τ v_a (7) When region A is the region in the first or last row of the image and is not an end region: λ1=1,λ2=0; When region A is the region in the first or last column of the image and is not the region at both ends: λ1=0,λ2=1; In other cases, if k h_a >k v_a ,λ1=0,λ2=1;if k h_a <k v_a ,λ1=1,λ2=0;if k h_a =k v_a , λ1=0.5,λ2=0.5; In formula (7), λ1 and λ2 represent weighting coefficients; According to step 1) to step 6), the average light transmittance of each rectangular area is finally obtained.

2. The flow field smoke transmittance distribution test data processing method according to claim 1, characterized in that: In the step 2, the grayscale data is converted into relative light intensity data using the grayscale-relative light intensity influence function of the image sensor.

3. The flow field smoke transmittance distribution test data processing method according to claim 1, characterized in that: The step 3 specifically synthesizes the reference light intensity image by averaging the relative light intensity data of multiple frames of smoke-free images, and the smoke-free relative light intensity images are no less than 10 frames.

Images