Image Dehazing Method for Eliminating Halo Effect

Abstract

This application relates to the field of image processing technology. Specifically, it relates to an image defogging method for eliminating the halo effect, which can solve the problem of halos in the image after defogging to a certain extent. The image defogging method includes the following steps: obtaining an input image; determining the atmospheric light value corresponding to the input image, where the atmospheric light value is a matrix with the same size as the input image; determining the image transmittance map corresponding to the input image by fusing the dark channel transmittance map and the bright channel transmittance map according to a preset ratio; substituting the atmospheric light value and the image transmittance map into the atmospheric scattering model to obtain an output image. This application can eliminate the halo effect of images taken in a haze environment, and eliminate halos through the method of obtaining the atmospheric light value and the method of determining the image transmittance map, improve the adaptability of the image, and facilitate subsequent processing.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and more specifically, to an image dehazing method for eliminating the halo effect. Background Art

[0002] In a hazy environment, the machine vision system will absorb or scatter light due to suspended particles in the air, resulting in visual interference in the captured image and reduced contrast.

[0003] Currently, the image dehazing method mainly establishes a model based on the Dark Channel Prior (DCP) to achieve image dehazing through the DCP model. However, after processing an image containing sky information and the like through the DCP model, halos will be generated in the light and dark boundary regions, resulting in the dehazed image being affected by halos and affecting subsequent processing. Summary of the Invention

[0004] To solve the problem of halos existing in the light and dark boundary regions after image dehazing, this application provides an image dehazing method for eliminating the halo effect.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] This application provides an image dehazing method for eliminating the halo effect, including the following steps:

[0007] Obtain the input image;

[0008] Determine the atmospheric light value corresponding to the input image, where the atmospheric light value is a matrix equal in size to the input image;

[0009] Determine the image transmittance map corresponding to the input image, where the image transmittance map is determined by fusing the dark channel transmittance map and the bright channel transmittance map;

[0010] Substitute the atmospheric light value and the image transmittance map into the atmospheric scattering model to obtain the output image.

[0011] In an implementable manner, the determination of the atmospheric light value includes:

[0012] Obtain a guidance image according to the input image;

[0013] Obtain the first corresponding image of the input image, where the first corresponding image is obtained by converting the input image from the RGB color space to the YUV color space

[0014] Determine the atmospheric light value according to the guidance image and the first corresponding image, where the first corresponding image is obtained by converting the input image from the RGB color space to the YUV color space.

[0015] In one realizable manner, the determination of the guiding image includes:

[0016] In the daytime environment, the guiding image is the minimum grayscale image of the input image;

[0017] In the nighttime environment, the guiding image is the maximum grayscale image of the input image minus the minimum grayscale image of the input image.

[0018] In one realizable manner, the determination of the atmospheric light value according to the guiding image and the first corresponding image includes:

[0019] Perform guided filtering on the Y channel of the first corresponding image, where the guiding image is used as the guiding map;

[0020] Perform Gaussian filtering on the U channel and V channel of the first corresponding image;

[0021] Determine the atmospheric light value through the Y channel map after guided filtering, the U channel map after Gaussian filtering, and the V channel map after Gaussian filtering.

[0022] In one realizable manner, the determination of the atmospheric light value according to the guiding image and the first corresponding image includes:

[0023] Perform guided filtering on the Y channel, U channel, and V channel of the first corresponding image to determine the atmospheric light value.

[0024] In one realizable manner, the determination of the image transmittance map includes:

[0025] Determine the dark channel transmittance map;

[0026] Determine the bright channel transmittance map, where the bright channel transmittance map is determined according to the input image and the atmospheric light value;

[0027] Fuse the dark channel transmittance map and the bright channel transmittance map according to a preset ratio to determine the image transmittance map.

[0028] In one realizable manner, the determination of the dark channel transmittance map includes:

[0029] Obtain the second corresponding image of the input image, where the second corresponding image is obtained by converting the input image from the RGB color space to the HSV color space;

[0030] According to the S channel and V channel of the second corresponding image, determine the dark channel transmittance map, where the dark channel transmittance map is calculated according to the following formula:

[0031] t1(x, y) = k1 * S(x, y) + k2 * V(x, y) + k3

[0032] Wherein, t1 is the dark channel transmittance map; k1, k2, and k3 are weight coefficients; S is the S-channel map of the second corresponding image; V is the V-channel map of the second corresponding image; (x, y) are the row and column values of the corresponding pixel points of the dark channel transmittance map, the S-channel map of the second corresponding image, or the V-channel map of the second corresponding image.

[0033] In one implementable manner, the determination of the bright channel transmittance map includes:

[0034] Determine the to-be-filtered bright channel transmittance map according to the input image and the atmospheric light value, and the to-be-filtered bright channel transmittance map is calculated according to the following formula:

[0035]

[0036] Wherein, t′2 is the to-be-filtered bright channel transmittance map; A c is the atmospheric light value; c is the channel of the image, including three channels of R, G, and B; (x, y) are the row and column values of the corresponding pixel points of the to-be-filtered bright channel transmittance map, the input image, or the atmospheric light value; max is the maximum gray value of the pixel point (x, y) in the three channels of R, G, and B;

[0037] Perform maximum filtering on the to-be-filtered bright channel transmittance map to determine the bright channel transmittance map, and the bright channel transmittance map is calculated according to the following formula:

[0038]

[0039] Wherein, t′2 is the to-be-filtered bright channel transmittance map; t2 is the bright channel transmittance map; (x, y) are the row and column values of the corresponding pixel points of the to-be-filtered bright channel transmittance map or the bright channel transmittance map; maxfilter is the maximum gray value in the [m, n] neighborhood centered on the pixel point (x, y).

[0040] In one implementable manner, the determination of the image transmittance map includes:

[0041] Fuse the dark channel transmittance map and the bright channel transmittance map according to a preset ratio to determine the to-be-filtered image transmittance map;

[0042] When the pixel points in the to-be-filtered image transmittance map are greater than 0.2, set the corresponding pixel points to 0.2;

[0043] When all the pixel points in the to-be-filtered image transmittance map are less than or equal to 0.2, determine the image transmittance map.

[0044] In one implementable manner, the output image is calculated according to the following formula:

[0045]

[0046] In the formula, is the output image; I c is the input image; A c is the atmospheric light value; t is the image transmittance map; c is the channel of the image, including three channels of R, G, and B; (x, y) is the row and column value of the corresponding pixel point of the input image, atmospheric light value, image transmittance map, or output image.

[0047] Beneficial effects of this application: Determine the atmospheric light value corresponding to the input image, where the atmospheric light value is a matrix with the same size as the input image; further fuse the dark channel transmittance map and the bright channel transmittance map according to a preset ratio to determine the image transmittance map corresponding to the input image; further substitute the atmospheric light value and the image transmittance map into the atmospheric scattering model to obtain the output image; it can eliminate the halo effect of the image taken in the haze environment, and eliminate the halo through the acquisition method of the atmospheric light value and the determination method of the image transmittance map, improve the adaptability of the image, and facilitate subsequent processing. Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0049] Figure 1 Shows the flowchart of the image dehazing method for eliminating the halo effect in the embodiments of the present application;

[0050] Figure 2 Shows the flowchart of determining the atmospheric light value in the embodiments of the present application;

[0051] Figure 3 Shows the flowchart of determining the dark channel transmittance map in the embodiments of the present application;

[0052] Figure 4 Shows the flowchart of determining the bright channel transmittance map in the embodiments of the present application;

[0053] Figure 5 Shows the flowchart of determining the image transmittance map in the embodiments of the present application. Detailed Embodiments

[0054] To enable those skilled in the art to better understand the solution of this application, the technical solution in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

[0055] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0056] In the machine vision system, suspended particles in the air will absorb light or scatter light, resulting in attenuation of the light irradiating the object to be measured, and further leading to a decrease in the contrast of the collected image and an increase in the recognition difficulty. Dehazing the image containing haze can improve the target recognition in the scene and facilitate scene monitoring, remote sensing monitoring, etc.

[0057] Currently, dehazing of images is achieved by establishing a model based on the dark channel prior, which will produce halos in the light and dark boundary regions for images containing sky information, etc., weakening the adaptability of the image during scene transformation, resulting in the dehazed image being affected by halos in terms of visual perception or subsequent processing.

[0058] To effectively remove the halo effect of images containing haze, the embodiments of this application provide an image dehazing method for eliminating the halo effect, as Figure 1 shown, the method includes the following steps:

[0059] S101. Obtain an input image.

[0060] The input image is an image taken in an environment such as haze. The input image is obtained by a camera. It can be a certain video frame in a captured video or a single captured image.

[0061] Among them, the input image has a three-channel RGB domain, and the input image is a color image.

[0062] S102. Determine the atmospheric light value corresponding to the input image, where the atmospheric light value is a matrix with the same size as the input image.

[0063] Each point in each channel of the input image corresponds to an element in the atmospheric light value. As Figure 2 shown, the process of determining the atmospheric light value includes the following steps:

[0064] S201. Obtain a guidance image based on the input image.

[0065] Among them, the obtaining of the guidance image is determined according to different scenarios. The specific scenarios are as follows:

[0066] In the daytime environment, the guidance image is the minimum grayscale image of the input image, and the guidance image is calculated according to the following formula:

[0067]

[0068] In the formula, I cd is the guidance image; I c is the input image; c is the channel of the input image, including three channels of R, G, and B; (x, y) is the row and column values of the corresponding pixel point of the guidance image or the input image; min is the minimum grayscale value of the pixel point (x, y) in the three channels of R, G, and B.

[0069] In the nighttime environment, the guidance image is the maximum grayscale image of the input image minus the minimum grayscale image of the input image, and the guidance image is calculated according to the following formula:

[0070]

[0071] In the formula, I cd is the guidance image; I c is the input image; c is the channel of the image, including three channels of R, G, and B; (x, y) is the row and column values of the corresponding pixel point of the guidance image or the input image; max is the maximum grayscale value of the pixel point (x, y) in the three channels of R, G, and B; min is the minimum grayscale value of the pixel point (x, y) in the three channels of R, G, and B.

[0072] The nighttime environment may have complex light sources or may have a single light source.

[0073] In some embodiments, the daytime environment and the nighttime environment can be divided according to time or can be divided according to the intensity of the ambient light.

[0074] S202. Determine the atmospheric light value according to the guidance image.

[0075] Determining the atmospheric light value according to the guidance image includes the following steps:

[0076] Obtain the first corresponding image of the input image, where the first corresponding image is obtained by converting the input image from the RGB color space to the YUV color space.

[0077] Use the guidance image as the guiding map, and perform guided filtering on the Y channel of the first corresponding image as the processing map; then perform Gaussian filtering on the U channel and the V channel; convert the Y channel image after guided filtering, the U channel image and the V channel image after Gaussian filtering into the RGB color space, which is the atmospheric light value.

[0078] In some embodiments, use the guidance image as the guiding map, and perform guided filtering on the Y, U, and V channels of the first corresponding image respectively as the processing maps; then perform Gaussian filtering on the U channel and the V channel; convert the Y channel image after guided filtering, the U channel image and the V channel image after Gaussian filtering into the RGB color space, which is the atmospheric light value.

[0079] In some embodiments, use the guidance image as the guiding map, and perform guided filtering on the Y, U, and V channels of the first corresponding image respectively as the processing maps, and convert the Y, U, and V channel images after guided filtering into the RGB color space, which is the atmospheric light value.

[0080] S103. Determine the image transmittance map corresponding to the input image.

[0081] Among them, the image transmittance map is determined by fusing the dark channel transmittance map and the bright channel transmittance map. Each point in the image has an independent transmittance. Therefore, the dark channel transmittance map is a matrix with the same size as the input image, and the bright channel transmittance map is also a matrix with the same size as the input image.

[0082] First, determine the dark channel transmittance map; the determination of the dark channel transmittance map is as Figure 3 shown. To determine the dark channel transmittance map, the following steps are included:

[0083] S301. Obtain the second corresponding image of the input image, where the second corresponding image is obtained by converting the input image from the RGB color space to the HSV color space.

[0084] In the HSV color space, the H channel represents hue information, the S channel represents saturation, and the V channel represents brightness.

[0085] S302. Determine the dark channel transmittance map according to the S channel and the V channel of the second corresponding image.

[0086] Among them, the dark channel transmittance map is calculated according to the following formula:

[0087] t1(x, y) = k1 * S(x, y) + k2 * V(x, y) + k3

[0088] In the formula, t1 is the dark channel transmittance map; k1, k2, and k3 are weight coefficients; S is the S-channel map of the second corresponding image; V is the V-channel map of the second corresponding image; (x, y) are the row and column values of the corresponding pixel points of the dark channel transmittance map, the S-channel map of the second corresponding image, or the V-channel map of the second corresponding image.

[0089] k1, k2, and k3 in the above formula can be preset values or values generated according to the characteristics of the input image. When k1, k2, and k3 are preset values, k1 + k2 + k3 = 0 can be set; other non-zero values can also be set.

[0090] Secondly, determine the bright channel transmittance map; the determination of the bright channel transmittance map is as Figure 4 shown, and the steps for determining the bright channel transmittance map are as follows:

[0091] S303. Determine the bright channel transmittance map to be filtered according to the input image and the atmospheric light value;

[0092] Among them, the bright channel transmittance map to be filtered is calculated according to the following formula:

[0093]

[0094] In the formula, t′2 is the bright channel transmittance map to be filtered; A c is the atmospheric light value; c is the channel of the image, including the three channels of R, G, and B; (x, y) are the row and column values of the corresponding pixel points of the bright channel transmittance map to be filtered, the input image, or the atmospheric light value; max is the maximum gray value of the pixel point (x, y) in the three channels of R, G, and B.

[0095] S304. Perform maximum filtering on the bright channel transmittance map to be filtered to determine the bright channel transmittance map.

[0096] Among them, the bright channel transmittance map is calculated according to the following formula:

[0097]

[0098] In the formula, t′2 is the bright channel transmittance map to be filtered; t2 is the bright channel transmittance map; (x, y) are the row and column values of the corresponding pixel points of the bright channel transmittance map to be filtered or the bright channel transmittance map; maxfilter is the maximum gray value in the [m, n] neighborhood centered on the pixel point (x, y).

[0099] In some embodiments, m can be equal to n, and its value can be set to 10.

[0100] Through the maximum filtering process in step S304, the halo effect can be further eliminated.

[0101] Finally, fuse the dark channel transmittance map and the bright channel transmittance map to determine the image transmittance map. As Figure 5 shown, the steps for determining the image transmittance map include the following:

[0102] S305. Fuse the dark channel transmittance map and the bright channel transmittance map according to a preset ratio to determine the transmittance map of the image to be filtered.

[0103] Among them, the transmittance map of the image to be filtered is obtained by calculating according to the following formula:

[0104] t′(x, y) = ω1 * t1(x, y) + ω2 * t2(x, y)

[0105] In the formula, t′ is the transmittance map of the image to be filtered; t1 is the dark channel transmittance map; t2 is the bright channel transmittance map; ω1 and ω2 are weight coefficients, and (x, y) are the row and column values of the corresponding pixel points of the transmittance map of the image to be filtered, the dark channel transmittance map, or the bright channel transmittance map.

[0106] ω1 and ω2 in the above formula can be preset values or values generated according to the characteristics of the input image.

[0107] S306. For the transmittance map limit of the image to be filtered greater than 0.2, determine the image transmittance map.

[0108] Among them, the filtering is obtained by calculating according to the following formula:

[0109] t(x, y) = min(t′(x, y), 0.2)

[0110] In the formula, t is the image transmittance map; t′ is the transmittance map of the image to be filtered; (x, y) are the row and column values of the corresponding pixel points of the transmittance map of the image to be filtered or the image transmittance map, and min is to take the smaller value of t′(x, y) and 0.2.

[0111] Through step S305, it is possible to prevent the local transmittance in the image transmittance map from being too large and limit it by a transmittance value greater than 0.2.

[0112] The above filtering formula realizes that when the pixel points in the transmittance map of the image to be filtered are greater than 0.2, the corresponding pixel points are set to 0.2; when all the pixel points in the transmittance map of the image to be filtered are less than or equal to 0.2, the image transmittance map is determined.

[0113] S104. Substitute the atmospheric light value and the image transmittance map into the atmospheric scattering model to obtain the output image.

[0114] The output image is the image after defogging processing. Among them, the output image is obtained by calculating according to the following formula:

[0115]

[0116] In the formula, is the output image; I c is the input image; A c is the atmospheric light value; t is the image transmittance map; c is the channel of the image, including three channels of R, G, and B; (x, y) is the row and column value of the corresponding pixel point of the input image, atmospheric light value, image transmittance map, or output image.

[0117] In some embodiments, the above image defogging method for eliminating the halo effect can process the data collected by the camera in real time, or can also store the video stream first and then process it.

[0118] It can be seen from the above technical solutions that the present application provides an image defogging method for eliminating the halo effect, including the following steps: obtaining an input image; determining the atmospheric light value corresponding to the input image, where the atmospheric light value is a matrix equal in size to the input image; determining the image transmittance map corresponding to the input image by fusing the dark channel transmittance map and the bright channel transmittance map according to a preset ratio; and substituting the atmospheric light value and the image transmittance map into the atmospheric scattering model to obtain the output image. The present application can eliminate the halo effect of the image taken in a haze environment, and eliminate the halo through the acquisition method of the atmospheric light value and the determination method of the image transmittance map, improve the adaptability of the image, and facilitate subsequent processing.

[0119] The above content is only for explaining the technical idea of the present application, and cannot be used to limit the protection scope of the present application. Any modification made on the basis of the technical solution according to the technical idea proposed by the present application falls within the protection scope of the claims of the present application.

[0120] In addition, unless clearly stated in the claims, the order of the processing elements and sequences, the use of numbers, letters, or other names described in the present application is not used to limit the order of the processes and methods of the present application. Although some currently considered useful embodiments are discussed through various examples in the above disclosure, it should be understood that such details only serve the purpose of explanation, and the appended claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that conform to the essence and scope of the embodiments of the present application. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only through software solutions, such as installing the described system on an existing server or mobile device.

[0121] Similarly, it should be noted that, in order to simplify the description disclosed in this application and thus help the understanding of one or more embodiments, in the foregoing description of the embodiments of this application, sometimes multiple features are merged into one embodiment, drawing, or the description thereof. However, this disclosure method does not mean that the features required by the subject matter of this application are more than those mentioned in the claims. In fact, the features of the embodiments are fewer than all the features of the single embodiment disclosed above.

[0122] For each patent, patent application, patent application publication, and other materials cited in this application, such as articles, books, specifications, publications, documents, etc., their entire contents are hereby incorporated into this application by reference. Except for the application history documents that are inconsistent with or conflict with the content of this application, and also except for the documents that limit the broadest scope of the claims of this application (currently or subsequently appended to this application). It should be noted that if there are any inconsistencies or conflicts between the descriptions, definitions, and / or uses of terms in the supplementary materials of this application and the content described in this application, the descriptions, definitions, and / or uses of terms in this application shall prevail.

Claims

1. An image defogging method for eliminating the halo effect, characterized in that, Including: Obtain an input image, where the input image has haze information; Determine the atmospheric light value corresponding to the input image, where the atmospheric light value is a matrix equal in size to the input image; In the step of determining the atmospheric light value corresponding to the input image, it includes: Obtain a guidance image according to the input image. In the daytime environment, the guidance image is the minimum grayscale image of the input image; in the nighttime environment, the guidance image is the maximum grayscale image of the input image minus the minimum grayscale image of the input image; Obtain a first corresponding image of the input image, where the first corresponding image is obtained by converting the input image from the RGB color space to the YUV color space; Determine the atmospheric light value according to the guidance image and the first corresponding image, where the first corresponding image is obtained by converting the input image from the RGB color space to the YUV color space; Determine the image transmittance map corresponding to the input image, where the image transmittance map is determined by fusing the dark channel transmittance map and the bright channel transmittance map; Substitute the atmospheric light value and the image transmittance map into the atmospheric scattering model to obtain an output image.

2. The image defogging method for eliminating the halo effect according to claim 1, characterized in that, In the step of determining the atmospheric light value according to the guidance image and the first corresponding image, it includes: Perform guided filtering on the Y channel of the first corresponding image, where the guidance image is used as the guidance map; Perform Gaussian filtering on the U channel and the V channel of the first corresponding image; Determine the atmospheric light value through the Y channel map after guided filtering, the U channel map after Gaussian filtering, and the V channel map after Gaussian filtering.

3. An image defogging method for eliminating halation effect according to claim 1, characterized in that, In the step of determining the atmospheric light value according to the guidance image and the first corresponding image, it includes: Perform guided filtering on the Y channel, U channel, and V channel of the first corresponding image to determine the atmospheric light value.

4. A method for image dehazing to eliminate the halo effect according to claim 1, characterized in that, In the step of determining the image transmittance map corresponding to the input image, it includes: Determine the dark channel transmittance map; Determine the bright channel transmittance map, where the bright channel transmittance map is determined according to the input image and the atmospheric light value; Fuse the dark channel transmittance map and the bright channel transmittance map according to a preset ratio to determine the image transmittance map.

5. A method for image dehazing to eliminate the halo effect according to claim 4, characterized in that, In the step of determining the dark channel transmittance map, it includes: Obtain a second corresponding image of the input image, where the second corresponding image is obtained by converting the input image from the RGB color space to the HSV color space; Determine the dark channel transmittance map according to the S channel and the V channel of the second corresponding image, where the dark channel transmittance map is calculated according to the following formula: t1(x, y) = k1 * S(x, y) + k2 * V(x, y) + k3 In the formula, t1 is the dark channel transmittance map; k1, k2, k3 are weight coefficients; S is the S channel map of the second corresponding image; V is the V channel map of the second corresponding image; (x, y) is the row and column values of the corresponding pixel points of the dark channel transmittance map, the S channel map of the second corresponding image, or the V channel map of the second corresponding image.

6. An image defogging method for eliminating the halo effect according to claim 4, characterized in that, In the step of determining the bright channel transmittance map, it includes: Determine the to-be-filtered bright-channel transmittance map according to the input image and the atmospheric light value, and the to-be-filtered bright-channel transmittance map is calculated according to the following formula: In the formula, t′2 is the transmission rate map of the bright channel to be filtered; A c is the atmospheric light value; c is the channel of the image, including three channels of R, G, and B; (x, y) is the row and column values of the corresponding pixel point of the transmission rate map of the bright channel to be filtered, the input image, or the atmospheric light value; max is the maximum value of the gray-scale values of the pixel point (x, y) in the three channels of R, G, and B; Perform maximum filtering on the to-be-filtered bright-channel transmittance map to determine the bright-channel transmittance map, and the bright-channel transmittance map is calculated according to the following formula: In the formula, t′2 is the to-be-filtered bright-channel transmittance map; t2 is the bright-channel transmittance map; (x, y) is the row and column value of the corresponding pixel point of the to-be-filtered bright-channel transmittance map or the bright-channel transmittance map; maxfilter is the maximum gray value in the [m, n] neighborhood centered on the pixel point (x, y).

7. An image defogging method for eliminating the halo effect according to claim 4, characterized in that, In the step of fusing the dark-channel transmittance map and the bright-channel transmittance map according to a preset ratio to determine the image transmittance map, it includes: Fuse the dark-channel transmittance map and the bright-channel transmittance map according to a preset ratio to determine the to-be-filtered image transmittance map; When the pixel point in the to-be-filtered image transmittance map is greater than 0.2, set the corresponding pixel point to 0.2; When all pixel points in the to-be-filtered image transmittance map are less than or equal to 0.2, determine the image transmittance map.

8. An image defogging method for eliminating the halo effect according to claim 1, characterized in that, The output image is calculated according to the following formula: In the formula, is the output image; I c is the input image; A c is the atmospheric light value; t is the image transmittance map; c is the channel of the image, including three channels of R, G, and B; (x, y) is the row and column value of the corresponding pixel point of the input image, atmospheric light value, image transmittance map, or output image.

Images