Image color correction method, electronic device and storage medium

Abstract

The present invention relates to an image color correction method, electronic equipment and storage medium. The image color correction method comprises: obtaining the quantization level, image shooting time and position of the space-based remote sensing image according to the historical space-based remote sensing image, and constructing a corresponding relationship model of the corresponding time and position; obtaining the atmospheric absorption and ground object color composition according to the historical space-based remote sensing image, and constructing a color correction model; after obtaining a new space-based remote sensing image, the space-based remote sensing image is corrected by using the corresponding relationship model and the color correction model in sequence. The present invention can avoid the problem of randomness of color correction caused by the uncertainty of the atmospheric macroscopic scale and the local microscopic scale, and does not need to consider the complex settings in the atmospheric model. It is easy to use, and the operator does not need to have professional knowledge of the atmospheric model, which is conducive to the color restoration of the image.

Description

Technical Field

[0001] The present invention relates to the technical field of space-based remote sensing image processing, and in particular to an image color correction method, electronic equipment and storage medium. Background Art

[0002] With the development of technology and space equipment, it is becoming easier to obtain space-based remote sensing images. Affected by satellite attitude, orbit, atmospheric motion and other factors, space-based remote sensing images need to be corrected geometrically and radiometrically before they can be used to improve data accuracy.

[0003] The color of space-based remote sensing images is a form of representation of the radiation results of remote sensing images.

[0004] Color-corrected images are widely used in various fields related to national economy and people's livelihood. There are two main types of application targets. One is applications that are sensitive to color quantification of remote sensing images: such as crop yield estimation, pest and disease analysis, vegetation classification, water body classification, etc., which involve quantitative inversion of remote sensing. The other is applications that are not sensitive to color quantification of remote sensing images: such as the display of general ground object information; and the detection of the state and changes of ground object targets, such as changes in planting areas, which involve color change trends, shape changes, or are used for image rendering and subjective judgment.

[0005] In the past, the color restoration process of space-based remote sensing images was often carried out through radiation correction, which was mainly due to the need for subsequent spectral numerical quantification or qualitative trend analysis of the images, and was common in fields such as agriculture and forestry remote sensing.

[0006] Radiation correction is aimed at the fact that when the remote sensing sensor receives electromagnetic radiation energy radiated or reflected from the ground target, it is affected by the characteristics of the remote sensing sensor itself, the atmospheric effect, and the lighting conditions (terrain and solar altitude angle), which makes the detection value of the remote sensing sensor different from the actual spectral radiation value of the ground object, causing grayscale distortion of the remote sensing image. In order to eliminate these effects and restore the original appearance of the image as much as possible, so as to more realistically reflect the reflection and radiation characteristics of the ground target, radiation correction is required. Radiation correction usually includes three aspects: sensor correction, atmospheric correction, solar altitude angle and terrain correction.

[0007] in:

[0008] Sensor correction, solar altitude angle and terrain correction, these three corrections are based on the fixed results of ground laboratory calibration or the fixed spatiotemporal relationship of time and position, which can form a unique one-to-one correspondence and have strong uniqueness.

[0009] Atmospheric correction is affected by the atmospheric dynamics of the Earth's large-scale environment and the local micro-environment of the region. Although there is historical statistical experience to follow, its uncertainty and randomness are relatively large. Therefore, the randomness of color correction mainly comes from the uncertainty of the macro-scale atmosphere and the local micro-scale. In order to obtain relatively stable data, a series of assumptions and a certain scope of application are required.

[0010] As a result, a variety of atmospheric radiation transfer models have been generated for different assumptions and scopes of application, among which the LOWTRAN and MODTRAN models are the most commonly used.

[0011] The LOWTRAN model was developed by the U.S. Air Force Geophysical Laboratory. It calculates the average transmittance and atmospheric path radiance of a given atmospheric path with a spectral resolution of 20 cm-1. It provides vertical temperature profiles of 6 reference atmospheric models, vertical profiles of mixing ratios of gases such as water vapor, ozone, and carbon dioxide, and vertical profiles of 13 trace gases. It also provides profiles of urban and rural atmospheric aerosols, fog, dust, clouds and rain, and radiation parameters such as extinction coefficients and absorption coefficients, as well as extraterrestrial solar spectra. Currently, the LOWTRAN7 version is the most popular. This version has fast calculations and a flexible and changeable structure. However, due to the limitations of the approximate molecular band model used, the accuracy of atmospheric correction is seriously reduced for atmospheric regions above 40 km.

[0012] The MODTRAN model is mainly a gradual improvement of the LOWTRAN model. It increases the spectral resolution of the LOWTRAN model from 20cm-1 to 2cm-1, adopts a molecular absorption algorithm with a spectral resolution of 2cm-1, and optimizes the processing of the pressure and temperature relationship of molecular absorption. At the same time, it maintains the basic program and usage structure of LOWTRAN7, has strong mode selectivity, and can choose LOWTRAN7 or MODTRAN radiation transmission mode at will. The radiation transmission process of this model takes into account the absorption, scattering and absorption and scattering effects of aerosols and clouds of almost all atmospheric molecules in the atmosphere, and can calculate the path transmittance of absorbing substances, atmospheric emissivity, single or multiple scattered solar radiation, and directly transmitted solar radiation.

[0013] In addition, space-based remote sensing images are also often used to analyze and identify ground objects, and are often used in areas such as obtaining information on ground object changes. This part of space-based remote sensing image information is mainly obtained through changes in ground object geometry, texture, etc. It has lower requirements for quantitative radiation correction, but some useful auxiliary information can still be obtained through color.

[0014] The existing technology uses atmospheric correction, which is affected by the atmospheric dynamics of the large-scale environment of the earth and the local micro-environment of the region. Although there is historical statistical experience to follow, its uncertainty and randomness are relatively large. Therefore, the randomness of color correction mainly comes from the uncertainty of the macro-scale atmosphere and the local micro-scale. In order to obtain relatively stable data, a series of assumptions and a certain scope of application are required, and the calculation process is complicated.

[0015] That is, the existing atmospheric model-based correction technology requires complex atmospheric parameter settings, is inconvenient to use, requires professional knowledge of atmospheric models, and is not conducive to the color restoration of images. Summary of the invention

[0016] In view of the above technical problems, the present invention proposes an image color correction method, electronic device and storage medium, which only requires sending the image to the color correction software without complex parameter settings, and is convenient and easy to use.

[0017] The technical solution to achieve the purpose of the present invention is: an image color correction method, comprising the following steps:

[0018] Step S1, according to historical space-based remote sensing images, obtaining the quantization level, image shooting time and location of the space-based remote sensing images, and constructing a corresponding relationship model between the corresponding time and location;

[0019] Step S2: Based on historical space-based remote sensing images, obtain atmospheric absorption and ground object color composition, and build a color correction model;

[0020] Step S3: After acquiring a new space-based remote sensing image, the space-based remote sensing image is corrected using the correspondence model and the color correction model in sequence.

[0021] According to one aspect of the present invention, in the step S1, it specifically includes:

[0022] Step S11, obtaining historical space-based remote sensing images;

[0023] Step S12: Determine the image shooting time, location, and quantization level based on historical space-based remote sensing images;

[0024] Step S13: Calculate the theoretical quantitative index range according to the quantization level of historical space-based remote sensing images.

[0025] According to one aspect of the present invention, in step S2, it specifically includes:

[0026] Step S21: Count the pixel values ​​of typical objects based on historical space-based remote sensing images, and obtain the near-infrared spectrum NIR values ​​of corresponding pixel points to confirm the historical atmospheric absorption of the point;

[0027] Step S22, calculating discrete values ​​according to the quantization level of the historical space-based remote sensing image;

[0028] Step S13, calculating the color combination of common ground objects in the RGB color mode and the relationship between the color combination and the near infrared spectrum NIR value;

[0029] Step S14, converting the RGB color space into the HSV color space value;

[0030] Step S15: Obtain historical local weather information, and construct a color correction model based on the saturation under different cloud cover conditions and the color relationship of different objects in the area.

[0031] According to one aspect of the present invention, in step S3, it specifically includes:

[0032] Step S31, removing clouds and fog from the new space-based remote sensing image;

[0033] Step S32: selecting a color correction model using a corresponding relationship model according to the shooting settings and shooting position of the new space-based remote sensing image;

[0034] Step S33: Perform color correction using the corresponding color correction model.

[0035] According to one aspect of the present invention, in step S14, the color space value includes at least: hue and hue value, saturation and color purity value, and lightness value.

[0036] According to one aspect of the present invention, a computer-readable storage medium is provided for storing computer instructions. When the computer instructions are executed by a processor, an image color correction method as described in any one of the above technical solutions is implemented.

[0037] According to the concept of the present invention, a method for image color correction, an electronic device and a computer-readable storage medium are proposed. Through the probability model of historical images, statistics of past same or similar areas are constructed, an artificial intelligence network is constructed, and color restoration is performed on the basis of existing image data. The problem of randomness of color correction caused by uncertainties in the macroscopic scale of the atmosphere and the local microscopic scale can be avoided. There is no need to consider the complex settings in the atmospheric model. The method is easy to use and the operator does not need to have professional knowledge of the atmospheric model, which is conducive to the color restoration of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram schematically showing a flow chart of an image color correction method according to an embodiment of the present invention;

[0039] Figure 2 The flowchart schematically shows a method for image color correction according to another embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments cannot be described one by one here, but the embodiments of the present invention are not therefore limited to the following embodiments.

[0042] like Figure 1 As shown, an image color correction method of the present invention comprises the following steps:

[0043] Step S1, according to historical space-based remote sensing images, obtaining the quantization level, image shooting time and location of the space-based remote sensing images, and constructing a corresponding relationship model between the corresponding time and location;

[0044] Step S2: Based on historical space-based remote sensing images, obtain atmospheric absorption and ground object color composition, and build a color correction model;

[0045] Step S3: After acquiring a new space-based remote sensing image, the space-based remote sensing image is corrected using the correspondence model and the color correction model in sequence.

[0046] In this embodiment, based on the historical space-based remote sensing images, the quantization level, image shooting time and location of the space-based remote sensing images are obtained to form a corresponding relationship between the shooting time and the shooting location with the new space-based remote sensing images. Based on the historical space-based remote sensing images, the atmospheric absorption conditions and the color composition of the ground objects are obtained, a color correction model is constructed, and a corresponding relationship between the historical space-based remote sensing images before and after correction is established. When a new space-based remote sensing image is obtained, a color correction model is selected according to the corresponding relationship model, and then the new space-based remote sensing image is corrected. The probability model of the historical images is used to count the same or similar areas in the past, an artificial intelligence network is constructed, and color restoration is performed based on the existing image data. This can avoid the randomness of color correction caused by the uncertainty of the atmospheric macroscale and the local microscale, and there is no need to consider the complex settings in the atmospheric model. It is easy to use, and the operator does not need to have professional knowledge of the atmospheric model, which is conducive to the color restoration of the image.

[0047] In one embodiment of the present invention, preferably, in the step S1, it is specifically included:

[0048] Step S11, obtaining historical space-based remote sensing images;

[0049] Step S12: Determine the image shooting time, location, and quantization level based on historical space-based remote sensing images;

[0050] Step S13: Calculate the theoretical quantitative index range according to the quantization level of historical space-based remote sensing images.

[0051] In one embodiment of the present invention, preferably, in the step S2, it is specifically included:

[0052] Step S21: Count the pixel values ​​of typical objects based on historical space-based remote sensing images, and obtain the near-infrared spectrum NIR values ​​of corresponding pixel points to confirm the historical atmospheric absorption of the point;

[0053] Step S22, calculating discrete values ​​according to the quantization level of the historical space-based remote sensing image;

[0054] Step S13, calculating the color combination of common ground objects in the RGB color mode and the relationship between the color combination and the near infrared spectrum NIR value;

[0055] Step S14, converting the RGB color space into the HSV color space value;

[0056] Step S15: Obtain historical local weather information, and construct a color correction model based on the saturation under different cloud cover conditions and the color relationship of different objects in the area.

[0057] In this embodiment, "quantization" is a process that refers to dividing the physical quantity (analog quantity) represented by the image into limited discrete levels (digital quantity) according to a certain functional relationship; according to the level of quantization, the maximum number of colors that the sensor can obtain is calculated, such as 8 bits for 256 colors (i.e. 256 colors) and 16 bits for 65536 colors (i.e. 65536 colors or color gradations). For example, 16 bits means that under the most accurate conditions, the sensor can only distinguish more than 60,000 colors.

[0058] Typical landforms include grasslands, roads, oceans, deserts, etc.; and the flag absorption conditions refer to water vapor, mist, haze, etc.

[0059] In one embodiment of the present invention, preferably, in the step S3, it is specifically included:

[0060] Step S31, removing clouds and fog from the new space-based remote sensing image;

[0061] Step S32: selecting a color correction model using a corresponding relationship model according to the shooting settings and shooting position of the new space-based remote sensing image;

[0062] Step S33: Perform color correction using the corresponding color correction model.

[0063] In one embodiment of the present invention, preferably, in the step S14, the color space value at least includes: hue and hue value, saturation and color purity value, and lightness value.

[0064] In step S14, the RGB color is converted into HSV, Hue (hue), Saturation (saturation, color purity), Value (brightness), etc. according to a general formula.

[0065] like Figure 2 As shown, the specific process of the image color correction method is as follows:

[0066] Based on the historical archived remote sensing image data before and after correction, the basic information of the historical archived remote sensing images is obtained, including the shooting time, shooting location and quantization level. The theoretical quantitative index range is calculated according to the quantization level. At the same time, the historical archived remote sensing image data is used for data splitting. The historical atmospheric absorption of the point is confirmed by using near infrared NIR, and the color composition is extracted by classifying the main ground objects using multi-spectrum. Then, the corresponding relationship model of the corresponding time and corresponding position under the de-clouding condition is constructed according to the color composition and the theoretical quantitative index range, and then the color correction model is obtained according to the color composition and atmospheric absorption.

[0067] After the color correction data is obtained, a color correction model is selected according to the corresponding relationship model, and then the color correction is performed using the selected color correction model.

[0068] According to one aspect of the present invention, an electronic device is provided, comprising: one or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are stored in the memory, and when the electronic device is running, the processor executes the one or more computer programs stored in the memory so that the electronic device performs the image color correction method as described in any one of the above technical solutions.

[0069] According to one aspect of the present invention, a computer-readable storage medium is provided for storing computer instructions. When the computer instructions are executed by a processor, an image color correction method as described in any one of the above technical solutions is implemented.

[0070] In summary, the present invention proposes an image color correction method, an electronic device and a computer-readable storage medium. According to historical space-based remote sensing images, the quantization level, image shooting time and position of the space-based remote sensing image are obtained, which are used to form a corresponding relationship between the shooting time and the shooting position with the new space-based remote sensing image. According to the historical space-based remote sensing image, the atmospheric absorption and the color composition of the ground object are obtained, a color correction model is constructed, and a corresponding relationship between the historical space-based remote sensing image before and after correction is established. When a new space-based remote sensing image is obtained, a color correction model is selected according to the corresponding relationship model, and then the new space-based remote sensing image is corrected. The probability model of the historical image is used to count the same or similar areas in the past, an artificial intelligence network is constructed, and color restoration is performed on the basis of the existing image data. The problem of randomness of color correction caused by the uncertainty of the atmospheric macroscale and the local microscale can be avoided. There is no need to consider the complex settings in the atmospheric model. It is easy to use, and the operator does not need to have professional knowledge of the atmospheric model, which is conducive to the color restoration of the image.

[0071] In addition, it should be noted that the present invention can be provided as a method, an apparatus or a computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0072] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0073] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0074] It should also be noted that, in this article, the terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0075] Finally, it should be noted that the above is a preferred embodiment of the present invention. It should be pointed out that although the preferred embodiment of the present invention has been described, for those skilled in the art, once the basic creative concept of the present invention is known, several improvements and modifications can be made without departing from the principles of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention. Therefore, the attached claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the present invention.

Claims

1. A method for image color correction, comprising the following steps: Step S1, according to historical space-based remote sensing images, obtaining the quantization level, image shooting time and location of the space-based remote sensing images, and constructing a corresponding relationship model between the corresponding time and location; Step S2: Based on historical space-based remote sensing images, obtain atmospheric absorption and ground object color composition, and build a color correction model; Step S3, after acquiring a new space-based remote sensing image, correcting the space-based remote sensing image using the correspondence model and the color correction model in sequence; In the step S2, it specifically includes: Step S21: Count the pixel values ​​of typical objects based on historical space-based remote sensing images, and obtain the near-infrared spectrum NIR values ​​of corresponding pixel points to confirm the historical atmospheric absorption of the point; Step S22, calculating discrete values ​​according to the quantization level of the historical space-based remote sensing image; Step S23, calculating the color combination of common ground objects in the RGB color mode and the relationship between the color combination and the near infrared spectrum NIR value; Step S24, converting the RGB color space into the HSV color space value; Step S25, obtaining historical local weather information, and constructing a color correction model based on the saturation under different cloud cover conditions and the color relationship of different objects in the area; In the step S3, it specifically includes: Step S31, removing clouds and fog from the new space-based remote sensing image; Step S32: selecting a color correction model using a corresponding relationship model according to the shooting settings and shooting position of the new space-based remote sensing image; Step S33: Perform color correction using the corresponding color correction model.

2. The method according to claim 1, characterized in that In the step S1, it specifically includes: Step S11, obtaining historical space-based remote sensing images; Step S12: Determine the image shooting time, location, and quantization level based on historical space-based remote sensing images; Step S13: Calculate the theoretical quantitative index range according to the quantization level of historical space-based remote sensing images.

3. The method according to claim 1, characterized in that In step S24, the color space value includes at least: hue and hue value, saturation and color purity value, and brightness value.

4. The method according to claim 1, characterized in that After the correction of the new space-based remote sensing image is completed, the new space-based remote sensing image is used as a historical space-based remote sensing image to correct the color correction model and the corresponding relationship model.

5. An electronic device, characterized in that: include: One or more processors, one or more memories, and one or more computer programs; wherein the processor is connected to the memory, and the one or more computer programs are stored in the memory. When the electronic device is running, the processor executes the one or more computer programs stored in the memory, so that the electronic device performs the image color correction method as described in any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, implement the image color correction method as claimed in any one of claims 1 to 4.

Images