Method for obtaining so2 concentration image by correcting optical dilution effect of camera image

Abstract

The application discloses a method for obtaining SO2 concentration image by correcting light dilution effect of camera image. The method comprises the following steps: taking an original image by an SO2 ultraviolet camera system, processing light intensity information contained in the original image, combining corresponding distance information and atmospheric radiation transmission formula, and inversely calculating atmospheric extinction coefficient; correcting light dilution effect of a smoke plume image collected by the SO2 ultraviolet camera system by using the extinction coefficient, and then inversely calculating the SO2 concentration image by using a standard bubble method calibration curve. The method overcomes the limitation that the standard bubble method calibration can only be used for monitoring pollution gas at a short distance, and fully utilizes information contained in a picture taken by the SO2 ultraviolet camera. The method is simple and independent in operation, high in calibration precision, and low in cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical remote sensing detection of polluted gas, in particular to the correction of light dilution effect when calibrating SO2 ultraviolet camera using standard bubble method, and specifically to a method for correcting light dilution effect of camera image to obtain SO2 concentration image. BACKGROUND

[0002] SO2 is one of the main polluted gases in the atmosphere, which is harmful to human health, can cause damage to the lungs when inhaled, and is the main precursor of acid rain, one of the three major hazards in the world. Industrial processes, vehicle and ship exhaust, and volcanic activity are the main emission pathways of SO2. Accurate monitoring of SO2 polluted gas can provide more accurate data support for atmospheric pollution prevention and monitoring, thereby better protecting the ecological environment and human production and life. In recent years, SO2 ultraviolet camera has been successfully applied to SO2 polluted gas monitoring due to its high spatial and temporal resolution, high detection sensitivity and high detection accuracy.

[0003] SO2 ultraviolet camera does not directly measure the SO2 concentration in the plume, but measures the optical thickness of SO2 in the plume. The measured optical thickness image needs to be converted into a concentration image through a calibration curve, and the accuracy of the calibration curve will directly affect the inversion of SO2 concentration results. The calibration curve can be obtained by standard bubble method and differential optical absorption spectroscopy (DOAS) method. The standard bubble method uses multiple SO2 standard bubbles with different concentrations to obtain a plurality of SO2 optical thickness and SO2 concentration relationship pairs to fit the calibration curve. This method is simple to operate, but it cannot consider the real atmospheric radiation transmission and cannot correct the light dilution effect, which will have a great impact on the accuracy of the calibration curve. The differential optical absorption spectroscopy method uses a high-resolution spectrometer co-located with the SO2 ultraviolet camera to obtain the relationship between the SO2 concentration in a small range of the plume and the SO2 optical thickness measured by the SO2 ultraviolet camera, and then deduces the calibration curve. This method overcomes the influence of light dilution effect on the measurement results, but it is difficult to completely match the field of view of the spectrometer and the camera in actual operation, and the calibration curve derived from a small range of the plume applied to the entire plume range will result in a large error. SUMMARY

[0004] The present application aims to provide a method for correcting light dilution effect of standard bubble calibration based on the original image shot by SO2 ultraviolet camera, specifically a method for correcting light dilution effect of camera image to obtain SO2 concentration image. The present application overcomes the problem of being unable to correct light dilution effect when using standard bubble method for calibration, thereby accurately obtaining SO2 concentration image.

[0005] To achieve the above object, a method for obtaining SO2 concentration image by camera image correction of light dilution effect is provided, comprising the following steps:

[0006] Step S10: calibrating light dilution effect, deriving atmospheric extinction coefficient by shooting original image through SO2 ultraviolet camera;

[0007] Step S20: shooting plume image and sky background image through SO2 ultraviolet camera, correcting plume image by light dilution effect through atmospheric extinction coefficient to obtain corrected plume image;

[0008] Step S30: calculating SO2 optical thickness image through plume image corrected in step S20;

[0009] Step S40: converting SO2 optical thickness image into SO2 concentration image by using standard bubble method to obtain calibration curve;

[0010] Wherein, step S10 comprises the following steps:

[0011] 1.1) quantifying influence of light dilution effect by atmospheric extinction coefficient, imaging two reference objects or terrain areas at different distances from the camera through SO2 ultraviolet camera, shooting two pairs of A, B channel original images, and the field of view of the camera should be as far away from the sun as possible during shooting.

[0012] 1.2) selecting rectangular pixel area in A channel original image, and taking mean value of light intensity of the rectangular pixel area to obtain reference point average light intensity value I Ori (λ) in original image; selecting pixel area in sky background area of A channel original image, and taking mean value of light intensity to obtain sky background intensity I OriW (λ);

[0013] 1.3) obtaining distance D Ori between camera and center point of rectangular pixel area along line of sight direction of SO2 ultraviolet camera through real-time positioning system in A channel original image;

[0014] 1.4) there is linear relationship between measured light intensity value I(λ) of SO2 ultraviolet camera and linear distance d between measured point and camera, which is represented by atmospheric radiation transmission equation as follows:

[0015] I(λ)=I0(λ)(e -σ(λ)d )+I A (λ)(1-e -σ(λ)d )

[0016] Wherein, I A(λ) is the atmospheric background light intensity, σ(λ) is the atmospheric extinction coefficient, I0(λ) is the real light intensity value after the light passes through the smoke plume or is reflected (i.e. the camera measures the light intensity value I(λ) at d = 0);

[0017] Therefore, the above two groups of average reference point light intensity I Ori (λ) and the corresponding distance information D Ori and the sky background intensity I OriW (λ) are substituted into the atmospheric radiation transfer equation after equation transformation, to obtain the A channel inverted atmospheric extinction coefficient σ Α ,

[0018]

[0019] In the formula, I Ori (λ) is the average light intensity value of the reference point in the original image. I OriW (λ) is the sky background intensity of the atmosphere, D Ori is the distance between the camera and the center point of the rectangular pixel region along the SO2 ultraviolet camera line of sight direction, I0(λ) is the initial intensity after the light passes through the smoke plume or is reflected (i.e. I Ori (λ) measured by the camera at D Ori = 0).

[0020] 1.5) B channel inverted atmospheric extinction coefficient σ B The method is the same as that of the A channel.

[0021] Wherein, the step S20 comprises the following steps:

[0022] 2.1) image the measured smoke plume through the A and B channels to obtain the smoke plume images I A , I B of the two channels; image the sky background through the A and B channels at the same time to obtain the sky background images I AW , I BW of the two channels;

[0023] 2.2) then, correct the light dilution effect of the A and B channel smoke plume images, and the steps of the light dilution effect correction of the A and B channel smoke plume images are the same, which are described by taking the A channel as an example.

[0024] 2.21) during the propagation of the photons passing through the smoke plume (or reflected) between the measured point and the camera, part of the photons will be scattered out of the camera field of view by the aerosol particles in the air, based on the atmospheric extinction coefficient σ Α and the A channel smoke plume image I A , the smoke plume image I′ crr,A corrected by the photon loss is obtained,

[0025]

[0026] where I A (i,j) is the light intensity of each pixel point in the A channel plume image, d i,j is the distance between each pixel point in the plume image and the camera (only the pixel points containing plume are considered here).

[0027] 2.22) In the background atmosphere, the photons outside the camera field of view are also scattered partially into the camera field of view. For this part of "pollution", based on the atmospheric extinction coefficient σ Α and the A channel background light intensity image I AW , the background sky error I" crr,A is calculated.

[0028]

[0029] where I AW (i,j) is the light intensity of each pixel point in the A channel background light intensity image,

[0030] 2.23) Finally, the plume image I' crr,A corrected for light depletion is subtracted by the "pollution" of the background sky I" crr,A to obtain the A channel plume image I crr,A， corrected for light depletion effect.

[0031] I crr,A (i,j) = I' crr,A (i,j) - I" crr,A (i,j)

[0032] 2.3) In the same way as step 2.2), the B channel plume image is corrected for light depletion effect to obtain the B channel plume image I crr,B corrected for light depletion effect.

[0033] wherein step S30 comprises the following steps:

[0034] 3.1) Calculate the optical thickness images τ A , τ B

[0035] τ A = -ln(I crr,A / I AW )

[0036] where I crr,A is the A channel plume image corrected for light depletion effect, I AW is the A channel sky background image, which has been photographed in step S20;

[0037] τ B = -ln(Icrr,B / I BW )

[0038] wherein, I crr,B is the plume image of B channel after correction of optical dilution effect, I BW is the sky background image of B channel, which has been taken in step S20;

[0039] 3.2) difference the optical thickness τ A , τ B of A channel and B channel, to obtain the SO2 optical thickness image after correction of optical dilution effect

[0040]

[0041] wherein, step S40 comprises the following steps:

[0042] 4.1) take the standard bubble images I cell,A , I cell,B of A channel and B channel respectively by SO2 standard bubble, take the sky background images I cell,AW , I cell,BW of A channel and B channel respectively;

[0043] 4.2) perform negative logarithm processing on the standard bubble images I cell,A , I cell,B and the corresponding sky background images I cell,AW , I cell,BW to obtain the optical thickness images τ cell,A , τ cell,B of A channel and B channel respectively,

[0044] τ cell,A = -ln(I cell,A / I cell,AW )

[0045] τ cell,B = -ln(I cell,B / I cell,BW )

[0046] 4.3) difference the optical thickness images τ cell,A , τ cell,B of A channel and B channel, to calculate the SO2 optical thickness corresponding to the SO2 concentration of standard bubble

[0047]

[0048] 4.4) replace the SO2 standard bubble with different concentration and repeat steps 4.1) - 4.3), to obtain the SO2 concentration and SO2 optical thickness The standard bubble method calibration curve is fitted as follows,

[0049] y=ax

[0050] wherein a is the calibration coefficient of the standard bubble method calibration curve, x is the SO2 optical thickness, and y is the SO2 concentration,

[0051] 4.5) The SO2 optical thickness image corrected for the light dilution effect in step S30 The SO2 concentration image corrected for the light dilution effect is obtained by substituting the calibration curve

[0052]

[0053] Advantages of the present application

[0054] High precision: After the SO2 concentration image is processed by the method, the SO2 concentration image is compared with the SO2 concentration image obtained by the traditional standard bubble method, and it is found that the method can effectively correct the light dilution effect.

[0055] Low cost: When using the SO2 ultraviolet camera to monitor SO2 in the plume, compared with the existing calibration method (DOAS method) which can correct the light dilution effect, the method has low cost, and the atmospheric extinction coefficient can be inverted without the aid of other external spectral instruments, and has high applicability.

[0056] Simple operation: Only the original image needs to be additionally shot, and the light dilution effect can be corrected. In addition, the original image can be shot at the beginning, middle and end of the monitoring process independently of the process of monitoring the SO2 concentration using the standard bubble method, and will not interfere with the monitoring process.

[0057] High instantaneity: The inversion process of the extinction coefficient for the original image is independent of the shooting process, the SO2 ultraviolet camera used can shoot the original image at a high frame rate of 1 Hz, and therefore the atmospheric extinction coefficient at any time can be obtained in real time.

[0058] The present application proposes a light dilution effect correction method based on SO2 ultraviolet camera image standard bubble calibration, and the main innovations are as follows:

[0059] Firstly, the light dilution effect correction is performed on the image collected by the SO2 ultraviolet camera, which overcomes the difficulty that the light dilution effect cannot be corrected when the traditional standard bubble method is used for calibration.

[0060] Secondly, the photodilution effect correction proposed in this invention for the standard bubble method does not require any additional equipment. It only requires capturing the original image using an SO2 ultraviolet camera to retrieve the atmospheric extinction coefficient, thereby correcting the photodilution effect. In contrast, an SO2 ultraviolet camera is an essential instrument for SO2 concentration monitoring, overcoming the significant cost of purchasing additional equipment required for traditional methods of addressing the photodilution effect. Attached Figure Description

[0061] Figure 1 This is a flowchart of a method for obtaining SO2 concentration images by correcting the light dilution effect in camera images, as proposed in this invention.

[0062] Figure 2 Nearby ship images acquired for channel A to retrieve extinction coefficients.

[0063] Figure 3 Nearby ship images acquired for channel B to retrieve extinction coefficients.

[0064] Figure 4 The distant ship image acquired for channel A, used to retrieve the extinction coefficient.

[0065] Figure 5 The distant ship image acquired for the B channel, used to retrieve the extinction coefficient.

[0066] Figure 6 Image of a ship at a distance of 2.5 km from the SO2 UV camera, obtained for channel A.

[0067] Figure 7 Image of a ship at a distance of 2.5 km from the SO2 UV camera, obtained for channel B.

[0068] Figure 8 An optical thickness image of SO2 in ship exhaust gas corrected for light dilution effect at a distance of 2.5 km from an SO2 ultraviolet camera.

[0069] Figure 9 The results show the SO2 content in the exhaust gas of a ship located 2.5 km away from the SO2 ultraviolet camera after correction for photodilution effect.

[0070] Figure 10 The results show the SO2 content in the exhaust gas of a ship located 2.5 km away from the SO2 ultraviolet camera, without considering the light dilution effect.

[0071] Figure 11 Image of a ship at a distance of 1.8 km from the SO2 UV camera, acquired for channel A.

[0072] Figure 12 Image of a ship at a distance of 1.8 km from the SO2 UV camera, obtained for channel B.

[0073] Figure 13 The SO2 optical thickness image of the ship's exhaust gas at a distance of 1.8 km from the SO2 ultraviolet camera after correction of the light dilution effect.

[0074] Figure 14 The SO2 content of the ship's exhaust gas at a distance of 1.8 km from the SO2 ultraviolet camera after correction of the light dilution effect.

[0075] Figure 15 The SO2 content of the ship's exhaust gas at a distance of 1.8 km from the SO2 ultraviolet camera without considering the light dilution effect.

[0076] Figure 16 Comparison of the SO2 content of the ship's exhaust gas at different distances without correction of the light dilution effect.

[0077] Figure 17 Comparison of the SO2 content of the ship's exhaust gas at different distances after correction of the light dilution effect. DETAILED DESCRIPTION

[0078] The light dilution effect not only affects the plume part of the image taken by the SO2 ultraviolet camera, but also affects the scene around the plume source. The following will further explain and illustrate the light dilution effect correction method based on the standard bubble calibration of the original image taken by the SO2 ultraviolet camera in combination with the accompanying drawings.

[0079] As shown in Figure 1 The method for correcting the light dilution effect of the camera image of the present application to obtain the SO2 concentration image comprises the following steps:

[0080] Step S10: Take the original image by the SO2 ultraviolet camera and derive the atmospheric extinction coefficient.

[0081] Specifically, the atmospheric extinction coefficient is derived from the information contained in the image taken by the SO2 ultraviolet camera to facilitate subsequent light dilution effect correction.

[0082] Two pairs of original images are additionally taken by the SO2 ultraviolet camera, two reference objects close to and far from the camera are selected, and the atmospheric extinction coefficient is derived according to the corresponding light intensity and distance information combined with the atmospheric radiation transfer equation.

[0083] The SO2 ultraviolet camera of the example is composed of two ultraviolet lenses with the same index, two ultraviolet sensitive CCD cameras with the same index, and two ultraviolet band-pass filters with the center wavelengths of 310 nm (filter A) and 330 nm (filter B) respectively. The SO2 ultraviolet camera A, the filter A, and the ultraviolet lens A are coaxially connected in sequence to form a signal channel (A channel); the SO2 ultraviolet camera B, the filter B, and the ultraviolet lens B are coaxially connected in sequence to form a reference channel (B channel).

[0084] The SO2 ultraviolet camera adopts a double-channel configuration, and when imaging the plume, the images of the A and B channels are obtained simultaneously. The inversion methods of the atmospheric extinction coefficient of the two channels are the same, and the A channel is taken as an example.

[0085] The SO2 ultraviolet camera is used to image two reference objects or terrain areas at different distances from the camera, and two pairs of A and B channel original images are obtained. When shooting, the field of view of the camera should be as far away from the sun as possible. For the A channel original image: in one of the original images containing a distant reference object, a rectangular pixel region is selected on the distant reference object, and the average light intensity of the rectangular pixel region is taken as the average light intensity of the distant reference point I Ori,f ; in one of the A channel original images containing a near reference object, a rectangular pixel region is selected on the near reference object, and the average light intensity of the rectangular pixel region is taken as the average light intensity of the near reference point I Ori,c .

[0086] The selection of the two rectangular pixel regions should follow the following three principles:

[0087] ① The rectangular pixel regions should be selected on terrain or objects that are exposed to the sun as much as possible.

[0088] ② If a terrain area is selected as a rectangular pixel region, the terrain should be selected as gently as possible and the same ground cover should be maintained as much as possible.

[0089] ③ The rectangular pixel regions should be avoided to be selected on terrain areas or objects covered by the plume shadow.

[0090] According to the ship real-time tracking website Shipfinder (http: / / www.shipfinder.com), the distances between the center points of the rectangular pixel regions selected in the two original images containing the distant and near reference objects in the A channel and the SO2 ultraviolet camera are denoted as D f and D c respectively. At the same time, in the background sky region of the A channel original image containing the distant reference object, a small pixel region is selected to take the average light intensity as the sky background intensity I OriAWThe above two groups of average reference point light intensity, corresponding distance information and sky background intensity are substituted into the following equation group, and the atmospheric extinction coefficient σ Α :

[0091]

[0092] I0 is the initial intensity of light passing through the plume or reflected by the solid (i.e. I f measured by the camera at D Ori,f = 0).

[0093] The above binary linear equation group is solved by the analytical expression method, and the atmospheric extinction coefficient σ Α of the A channel is obtained. B The method is the same as that of the A channel.

[0094] Step S20: The plume image and the sky background image are photographed by the SO2 ultraviolet camera, and the plume image is corrected for the light dilution effect by the atmospheric extinction coefficient, to obtain the corrected plume image.

[0095] Specifically, the influence of the light dilution effect can be quantified by the atmospheric extinction coefficient, and the plume image photographed by the SO2 ultraviolet camera is corrected for the light dilution effect in combination with the atmospheric extinction coefficient obtained in step S10.

[0096] The A and B channels are used to simultaneously image the measured plume source according to step S10, to obtain the plume images I A and I B of the two channels; and the sky background is imaged to obtain the sky background images I AW and I BW of the two channels.

[0097] The steps of the A and B channels for correcting the light dilution effect of the plume image are the same, and the A channel is taken as an example.

[0098] Due to the propagation of photons passing through the plume (or reflected) between the measured point and the camera, part of the photons will be scattered out of the camera field of view by aerosol particles in the air. Based on the atmospheric extinction coefficient σ Α and the A channel plume image I A , the corrected plume image I′ crr,A ,

[0099]

[0100] wherein I A (i,j) is the light intensity of each pixel point in the A channel plume image, d i,j is the distance between each pixel point in the plume image and the camera (only the pixel points containing the plume are considered here).

[0101] 2.22) Photons outside the camera's field of view in the background atmosphere will be partially scattered into the camera's field of view, based on the atmospheric extinction coefficient σ. Α and A channel background light intensity image I AW Calculate the background sky error I″ crr,A ,

[0102]

[0103] Optical dilution correction is performed on each pixel of the plume image in channel A, i.e., the plume image I′ is corrected for photon loss. crr,A Subtracting background sky error I″ crr,A Calculate the A-channel plume image I after optical dilution correction. crr,A ,

[0104] I crr,A (i,j)=I′ crr,A (i,j)-I″ crr,A (i,j)

[0105] Using the same method as for channel A, optical dilution correction is applied to the plume image of channel B to obtain the optically dilution-corrected plume image I of channel B. crr,B ;

[0106] Step S30: Calculate the SO2 optical thickness image corrected for optical dilution effect.

[0107] Specifically: The SO2 ultraviolet camera uses a dual-channel imaging method, and the A-channel plume image I obtained in step S20 after light dilution effect correction is... crr,A Compared with the A-channel sky background image I captured in step S20 AW By performing negative logarithmic processing, the optical thickness image τ of channel A is obtained. A .

[0108] τ A =-ln(I crr,A / I AW )

[0109] The B-channel plume image I obtained in step S20 after correction for light dilution effect. crr,B Compared with the B channel sky background image I captured in step S20 BW By performing negative logarithmic processing, the optical thickness image τ of channel B is obtained. B .

[0110] τ B =-ln(I crr,B / I BW )

[0111] The optical thickness of the A and B channels obtained by calculation is subtracted, that is, the SO2 optical thickness image after light dilution effect correction is obtained

[0112]

[0113] Step S40: obtaining a calibration curve by the standard bubble method and inverting the SO2 concentration image.

[0114] Specifically, the SO2 camera measures the SO2 optical thickness, and there is a certain conversion relationship between the SO2 optical thickness and the SO2 concentration, and the calibration curve gives such a conversion relationship. The calibration curve obtained by the standard bubble method does not consider the influence of the light dilution effect, so the image photographed by the SO2 camera is first corrected for the light dilution effect, and then the true concentration of SO2 in the plume can be inverted by the calibration curve of the standard bubble method. Since the standard bubble method for obtaining the calibration curve is not within the protection scope of the present patent, only a brief description is given for this part.

[0115] A standard bubble filled with a certain concentration of SO2 is placed in front of the two lenses of the camera, so as to cover the entire field of view of the camera. The standard bubble images of the A and B channels are photographed respectively. Then the standard bubble is removed, and the sky background images of the A and B channels are photographed respectively. The standard bubble images I cell and the sky background images I W are taken. According to the following formula, the optical thickness images τ cell of the respective channels are obtained by negative logarithmic processing.

[0116] τ cell = -ln(I cell / I W )

[0117] The optical thickness images τ cell,A and τ cell,B of the two channels obtained are subtracted, and the SO2 optical thickness under the SO2 concentration of the standard bubble can be obtained

[0118]

[0119] Different concentrations of SO2 standard bubbles are replaced and the above operation is repeated at least 3 times, and the standard bubble method calibration curve is fitted by the relationship between different SO2 concentrations and the corresponding SO2 optical thickness. The fitting results of the calibration curve are as follows:

[0120]

[0121] That is, the SO2 concentration image after light dilution effect correction, a is the parameter of the calibration curve obtained by the standard bubble method, The optical thickness image of SO2 corrected by the light dilution effect.

[0122] Case implementation and verification experiment results

[0123] In step S10, the two pairs of A, B channel original images taken by the SO2 ultraviolet camera are respectively as shown in Figure 2 , 3 , 4, 5. In this case, the average light intensity of the two selected reference points in the A channel is I Ori,c =8777; I Ori,f =10421, and the average light intensity of the two selected reference points in the B channel is I Ori,c =36197; I Ori,f =43284. Through the ship real-time positioning system, the distance information between the center points of the selected rectangular pixels and the camera is obtained, which is D c =3.01km; D f =4.00km, which is the same for the A and B channels. The sky background light intensity of the A and B channels is I OriAW =15000; I OriBW =55000, and the finally inverted atmospheric extinction coefficients of the A and B channels are σ Α =0.13318, σ B =0.20546.

[0124] In step S20, the A, B channel plume source images taken by the SO2 ultraviolet camera: the A, B channel exhaust images taken at a distance of 2.5km from the SO2 ultraviolet camera are respectively as shown in Figure 6 , 7 , in this case, Figure 6 , 7 , the distance of each pixel point of the ship plume in the above two images from the camera is an average distance d i,j =2.5km. The A, B channel exhaust images of the same ship taken at a distance of 1.8km from the SO2 ultraviolet camera are respectively as shown in Figure 11 , 12 , in this case, Figure 11 , 12 , the distance of each pixel point of the ship plume in the above two images from the camera is an average distance d i,j =1.8km.

[0125] In step S30, the plume images corrected by the light dilution effect and the respective sky background images are subjected to negative logarithmic processing to obtain the respective optical thickness images of the A and B channels, and then the optical thickness images of the A and B channels are subtracted to obtain the SO2 optical thickness image corrected by the light dilution effect as shown in Figure 8 , 13 .

[0126] In step S40, the coefficient a = 2494 of the standard bubble method calibration curve obtained by final fitting, the SO2 content inversion results of the ship's exhaust gas at a distance of 2.5 km from the SO2 ultraviolet camera, the SO2 concentration image after light dilution effect correction is as shown in Figure 9 , the SO2 concentration image without light dilution effect correction is as shown in Figure 10 ; The SO2 content inversion results of the ship's exhaust gas at a distance of 1.8 km from the SO2 ultraviolet camera, the SO2 concentration image after light dilution effect correction is as shown in Figure 14 , the SO2 concentration image without light dilution effect correction is as shown in Figure 15 .

[0127] Since the light dilution effect will make the measured concentration value lower than the actual value, there is a difference between the ship's exhaust concentration values calculated before and after the light dilution effect correction above Figure 9 , Figure 10 ; Figure 14 , Figure 15 ), which is reflected in the black and white images that the brightness of the ship's exhaust is different, and the higher the brightness, the greater the concentration; after correction, the brightness of the ship's plume is higher, that is, the concentration is greater, and the measurement is more accurate; that is, the SO2 concentration after light dilution effect correction is significantly higher than that before light dilution effect correction, and the image mainly shows that the brightness of the ship's exhaust after light dilution effect correction is higher than that before light dilution effect correction, which shows that the SO2 concentration image after light dilution effect correction is more accurate.

[0128] At the same time, in this example, the same ship is imaged when it travels to different distances from the camera, and the ship's exhaust concentration is inverted using the method of the present application. By comparing the results of the ship traveling to a distance of 1.8 km from the camera with the results of the ship traveling to a distance of 2.5 km from the camera Figure 16 , Figure 17 ), there is a large difference between the results before light dilution effect correction, with an average difference of about 24%; after light dilution effect correction, the results between the two have good consistency, with an average difference of only 0.014%. The above shows that the present application can effectively correct the influence of light dilution effect on the experimental results and has better accuracy.

[0129] The above specific embodiments have shown and described the embodiments of the present application, but those skilled in the art can understand that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirits of the present patent, and the scope of the present patent is defined by the appended claims and their equivalents.

Claims

1. A method for obtaining SO2 concentration images by correcting the optical dilution effect of camera images, characterized in that... Includes the following steps: step S10: Calibration of light dilution effect, deriving atmospheric extinction coefficient from raw images captured by SO2 ultraviolet camera; Step S20: Take images of the plume and the sky background using an SO2 ultraviolet camera, and correct the light dilution effect of the plume image using the atmospheric extinction coefficient to obtain the corrected plume image; Step S30: Calculate the SO2 optical thickness image using the plume image corrected in step S20; Step S40: Use the standard bubble method to obtain a calibration curve and convert the SO2 optical thickness image into an SO2 concentration image; Step S10 includes the following steps: 1.1) Using an SO2 ultraviolet camera, images were taken of two reference objects or terrain areas at different distances from the camera, resulting in two pairs of raw images in channels A and B. 1.2) Select a rectangular pixel region in the original image of channel A, and take the average light intensity of the rectangular pixel region to obtain the average light intensity value I of the reference point in the original image. Ori (λ); Select a pixel region in the background sky area of ​​the original image in channel A, take the average light intensity, and calculate the sky background intensity I. OriW (λ); 1.3) Obtain the distance D between the camera and the center point of the rectangular pixel region in the original image of channel A, along the line of sight of the SO2 ultraviolet camera, using a real-time positioning system. Ori ; 1.4) Since there is a linear relationship between the measured light intensity value I(λ) of the SO2 ultraviolet camera and the straight-line distance d between the measured point and the camera, the atmospheric radiative transfer equation is transformed, and the average light intensity value I of the reference point in the original image is used. Ori (λ), Sky background intensity I OriW (λ) and the distance D between the camera and the center point of the rectangular pixel region along the line of sight of the SO2 ultraviolet camera. Ori Find the atmospheric extinction coefficient σ of channel A. A , In the formula, I0(λ) is the initial intensity of the light after passing through the plume or being reflected by the solid, and λ is the wavelength; 1.5) Calculate the atmospheric extinction coefficient σ of channel B using the same method as channel A. B ; The detailed steps of step S20 are as follows: 2.1) Image the plume under test through channels A and B to obtain two-channel plume images I. A I B The sky background is imaged simultaneously using channels A and B to obtain a two-channel sky background image I. AW I BW ; 2.2) Perform optical dilution correction on the plume image of channel A. The specific steps are as follows: 2.21) During the propagation of photons passing through the plume or after reflection between the measurement point and the camera, some photons are scattered out of the camera's field of view by aerosol particles in the air. Based on the atmospheric extinction coefficient σ A and A-channel plume image I A Obtain the plume image I′ with photon loss corrected. crr,A , Among them, I A (i,j) represents the light intensity of each pixel in the A-channel plume image, d i,j This represents the distance between each pixel in the plume image and the camera. 2.22) Photons outside the camera's field of view in the background atmosphere will be partially scattered into the camera's field of view, based on the atmospheric extinction coefficient σ. A and A channel background light intensity image I AW Calculate the background sky error I″ crr,A , Among them, I AW (i,j) represents the light intensity of each pixel in the background light intensity image of channel A; 2.23) The plume image I′ corrected for photon loss crr,A Subtracting background sky error I″ crr,A That is, the A-channel plume image I after light dilution effect correction. crr,A , I crr,A (i,j)=I′ crr,A (i,j)-I″ crr,A (i,j) 2.3) Using the same method as step 2.2), perform optical dilution correction on the B-channel plume image to obtain the optically dilution-corrected B-channel plume image I. crr,B ; Step S30 includes the following steps: 3.1) Define the optical thickness τ of channel A. A , The sum of the optical thicknesses of SO2 and aerosol particles, corrected for optical dilution effect, is shown in the A-channel plume image I. crr,A And the sky background image of channel A AW In other words, the optical thickness τ of channel A is... A As shown below, τ A =-ln(I crr,A / I AW ) 3.2) Define the optical thickness τ of channel B. B , Channel B is affected by aerosol particles; therefore, the optical thickness of the aerosol particles is determined by the plume image I after optical dilution correction via channel B. crr,B Sky background image with B channel I BW This indicates that the optical thickness τ of channel B is... B As shown below, τ B =-ln(I crr,B / I BW ) 3.3) The optical thickness τ of channel A and channel B A τ B Difference processing is performed to eliminate the influence of aerosol particles on the results, thus obtaining the SO2 optical thickness image corrected for photodilution effect. Step S40 includes the following steps: 4.1) Take standard bubble images (A and B channels) using the SO2 standard bubble. cell,A I cell,B Capture the sky background images for channels A and B respectively. cell,AW I cell,BW ; 4.2) Standard bubble image I cell,A I cell,B Corresponding sky background image I cell,AW I cell,BW Negative logarithmic processing is performed to obtain the optical thickness images τ of channels A and B. cell,A τ cell,B , τ cell,A =-ln(I cell,A / I cell,AW ) τ cell,B =-ln(I cell,B / I cell,BW ) 4.3) The optical thickness images τ of channels A and B cell,A τ cell,B Calculate the optical thickness of SO2 at the standard bubble SO2 concentration by taking the difference. 4.4) Replace the SO2 standard bubbles with different concentrations and repeat steps 4.1)–4.3) to obtain the SO2 concentration and SO2 optical thickness. The relationship was used to fit the standard bubble calibration curve, as shown below. y = ax In the formula, 'a' is the calibration coefficient of the calibration curve obtained by fitting the standard bubble method, 'x' is the optical thickness of SO2, and 'y' is the SO2 concentration. 4.5) SO2 optical thickness image corrected for optical dilution effect in step S30 Substitute the values ​​into the calibration curve to obtain the SO2 concentration image after photodilution correction.

Images