Method for Evaluating Synthetic Aperture Radar Image Quality under Inaccurate Imaging Parameters

Abstract

The present invention relates to a method for evaluating the quality of synthetic aperture radar images under inaccurate imaging parameters, comprising the following steps: selecting a variety of image quality evaluation functions and performing weighted summation to obtain a weighted minimum deviation evaluation function; according to different imaging scenarios to be measured, using the weighted minimum deviation evaluation function with the corresponding optimal weighting coefficient to evaluate the quality of synthetic aperture radar images with inaccurate imaging parameters under the imaging scenarios to be measured. Compared with the prior art, the present invention has the advantages of high flexibility, strong scene pertinence, and better reflection of image quality, etc.

Description

Technical Field

[0001] The present invention relates to the technical field of synthetic aperture radar (SAR) image evaluation, and particularly to a method for evaluating the quality of SAR images under inaccurate imaging parameters. Background Art

[0002] The quality of synthetic aperture radar (SAR) images reflects to a certain extent the performance of the SAR system, and also largely determines the application effectiveness of the SAR. Therefore, it is of great significance to evaluate the quality of SAR images. SAR images have unique problems such as speckle noise and ghosting, which are significantly different from optical images. Therefore, it is necessary to specifically design an image quality evaluation scheme for SAR. During the SAR imaging process, numerous factors can lead to inaccurate imaging parameters. For example, the movement of the platform carrying the SAR is unstable, or the characteristics of the electromagnetic wave conduction medium change due to changes in meteorological conditions. Imaging with inaccurate parameters will lead to the deterioration of the imaging result, thereby affecting the application of SAR images. Currently, there is no specific method for evaluating the quality of SAR images under inaccurate imaging parameters. Therefore, when the imaging parameters are inaccurate, the imaging parameters cannot be calibrated to improve the imaging effect.

[0003] The existing methods for evaluating the quality of SAR images are mainly divided into two categories: subjective evaluation methods and objective evaluation methods. Subjective evaluation methods evaluate the quality of images based on the subjective feelings of viewers. The advantage is that it can directly reflect the human eye's perception and is more in line with reality. The disadvantages are that it is time-consuming and laborious, not easy to operate, and may be affected by personal factors of the viewers. Objective evaluation methods evaluate by calculating the objective quality indicators of the images, aiming to meet the requirements consistent with human subjective feelings. Traditional objective evaluation methods mainly measure the quality of images from two aspects: point targets and area targets. The evaluation indicators for point targets include: spatial resolution, peak sidelobe ratio, integrated sidelobe ratio, ambiguity, etc. The evaluation indicators for area targets include: mean and variance, entropy, radiometric resolution, equivalent number of looks, dynamic range, etc. These indicators can reflect the system performance, but they may not be consistent with the human eye's perception. Subsequently, image quality evaluation methods based on contrast, texture features, edge features, the human visual system, etc. have been developed, such as the Tenengrad criterion, HVSNR (Human Visual System Signal-to-Noise Ratio), etc. However, these methods are not proposed for the case of inaccurate imaging parameters and do not consider the impact of the deviation of imaging parameters from the standard values on the imaging result. Some image quality evaluation methods were selected and their evaluation performances under inaccurate imaging parameters were tested. The results show that the performances of these methods are not ideal. The specific work is as follows:

[0004] Four existing image quality assessment methods are selected: variance, entropy, equivalent number of looks, and Tenengrad criterion for testing. A piece of SAR raw echo data with known imaging parameters is found and imaged using the simplest two-dimensional matched filtering algorithm. The parameters of the matched filter are deliberately deviated from the standard values, and images under different deviation degrees are obtained. Then, the function values corresponding to these four evaluation functions are calculated. The results are as Figure 1 shown. Except for entropy, the larger the value of the other three evaluation functions, the better the image quality. To maintain this consistency, the opposite of entropy is taken. For easy comparison, these values are scaled so that all values fall between 95 and 100. An ideal evaluation function should achieve the maximum value when the imaging parameter deviation is 0, the minimum value when the deviation is the largest, and the value decreases as the deviation increases, as shown by the blue line in Figure 1 . However, from Figure 1 it can be seen that none of these four evaluation functions have these characteristics. Other data are also tried, and the results are not very satisfactory. As described above, traditional image quality assessment methods cannot correctly reflect the image quality when the imaging parameters are inaccurate. Therefore, it is necessary to propose an effective SAR image quality assessment method for the case of inaccurate imaging parameters. Summary of the Invention

[0005] The purpose of the present invention is to provide a synthetic aperture radar image quality assessment method for the case of inaccurate imaging parameters to better reflect the image quality, overcoming the defects of the above-mentioned existing technologies.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A synthetic aperture radar image quality assessment method for the case of inaccurate imaging parameters includes the following steps:

[0008] Select a variety of image quality evaluation functions and perform weighted summation to obtain a weighted minimum deviation evaluation function; according to different imaging scenarios to be measured, use the weighted minimum deviation evaluation function with the corresponding optimal weighting coefficient to evaluate the quality of synthetic aperture radar images with inaccurate imaging parameters under the imaging scenarios to be measured;

[0009] The process of obtaining the optimal weighting coefficient includes: obtaining the synthetic aperture radar raw echo and the corresponding imaging parameters under the current imaging scenario, applying different deviation degrees to the imaging parameters, obtaining multiple imaging images and the corresponding standard function values;

[0010] Randomly generate the weighting coefficients in the weighted minimum deviation evaluation function, evaluate each imaging image according to the weighted minimum deviation evaluation function, obtain multiple function values, and calculate the mean square error with the corresponding standard function value. Then, continuously iterate and optimize the weighting coefficients using an optimization algorithm to minimize the mean square error and obtain the optimal weighting coefficients under the current imaging scenario.

[0011] Furthermore, the weighted minimum deviation evaluation function is a weighted sum of multiple functions among the image quality evaluation function based on variance, the image quality evaluation function based on entropy, the image quality evaluation function based on equivalent number of looks, and the image quality evaluation function based on the Tenengrad criterion.

[0012] Furthermore, the expression of the weighted minimum deviation evaluation function is:

[0013] WMD = c1·V + c2·H + c3·ENL + c4·TEN

[0014] In the formula, WMD is the evaluation result of the weighted minimum deviation evaluation function, V is the evaluation result of the image quality evaluation function based on variance, H is the evaluation result of the image quality evaluation function based on entropy, ENL is the evaluation result of the image quality evaluation function based on equivalent number of looks, TEN is the evaluation result of the image quality evaluation function based on the Tenengrad criterion, c1 is the first weighting coefficient, c2 is the second weighting coefficient, c3 is the third weighting coefficient, and c4 is the fourth weighting coefficient.

[0015] Furthermore, the weighting coefficients of the weighted minimum deviation evaluation function are different under different imaging scenarios.

[0016] Furthermore, the weighting coefficients in the weighted minimum deviation evaluation function are randomly generated within the interval [-1, 1].

[0017] Furthermore, use the particle swarm optimization algorithm to continuously iterate and optimize the weighting coefficients according to the calculated mean square error.

[0018] Furthermore, the specific way of imposing different deviation degrees on the imaging parameters is: making the imaging parameters deviate from the accurate value by m%; the corresponding standard function value is 100 - |m|.

[0019] Furthermore, the method also includes guiding the selection of corresponding imaging parameters according to the quality evaluation result of the obtained synthetic aperture radar image to improve the imaging effect.

[0020] Furthermore, use the imaging parameters corresponding to the maximum evaluation function value of the weighted minimum deviation evaluation function as the true imaging parameters.

[0021] Furthermore, the selected multiple image quality evaluation functions are all objective evaluation methods for synthetic aperture radar image quality.

[0022] Compared with the prior art, the present invention has the following advantages:

[0023] (1) The present invention uses the weighted sum of several image quality evaluation functions as the evaluation index and uses an optimization algorithm to determine the weighting coefficients. In the case of inaccurate imaging parameters, compared with variance, entropy, equivalent number of looks, and Tenengrad criterion, the weighted minimum deviation evaluation function WMDEF can better reflect the image quality.

[0024] The advantages of this method are as follows: The design idea is simple and has strong flexibility; it has strong scene specificity. Different image quality evaluation functions can be selected to calculate the weighted sum, different optimization algorithms can be used to determine the weighting coefficients, and different standard function values can be selected to make it applicable to different imaging scenes. Specific weighting coefficients will be obtained for a specific scene to make it well match the characteristics of that scene.

[0025] (2) The weighted minimum deviation evaluation function WMDEF of the present invention provides a new solution for the SAR image quality evaluation under inaccurate imaging parameters. WMDEF can be used for imaging parameter optimization and estimation under inaccurate imaging parameters. When the true imaging parameters are unknown, the currently used imaging parameters can be adjusted according to the evaluation results of WMDEF to achieve a better imaging effect. The true imaging parameters can be estimated by finding the imaging parameters corresponding to the maximum function value of WMDEF. Inspired by WMDEF, new image quality evaluation methods may be born in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram showing the variation of the values of different evaluation functions provided in the background art of the present invention with the deviation degree of imaging parameters;

[0027] Figure 2 It is a schematic flow diagram of a method for evaluating the synthetic aperture radar image quality under inaccurate imaging parameters provided in the embodiment of the present invention;

[0028] Figure 3 It is a schematic diagram showing the calculation process of the WMDEF coefficient provided in the embodiment of the present invention;

[0029] Figure 4 It is a schematic diagram showing the variation of WMDEF with the deviation of imaging parameters provided in the embodiment of the present invention;

[0030] Figure 5 It is a schematic comparison diagram of WMDEF and four other evaluation functions provided in the embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Components of the embodiments of the present invention described and illustrated herein generally may be arranged and designed in a variety of different configurations.

[0032] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0033] It should be noted that like reference numerals and letters denote like items in the following drawings. Therefore, once an item is defined in one drawing, it does not require further definition and explanation in subsequent drawings.

[0034] It should be noted that the terms "first" and "second" are used for descriptive purposes only and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, "a plurality of" means two or more unless otherwise specifically defined.

[0035] Embodiment 1

[0036] As Figure 2 shown, this embodiment provides a method for evaluating the quality of synthetic aperture radar images in the case of inaccurate imaging parameters, including the following steps:

[0037] S1: Select a variety of image quality evaluation functions and perform weighted summation to obtain a weighted minimum deviation evaluation function (WMDEF);

[0038] S2: Obtain the original echo of the synthetic aperture radar and the corresponding imaging parameters in the current imaging scene, apply different degrees of deviation to the imaging parameters, and obtain multiple imaging images and the corresponding standard function values;

[0039] S3: Randomly generate the weighting coefficients in the weighted minimum deviation evaluation function, evaluate each imaging image according to the weighted minimum deviation evaluation function, obtain multiple function values, and calculate the mean square error with the corresponding standard function value. Then, continuously iterate and optimize the weighting coefficients using an optimization algorithm to minimize the mean square error and obtain the optimal weighting coefficients for the current imaging scenario.

[0040] S4: According to different imaging scenarios to be measured, use the weighted minimum deviation evaluation function with the corresponding optimal weighting coefficients to evaluate the quality of synthetic aperture radar images with inaccurate imaging parameters in the imaging scenarios to be measured.

[0041] The weighted minimum deviation evaluation function WMDEF can select different image quality evaluation functions to calculate the weighted sum. For example, multiple functions can be selected from the variance-based image quality evaluation function, entropy-based image quality evaluation function, equivalent number of looks-based image quality evaluation function, and Tenengrad criterion-based image quality evaluation function. Different optimization algorithms can also be used to determine the weighting coefficients.

[0042] In this embodiment, the weighted minimum deviation evaluation function WMDEF is defined as the weighted sum of four image quality evaluation functions: variance (V), entropy (H), equivalent number of looks (ENL), and Tenengrad criterion (TEN). The corresponding calculation expression is:

[0043] WMD = c1·V + c2·H + c3·ENL + c4·TEN

[0044] In the formula, WMD is the evaluation result of the weighted minimum deviation evaluation function, V is the evaluation result of the variance-based image quality evaluation function, H is the evaluation result of the entropy-based image quality evaluation function, ENL is the evaluation result of the equivalent number of looks-based image quality evaluation function, TEN is the evaluation result of the Tenengrad criterion-based image quality evaluation function, c1 is the first weighting coefficient, c2 is the second weighting coefficient, c3 is the third weighting coefficient, and c4 is the fourth weighting coefficient. Among them, the weighting coefficients c1, c2, c3, and c4 are obtained by the particle swarm algorithm.

[0045] The weighting coefficients in WMDEF vary according to different application scenarios. If you want to use WMDEF in a certain scenario, first, you need a set of imaging results with different deviation degrees of imaging parameters in this scenario and select a standard function with the deviation degree as the independent variable to reflect the impact of the deviation result on imaging.

[0046] For example, as Figure 3As shown, the selected standard deviation function is 100 - |m|, where m is 100 times the degree of deviation, and the accurate value of the imaging parameter is a. The imaging parameter deviates from the accurate value by -5%, -4.5%,..., m% respectively to obtain n images. The specific method for calculating the weighting coefficient is as follows: First, randomly generate four weighting coefficients within the range of [-1, 1]. Then, calculate the function values of WMDEF under different degrees of deviation, and calculate the mean square error between these function values and the standard function value (normalization of these values can be performed if necessary). Use the particle swarm algorithm to continuously iterate the coefficients to minimize the mean square error. The coefficients obtained at this time are the coefficients of WMDEF in this scenario.

[0047] This embodiment also verifies the effectiveness of this method through experiments and draws the conclusion that this method provides an effective solution for image quality evaluation in inaccurate situations and is beneficial to the improvement of image quality in such situations.

[0048] The specific process is as follows:

[0049] Use a set of data to test the performance of WMDEF, and the variation of the WMDEF function value with the deviation of the imaging parameter is shown as Figure 4 shown, and the variation trend of the function meets the expectation.

[0050] To verify the general applicability of WMDEF, this embodiment also uses several groups of SAR raw echo data, performs the same imaging processing, and draws the variation of the function values of WMDEF and other four image quality evaluation functions with the deviation of the imaging parameter, and finds that WMDEF can accurately reflect the variation trend of image quality. One set of results is shown as Figure 5 shown.

[0051] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art should fall within the protection scope determined by the claims.

Claims

1. A method for evaluating the quality of synthetic aperture radar images under inaccurate imaging parameters, characterized in that It includes the following steps: Select multiple image quality evaluation functions, and perform weighted summation to obtain a weighted minimum deviation evaluation function; according to different imaging scenarios to be measured, use the weighted minimum deviation evaluation function with the corresponding optimal weighted coefficient to evaluate the synthetic aperture radar image with inaccurate imaging parameters in the imaging scenario to be measured; The process of obtaining the optimal weighted coefficient includes: obtaining the original echo of the synthetic aperture radar and the corresponding imaging parameters in the current imaging scenario, applying different deviation degrees to the imaging parameters, and obtaining multiple imaging images and the corresponding standard function values; the specific method of applying different deviation degrees to the imaging parameters is: making the imaging parameters deviate from the accurate value by m%; the corresponding standard function value is 100 - |m|; Randomly generate the weighted coefficients in the weighted minimum deviation evaluation function, evaluate each imaging image according to the weighted minimum deviation evaluation function, obtain multiple function values, and calculate the mean square error with the corresponding standard function values, so as to continuously iterate and optimize the weighted coefficients using an optimization algorithm to minimize the mean square error and obtain the optimal weighted coefficient in the current imaging scenario.

2. The synthetic aperture radar image quality evaluation method under inaccurate imaging parameters according to claim 1, characterized in that The weighted minimum deviation evaluation function is the weighted sum of multiple functions among the image quality evaluation function based on variance, the image quality evaluation function based on entropy, the image quality evaluation function based on equivalent number of looks, and the image quality evaluation function based on Tenengrad criterion.

3. The synthetic aperture radar image quality evaluation method under inaccurate imaging parameters according to claim 2, characterized in that The expression of the weighted minimum deviation evaluation function is: WMD = c1·V + c2·H + c3·ENL + c4·TEN In the formula, WMD is the evaluation result of the weighted minimum deviation evaluation function, V is the evaluation result of the image quality evaluation function based on variance, H is the evaluation result of the image quality evaluation function based on entropy, ENL is the evaluation result of the image quality evaluation function based on equivalent number of looks, TEN is the evaluation result of the image quality evaluation function based on Tenengrad criterion, c1 is the first weighted coefficient, c2 is the second weighted coefficient, c3 is the third weighted coefficient, and c4 is the fourth weighted coefficient.

4. The synthetic aperture radar image quality evaluation method under inaccurate imaging parameters according to claim 1, wherein The weighted coefficients of the weighted minimum deviation evaluation function in different imaging scenarios are different.

5. A method for evaluating the quality of synthetic aperture radar images in the case of inaccurate imaging parameters according to claim 1, characterized in that, The weighted coefficients in the weighted minimum deviation evaluation function are randomly generated within the interval [-1, 1].

6. The synthetic aperture radar image quality evaluation method under inaccurate imaging parameters according to claim 1, wherein Use the particle swarm algorithm to continuously iterate and optimize the weighted coefficients according to the calculated mean square error.

7. A method for evaluating the quality of synthetic aperture radar images in the case of inaccurate imaging parameters according to claim 1, characterized in that, The method further includes guiding the selection of the corresponding imaging parameters according to the quality evaluation result of the obtained synthetic aperture radar image to improve the imaging effect.

8. A method for evaluating the quality of synthetic aperture radar images in the case of inaccurate imaging parameters according to claim 1, characterized in that Take the imaging parameters corresponding to the maximum evaluation function value of the weighted minimum deviation evaluation function as the true imaging parameters.

9. A method for evaluating the quality of synthetic aperture radar images in the case of inaccurate imaging parameters according to claim 1, characterized in that The multiple selected image quality evaluation functions are all objective evaluation functions for the quality of synthetic aperture radar images.

Images