Intelligent defogging method, system and storage medium for remote sensing images

Abstract

The present application relates to the field of image data processing technology, and specifically to an intelligent defogging method, system and storage medium for remote sensing images, the method comprising: obtaining a remote sensing image to be defogged; performing brightness analysis on multiple areas to be processed in the defogged remote sensing image to obtain the clarity of each area to be processed; performing color change analysis on pixels in each area to be processed to obtain the degree of detail expression corresponding to each pixel; obtaining detail features of each area to be processed based on the degree of detail expression, and performing detail difference analysis on multiple areas to be processed based on the detail features to obtain the detail richness of each area to be processed; obtaining the defogging weight of each area to be processed based on the clarity and detail richness, and performing intelligent defogging processing on the defogged remote sensing image based on the defogging weight. The present application performs image data processing on the defogged remote sensing image, so that the processed image is both clear and retains detail features.

Description

Technical Field

[0001] The present application relates to the technical field of image data processing, and in particular to an intelligent defogging method, system and storage medium for remote sensing images. Background Art

[0002] Remote sensing images are images of the earth's surface or other targets obtained through remote sensing technology. Clouds and other particles in the atmosphere can cause light scattering and absorption, which reduces the clarity of remote sensing images and affects the details of the images. Remote sensing images taken from a long distance are usually of poor quality, with low clarity and readability of details, which affects subsequent data analysis. Due to different meteorological conditions, clouds and fog may appear unevenly distributed in remote sensing images. This unevenness can lead to differences in clarity and quality in different areas of the image, increasing the difficulty of subsequent data analysis.

[0003] Remote sensing image defogging refers to removing fog caused by atmospheric scattering effects in remote sensing images, thereby improving image clarity and making surface features more obvious. The inconsistent thickness of fog in different areas has different effects on details. Therefore, if remote sensing images are directly defogged, important features and details will be lost in areas with more details, thereby reducing the quality of remote sensing images.

[0004] Therefore, how to defog remote sensing images without reducing the quality of remote sensing images is a problem that needs to be solved urgently. Summary of the invention

[0005] In order to solve the technical problem of how to defog remote sensing images without reducing the quality of remote sensing images, the purpose of this application is to provide an intelligent defogging method, system and storage medium for remote sensing images. The technical solutions adopted are as follows:

[0006] The present application provides an intelligent defogging method for remote sensing images, the method comprising:

[0007] Acquire the remote sensing image to be defogged;

[0008] Performing brightness analysis on a plurality of areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed;

[0009] Performing color change analysis on the pixel points in each of the areas to be processed to obtain the degree of detail expression corresponding to each of the pixel points;

[0010] According to the detail expression degree corresponding to each pixel point, the detail feature of each area to be processed is obtained, and according to the detail feature of each area to be processed, detail difference analysis is performed on the multiple areas to be processed to obtain the detail richness degree of each area to be processed;

[0011] According to the clarity and detail richness of each of the to-be-processed areas, a defogging weight of each of the to-be-processed areas is obtained, and intelligent defogging processing is performed on the to-be-defogged remote sensing image according to the defogging weight.

[0012] In some embodiments, after acquiring the remote sensing image to be defogged, the method further includes:

[0013] According to a plurality of preset image enhancement coefficients, image enhancement processing is performed on the remote sensing image to be defogged to obtain an enhanced image corresponding to each of the preset image enhancement coefficients;

[0014] Obtaining a plurality of differential images according to pixel differences between adjacent enhanced images;

[0015] Obtaining a fog density change area according to pixel change information in each of the differential images;

[0016] Comparative analysis is performed on the fog concentration change areas in the adjacent difference images to determine a plurality of areas to be processed in the remote sensing image to be defogged.

[0017] In some embodiments, performing brightness analysis on a plurality of areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed includes:

[0018] Acquire brightness information of each pixel in the area to be processed;

[0019] Determine the average brightness of neighboring pixels of each pixel within a preset range according to the brightness information of each pixel;

[0020] Determine the brightness mean and brightness variance of all pixels in the area to be processed according to the brightness information of each pixel;

[0021] Determine the neighborhood brightness difference according to the brightness information of each pixel and the average brightness of the neighborhood pixels;

[0022] Determine a regional characteristic according to the brightness mean and the brightness variance, wherein the regional characteristic is used to indicate uniformity and concentration of brightness in the area to be processed;

[0023] The clarity of each of the to-be-processed regions is obtained according to the neighborhood brightness difference and the regional characteristics.

[0024] In some embodiments, the color change analysis of the pixels in each of the to-be-processed areas to obtain the degree of detail expression corresponding to each of the pixels includes:

[0025] Acquire color change information of each pixel in the area to be processed;

[0026] Determining detail expression possibilities of each pixel point according to the color change information of each pixel point;

[0027] Comparing the brightness information of each pixel point to obtain the number of pixels corresponding to different brightness information;

[0028] The detail expression degree corresponding to each pixel point is determined by combining the detail expression possibility and the number of pixel points corresponding to the different brightness information.

[0029] In some embodiments, obtaining the detail features of each of the to-be-processed areas according to the detail expression level corresponding to each of the pixel points includes:

[0030] Determining brightness distribution information of the area to be processed according to the brightness information of each pixel;

[0031] The detail features of each of the areas to be processed are obtained by combining the brightness distribution information and the detail expression degree corresponding to each of the pixel points.

[0032] In some embodiments, performing detail difference analysis on the multiple areas to be processed according to the detail features of each area to be processed to obtain the detail richness of each area to be processed includes:

[0033] Comparing the detail features of each of the to-be-processed regions with the detail features of the adjacent to-be-processed regions to obtain a first detail difference degree;

[0034] Comparing the detail features of each of the to-be-processed regions with the detail features of a plurality of to-be-processed regions to obtain a second detail difference degree;

[0035] The detail richness of each of the to-be-processed regions is determined by combining the first detail difference degree and the second detail difference degree.

[0036] In some embodiments, obtaining the defogging weight of each of the to-be-processed areas according to the clarity and detail richness of each of the to-be-processed areas includes:

[0037] Acquire a first change trend of the clarity and a second change trend of the detail richness;

[0038] According to the first change trend and the second change trend, the defogging weight of each of the to-be-processed areas is adjusted.

[0039] In some embodiments, performing intelligent defogging processing on the remote sensing image to be defogged according to the defogging weight includes:

[0040] Obtaining an initial atmospheric light estimation result for each of the areas to be processed;

[0041] Adjusting the initial atmospheric light estimation result based on the defogging weight to obtain a target atmospheric light estimation result for each of the areas to be processed;

[0042] The target atmospheric light estimation result is transmitted to a pre-built atmospheric scattering model to perform intelligent defogging processing on each of the areas to be processed.

[0043] The present application also provides an intelligent defogging system for remote sensing images, the system comprising:

[0044] An image acquisition module, used for acquiring remote sensing images to be defogged;

[0045] A brightness analysis module, used for performing brightness analysis on a plurality of areas to be processed in the remote sensing image to be defogged, to obtain the clarity of each area to be processed;

[0046] A color change analysis module, used to perform color change analysis on the pixels in each of the areas to be processed to obtain the degree of detail expression corresponding to each of the pixels;

[0047] A detail analysis module, used to obtain detail features of each of the to-be-processed areas according to the detail expression degree corresponding to each of the pixel points, and to perform detail difference analysis on the multiple to-be-processed areas according to the detail features of each of the to-be-processed areas to obtain the detail richness degree of each of the to-be-processed areas;

[0048] The defogging processing module is used to obtain the defogging weight of each area to be processed according to the clarity and detail richness of each area to be processed, and perform intelligent defogging processing on the remote sensing image to be defogged according to the defogging weight.

[0049] An embodiment of the present application also provides a computer-readable storage medium, which stores multiple instructions, and the instructions are suitable for a processor to load to execute the steps in any one of the intelligent defogging methods for remote sensing images provided in the embodiments of the present application.

[0050] This application has the following beneficial effects:

[0051] First, a remote sensing image to be defogged is acquired; then, a brightness analysis is performed on a plurality of areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed; then, a color change analysis is performed on the pixels in each area to be processed to obtain the degree of detail expression corresponding to each pixel; then, according to the degree of detail expression corresponding to each pixel, the detail features of each area to be processed are obtained, and according to the detail features of each area to be processed, a detail difference analysis is performed on the plurality of areas to be processed to obtain the detail richness of each area to be processed; finally, according to the clarity and detail richness of each area to be processed, the defogging weight of each area to be processed is obtained, and according to the defogging weight, the remote sensing image to be defogged is subjected to intelligent defogging processing. In this application, the clarity of each area to be processed is obtained through brightness analysis to determine the impact of fog on the remote sensing image to be defogged and the degree of preliminary defogging; the degree of detail expression obtained through color change analysis reflects the color change around the pixel point, and the area with larger color change usually has more detail information; the defogging weight is calculated by the clarity and detail richness to ensure that the defogging processing of each area is more reasonable; by performing regional analysis on remote sensing images affected by uneven fog and comparing the detail differences in different areas, the processed image is clear and retains the detail features. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0053] Figure 1 A schematic diagram of an implementation environment of an intelligent defogging method for remote sensing images provided by an embodiment of the present application;

[0054] Figure 2 A schematic diagram of a flow chart of an intelligent defogging method for remote sensing images provided in one embodiment of the present application;

[0055] Figure 3 A schematic diagram of the structure of an intelligent defogging system for remote sensing images provided in one embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, describes in detail the specific implementation methods, structures, features and effects of an intelligent defogging method, system and storage medium for remote sensing images proposed in the present application. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0057] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0058] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0059] The specific scheme of an intelligent defogging method, system and storage medium for remote sensing images provided by the present application is described in detail below in conjunction with the accompanying drawings.

[0060] See also Figure 1 , Figure 1 A schematic diagram of an implementation environment of an intelligent defogging method for remote sensing images provided by an embodiment of the present application. Figure 1As shown, the implementation environment includes an intelligent defogging processing terminal 101 and an image acquisition terminal 102. The intelligent defogging processing terminal 101 can be a terminal device equipped with an intelligent defogging system, including but not limited to a laptop computer, tablet computer, handheld computer, PAD, desktop computer, etc. with local computing capabilities; the intelligent defogging system can be implemented in the form of a target client, and the target client can be a video client, an instant messaging client, a browser client, etc. that supports intelligent defogging; the intelligent defogging processing terminal 101 can, but is not limited to, communicate with the image acquisition terminal 102 through a network, and the above network can include but is not limited to: a wired network, a wireless network, wherein the wired network includes: a local area network, a metropolitan area network, and a wide area network, and the wireless network includes: Bluetooth, WIFI, and other networks that realize wireless communication. The above intelligent defogging processing terminal 101 can, but is not limited to, include a human-computer interaction screen, a processor, and a memory. The above human-computer interaction screen can, but is not limited to, be used to display remote sensing images after intelligent defogging processing. The above processor can, but is not limited to, be used to respond to human-computer interaction operations, perform corresponding operations, or generate corresponding instructions.

[0061] As an optional method, remote sensing images can be collected through the image acquisition terminal 102, and the remote sensing images include at least one foggy area. The image acquisition terminal 102 may include, for example: satellite sensors, aerial sensors, and ground sensors. Exemplarily, when the image acquisition terminal 102 is a satellite sensor, satellite sensors such as optical cameras, multispectral scanners, and synthetic aperture radars carried by satellites in Earth orbit can be used to obtain remote sensing images. These satellites can provide data with different resolutions and different spectral ranges, which are suitable for global or large-scale regional monitoring; when the image acquisition terminal 102 is an aerial sensor, high-resolution cameras, infrared scanners and other aerial sensors can be installed on aircraft or drones to collect surface images. Aerial sensors have the characteristics of high resolution and flexibility, and are suitable for detailed observation of small areas or specific targets; when the image acquisition terminal 102 is a ground sensor, the ground remote sensing station can be equipped with ground sensors such as spectrometers and radars for ground observation and data acquisition.

[0062] As an optional method, the intelligent defogging processing terminal 101 can also be a server, which can be a single server, a server cluster composed of multiple servers, or a cloud server. The above is only an example, and this embodiment does not make any limitation to this.

[0063] As an optional method, the following steps of the intelligent defogging method for remote sensing images may be performed on the intelligent defogging processing terminal 101:

[0064] Acquire the remote sensing image to be defogged;

[0065] Performing brightness analysis on a plurality of areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed;

[0066] Performing color change analysis on the pixel points in each of the areas to be processed to obtain the degree of detail expression corresponding to each of the pixel points;

[0067] According to the detail expression degree corresponding to each pixel point, the detail feature of each area to be processed is obtained, and according to the detail feature of each area to be processed, detail difference analysis is performed on the multiple areas to be processed to obtain the detail richness degree of each area to be processed;

[0068] According to the clarity and detail richness of each of the to-be-processed areas, a defogging weight of each of the to-be-processed areas is obtained, and intelligent defogging processing is performed on the to-be-defogged remote sensing image according to the defogging weight.

[0069] Based on the above method, the clarity of each area to be processed is obtained through brightness analysis to determine the impact of fog on the remote sensing image to be defogged and the degree of preliminary defogging; the degree of detail expression obtained through color change analysis reflects the color changes around the pixel points, and areas with larger color changes usually have more detail information; the defogging weight is calculated by clarity and detail richness to ensure that the defogging processing of each area is more reasonable; by performing regional analysis on remote sensing images affected by uneven fog and comparing the detail differences in different areas, the processed image is clear and retains detail features.

[0070] As an optional example, this embodiment does not limit the execution subject of the above-mentioned intelligent defogging method for remote sensing images. The above-mentioned intelligent defogging method for remote sensing images can be executed on the intelligent defogging processing terminal 101. For example, when the intelligent defogging processing terminal 101 is a desktop computer, some or all steps of the above-mentioned intelligent defogging method for remote sensing images can be executed on the desktop computer.

[0071] The above section introduces the contents of an exemplary implementation environment for applying the technical solution of the present application. Next, the intelligent defogging method for remote sensing images of the present application will be introduced.

[0072] In order to solve the problem in the prior art of how to defog remote sensing images without reducing the quality of remote sensing images, the embodiments of the present application respectively propose an intelligent defogging method for remote sensing images, an intelligent defogging system for remote sensing images, an electronic device, a computer-readable storage medium and a computer program product. These embodiments will be described in detail below.

[0073] See also Figure 2 , Figure 2A schematic diagram of a flow chart of an intelligent defogging method for remote sensing images provided in an embodiment of the present application, which can be applied to Figure 1 It should be understood that the method can also be applied to other exemplary implementation environments and specifically executed by devices in other implementation environments, and this embodiment does not limit the implementation environment to which the method is applicable.

[0074] like Figure 2 As shown, in an exemplary embodiment, the intelligent defogging method for remote sensing images includes at least steps S210 to S250, which are described in detail as follows:

[0075] In step S210, a remote sensing image to be defogged is obtained.

[0076] Among them, remote sensing images to be dehazed refer to remote sensing images whose image quality has been degraded due to atmospheric conditions (such as haze, dust, water vapor, etc.). Such images are usually visually manifested as reduced contrast, blurred colors, unclear details, etc., which directly affects the interpretability and application value of the images.

[0077] In step S220, brightness analysis is performed on a plurality of areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed.

[0078] Among them, the area to be processed refers to the specific parts of the remote sensing image to be defogged that need special attention and processing. These areas may be the most blurred and lowest contrast areas in the image; the area to be processed may be distributed in different positions of the image, and the characteristics of each area (such as fog concentration, lighting conditions, etc.) may be different. Detailed analysis and processing of the area to be processed can improve the image quality more specifically.

[0079] Among them, brightness analysis refers to the process of quantitatively analyzing the brightness of the area to be processed, which is usually used to evaluate the clarity, contrast and other light-related properties of the area to be processed. In remote sensing image processing, brightness analysis can help identify which areas in the image are more seriously affected by atmospheric conditions such as fog, thereby providing a basis for subsequent defogging processing.

[0080] Among them, clarity is an important indicator of image quality, which reflects whether the details in the image can be clearly distinguished. By evaluating the clarity of different areas, it is possible to determine which areas require more dehazing. For example, if the clarity of an area is low, then it may be necessary to increase the dehazing efforts in that area to improve its clarity.

[0081] In step S230, color change analysis is performed on the pixel points in each of the to-be-processed areas to obtain the detail expression degree corresponding to each of the pixel points.

[0082] Among them, pixels are the most basic unit of image composition. A remote sensing image to be dehazed is actually a matrix composed of many pixels. Each pixel represents a part of the image and carries specific color information.

[0083] Color change analysis refers to evaluating the distribution and changes of colors in an image to identify detailed features in the image. By performing color change analysis on the pixels in the area to be processed, detailed information in the image can be more accurately identified, thereby guiding the dehazing process and improving the clarity and usability of the image.

[0084] In step S240, the detail features of each of the areas to be processed are obtained according to the detail expression levels corresponding to each of the pixel points, and detail difference analysis is performed on the multiple areas to be processed according to the detail features of each of the areas to be processed to obtain the detail richness of each of the areas to be processed.

[0085] Among them, detail features refer to the characteristics of the image that reflect the object boundary, texture, color change, etc. By extracting the detail features of each area to be processed and performing dehazing processing based on these features, the quality of the image can be significantly improved, making the detail information in the image clearer, thereby increasing the value of the image in practical applications.

[0086] Among them, detail difference analysis is the process of evaluating the detail richness between different areas of the image, which can help identify which areas require more processing to restore details.

[0087] In step S250, the defogging weight of each of the to-be-processed areas is obtained according to the clarity and detail richness of each of the to-be-processed areas, and intelligent defogging processing is performed on the to-be-defogged remote sensing image according to the defogging weight.

[0088] The defogging weight refers to a value assigned according to the characteristics of different areas in the image (such as clarity, detail richness, etc.) to guide the degree of defogging. The larger the defogging weight, the more defogging is required for the area; the smaller the defogging weight, the less defogging is required for the area.

[0089] It can be seen from the above steps S210 to S250 that the solution proposed in this embodiment obtains the clarity of each area to be processed through brightness analysis to determine the impact of fog on the remote sensing image to be defogged and the degree of preliminary defogging; the degree of detail expression obtained through color change analysis reflects the color changes around the pixel points, and areas with larger color changes usually have more detail information; the defogging weight is calculated by clarity and detail richness to ensure that the defogging processing of each area is more reasonable; by performing regional analysis on remote sensing images affected by uneven fog and comparing the detail differences in different areas, the processed image is clear and retains detail features.

[0090] In one embodiment of the present application, after acquiring the remote sensing image to be defogged, the method further includes:

[0091] According to a plurality of preset image enhancement coefficients, image enhancement processing is performed on the remote sensing image to be defogged to obtain an enhanced image corresponding to each of the preset image enhancement coefficients;

[0092] Obtaining a plurality of differential images according to pixel differences between adjacent enhanced images;

[0093] Obtaining a fog density change area according to pixel change information in each of the differential images;

[0094] Comparative analysis is performed on the fog concentration change areas in the adjacent difference images to determine a plurality of areas to be processed in the remote sensing image to be defogged.

[0095] Among them, the remote sensing images to be defogged can be obtained through sensors installed on satellite platforms. Due to different meteorological conditions, clouds of different thicknesses may appear and appear differently in the remote sensing images to be defogged. Due to different ground occlusion conditions, different fog concentrations have different effects on image details. It is necessary to divide the different fog concentrations in the remote sensing images to be defogged into regions, and different degrees of image enhancement processing can be used to obtain areas to be processed with different fog concentrations.

[0096] Exemplarily, the preset image enhancement coefficient may be a gamma coefficient, and gamma transformation may be performed through multiple gamma coefficients to achieve image enhancement processing. Performing gamma transformation on an image through multiple gamma coefficients may effectively adjust the brightness and contrast of the image, thereby enhancing detail information in the image.

[0097] Exemplarily, the remote sensing image is first subjected to a gamma transform, where the gamma coefficient may be, for example, 5, 10, or 15. After the gamma transform, the fog display areas are different between different enhanced images. When obtaining areas with different fog concentrations between adjacent enhanced images, an image difference algorithm is used for the fog-changed areas of adjacent enhanced images to determine pixel differences, thereby obtaining multiple differential images. At this time, each differential image can display the entire area of ​​the part where the fog concentration changes. Finally, by subtracting adjacent differential images, a comparative analysis of the fog concentration change areas in adjacent differential images is achieved, thereby determining multiple areas to be processed.

[0098] It can be seen from the above embodiments that different gamma coefficients can enhance different details of the image, which is helpful for the subsequent determination of pixel differences; through the image difference algorithm, the part of the image where the fog concentration changes can be captured, and at the same time, the pixel changes caused by other factors such as lighting can be reduced, focusing on the impact of fog; through comparative analysis, the key areas that require defogging can be identified more accurately, while ensuring that the defogging process is concentrated where it is most needed, thereby improving processing efficiency.

[0099] In one embodiment of the present application, the brightness analysis of the multiple areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed includes:

[0100] Acquire brightness information of each pixel in the area to be processed;

[0101] Determine the average brightness of neighboring pixels of each pixel within a preset range according to the brightness information of each pixel;

[0102] Determine the brightness mean and brightness variance of all pixels in the area to be processed according to the brightness information of each pixel;

[0103] Determine the neighborhood brightness difference according to the brightness information of each pixel and the average brightness of the neighborhood pixels;

[0104] Determine a regional characteristic according to the brightness mean and the brightness variance, wherein the regional characteristic is used to indicate uniformity and concentration of brightness in the area to be processed;

[0105] The clarity of each of the to-be-processed regions is obtained according to the neighborhood brightness difference and the regional characteristics.

[0106] It should be noted that there are many details in the remote sensing image to be defogged that need to be retained while defogging. For remote sensing images with forest cover to be defogged, they usually contain a lot of details, such as roads, different types of vegetation and other ground features. The presence of fog blocks the contours and features of roads or different types of vegetation, affecting the overall quality of the remote sensing image to be defogged, and may also cover up important geographic information and ecological features. However, direct defogging will affect the performance of these details. It is necessary to perform appropriate defogging according to the fog occlusion in different areas to avoid the loss of details. Due to the uneven distribution of fog, the density of fog in different areas of the remote sensing image to be defogged is different, and the degree of influence on the performance of ground information is different. In areas with thicker fog, a greater degree of defogging is required, and in areas with thinner fog, the degree of influence on the ground information in the remote sensing image is small, and a smaller degree of defogging is required. Therefore, it is necessary to perform brightness analysis on multiple areas to be processed in the remote sensing image to obtain the clarity of each area to be processed.

[0107] Exemplarily, the brightness information may be a grayscale value. Due to the influence of fog in some areas to be processed, the contrast of the image is reduced, the grayscale value difference between pixels is reduced, and the pixels of the image are distributed in a similar grayscale value range. Since fog is white-gray in remote sensing images, the grayscale value of pixels in foggy areas is high. The average brightness of neighborhood pixels can be the average grayscale value of pixels in 4 or 8 neighborhoods around the pixel. The neighborhood brightness difference can be the difference between the brightness of a pixel and the average brightness of its neighborhood pixels.

[0108] For example, the clarity can be expressed as follows:

[0109] ;

[0110] in, is the clarity of the i-th area to be processed; is the gray value of the u-th pixel in the i-th area to be processed; is the mean gray value of the pixels in the 8-neighborhood of the u-th pixel in the i-th area to be processed; is the mean gray value of all pixels; is the gray value variance of all pixels in the i-th area to be processed.

[0111] in, is the difference in grayscale value between the u-th pixel in the i-th area to be processed and the pixels in the neighborhood. The greater the difference, the more obvious the contrast in the image is and the higher the clarity of the image. It means that when the grayscale values ​​of all pixels in the i-th area to be processed are generally high, the more concentrated the grayscale value distribution is, the lower the clarity of the image is.

[0112] Among them, the greater the clarity, the less the image is affected by fog, and the clearer the image is; the smaller the clarity, the greater the image may be affected by fog, and the less clear the image is.

[0113] In one embodiment of the present application, the color change analysis of the pixels in each of the to-be-processed areas to obtain the detail expression level corresponding to each of the pixels includes:

[0114] Acquire color change information of each pixel in the area to be processed;

[0115] Determining detail expression possibilities of each pixel point according to the color change information of each pixel point;

[0116] Comparing the brightness information of each pixel point to obtain the number of pixels corresponding to different brightness information;

[0117] The detail expression degree corresponding to each pixel point is determined by combining the detail expression possibility and the number of pixel points corresponding to the different brightness information.

[0118] It should be noted that when defogging each area, due to the uneven distribution of fog, each area in the remote sensing image is affected differently by the fog. Areas that are more affected by fog require stronger defogging processing, but this defogging processing must ensure that important details in the image are not damaged. However, due to the different details contained in different areas, direct defogging in areas with more details may cause the loss of important features and details of the image, affecting the final image quality. Therefore, when defogging, it is necessary to appropriately adjust the defogging degree of each area, retain the detailed information of the image while defogging, and retain the detailed information of the image to the maximum extent while ensuring the defogging effect.

[0119] Exemplarily, the color change information may be a gradient amplitude. The detail differences presented in each region are derived from the different performances of the pixels in the region. Each pixel may represent different detail information of the image. The gradient amplitude reflects the sensitivity of the pixel to color changes. A high gradient amplitude means that the grayscale or color around the point has changed. When the gradient amplitude of the pixel is large, it means that these pixels are at the edge of the image or in an area where the texture changes significantly, and are more likely to show the details of the image. In foggy areas, the visual clarity of the image is reduced, and most of the pixels may show similar grayscale values. The detail information of the image becomes blurred, resulting in unclear color changes. Although the differences between the pixels in these areas are small, some special pixels may show different colors from the surroundings, representing important details or features in the image. Therefore, the pixels with fewer numbers of the same grayscale values ​​may be special details in the image. By analyzing the pixels with fewer numbers of the same grayscale values, the detail information that is still retained under foggy conditions can be identified, thereby showing the potential important features in the image. Therefore, it is necessary to perform color change analysis on the pixels in each area to be processed to obtain the degree of detail expression corresponding to each pixel.

[0120] For example, the degree of detail may be expressed as follows:

[0121] ;

[0122] in, is the detail expression level corresponding to the u-th pixel in the i-th area to be processed; is the gradient amplitude of the u-th pixel in the i-th area to be processed; is the mean value of the gradient amplitude of all pixels in the i-th area to be processed; is the number of pixels with the same grayscale value as the u-th pixel.

[0123] in, is the possibility of detail expression, which represents the ratio of the gradient amplitude of the u-th pixel point in the i-th area to be processed to all the pixels. The larger the ratio is, the more likely the u-th pixel point in the i-th area to be processed is to be the pixel point that expresses the details of the i-th area to be processed. The smaller it is, the fewer the number of pixels with the same grayscale value in the i-th area to be processed is, and the more likely the u-th pixel is to reflect the details of the i-th area to be processed.

[0124] Among them, combining the possibility of detail expression and the number of pixels corresponding to different brightness information, the greater the degree of detail expression corresponding to each pixel, the more likely the u-th pixel is to be a pixel representing the detail of the i-th area to be processed, and the more likely it is to be the embodiment of the detail of the i-th area to be processed, representing important details or feature information in the image.

[0125] In one embodiment of the present application, obtaining the detail features of each of the to-be-processed areas according to the detail expression degree corresponding to each of the pixel points includes:

[0126] Determining brightness distribution information of the area to be processed according to the brightness information of each pixel;

[0127] The detail features of each of the areas to be processed are obtained by combining the brightness distribution information and the detail expression degree corresponding to each of the pixel points.

[0128] Among them, the detail information analysis of each pixel in each area together constitutes the detail features of the area where each pixel is located, which may be the key component of the image. Since different areas are affected by fog to different degrees and have different levels of detail information, different areas need to be processed during defogging. For each area, each pixel represents different regional features in the area. If the pixels in the area are distributed on multiple grayscale values, it means that the area contains rich color information and detail features. The greater the detail expression of each pixel, the more obvious the detail features in this area.

[0129] Exemplarily, the detailed features may be represented as follows:

[0130] ;

[0131] in, is the detailed feature of the i-th area to be processed; is the number of grayscale pixels with the largest grayscale value distribution; is the number of grayscale pixels with the least grayscale value distribution; is the number of gray levels of the pixel points in the i-th area to be processed; is the detail level corresponding to the u-th pixel in the i-th area to be processed, is the number of pixels in the i-th area to be processed.

[0132] in, is the brightness distribution information, which means that when the grayscale distribution of the pixels in the i-th area to be processed is more, and the difference between the grayscale of the most pixels and the grayscale of the least pixels is smaller, the grayscale value distribution in the i-th area to be processed is more dispersed and more even, and the i-th area to be processed shows more details; It represents the sum of the detail expression levels of all pixels in the i-th area to be processed. The larger the value is, the more details of the pixels in the i-th area to be processed are, and the more obvious the detail features can be.

[0133] Among them, the larger the detail feature, the more details of the pixels in the i-th area to be processed, the more obvious the detail feature that can be expressed, the more it needs to be retained, and the more appropriate defogging is required.

[0134] In one embodiment of the present application, performing detail difference analysis on the multiple areas to be processed according to the detail features of each area to be processed to obtain the detail richness of each area to be processed includes:

[0135] Comparing the detail features of each of the to-be-processed regions with the detail features of the adjacent to-be-processed regions to obtain a first detail difference degree;

[0136] Comparing the detail features of each of the to-be-processed regions with the detail features of a plurality of the to-be-processed regions to obtain a second detail difference degree;

[0137] The detail richness of each of the to-be-processed regions is determined by combining the first detail difference degree and the second detail difference degree.

[0138] Among them, the detail differences between different areas in the entire remote sensing image are different. In an image with fewer detail features, if the detail features in a single area are obvious, it means that it is a more important area in the image; although the distribution of fog in the image is uneven, the fog has no clear boundary line, and the change is gradual rather than sudden. The difference in the impact of fog on adjacent areas is small, and the impact on the detail performance of the image is similar. Therefore, when the difference in the detail performance of a region is large, it means that the detail information of the region is richer. Therefore, it is necessary to perform detail difference analysis on multiple areas to be processed according to the detail features of each area to be processed to obtain the detail richness of each area to be processed.

[0139] Exemplarily, the richness of details may be represented by:

[0140] ;

[0141] in, is the detail richness of the i-th area to be processed; is the detailed feature of the i-th area to be processed; is the detailed feature of the i+1th area to be processed; is the mean value of the detail features of all the areas to be processed in the image.

[0142] in, is the first detail difference degree, which indicates the difference in detail features between the i-th region to be processed and the adjacent region. The larger the first detail difference degree is, the higher the detail richness of the i-th region to be processed is. is the second detail difference degree, which represents the ratio of the detail features of the ith area to be processed to the mean of the detail features of the entire image. The larger the ratio is, the more important the details represented by the ith area to be processed are in the image, the more obvious the detail features are, and the higher the detail richness is.

[0143] The greater the detail richness, the more details the ith area to be processed contains in the adjacent area of ​​the ith area to be processed, the more important the details are in the image, and the more obvious the detail features are.

[0144] In one embodiment of the present application, obtaining the defogging weight of each of the to-be-processed areas according to the clarity and detail richness of each of the to-be-processed areas includes:

[0145] Acquire a first change trend of the clarity and a second change trend of the detail richness;

[0146] According to the first change trend and the second change trend, the defogging weight of each of the to-be-processed areas is adjusted.

[0147] Among them, the detail information of different areas may contain important data of the image. Simply applying a unified dehazing process to the entire image may lead to the loss of important details, thus affecting the overall quality of the image. While dehazing, detail features must also be retained. Moderate dehazing processing should be adopted to reduce the loss of details that may be caused in the dehazing process, while retaining important visual and structural features in the image to maximize the retention of these details, thereby ensuring that the processed image is both clear and has sufficient detail expression. Therefore, it is necessary to obtain the dehazing weight of each area to be processed based on the clarity and detail richness of each area to be processed.

[0148] For example, the clarity can be compared with the preset clarity to obtain a first change trend, and the detail richness can be compared with the preset detail richness to obtain a second change trend, and then the defogging weight of each area to be processed is adjusted according to the first change trend and the second change trend. The expression method of adjusting the defogging weight of each area to be processed can be, for example:

[0149] ;

[0150] in, is the defogging weight of the i-th area to be processed; is the clarity of the i-th area to be processed; is the detail richness of the i-th area to be processed, and exp represents an exponential function with a natural constant as the base.

[0151] in, It means that when the clarity of the i-th area to be processed is higher, it is less affected by fog and the degree of defogging is less. When the detail richness is high, it means that the i-th area to be processed contains more detail information, and it is necessary to reduce the degree of defogging and the defogging weight to avoid losing details during defogging.

[0152] Among them, the smaller the defogging weight is, the more detail information the i-th area to be processed contains. When defogging, it is necessary to reduce the degree of defogging and reduce the defogging weight to avoid losing the details during defogging.

[0153] In one embodiment of the present application, the intelligent defogging process is performed on the remote sensing image to be defogged according to the defogging weight, including:

[0154] Obtaining an initial atmospheric light estimation result for each of the areas to be processed;

[0155] Adjusting the initial atmospheric light estimation result based on the defogging weight to obtain a target atmospheric light estimation result for each of the areas to be processed;

[0156] The target atmospheric light estimation result is transmitted to a pre-built atmospheric scattering model to perform intelligent defogging processing on each of the areas to be processed.

[0157] Exemplarily, the dark channel prior algorithm can be adjusted by obtaining different defogging weights for each area to adjust the initial atmospheric light estimation result. First, according to the dark channel prior algorithm, the initial atmospheric light estimation result of each area to be processed is obtained; the initial atmospheric light estimation result of the area to be processed is multiplied by the defogging weight to obtain the target atmospheric light estimation result of each area to be processed; finally, the target atmospheric light estimation result is substituted into the atmospheric scattering model to defog each area and then merge all areas to obtain the remote sensing image after defogging.

[0158] Figure 3 A schematic diagram of the structure of an intelligent defogging system for remote sensing images provided in one embodiment of the present application. The system can be applied to Figure 1 The system can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the system is applied.

[0159] like Figure 3 As shown, the exemplary intelligent defogging system for remote sensing images includes:

[0160] An image acquisition module 301 is used to acquire a remote sensing image to be defogged;

[0161] A brightness analysis module 302 is used to perform brightness analysis on a plurality of to-be-processed areas in the remote sensing image to be defogged, and obtain the clarity of each of the to-be-processed areas;

[0162] A color change analysis module 303 is used to perform color change analysis on the pixels in each of the areas to be processed to obtain the detail expression level corresponding to each of the pixels;

[0163] A detail analysis module 304 is used to obtain detail features of each of the to-be-processed regions according to the detail expression degree corresponding to each of the pixel points, and to perform detail difference analysis on the multiple to-be-processed regions according to the detail features of each of the to-be-processed regions to obtain the detail richness degree of each of the to-be-processed regions;

[0164] The defogging processing module 305 is used to obtain the defogging weight of each of the to-be-processed areas according to the clarity and detail richness of each of the to-be-processed areas, and perform intelligent defogging processing on the to-be-defogged remote sensing image according to the defogging weight.

[0165] In this exemplary intelligent defogging system for remote sensing images, the clarity of each area to be processed is obtained through brightness analysis to determine the impact of fog on the remote sensing image to be defogged and the degree of preliminary defogging; the degree of detail expression obtained through color change analysis reflects the color changes around the pixels, and areas with larger color changes usually have more detail information; the defogging weight is calculated by the clarity and detail richness to ensure that the defogging processing of each area is more reasonable; by performing regional analysis on remote sensing images affected by uneven fog and comparing the detail differences in different areas, the processed image is clear and retains detail features.

[0166] It should be noted that the intelligent defogging system for remote sensing images provided in the above embodiment and the intelligent defogging method for remote sensing images provided in the above embodiment belong to the same concept, and the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In practical applications, the intelligent defogging system for remote sensing images provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0167] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device implements the intelligent defogging method for remote sensing images provided in the above-mentioned embodiments.

[0168] It should be noted that the computer-readable medium shown in the embodiment of the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier, which carries a computer-readable computer program. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0169] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0170] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0171] Another aspect of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for a processor to load to execute the steps in any one of the intelligent defogging methods for remote sensing images provided in the embodiments of the present application. The computer-readable storage medium may be included in the electronic device described in the above embodiments, or may exist independently without being assembled into the electronic device.

[0172] It should be noted that the sequence of the above embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0173] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent defogging method for remote sensing images, characterized in that: The method comprises: Acquire the remote sensing image to be dehazed; Performing brightness analysis on a plurality of areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed; Performing color change analysis on the pixel points in each of the areas to be processed to obtain the degree of detail expression corresponding to each of the pixel points; According to the detail expression degree corresponding to each pixel point, the detail feature of each area to be processed is obtained, and according to the detail feature of each area to be processed, detail difference analysis is performed on the multiple areas to be processed to obtain the detail richness degree of each area to be processed; According to the clarity and detail richness of each of the to-be-processed areas, a defogging weight of each of the to-be-processed areas is obtained, and intelligent defogging processing is performed on the to-be-defogged remote sensing image according to the defogging weight; The color change analysis of the pixels in each of the to-be-processed areas to obtain the detail expression level corresponding to each of the pixels includes: Acquire color change information of each pixel in the area to be processed; Determining detail expression possibilities of each pixel point according to the color change information of each pixel point; Comparing the brightness information of each pixel point to obtain the number of pixels corresponding to different brightness information; Determine the degree of detail expression corresponding to each pixel point by combining the detail expression possibility and the number of pixels corresponding to the different brightness information; The performing detail difference analysis on the multiple areas to be processed according to the detail features of each area to be processed to obtain the detail richness of each area to be processed includes: Comparing the detail features of each of the to-be-processed regions with the detail features of the adjacent to-be-processed regions to obtain a first detail difference degree; Comparing the detail features of each of the to-be-processed regions with the detail features of a plurality of to-be-processed regions to obtain a second detail difference degree; The detail richness of each of the to-be-processed regions is determined by combining the first detail difference degree and the second detail difference degree.

2. The intelligent defogging method for remote sensing images according to claim 1, characterized in that: After obtaining the remote sensing image to be defogged, the method further includes: According to a plurality of preset image enhancement coefficients, image enhancement processing is performed on the remote sensing image to be defogged to obtain an enhanced image corresponding to each of the preset image enhancement coefficients; Obtaining a plurality of differential images according to pixel differences between adjacent enhanced images; Obtaining a fog density change area according to pixel change information in each of the differential images; Comparative analysis is performed on the fog concentration change areas in the adjacent difference images to determine a plurality of areas to be processed in the remote sensing image to be defogged.

3. The intelligent defogging method for remote sensing images according to claim 1, characterized in that: The performing brightness analysis on the multiple areas to be processed in the remote sensing image to be defogged to obtain the clarity of each area to be processed includes: Acquire brightness information of each pixel in the area to be processed; Determine the average brightness of neighboring pixels of each pixel within a preset range according to the brightness information of each pixel; Determine the brightness mean and brightness variance of all pixels in the area to be processed according to the brightness information of each pixel; Determine the neighborhood brightness difference according to the brightness information of each pixel and the average brightness of the neighborhood pixels; Determine a regional characteristic according to the brightness mean and the brightness variance, wherein the regional characteristic is used to indicate uniformity and concentration of brightness in the area to be processed; The clarity of each of the to-be-processed regions is obtained according to the neighborhood brightness difference and the regional characteristics.

4. The intelligent defogging method for remote sensing images according to claim 1, characterized in that: The obtaining of detail features of each of the regions to be processed according to the detail expression degree corresponding to each of the pixel points includes: Determining brightness distribution information of the area to be processed according to the brightness information of each pixel; The detail features of each of the areas to be processed are obtained by combining the brightness distribution information and the detail expression degree corresponding to each of the pixel points.

5. The intelligent defogging method for remote sensing images according to claim 1, characterized in that: The defogging weight of each of the areas to be processed is obtained according to the clarity and detail richness of each of the areas to be processed, including: Acquire a first change trend of the clarity and a second change trend of the detail richness; According to the first change trend and the second change trend, the defogging weight of each of the to-be-processed areas is adjusted.

6. The intelligent defogging method for remote sensing images according to claim 1, characterized in that: The intelligent defogging process is performed on the remote sensing image to be defogged according to the defogging weight, comprising: Obtaining an initial atmospheric light estimation result for each of the areas to be processed; Adjusting the initial atmospheric light estimation result based on the defogging weight to obtain a target atmospheric light estimation result for each of the areas to be processed; The target atmospheric light estimation result is transmitted to a pre-built atmospheric scattering model to perform intelligent defogging processing on each of the areas to be processed.

7. An intelligent defogging system for remote sensing images, characterized in that: The system comprises: An image acquisition module, used for acquiring remote sensing images to be defogged; A brightness analysis module, used for performing brightness analysis on a plurality of areas to be processed in the remote sensing image to be defogged, to obtain the clarity of each area to be processed; A color change analysis module, used to perform color change analysis on the pixels in each of the areas to be processed to obtain the degree of detail expression corresponding to each of the pixels; A detail analysis module, used to obtain detail features of each of the to-be-processed areas according to the detail expression degree corresponding to each of the pixel points, and to perform detail difference analysis on the multiple to-be-processed areas according to the detail features of each of the to-be-processed areas to obtain the detail richness degree of each of the to-be-processed areas; A defogging processing module is used to obtain the defogging weight of each of the to-be-processed areas according to the clarity and detail richness of each of the to-be-processed areas, and to perform intelligent defogging processing on the to-be-defogged remote sensing image according to the defogging weight; The color change analysis of the pixels in each of the to-be-processed areas to obtain the detail expression level corresponding to each of the pixels includes: Acquire color change information of each pixel in the area to be processed; Determining detail expression possibilities of each pixel point according to the color change information of each pixel point; Comparing the brightness information of each pixel point to obtain the number of pixels corresponding to different brightness information; Determine the degree of detail expression corresponding to each pixel point by combining the detail expression possibility and the number of pixels corresponding to the different brightness information; The performing detail difference analysis on the multiple areas to be processed according to the detail features of each area to be processed to obtain the detail richness of each area to be processed includes: Comparing the detail features of each of the to-be-processed regions with the detail features of the adjacent to-be-processed regions to obtain a first detail difference degree; Comparing the detail features of each of the to-be-processed regions with the detail features of a plurality of to-be-processed regions to obtain a second detail difference degree; The detail richness of each of the to-be-processed regions is determined by combining the first detail difference degree and the second detail difference degree.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, which are suitable for loading by a processor to execute the steps in the intelligent defogging method for remote sensing images as described in any one of claims 1 to 6.

Images