Method, System, Storage Medium and Device for Removing Image Meteorological Noise Information

Abstract

The present invention provides a method, a system, a storage medium and a device for removing image meteorological noise information, including the following steps: dividing an original image with meteorological noise information into a foreground image and a background image; respectively preprocessing the foreground image and the background image; constructing a mixed dark channel image according to the preprocessed foreground image and background image; estimating the atmospheric light intensity value by using the background image; solving the light transmittance according to the estimated atmospheric light intensity value; and calculating the image after removing the meteorological noise information from the original image by using a dark channel prior model and an atmospheric scattering model according to the mixed dark channel image, the atmospheric light intensity value and the light transmittance. The present invention effectively restores the image quality.

Description

Technical Field

[0001] The present invention belongs to the technical field of image processing, and particularly relates to a method, a system, a storage medium and a device for removing image meteorological noise information. Background Technique

[0002] For images taken in weather such as light rain and fog, meteorological noise information such as rain and fog in the images causes the images to be blurred, affecting the recognition of the image content and subsequent image recognition and extraction processes. Existing technologies often use the dark channel prior method to restore images. Among them, the atmospheric light intensity value used in the dark channel prior model is an estimated value. One day is divided into several time periods, and each time period corresponds to an atmospheric light intensity value. The atmospheric light intensity value used for image restoration is the atmospheric light intensity value corresponding to the time period when the image was taken. Using the estimated value as a model parameter results in low model accuracy, large computational complexity, unsatisfactory operating efficiency, and low image restoration degree. Summary of the Invention

[0003] The purpose of the present invention is to solve the deficiencies existing in the above background technique and provide a method for removing image meteorological noise information to effectively restore the image quality.

[0004] The technical solution adopted by the present invention is: a method for removing image meteorological noise information, comprising the following steps:

[0005] Dividing the original image with meteorological noise information into a foreground image and a background image;

[0006] Preprocessing the foreground image and the background image respectively;

[0007] Constructing a hybrid dark channel image according to the preprocessed foreground image and background image;

[0008] Estimating the atmospheric light intensity value using the un-preprocessed background image;

[0009] Solving the light transmittance according to the estimated atmospheric light intensity value;

[0010] Using the dark channel prior model and the atmospheric scattering model, calculating the image after removing the meteorological noise information from the original image according to the hybrid dark channel image, the atmospheric light intensity value, and the light transmittance.

[0011] In the above technical solution, the specific steps for dividing the original image with meteorological noise information into a foreground image and a background image are as follows: First, convert the original image with meteorological noise information into a grayscale image, and calculate the grayscale value of each pixel among all pixels in the grayscale image and the root mean square value of the grayscale values of all pixels; then, use the pixels with grayscale values less than the root mean square value among all pixels of the grayscale image as the pixels of the foreground image, and use the pixels with grayscale values greater than the root mean square value among all pixels of the grayscale image as the pixels of the background image.

[0012] In the above technical solution, the preprocessing of the foreground image includes the following steps: Convert each pixel of the foreground image from the RGB image space to the HSL space, and obtain the brightness value of each pixel in the HSL space; compare the brightness values of each pixel and obtain the minimum brightness value L min and set the brightness values of all pixels in the HSL space to the minimum brightness value L min to obtain a single-channel image as the preprocessed foreground image.

[0013] In the above technical solution, the preprocessing of the background image includes the following steps: Filter the background image after grayscale conversion to obtain the preprocessed background image.

[0014] In the above technical solution, the hybrid dark channel image I COME is calculated using the following formula:

[0015] I COMB = α1I1 + α2I2;

[0016] where α1 and α2 are adjustment coefficients, α1 + α2 = 1, and the value ranges of α1 and α2 are (0, 1); I1 is the preprocessed foreground image; I2 is the preprocessed background image.

[0017] In the above technical solution, using the dark channel prior model and the atmospheric scattering model, the image after removing the meteorological noise information is expressed as:

[0018]

[0019] where, A * is the atmospheric light intensity value, max[] is the maximum value function; t(x) is the light transmittance, t0 is the minimum value of the light transmittance t(x); I COME (x) is the x function corresponding to the hybrid dark channel image, and x is the pixel coordinate in the original image.

[0020] In the above technical solution, estimating the atmospheric light intensity value using the unpreprocessed background image specifically includes the following steps: obtaining the color channel brightness values of each pixel of the background image in the RGB image space, removing the brightest and darkest pixels, and then performing weighted processing on the color channel brightness values of the remaining pixels to obtain the brightness mean value of the background image for estimating the atmospheric light intensity value A * , where:

[0021] A * = [(0.3A R + 0.587A G + 0.113A B ) / N] + A0;

[0022] Where A R , A G and A B are the brightness values corresponding to the R, G, and B channels of the remaining pixels after removing the brightest and darkest partial pixels of the background image; N is 90% of the number of pixels of the background image; A0 is an adjustment term used to optimize the estimated atmospheric light intensity value A * .

[0023] The present invention provides an image meteorological noise information removal system, including an image separation module, a foreground image preprocessing module, a background image preprocessing module, an atmospheric light intensity value estimation module, a light transmittance generation module, a mixed dark channel image generation module, and a meteorological noise information removal module;

[0024] Among them, the image separation module divides the original image with meteorological noise information into a foreground image and a background image, and sends the generated foreground image to the foreground image preprocessing module, and sends the generated background image to the background image preprocessing module and the atmospheric light intensity value estimation module;

[0025] The foreground image preprocessing module preprocesses the received foreground image and sends the preprocessed foreground image to the mixed dark channel image generation module;

[0026] The background image preprocessing module preprocesses the received background image and sends the preprocessed background image to the mixed dark channel image generation module;

[0027] The mixed dark channel image generation module is used to construct a mixed dark channel image based on the preprocessed foreground image and background image, and send the generated mixed dark channel image to the meteorological noise information removal module;

[0028] The atmospheric light intensity value estimation module is used to estimate the atmospheric light intensity value using the unpreprocessed background image, and send the estimation result to the light transmittance generation module;

[0029] The light transmittance generation module is used to solve the light transmittance according to the atmospheric light intensity value and send the calculation result to the meteorological noise information removal module;

[0030] The meteorological noise information removal module adopts the dark channel prior model and calculates the image after removing the meteorological noise information from the original image according to the mixed dark channel image, the atmospheric light intensity value and the light transmittance.

[0031] In the above technical solution, it is characterized in that an image meteorological noise information removal method program is stored on the computer-readable storage medium, and when the image meteorological noise information removal method program is executed by a processor, the steps of the image meteorological noise information removal method described in the above technical solution are realized.

[0032] In the above technical solution, the computer device includes: a memory, a processor, and an image meteorological noise information removal method program stored on the memory and operable on the processor, and the image meteorological noise information removal method program is configured to realize the steps of the image meteorological noise information removal method described in the above technical solution.

[0033] The beneficial effects of the present invention are:

[0034] (1) By dividing the original image into a foreground image and a background image, the present invention constructs a mixed dark channel image from the preprocessed foreground image and background image. After preprocessing, the foreground image and background image can reduce the overall impact of atmospheric parameters such as the estimated atmospheric light intensity and light transmittance on the global original image, improve the effect of removing meteorological noise information, and reduce the time complexity and computational amount of noise reduction;

[0035] (2) According to the image obtained by converting the original image to grayscale, the global root mean square value of grayscale is obtained, and the foreground and background images are divided according to the size of the grayscale value and the root mean square value, which is convenient for further differential processing;

[0036] (3) The image space of the foreground image is transformed to obtain an optimized foreground image with the brightness information of each pixel; the pixels of the background image are window-filtered to obtain an optimized background image; the mixed dark channel image constructed according to the optimized foreground image and background image is closer to the real image and can further improve the effect of image processing;

[0037] (4) The present invention estimates the atmospheric light intensity value based on the obtained background image. Compared with the model parameter setting method of obtaining parameter empirical values through the shooting time in the prior art, the parameter setting method of the present invention is more accurate; the light transmittance is further obtained through the atmospheric light intensity value estimated by the present invention, which is more accurate than the model parameter estimation and given value in the prior art. Description of the Drawings

[0038] Figure 1 This is a schematic flowchart of the present invention. Detailed implementation manners

[0039] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments for a clear understanding of the present invention, but they do not limit the present invention.

[0040] As Figure 1 shown, the present invention provides a method for removing image meteorological noise information, including the following steps:

[0041] S1: Divide the original image I(x) with meteorological noise information into a foreground image and a background image;

[0042] Specifically, first, convert the original image I(x) with meteorological noise information into a grayscale image, and calculate the grayscale value of each pixel among all pixels of the grayscale image and the root mean square value of the grayscale values of all pixels; then, among all pixels of the grayscale image, pixels with grayscale values less than the root mean square value are used as pixels of the foreground image, that is, this type of pixels constitutes the foreground image; pixels with grayscale values greater than the root mean square value are used as pixels of the background image, that is, this type of pixels constitutes the background image; pixels with grayscale values equal to the root mean square value are used as pixels of the common boundary between the background image and the foreground image, and this type of pixels constitutes the boundary between the background image and the foreground image.

[0043] Among them, the imaging model of the original image I(x) based on atmospheric scattering is:

[0044] I(x) = J(x)t(x) + A[1 - t(x)];

[0045] Among them, I(x) is the original image with meteorological noise information acquired by the imaging device, and the original image is based on the RGB image space; J(x) is the image after removing meteorological noise information, t(x) is the light transmittance, and the value range of t(x) is (0, 1); A is the atmospheric light intensity value, and x is the pixel coordinate in the original image.

[0046] S2: Preprocess the foreground image and the background image respectively to obtain the preprocessed foreground image I1 and the preprocessed background image I2;

[0047] (1) The preprocessing of the foreground image is specifically as follows: Convert each pixel of the foreground image from the RGB image space to the HSL space, obtain the brightness value of each pixel in the HSL space, compare the brightness values of each pixel, and obtain the minimum brightness value L min ; Set the brightness value of all pixels in the HSL space to L min , to obtain a single-channel image, which is used as the preprocessed foreground image I1;

[0048] Further, for the processing of the foreground image, let the value of a certain pixel x1 (x1 ∈ x) in the foreground image in the RGB image space be {r, g, b}, where r, g, and b respectively correspond to the red, green, and blue intensity values of the pixel in the RGB image space. The value ranges of r, g, and b are [0, 1], which are obtained by normalizing the pixel value range [0, 255]. Define max{r, g, b} as the maximum value among the intensity values of the three color channels (red, green, and blue), and min{r, g, b} as the minimum value among the intensity values of the three color channels (red, green, and blue); the value of the pixel x1 in the HSL space corresponds to {H, S, L}, where H, S, and L respectively correspond to the hue, saturation, and lightness in the HSL space; then:

[0049] H ∈ [0, 360°]; L, S ∈ [0, 1];

[0050] L = 0.5[max{r, g, b} + min{r, g, b}];

[0051]

[0052] Calculate and compare the lightness values L of each pixel in the foreground image to obtain the minimum lightness value L min , set the lightness values of all pixels in the HSL space to L min , to obtain a single-channel image I1, that is, the preprocessed foreground image.

[0053] (2) The preprocessing of the background image is specifically as follows: After graying the obtained background image, filter it with a 3×3 pixel window to obtain the preprocessed background image I2;

[0054] Further, after graying the background image, divide the background image into several non-overlapping nine-square grids with a 3×3 pixel window; the gray value of the pixel at the center of the nine-square grid is used as a reference value, and the gray values of the remaining 8 pixels are compared with the reference value. If the gray values of these 8 pixels are greater than or equal to the gray value of the pixel at the center of the nine-square grid, set them to 1; if the gray values are less than the gray value of the pixel at the center of the nine-square grid, set them to 0. Denoise the background image through the above method to obtain the background image I2 with brightness noise filtered out.

[0055] Generally, an odd-class 3*3 window is used, or a 2*2 window can be selected according to needs for more accurate calculation; the longer the window radius, the more dark pixels the window covers, the lower the obtained dark pixel value, and the coarser the texture of the generated dark channel map of the image.

[0056] S3: Construct a mixed dark channel image according to the preprocessed foreground image I1 and background image I2;

[0057] Construct a mixed dark channel image based on the minimum value of the brightness values of each pixel in the foreground image and the gray values of each pixel in the background image;

[0058] The mixed dark channel image I COMB is calculated using the following formula:

[0059] I COMB =α1I1 + α2I2;

[0060] where, I COMB is the mixed dark channel image, α1 and α2 are adjustment coefficients, and α1 + α2 = 1; the values of α1 and α2 range from [0, 1] and are set manually;

[0061] The weather and air quality determine the trend of the values of α1 and α2. When α1 > α2, mainly perform meteorological noise information removal processing on the foreground image; when α1 < α2, mainly perform meteorological noise information removal processing on the background image; different settings are used for α1 and α2 of different images.

[0062] S4: Estimate the atmospheric light intensity value A using the background image not preprocessed in step S1 * , and solve the light transmittance t(x) according to the estimated atmospheric light intensity value A * ;

[0063] Specifically, obtain the color channel brightness values in the RGB image space of each pixel in the background image, perform weighted processing after removing the 5% of the pixels with the brightest brightness values and the 5% of the pixels with the darkest brightness values, obtain the mean value of the brightness values of the background image, and further obtain the estimated atmospheric light intensity value A * :

[0064] A * =[(0.3A R +0.587A G +0.113A B ) / N]+A0;

[0065] where A R , A G and A B are the brightness values corresponding to the R, G, and B channels of the remaining pixels after removing the brightest and darkest 5% of the background image, and the brightness values are obtained by normalizing the pixel value range [0, 255] of the remaining pixels; N is the number of 90% pixels of the background image; A0 is an adjustment term. Among them, 5% can be adjusted to other values as needed.

[0066] When the original image I(x) has a sky area, A0 = 0.

[0067] When there is no sky area in the original image I(x), after converting the original image I(x) into a grayscale image, the brightest 3% of the pixels and the darkest 3% of the pixels are removed. Among the remaining pixels, the 0.1% of the pixels with the largest grayscale values are traversed. After averaging the grayscale values of the above remaining pixels, the values corresponding to the range of 0-255 are mapped to the range of 0-1 as the adjustment term A0. Among them, the above value of 3% can be adjusted as needed; 0.1% is specified according to conventional means for reference.

[0068] S5: Solve the light transmittance according to the estimated atmospheric light intensity value.

[0069] For any input image J, there is always a color channel value that is relatively low. The dark channel is obtained through step S3, that is, the dark channel prior model is expressed as:

[0070]

[0071] where J dark represents the dark channel, J c (y) represents any color channel of J, Ω(x) represents a window centered on the pixel point x, y represents the pixel points within the window, c represents the color channel, and x is the pixel coordinate in the original image.

[0072] Then, the grayscale of the dark channel image is relatively low and satisfies:

[0073] J dark →0 (2)

[0074] According to the atmospheric scattering model I(x) = J(x)t(x) + A[1 - t(x)] (3), through deformation processing, we get Taking the minimum value twice, then:

[0075]

[0076] It can be deduced that:

[0077]

[0078] S6: According to the dark channel prior model and the atmospheric scattering model, obtain the image after removing meteorological noise information, which is expressed as J(x), that is, obtained according to formulas (1) to (4):

[0079]

[0080] In the formula, max[] is the maximum value function. When the light transmittance t value is too small, it will cause the J value to be too large and the overall image to transition to the white field. Therefore, a threshold t0 is set. When the t value is less than t0, take t = t0, and generally t0 = 0.1 can be taken.

[0081] In expression (3), I(x) takes the mixed dark channel image I of step S3 COMB , A takes the atmospheric light intensity value A obtained in step S4 * , and t(x) is the light transmittance t(x) obtained in step S5. Substituting into expression (5), we get:

[0082]

[0083] Perform calculations and output the image J(x) with meteorological noise information removed.

[0084] The present invention provides an image meteorological noise information removal system, including an image separation module, a foreground image preprocessing module, a background image preprocessing module, an atmospheric light intensity value estimation module, a light transmittance generation module, a mixed dark channel image generation module, and a meteorological noise information removal module.

[0085] Among them, the image separation module divides the original image with meteorological noise information into a foreground image and a background image, and sends the generated foreground image to the foreground image preprocessing module, and sends the generated background image to the background image preprocessing module and the atmospheric light intensity value estimation module.

[0086] The image separation module converts the original image with meteorological noise information into a grayscale image, and obtains the grayscale value of each pixel among all pixels in the grayscale image and the root mean square value of the grayscale values of all pixels; then, the pixels in the grayscale image with grayscale values less than the root mean square value are used as the pixels of the foreground image, and the pixels in the grayscale image with grayscale values greater than the root mean square value are used as the pixels of the background image.

[0087] The foreground image preprocessing module preprocesses the received foreground image and sends the preprocessed foreground image to the mixed dark channel image generation module.

[0088] The foreground image preprocessing module converts each pixel of the foreground image from the RGB image space to the HSL space, and obtains the brightness value of each pixel in the HSL space; compares the brightness values of each pixel and obtains the minimum brightness value L min , and sets the brightness values of all pixels in the HSL space to the minimum brightness value L min , and obtains a single-channel image as the preprocessed foreground image.

[0089] The background image preprocessing module preprocesses the received background image and sends the preprocessed background image to the mixed dark channel image generation module.

[0090] The background image preprocessing module filters the grayscale background image to obtain the preprocessed background image.

[0091] The hybrid dark channel image generation module is used to construct a hybrid dark channel image based on the preprocessed foreground image and background image, and send the generated hybrid dark channel image to the meteorological noise information removal module.

[0092] The hybrid dark channel image generation module calculates the hybrid dark channel image I using the following formula COME :

[0093] I COMB = α1I1 + α2I2;

[0094] Where α1 and α2 are adjustment coefficients, α1 + α2 = 1, and the value ranges of α1 and α2 are (0, 1); I1 is the preprocessed foreground image; I2 is the preprocessed background image.

[0095] The atmospheric light intensity value estimation module is used to estimate the atmospheric light intensity value using the un-preprocessed background image, and send the estimation result to the light transmittance generation module.

[0096] The light transmittance generation module is used to solve the light transmittance based on the atmospheric light intensity value, and send the calculation result to the meteorological noise information removal module;

[0097] The meteorological noise information removal module uses the dark channel prior model to calculate the image after removing the meteorological noise information from the original image based on the hybrid dark channel image, the atmospheric light intensity value, and the light transmittance.

[0098] The meteorological noise information removal module uses the dark channel prior model and the atmospheric scattering model to represent the image after removing the meteorological noise information as:

[0099]

[0100] Where, A * is the atmospheric light intensity value, max[] is the maximum value function; t(x) is the light transmittance, t0 is the minimum value of the light transmittance t(x); I COME (x) is the x function corresponding to the hybrid dark channel image, and x is the pixel coordinate in the original image.

[0101] The present invention provides a computer-readable storage medium, on which an image meteorological noise information removal method program is stored. When the image meteorological noise information removal method program is executed by a processor, the steps of the image meteorological noise information removal method described in the above technical solution are implemented.

[0102] The present invention provides a computer device, which includes: a memory, a processor, and an image meteorological noise information removal method program stored on the memory and executable on the processor. The image meteorological noise information removal method program is configured to implement the steps of the image meteorological noise information removal method described in the above technical solution.

[0103] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or multiple flows and / or blocks.

[0105] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or multiple flows and / or blocks.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the flows Figure 1 or blocks or multiple flows and / or blocks.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the scope of its protection. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that after reading the present invention, various changes, modifications or equivalent substitutions can still be made to the specific implementation manners of the invention, but these changes, modifications or equivalent substitutions are all within the scope of protection of the claims pending for the invention.

Claims

1. A method for removing image meteorological noise information, characterized in that: It includes the following steps: Divide the original image with meteorological noise information into a foreground image and a background image; Preprocess the foreground image and the background image respectively; Construct a hybrid dark channel image based on the preprocessed foreground image and background image; Estimate the atmospheric light intensity value using the un-preprocessed background image; Solve for the light transmittance based on the estimated atmospheric light intensity value; Using the dark channel prior model and the atmospheric scattering model, calculate the image after removing the meteorological noise information from the original image based on the hybrid dark channel image, the atmospheric light intensity value, and the light transmittance; Dividing the original image with meteorological noise information into a foreground image and a background image specifically includes the following steps: First, convert the original image with meteorological noise information into a grayscale image, and calculate the grayscale value of each pixel among all pixels in the grayscale image and the root mean square value of the grayscale values of all pixels; Then, use the pixels in the grayscale image with grayscale values less than the root mean square value as the pixels of the foreground image, and use the pixels in the grayscale image with grayscale values greater than the root mean square value as the pixels of the background image; The preprocessing of the foreground image includes the following steps: Convert each pixel of the foreground image from the RGB image space to the HSL space, and obtain the brightness value of each pixel in the HSL space; Compare the brightness values of each pixel and obtain the minimum brightness value L min , and set the brightness value of all pixels in the HSL space to the minimum brightness value L min , and obtain a single-channel image as the preprocessed foreground image; The preprocessing of the background image includes the following steps: Filter the grayscale background image to obtain the preprocessed background image.

2. The method for removing image meteorological noise information according to claim 1, wherein: Mixed dark channel image I COMB It is calculated by the following formula: I COMB = α1I1 + α2I2; Where α1 and α2 are adjustment coefficients, α1 + α2 = 1, and the value ranges of α1 and α2 are (0, 1); I1 is the preprocessed foreground image; I2 is the preprocessed background image.

3. The method for removing image meteorological noise information according to claim 1, characterized in that: Using the dark channel prior model and the atmospheric scattering model, the image J(x) after removing the meteorological noise information is expressed as: Among them, A * is the atmospheric light intensity value, max[] is the maximum value function; t(x) is the light transmittance, and t0 is the minimum value of the light transmittance t(x); I COMB (x) is the x function corresponding to the mixed dark channel image, and x is the pixel coordinate in the original image.

4. The method for removing image meteorological noise information according to claim 1, characterized in that: Estimating the atmospheric light intensity value using an unpreprocessed background image specifically includes the following steps: obtaining the color channel brightness values of each pixel in the RGB image space of the background image, excluding the brightest and darkest pixels, and then performing weighted processing on the color channel brightness values of the remaining pixels to obtain the brightness mean value of the background image for estimating the atmospheric light intensity value A * , where: A * = [(0.3A R + 0.587A G + 0.113A B ) / N] + A0; Where A R , A G and A B are the brightness values corresponding to the R, G, and B channels of the remaining pixels after removing the brightest and darkest partial pixels from the background image; N is 90% of the number of pixels in the background image; A0 is an adjustment term used to optimize the estimated atmospheric light intensity value A * .

5. An image meteorological noise information removal system, characterized in that: For implementing the steps of the method for removing meteorological noise information from an image as described in any one of claims 1 to 4; It includes an image separation module, a foreground image preprocessing module, a background image preprocessing module, an atmospheric light intensity value estimation module, a light transmittance generation module, a hybrid dark channel image generation module, and a meteorological noise information removal module; Among them, the image separation module divides the original image with meteorological noise information into a foreground image and a background image, and sends the generated foreground image to the foreground image preprocessing module, and sends the generated background image to the background image preprocessing module and the atmospheric light intensity value estimation module; The foreground image preprocessing module preprocesses the received foreground image and sends the preprocessed foreground image to the hybrid dark channel image generation module; The background image preprocessing module preprocesses the received background image and sends the preprocessed background image to the hybrid dark channel image generation module; The hybrid dark channel image generation module is used to construct a hybrid dark channel image based on the preprocessed foreground image and background image, and send the generated hybrid dark channel image to the meteorological noise information removal module; The atmospheric light intensity value estimation module is used to estimate the atmospheric light intensity value using the un-preprocessed background image, and send the estimation result to the light transmittance generation module; The light transmittance generation module is used to solve for the light transmittance based on the atmospheric light intensity value, and send the calculation result to the meteorological noise information removal module; The meteorological noise information removal module uses the dark channel prior model to calculate the image after removing the meteorological noise information from the original image based on the hybrid dark channel image, the atmospheric light intensity value, and the light transmittance.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for the method for removing meteorological noise information from an image, and when the program for the method for removing meteorological noise information from an image is executed by a processor, it implements the steps of the method for removing meteorological noise information from an image as described in any one of claims 1 to 4.

7. A computer device, characterized in that, The computer device includes: a memory, a processor, and a program for the method for removing meteorological noise information from an image stored on the memory and executable on the processor, and the program for the method for removing meteorological noise information from an image is configured to implement the steps of the method for removing meteorological noise information from an image as described in any one of claims 1 to 4.

Images