Multi-target color correction method and device for SAR (Synthetic Aperture Radar) image

Abstract

The invention provides a multi-target color correction method and device for an SAR image. The method comprises the steps that an original image is preprocessed; performing linear stretching transformation based on the image gray value to construct a multi-target optimization model; solving linear stretching parameters of all the images based on a multi-objective optimization model; and carrying out stretching processing on the original image by using the obtained linear stretching parameter, and then embedding. According to the method, the linear stretching coefficient of the SAR image is solved based on the multi-target optimization idea, gray value information on the SAR image is effectively utilized, and the inlaid SAR image shows color consistency.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a multi-target color correction method and device for SAR images. Background Art

[0002] Seasonal changes, radar signal attenuation, noise and other factors can reduce the radiometric quality of images, resulting in significant brightness differences in SAR images even after calibration. If there is an abnormal brightness in a SAR image, the image will appear dark as a whole. Most radiometric correction methods for SAR images are based on optical images. Depending on the scope of application, radiometric correction methods for multiple images can be divided into global, local or combined models. Among them, the global model is a mainstream method. The global model assumes that the radiometric relationship between image pairs can be fitted by a linear or nonlinear function. The linear model assumes that the intensity difference between different images is determined by a linear relationship. In order to describe the linear relationship between images, some researchers have constructed linear models based on pseudo-invariant features (PIFs). Histogram matching is a nonlinear model that matches the histograms of two images so that the explicit distribution of pixel values ​​in the two images is as close as possible. However, when the objects in the image change or the same object has significant differences due to factors such as seasons, the histogram matching method cannot identify these differences. Summary of the invention

[0003] The purpose of the present invention is to solve the technical problems in the prior art and provide a multi-target color correction method for SAR images, which simultaneously optimizes the two goals of maximum color consistency and minimum information loss of stretched images.

[0004] In order to achieve the above object, the present invention proposes a multi-target color correction method for SAR images, including the following processing:

[0005] Preprocess the original image;

[0006] Construct a multi-objective optimization model based on linear stretching transformation of image grayscale values;

[0007] Solve the linear stretching parameters of all images based on a multi-objective optimization model;

[0008] The original image is stretched using the obtained linear stretching parameters and then mosaicked.

[0009] Moreover, the preprocessing of the original images includes obtaining the total number of all images, the maximum pixel grayscale value of each image, the minimum pixel grayscale value of each image, the grayscale mean of each image, the standard deviation of each image, the number of pixels in each image, the grayscale mean of image pairs with overlapping areas, the standard deviation of image pairs with overlapping areas and the number of valid pixels of image pairs with overlapping areas.

[0010] Moreover, the multi-objective optimization model is constructed based on the linear stretching transformation of the image grayscale value, including taking the linear stretching parameter as the decision variable, setting two objective functions of simultaneously optimizing the maximum color consistency of the image after stretching and the minimum information loss, setting an equality constraint based on maintaining the overall brightness and contrast of the image before and after correction, and setting an inequality constraint based on the grayscale value of the pixel after correction satisfying the quantization bit limit.

[0011] Moreover, when setting the objective function with the highest color consistency, the color consistency optimization target is constructed with the mean and standard deviation of the indicators in the overlapping area as observation values.

[0012] Moreover, when setting the objective function of minimizing the amount of information loss, the minimum total number of pixels whose grayscale values ​​exceed the limit after the original image is stretched is taken as the optimization target of the amount of information loss.

[0013] Furthermore, the equation constraint is set based on maintaining the overall brightness and contrast of the image before and after correction, and is implemented by setting a condition requiring that the sum of grayscale means and the sum of standard deviations of all overlapping areas of a single image remain unchanged before and after stretching.

[0014] Furthermore, the inequality constraint is set based on the grayscale value of the pixel after correction satisfying the limit of the number of quantization bits, and is implemented by setting a condition requiring the maximum grayscale value of the image after stretching to satisfy the value range before transformation.

[0015] Moreover, the multi-objective optimization model-based solution of the linear stretching parameters of all images includes the fusion of the meta-heuristic algorithm NSAG-II and the obstacle function interior point method for solution, wherein the obstacle function interior point method is used to calculate the objective function value of the color consistency optimization goal.

[0016] On the other hand, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the multi-target color correction method for SAR images as described above is implemented.

[0017] On the other hand, the present invention further provides a computer program product, comprising a computer program, characterized in that: when the computer program is executed by a processor, the multi-target color correction method for SAR images as described above is implemented.

[0018] The present invention performs color correction processing on SAR images based on a multi-objective optimization method, effectively utilizes the pixel information on the SAR images, and makes the corrected SAR images present more consistent color tones.

[0019] The solution of the present invention is simple and convenient to implement, has strong practicality, solves the problems of low practicality and inconvenience in actual application existing in related technologies, can improve user experience, and has important market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a specific flow chart of the multi-target color correction method for SAR images according to an embodiment of the present invention;

[0021] Figure 2 This is a mosaic effect diagram of the original SAR image according to an embodiment of the present invention;

[0022] Figure 3 This is a rendering of a multi-target color correction processing method for SAR images according to an embodiment of the present invention;

[0023] Figure 4 This is a diagram showing the effect of the comparison method in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0025] See also Figure 1 ,The embodiment of the present invention provides a multi-target color correction method for SAR images, including the following processes, S1) preprocessing the original image;

[0026] In the embodiment, the preprocessing of the original image is implemented as follows:

[0027] Get the following data: total number of all images n, maximum pixel gray value of each image Minimum pixel gray value for each image The mean gray value of each image μ i , standard deviation of each image σ i , the number of pixels per image s i , the grayscale mean μ of the image pair with overlapping areas ij , μ ji , the standard deviation of the image pairs with overlapping areas σ ij , σ ji , the number of valid pixels S of the image pairs with overlapping areas ij . Where i and j are image labels, i≠j.

[0028] S2) Performing linear stretching transformation based on the grayscale value of the image and constructing a multi-objective optimization model;

[0029] In the embodiment, the linear stretching transformation is performed based on the gray value of the image to construct a multi-objective optimization model, which is implemented as follows:

[0030] With linear stretch parameter a i 、b iAs the decision variable, the two goals of maximum color consistency and minimum information loss of the stretched image are optimized simultaneously. The overall brightness and contrast of the image before and after correction remain unchanged as an equality constraint, and the grayscale value of the pixel after correction satisfies the quantization bit limit as an inequality constraint. The optimization model is expressed as:

[0031]

[0032] A={Y i ∈N * |a i Y i +b i >2 m -1,a i Y i +b i <0}

[0033]

[0034] stmin(y′)≥min(Y1);

[0035] max(y′)≤max(Y1);

[0036]

[0037]

[0038] Among them, n is the total number of images, m is the number of image quantization bits, μ ij , μ ji are the grayscale mean of the overlapping areas of i and j, σ ij , σ ji are the standard deviations of the overlapping areas of i and j, S ij is the number of pixels in the overlapping area of ​​images i and j, a i 、b i is the linear stretching coefficient to be determined, y′ is the gray value of the image after linear stretching, min(Y1) and max(Y1) are the minimum and maximum gray values ​​allowed for this type of image, respectively, s i is the number of pixels in image i, μ i is the grayscale mean of image i, σ i is the standard deviation of image i; Y i is the gray value set of the i-th original image.

[0039] The specific construction instructions of the optimization model are as follows:

[0040] The color consistency optimization objective is constructed with the mean and standard deviation of the indicators in the overlapping area as observation values, that is, objective function 1, which can be expressed as:

[0041]

[0042] Where n is the total number of images, μ ij , μ ji are the grayscale mean of the overlapping areas of i and j, σ ij , σ ji are the standard deviations of the overlapping areas of i and j, S ij is the number of pixels in the overlapping area of ​​images i and j.

[0043] The optimization target of information loss is to minimize the total number of pixels whose grayscale values ​​exceed the limit after the original image is stretched, that is, objective function 2, which can be expressed as:

[0044] A={Y i ∈N * |a i Y i +b i >2 m -1, a i Y i +b i <0}

[0045]

[0046] Among them, a i 、b i is the linear stretch coefficient to be determined, Y i is the gray value set of the original image of the i-th scene, m is the number of image quantization bits, N * is a positive integer, num(A) is the number of pixels that exceed the image quantization range, and minimize is the minimization objective function.

[0047] In order to use the grayscale value of the corrected pixel to satisfy the quantization bit limit as an inequality constraint, the embodiment further proposes that the grayscale value of the model after stretching needs to satisfy the value range allowed by the image category, which can be expressed as:

[0048] min(y′)≥min(Y1)

[0049] max(y′)≤max(Y1)

[0050] Wherein, y′ is the gray value of the image after linear stretching, min(y′) and max(y′) are the minimum and maximum gray values ​​of the image after linear stretching, min(Y1) and max(Y1) are the minimum and maximum gray values ​​allowed for such images, respectively. The scheme of the present invention is mainly designed for SAR images, and the value range of SAR images is 0 to 65535 (16 bits);

[0051] In order to set the equation constraint based on maintaining the overall brightness and contrast of the image before and after correction, the embodiment further proposes the equation constraint condition requirement of the model, that the sum of the grayscale means and the sum of the standard deviations of all overlapping areas of a single image remain unchanged before and after stretching.

[0052]

[0053] Where n is the total number of images, s i is the number of pixels in the overlapping area of ​​image i, a i , b i is the decision variable to be determined, μ i is the grayscale mean of the overlapping area of ​​image i;

[0054]

[0055] Where n is the total number of images, s i is the number of pixels in the overlapping area of ​​image i, a i is the decision variable to be determined, σ i is the standard deviation of the overlapping area of ​​image i.

[0056] S3) solving the multi-objective optimization model established in S2) to obtain linear stretching parameters of all images;

[0057] In the embodiment, in the case of a model with multiple constraints, a continuous objective function, and a discrete objective function, a fusion algorithm is preferably proposed that fully combines the global search capability of the meta-heuristic algorithm NSAG-II and the ability of the barrier function interior point method to accurately solve the quadratic programming model. The implementation process includes the following steps:

[0058] Step 1: Generate the initial population, i.e. the linear stretch coefficient of the decision variable;

[0059] In the specific implementation, it is preferably recommended to use the NSGA-II algorithm to generate the initial population. The detailed implementation description of the NSGA-II algorithm can be found in the prior art literature, which will not be repeated in the present invention: W. Zheng and B. Doerr, "Approximation Guarantees for the Non-Dominated Sorting Genetic Algorithm II (NSGA-II)," IEEE Trans. Evol. Computat., pp. 1–1, 2024, doi: 10.1109 / TEVC.2024.3402996.

[0060] Step 2: Calculate the fitness value of each individual in the initial population

[0061] (1) Calculate the objective function value 1 for each individual in the initial population;

[0062] In the specific implementation, it is preferably recommended to use the obstacle function interior point method to calculate the objective function value of the color consistency optimization target. The specific description of the obstacle function interior point method can be found in the prior art literature, which will not be described in detail in the present invention: P. Malisani, "Interior Point Methods in Optimal Control", doi: 10.1051 / cocv / 2024049.

[0063] (2) Calculate the objective function value 2 of each individual in the initial population; in specific implementation, it is preferably recommended to use a statistical method (i.e., counting accumulation) to calculate the objective function value of the information loss optimization target.

[0064] Step 3: Perform fast non-dominated sorting on the two objective function values ​​(the objective function values ​​of the color consistency optimization objective and the information loss optimization objective), and generate a new offspring population using selection, crossover and mutation operations;

[0065] Step 4: Merge the parent population and the offspring population;

[0066] Step 5: Perform fast non-dominated sorting and crowding distance calculation on the merged population, and select excellent individuals to form a new parent population;

[0067] Step 6: Use selection, crossover, mutation and other operations to generate a new offspring population, merge the new parent population with the new offspring population to form an updated population;

[0068] Step 7: Repeat steps 5 and 6. Stop the iteration when the number of iterations exceeds the maximum limit;

[0069] Step 8: Output the final Pareto front solution and terminate the algorithm. Each solution on the front surface represents a set of stretching parameters for all images.

[0070] S4) stretching the original image using the obtained linear stretching parameters and then mosaicking the original image.

[0071] In specific implementation, if the longitude and latitude of adjacent SAR images overlap, there will generally be a certain overlap area. In this embodiment, the original image with the overlapping area is stretched using the obtained linear stretching parameter, and the implementation method is to use the obtained decision variable a i 、b i The grayscale value of the original image is linearly stretched using the following formula to obtain a color-corrected image.

[0072] y′=a i Y i +b i

[0073] To facilitate the description of the technical effects of the present invention, Table 1 provides a comparison of the results of applying the present invention and the prior art method. Figure 2 This is a mosaic effect diagram of the original SAR image according to an embodiment of the present invention; Figure 3 This is a rendering of a multi-target color correction processing method for SAR images according to an embodiment of the present invention; Figure 4 The figure is a diagram showing the effect of the comparison algorithm method of the embodiment of the present invention. It can be seen that the results of the present invention are better in many aspects.

[0074] Table 1

[0075]

[0076] For the implementation of the comparison method, please refer to the literature: Cresson R, Saint-Geours N. Natural Color Satellite Image Mosaicking Using Quadratic Programming in Decorrelated Color Space[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 8(8): 4151-4162. DOI: 10.1109 / JSTARS.2015.2449233.

[0077] During specific implementation, the above process can be automatically executed using computer software technology.

[0078] In addition, the present application also relates to an electronic device, comprising at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned multi-target color correction method for SAR images.

[0079] The present application also relates to a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable the computer to execute the above-mentioned multi-target color correction method for SAR images.

[0080] The present application also relates to a computer program product, including a computer program, characterized in that when the computer program is executed by a processor, the multi-target color correction method for SAR images is implemented.

[0081] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0082] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-target color correction method for SAR images, characterized by: Including the following processing, Preprocess the original image; Construct a multi-objective optimization model based on linear stretching transformation of image grayscale values; Solve the linear stretching parameters of all images based on a multi-objective optimization model; The original image is stretched using the obtained linear stretching parameters and then mosaicked.

2. The multi-target color correction method for SAR images according to claim 1, characterized in that: The preprocessing of the original images includes obtaining the total number of all images, the maximum pixel grayscale value of each image, the minimum pixel grayscale value of each image, the grayscale mean of each image, the standard deviation of each image, the number of pixels in each image, the grayscale mean of image pairs with overlapping areas, the standard deviation of image pairs with overlapping areas, and the number of valid pixels of image pairs with overlapping areas.

3. The multi-target color correction method for SAR images according to claim 1, characterized in that: The multi-objective optimization model is constructed based on linear stretching transformation of image grayscale values, including taking linear stretching parameters as decision variables, setting two objective functions of simultaneously optimizing the maximum color consistency of the image after stretching and the minimum information loss, setting equality constraints based on maintaining the overall brightness and contrast of the image before and after correction, and setting inequality constraints based on the grayscale values ​​of the pixels after correction satisfying the quantization bit limit.

4. The multi-target color correction method for SAR images according to claim 3, characterized in that: When setting the objective function for the highest color consistency, the color consistency optimization target is constructed using the mean and standard deviation of the indicators in the overlapping area as observation values.

5. The multi-target color correction method for SAR images according to claim 3, characterized in that: When setting the objective function of minimizing the amount of information loss, the minimum total number of pixels whose grayscale values ​​exceed the limit after the original image is stretched is taken as the optimization target of the amount of information loss.

6. The multi-target color correction method for SAR images according to claim 3, characterized in that: The equation constraint is set based on maintaining the overall brightness and contrast of the image before and after correction, and is implemented by setting a condition requiring that the sum of the grayscale means and the sum of the standard deviations of all overlapping areas of a single image remain unchanged before and after stretching.

7. The multi-target color correction method for SAR images according to claim 3, characterized in that: The inequality constraint is set based on the grayscale value of the pixel after correction satisfying the limit of the number of quantization bits, and is implemented by setting a condition requiring the maximum grayscale value of the image after stretching to satisfy the value range before transformation.

8. The multi-target color correction method for SAR images according to claim 1, characterized in that: The linear stretching parameters of all images are solved based on the multi-objective optimization model, including the fusion of the meta-heuristic algorithm NSAG-II and the barrier function interior point method for solution, wherein the barrier function interior point method is used to calculate the objective function value of the color consistency optimization goal.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the multi-target color correction method for SAR images as claimed in any one of claims 1 to 8 is implemented.

10. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by a processor, the multi-target color correction method for SAR images as claimed in any one of claims 1 to 8 is implemented.

Images