A method for removing oblique stripe noise from SAR images and related equipment

Abstract

The present invention relates to the field of computer technology, and in particular to a method for removing oblique stripe noise from SAR images and related equipment. The method comprises: performing two-dimensional discrete Fourier transform on a SAR image to be processed and an oblique stripe noise region in the SAR image, respectively, to generate a SAR image frequency domain spectrum and a noise region frequency domain spectrum; determining a feature domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and position information of the feature domain relative to the SAR image frequency domain spectrum; performing consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information to generate a processed target SAR image frequency domain spectrum; performing two-dimensional discrete Fourier inverse transform on the target SAR image frequency domain spectrum to generate a target SAR image corresponding to the SAR image. Based on the above method, the oblique stripe noise of the SAR image can be effectively removed, and the loss of original information in the SAR image can be avoided.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method for removing oblique stripe noise from SAR images and related equipment. Background Art

[0002] In the field of synthetic aperture radar (SAR) image processing, the existence of diagonal stripe noise is a common and challenging problem. During the acquisition and transmission of SAR images, diagonal stripe noise is often easily affected by various factors, which seriously affects the image quality and subsequent analysis and application.

[0003] Traditional SAR image denoising methods may have certain limitations when dealing with oblique stripe noise and cannot effectively remove this specific type of noise, or may damage other important features of the image during the removal process.

[0004] At present, there is still a great demand for efficient and accurate methods to remove oblique stripe noise in SAR images. How to accurately identify the oblique stripe noise area and its characteristic domain, and repair the image frequency domain spectrum in an appropriate way to generate high-quality target SAR images has become a technical problem that needs to be solved in the field of SAR image processing. Summary of the invention

[0005] The embodiments of the present invention provide a method for removing oblique stripe noise for SAR images and related devices thereof, which at least solve the problem in the related art that this specific type of noise cannot be effectively removed, or other important features of the image may be damaged during the removal process.

[0006] In a first aspect, an embodiment of the present invention provides a method for removing oblique stripe noise for a SAR image, comprising:

[0007] Performing two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise region in the SAR image respectively to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region, wherein the oblique stripe noise region includes oblique stripe noise;

[0008] Determining a characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and position information of the characteristic domain relative to the SAR image frequency domain spectrum;

[0009] Performing consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information to generate a processed target SAR image frequency domain spectrum;

[0010] Performing a two-dimensional discrete Fourier inverse transform on the target SAR image frequency domain spectrum to generate a target SAR image corresponding to the SAR image.

[0011] According to an embodiment of the present invention, determining the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and the position information of the characteristic domain relative to the SAR image frequency domain spectrum, includes:

[0012] Morphological processing is performed on the frequency domain spectrum of the noise region, and the characteristic domain corresponding to the oblique stripe noise is determined in the frequency domain spectrum of the noise region.

[0013] According to an embodiment of the present invention, determining the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and the position information of the characteristic domain relative to the SAR image frequency domain spectrum, includes:

[0014] Based on the pixel ratio of the SAR image frequency domain spectrum, the feature domain is interpolated by a cubic spline interpolation method to determine the position information of the feature domain relative to the SAR image frequency domain spectrum. According to an embodiment of the present invention, the consistent diffusion interpolation processing is performed on the feature domain to repair the SAR image frequency domain spectrum based on the position information to generate a processed target SAR image frequency domain spectrum, including:

[0015] Generate a target binary image based on the image size of the SAR image and the position information of the feature domain relative to the frequency domain spectrum of the SAR image, wherein pixels in the binary image corresponding to the position information are marked as non-zero values;

[0016] Determine a target repair area in the SAR image frequency domain spectrum based on the value of each pixel in the target binary image;

[0017] Determine a Euclidean distance between a region boundary of the target repair region and a plurality of pixels in the target repair region, so as to determine a filling order corresponding to each of the plurality of pixels based on the Euclidean distance;

[0018] Based on the filling order, the multiple pixels in the target repair area are processed in sequence to generate the target SAR image frequency domain spectrum.

[0019] According to an embodiment of the present invention, sequentially processing the multiple pixels in the target repair area based on the filling order to generate the target SAR image frequency domain spectrum includes:

[0020] For a target pixel among the multiple pixels, taking the coordinates of the target pixel as the center, determining a weighted average value of all pixel values ​​within a preset radius in the SAR image frequency domain spectrum;

[0021] The target pixels are mapped based on the weighted average value to generate the target SAR image frequency domain spectrum.

[0022] According to an embodiment of the present invention, after performing a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image, the method further includes:

[0023] Detecting the oblique stripe noise area in the target SAR image and outputting the detection result;

[0024] When the detection result indicates that the oblique stripe noise does not exist in the oblique stripe noise area, the target SAR image is output.

[0025] According to an embodiment of the present invention, it also includes:

[0026] Acquire initial SAR image;

[0027] Judging the initial SAR image based on a preset judgment method, and outputting a judgment result;

[0028] If the judgment result indicates that there is a null value area in the initial SAR image, filling processing is performed on the null value area in the initial SAR image;

[0029] The initial SAR image after the filling process is subjected to boundary extension process to generate the SAR image to be processed.

[0030] In a first aspect, an embodiment of the present invention provides a device for removing oblique stripe noise for a SAR image, comprising:

[0031] a transform module, configured to perform two-dimensional discrete Fourier transform on a to-be-processed SAR image and an oblique stripe noise region in the SAR image, respectively, to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region, wherein the oblique stripe noise region includes oblique stripe noise;

[0032] A determination module, configured to determine a characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and position information of the characteristic domain relative to the SAR image frequency domain spectrum;

[0033] A processing module, configured to perform a consistent diffusion interpolation process on the feature domain, so as to repair the SAR image frequency domain spectrum based on the position information, and generate a processed target SAR image frequency domain spectrum;

[0034] The generating module is used to perform a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image.

[0035] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a processor, and a memory storing a program, wherein the program comprises instructions, and when the instructions are executed by the processor, the processor executes the method described in the first aspect.

[0036] In a fourth aspect, an embodiment of the present invention provides a non-transitory machine-readable medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method described in the first aspect.

[0037] Beneficial effects of the embodiments of the present invention:

[0038] The embodiment of the present invention provides a method for removing oblique stripe noise for SAR images. For a SAR image containing an oblique stripe noise region including oblique stripe noise, firstly, a two-dimensional discrete Fourier transform is performed on the SAR image to be processed and the oblique stripe noise region respectively to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region. By performing two-dimensional discrete Fourier transform on the SAR image and the oblique stripe noise region respectively, the characteristics of the noise can be accurately located and analyzed. Then, the feature domain corresponding to the oblique stripe noise and the position information of the feature domain relative to the SAR image frequency domain spectrum are determined in the noise region frequency domain spectrum. In this way, the feature domain can be processed by consistent diffusion interpolation to repair the SAR image frequency domain spectrum based on the position information to generate a processed target SAR image frequency domain spectrum. The image frequency domain spectrum is repaired by consistent diffusion interpolation to avoid the loss of original information in the SAR image. Finally, a two-dimensional discrete Fourier inverse transform is performed on the target SAR image frequency domain spectrum to generate a target SAR image corresponding to the SAR image. Based on the above method, the oblique stripe noise of the SAR image can be effectively removed, and at the same time, the loss of original information in the SAR image can be avoided, and the clarity of the SAR image can be maintained.

[0039] The details of one or more embodiments of the invention are set forth in the following drawings and description so that other features, objects, and advantages of the invention are more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0041] Figure 1 A schematic flow chart of a method for removing oblique stripe noise from a SAR image provided by an exemplary embodiment of the present invention.

[0042] Figure 2 The present invention provides a flowchart of a method for repairing a SAR image frequency domain spectrum according to an exemplary embodiment of the present invention.

[0043] Figure 3 A schematic flow chart of a method for removing oblique stripe noise from a SAR image provided by an exemplary embodiment of the present invention.

[0044] Figure 4 (a) is a schematic diagram of a sample SAR image provided by an exemplary embodiment of the present invention.

[0045] Figure 4 (b) is a schematic diagram of a sample SAR image with added lateral simulated stripes provided by an exemplary embodiment of the present invention.

[0046] Figure 5 (a) is a schematic diagram of a target SAR image obtained by using the LF processing method provided by an exemplary embodiment of the present invention.

[0047] Figure 5 (b) is a schematic diagram of a target SAR image obtained by using the LRISD processing method provided by an exemplary embodiment of the present invention.

[0048] Figure 5 (c) is a schematic diagram of a target SAR image obtained by using the ASL processing method provided by an exemplary embodiment of the present invention.

[0049] Figure 5 (d) is a schematic diagram of a target SAR image obtained by using a method for removing oblique stripe noise for SAR images provided by an exemplary embodiment of the present invention.

[0050] Figure 6 (a) is a schematic diagram of a sample SAR image provided by an exemplary embodiment of the present invention.

[0051] Figure 6 (b) is a schematic diagram of a sample SAR image with added oblique simulated strips provided by an exemplary embodiment of the present invention.

[0052] Figure 7 (a) is a schematic diagram of a target SAR image obtained by using the LF processing method provided by an exemplary embodiment of the present invention.

[0053] Figure 7(b) is a schematic diagram of a target SAR image obtained by using the ASL processing method provided by an exemplary embodiment of the present invention.

[0054] Figure 7 (c) is a schematic diagram of a target SAR image obtained by using a method for removing oblique stripe noise for SAR images provided by an exemplary embodiment of the present invention.

[0055] Figure 8 A schematic diagram of a SAR image to be processed provided by an exemplary embodiment of the present invention.

[0056] Fig. 9 (a) is a schematic diagram of a target SAR image obtained by using the LRISD processing method provided by an exemplary embodiment of the present invention.

[0057] Fig. 9 (b) is a schematic diagram of a target SAR image obtained by using the ASL processing method provided by an exemplary embodiment of the present invention.

[0058] Fig. 9 (c) is a schematic diagram of a target SAR image obtained by using a method for removing oblique stripe noise for SAR images provided by an exemplary embodiment of the present invention.

[0059] Fig.10 The present invention is a schematic structural diagram of a device for removing oblique stripe noise from SAR images provided by an exemplary embodiment of the present invention.

[0060] Fig.11 The present invention provides a schematic structural diagram of an electronic device according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0061] Embodiments of the present embodiment will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present embodiment are shown in the accompanying drawings, it should be understood that the present embodiment can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein, which are instead provided for a more thorough and complete understanding of the present embodiment. It should be understood that the drawings and embodiments of the present embodiment are only for exemplary purposes and are not intended to limit the scope of protection of the present embodiment.

[0062] The application of SAR images has become a vital component in the fields of geological and mineral surveys, military reconnaissance, military mapping, scene simulation, etc. In the imaging process of the SAR system, due to the differences in the responses between the charge-coupled device detectors, the image will present stripe noise with periodic characteristics after imaging, which will affect the image quality and seriously limit the application of SAR image interpretation and information extraction. At present, most SAR image stripe noise removal algorithms only focus on how to remove the existing vertical and horizontal stripe noise in the image, while the removal methods for oblique stripe noise in remote sensing images are relatively rare.

[0063] The technical solutions provided by various embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0064] Figure 1 The following is a flow chart of a method for removing oblique stripe noise from SAR images provided by an exemplary embodiment of the present invention. Figure 1 , the method comprises the following steps.

[0065] Step S101, performing two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise region in the SAR image, respectively, to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region, wherein the oblique stripe noise region includes oblique stripe noise.

[0066] Step S102: determining a feature domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and position information of the feature domain relative to the SAR image frequency domain spectrum.

[0067] Step S103 , performing consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information, and generating a processed target SAR image frequency domain spectrum.

[0068] Step S104, performing a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image.

[0069] In practical applications, oblique stripe noise in SAR images is a common type of noise with specific characteristics. Oblique stripe noise usually appears as a stripe distribution with a certain tilt angle in SAR images. It often exists in system-level geometric correction remote sensing products obtained after geometric positioning, map projection, and resampling. Oblique stripe noise can seriously affect the quality of SAR images, reducing the readability and analyzability of the images. It can cover up the real features and details in the image and interfere with operations such as target recognition, detection, and classification.

[0070] In order to eliminate the oblique stripe noise in the SAR image, firstly, two-dimensional discrete Fourier transform is performed on the processed SAR image and the oblique stripe noise area in the SAR image respectively to generate the SAR image frequency domain spectrum corresponding to the SAR image and the noise area frequency domain spectrum corresponding to the oblique stripe noise area.

[0071] In this embodiment, before performing two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise region in the SAR image, it is necessary to determine the oblique stripe noise region containing oblique stripe noise in the SAR image. Specifically, the oblique stripe noise region may include one or more.

[0072] In practical applications, the oblique stripe noise region containing oblique stripe noise can be determined in the SAR image by visual analysis, frequency analysis, statistical feature analysis, etc. For example, when determining the oblique stripe noise region by visual analysis, the SAR image can be enlarged in advance, and the oblique stripe noise region can be determined after visual analysis of the enlarged SAR image.

[0073] In SAR images, systematic stripe noise is presented in the form of symmetrical points on its two-dimensional frequency domain spectrum, and the same is true for diagonal stripe noise. Based on the SAR image frequency domain spectrum corresponding to the SAR image and the noise area frequency domain spectrum corresponding to the diagonal stripe noise area, we can further understand the characteristics and distribution of diagonal stripe noise, thereby improving the positioning accuracy and removal effect of diagonal stripe noise.

[0074] In an optional embodiment, before performing two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise area in the SAR image, the acquired SAR image may be preprocessed. In practical applications, first, an initial SAR image may be acquired, and the initial SAR image may be judged based on a preset judgment method, and a judgment result may be output. When the judgment result indicates that there is a null value area in the initial SAR image, the null value area of ​​the initial SAR image may be filled, and finally the initial SAR image after the filling process may be subjected to boundary extension processing to generate the SAR image to be processed.

[0075] In this embodiment, due to different ways of acquiring the initial SAR image, the initial SAR image may have multiple image types, so it is necessary to first determine the image type of the initial SAR image, and then process the initial SAR image according to the image type of the initial SAR image to obtain the above-mentioned SAR image to be processed.

[0076] In practical applications, since oblique stripe noise often exists in system-level geometrically corrected remote sensing products obtained after geometric positioning, map projection, and resampling, the image type of the initial SAR image can be determined by judging whether the initial SAR image has undergone geometric correction.

[0077] If the judgment result shows that the initial SAR image is a geometrically corrected image type, the initial SAR image needs to be preprocessed. Since the initial SAR image has been geometrically corrected, it means that the initial SAR image has been deflected. In this process, the size of the initial SAR image is enlarged and there are null values ​​around it.

[0078] Subsequently, the empty value areas of the initial SAR image need to be filled. In practical applications, the empty value areas of the initial SAR image can be filled by the mirror filling method. Specifically, for edge pixel positions, the pixel value after mirror filling can be obtained by calculating its offset relative to the edge of the image. If the offset is positive, the pixel value of the pixel position coordinate minus the offset can be filled; correspondingly, if the offset is negative, the pixel value of the pixel position coordinate plus the offset can be filled.

[0079] The empty value area of ​​the initial SAR image is filled by the mirror filling method, that is, the boundary line between the empty value area and the non-empty value area is used as the mirror axis for mirror filling.

[0080] Finally, the initial SAR image after the filling process is subjected to boundary extension process to generate a SAR image to be processed. In this embodiment, based on the above steps, the mirror filling can be continued to expand the boundary range around the image. That is, the edges around the image are used as the mirror axis to perform mirror filling. Optionally, the extended boundary range can be one eighth of the initial SAR image.

[0081] In this embodiment, by expanding the image range, the problem of blurred edge details after image processing is reduced.

[0082] In an optional embodiment, when determining the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum and the position information of the characteristic domain relative to the SAR image frequency domain spectrum, the noise region frequency domain spectrum can be morphologically processed to determine the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum.

[0083] In this embodiment, the morphological processing method may include dilation, erosion, and edge extraction. By performing morphological processing on the frequency domain spectrum of the noise region, a binary image of the region where the oblique stripe noise is located can be obtained, that is, a characteristic domain of the oblique stripe noise in the frequency domain spectrum of the noise region. It should be noted that in the above binary image, the region with a pixel value of 1 represents the pixel position of the oblique stripe noise, and the pixel value of the region outside the region is 0.

[0084] After performing morphological processing on the frequency domain spectrum of the noise region and determining the characteristic domain corresponding to the oblique stripe noise in the frequency domain spectrum of the noise region, the position information of the characteristic domain relative to the frequency domain spectrum of the SAR image can be determined.

[0085] Specifically, based on the pixel ratio of the SAR image frequency domain spectrum, the feature domain may be interpolated by a cubic spline interpolation method to determine the position information of the feature domain relative to the SAR image frequency domain spectrum.

[0086] In this embodiment, the distribution of the oblique stripe noise in the local area image in the frequency domain spectrum of the noise area corresponds to the distribution of the oblique stripe noise in the entire image. Therefore, by processing the feature domain so that the feature domain has the same pixel ratio as the frequency domain spectrum of the SAR image, the corresponding point of the oblique stripe noise in the SAR image in the frequency domain spectrum of the SAR image can be obtained by interpolating the oblique stripe noise area according to the pixel ratio of the SAR image, that is, the position information of the feature domain relative to the frequency domain spectrum of the SAR image.

[0087] In an optional embodiment, consistent diffusion interpolation processing may be performed on the feature domain based on the following method to repair the SAR image frequency domain spectrum based on the position information and generate a processed target SAR image frequency domain spectrum.

[0088] Figure 2 A flowchart of a method for repairing a SAR image frequency domain spectrum provided by an exemplary embodiment of the present invention. Figure 2 , the method comprises the following steps.

[0089] Step S201 : generating a target binary image based on the image size of the SAR image and the position information of the feature domain relative to the frequency domain spectrum of the SAR image, wherein the pixels in the binary image corresponding to the position information are marked as non-zero values.

[0090] Step S202 : determining a target restoration area in the SAR image frequency domain spectrum based on the value of each pixel in the target binary image.

[0091] Step S203 , determining the Euclidean distance between the region boundary of the target repair region and a plurality of pixels in the target repair region, so as to determine the filling order corresponding to each of the plurality of pixels based on the Euclidean distance.

[0092] Step S204 , based on the filling order, sequentially process multiple pixels in the target repair area to generate a target SAR image frequency domain spectrum.

[0093] In this embodiment, consistent diffusion is highly similar to interpolation and can be used in the field of image restoration. Originally, it is a method used to remove objects and fill areas in images. Its core principle is to draw from the boundary pixels of the target area inward. The drawing value of a pixel is estimated from its related adjacent pixels with known values.

[0094] Assume that the set D identifies the area to be interpolated and repaired, which is the target repair area in this embodiment. Pixels in the area D are sorted according to the size of the Euclidean distance of the area boundary, and the sorted set D is obtained as shown in Formula 1.

[0095] D={x1,……,x N} (1)

[0096] Among them, 1 to N is the filling order of pixels in area D.

[0097] In an optional embodiment, when multiple pixels in the target repair area are processed in sequence based on the filling order to generate a target SAR image frequency domain spectrum, for the target pixel among the multiple pixels, a weighted average value of all pixel values ​​within a preset radius is determined in the SAR image frequency domain spectrum with the coordinates of the target pixel as the center, and then the target pixel is drawn based on the weighted average value to generate the target SAR image frequency domain spectrum.

[0098] In this embodiment, the preset radius range can be represented by the Euclidean distance from the target pixel. Assuming that the preset radius range is the Euclidean distance ε, the set of all pixels within the preset radius range (the neighborhood set in this embodiment) is as shown in the following formula 2.

[0099] B ε (x)={y∈Ω:|yx|≤s} (2)

[0100] Among them, x is the coordinate of the target pixel, and Ω is the frequency domain spectrum of the SAR image.

[0101] The pixel sets that have been repaired and those that do not need to be repaired in the neighborhood set are shown in the following formula (3).

[0102]

[0103] Among them, B ε (x k )\{x1……x N} is divided by x1...x N Other than B ε (x k )'s collection.

[0104] The pixel set that needs to be repaired is shown in the following formula (4).

[0105]

[0106] Among them, w(x,y) represents its weight value, and u(y) represents the pixel value of the point that has been repaired and does not need to be repaired.

[0107] Assumptions is the position information of the image area to be repaired, and the weight function of the image repair algorithm is shown in the following formula (5).

[0108]

[0109] in, It is the unit normal vector pointing to the direction of the repair point, and y is the pixel point in the area adjacent to x.

[0110] Based on the above method, it is ensured that the filled image makes more use of the known background information near the defect to effectively repair, so as to ensure the authenticity and neighborhood consistency of the repaired image.

[0111] In an optional embodiment, after performing a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image, the oblique stripe noise region in the target SAR image may be detected and the detection result may be output; if the detection result indicates that there is no oblique stripe noise in the oblique stripe noise region, the target SAR image may be output. On the contrary, if the detection result indicates that there is still oblique stripe noise in the oblique stripe noise region, the oblique stripe noise removal operation may be performed again on the target SAR image based on the above method.

[0112] A method for removing oblique stripe noise for SAR images provided by an embodiment of the invention is described below in conjunction with specific embodiments and drawings.

[0113] Figure 3 The following is a flow chart of a method for removing oblique stripe noise from SAR images provided by an exemplary embodiment of the present invention. Figure 3 , the method comprises the following steps.

[0114] Step S301: first, obtain an initial SAR image.

[0115] In this embodiment, it is assumed that the initial SAR image is an image that has undergone geometric correction. Since the initial SAR image has undergone geometric correction, it means that the initial SAR image is deflected. During this process, the size of the initial SAR image is enlarged and there are null values ​​around it.

[0116] Step S302: Filling processing is performed on the empty value area of ​​the initial SAR image.

[0117] In this embodiment, the empty value area of ​​the initial SAR image is filled by a mirror filling method.

[0118] Step S303: performing boundary extension processing on the initial SAR image after the filling processing to generate a SAR image to be processed.

[0119] In this embodiment, the extended boundary range is one eighth of the initial SAR image.

[0120] Step S304: perform two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise region in the SAR image respectively.

[0121] In this embodiment, a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region are generated through two-dimensional discrete Fourier transform processing.

[0122] Step S305 , performing morphological processing on the frequency domain spectrum of the noise region, and determining a feature domain corresponding to the oblique stripe noise in the frequency domain spectrum of the noise region.

[0123] In this embodiment, the morphological processing method is an edge extraction method.

[0124] Step S306: interpolate the feature domain using a cubic spline interpolation method to determine position information of the feature domain relative to the frequency domain spectrum of the SAR image.

[0125] In this embodiment, the feature domain is interpolated using the pixel ratio of the SAR image frequency domain spectrum.

[0126] Step S307 , performing consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information, and generating a processed target SAR image frequency domain spectrum.

[0127] Step S308, performing a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image.

[0128] Step S309: determine the oblique stripe noise area in the target SAR image.

[0129] In this embodiment, if the determination result indicates that there is no stripe noise in the oblique stripe noise region, step S310 is executed; conversely, if the determination result indicates that there is still oblique stripe noise in the stripe noise region, step S304 is executed.

[0130] Step S310: output the target SAR image.

[0131] The following further describes a method for removing oblique stripe noise for SAR images provided by an embodiment of the present invention in combination with comparative experiments.

[0132] In this embodiment, the SAR image is processed by low-pass filtering (LF), based on the Low-Rank Single Image Decomposition model method (LRISD) and the shearing low-rank model method suitable for oblique stripes (ASL), and compared with the oblique stripe noise removal method for SAR images provided in an embodiment of the present invention.

[0133] First, a simulation experiment was conducted to Figure 4 (a) is a sample SAR image without stripes. After processing, simulated stripes are added (in this embodiment, horizontal simulated stripes are taken as an example). The processed image is as follows: Figure 4 As shown in (b), in this embodiment, Figure 4 (b) Taking the SAR image to be processed as an example, LF, LRISD, ASL and the oblique stripe noise removal method for SAR image provided by the embodiment of the present invention are used to remove the oblique stripe noise of the SAR image. Figure 4 (b) is processed, and the target SAR images obtained are as follows Figure 5 (a) Figure 5 (b) Figure 5 (c) and Figure 5 (d) as shown.

[0134] After analysis, it can be seen that after being processed by the oblique stripe noise removal method for SAR images provided by the embodiment of the present invention, Figure 5 (d) Closer to the sample SAR image Figure 4 (a), indicating that the oblique stripe noise removal method for SAR images provided by the embodiment of the present invention has a better processing effect. The evaluation index summary of the conventional stripe removal result is shown in Table 1.

[0135] Table 1 Summary of evaluation indicators for conventional band removal results

[0136]

[0137] by Figure 6 (a) is a sample SAR image without stripes. After processing, simulated stripes are added (in this embodiment, oblique simulated stripes are used as an example). The processed image is as follows: Figure 6 As shown in (b), in this embodiment, Figure 6 (b) Taking the SAR image to be processed as an example, LF, ASL and the oblique stripe noise removal method for SAR image provided by the embodiment of the present invention are used to remove the oblique stripe noise of the SAR image. Figure 6 (b) is processed, and the target SAR images obtained are as follows Figure 7 (a) Figure 7 (b) Figure 7 (c) as shown.

[0138] After analysis, it can be seen that after being processed by the oblique stripe noise removal method for SAR images provided by the embodiment of the present invention, Figure 7 (c) Closer to the sample SAR image Figure 6 (a), indicating that the oblique stripe noise removal method for SAR images provided by the embodiment of the present invention has a better processing effect. The evaluation index summary of the oblique stripe removal result is shown in Table 2.

[0139] Table 2 Summary of evaluation indexes for oblique stripe removal results

[0140]

[0141] Then, a real experiment was conducted with real samples containing regular bands and oblique bands. Figure 8 Taking the SAR image to be processed as an example, LRISD, ASL and the oblique stripe noise removal method for SAR image provided by the embodiment of the present invention are respectively used to remove the oblique stripe noise of the SAR image. Figure 8 After processing, the target SAR images obtained are as follows Fig. 9 (a) Fig. 9 (b) Fig. 9 (c) as shown.

[0142] After analysis, it can be seen that after being processed by the oblique stripe noise removal method for SAR images provided by the embodiment of the present invention, Fig. 9 (c) Has better processing effect.

[0143] Based on the above-mentioned method for removing oblique stripe noise for SAR images provided by the embodiment of the present invention, the embodiment of the present invention also provides a device for removing oblique stripe noise for SAR images, such as Fig.10 As shown, the device for removing oblique stripe noise of SAR images includes a transform module 1001 , a determination module 1002 , a processing module 1003 and a generation module 1004 .

[0144] The transformation module 1001 is used to perform two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise area in the SAR image, respectively, to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise area frequency domain spectrum corresponding to the oblique stripe noise area, wherein the oblique stripe noise area includes the oblique stripe noise.

[0145] The determination module 1002 is used to determine the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and the position information of the characteristic domain relative to the SAR image frequency domain spectrum.

[0146] The processing module 1003 is used to perform consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information and generate a processed target SAR image frequency domain spectrum.

[0147] The generating module 1004 is used to perform a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image.

[0148] Optionally, the determination module 1002 is specifically configured to perform morphological processing on the frequency domain spectrum of the noise region, and determine a feature domain corresponding to the oblique stripe noise in the frequency domain spectrum of the noise region.

[0149] Optionally, the determination module 1002 is specifically used to obtain the pixel ratio of the SAR image frequency domain spectrum respectively. Correspondingly, the processing module 1003 is specifically used to interpolate the feature domain by a cubic spline interpolation method based on the pixel ratio of the SAR image frequency domain spectrum to determine the position information of the feature domain relative to the SAR image frequency domain spectrum.

[0150] Optionally, the processing module 1003 is further used to generate a target binary image based on the image size of the SAR image and the position information of the feature domain relative to the SAR image frequency domain spectrum, and the pixels corresponding to the position information in the binary image are marked as non-zero values. The determination module 1002 is further used to determine the target repair area in the SAR image frequency domain spectrum based on the values ​​of each pixel in the target binary image; determine the Euclidean distance between the region boundary of the target repair area and multiple pixels in the target repair area, so as to determine the filling order corresponding to each of the multiple pixels based on the Euclidean distance. The processing module 1003 is further used to process the multiple pixels in the target repair area in sequence based on the filling order to generate the target SAR image frequency domain spectrum.

[0151] Optionally, the processing module 1003 is further specifically used to determine, for a target pixel among multiple pixels, a weighted average value of all pixel values ​​within a preset radius in the SAR image frequency domain spectrum with the coordinates of the target pixel as the center; and draw the target pixel based on the weighted average value to generate a target SAR image frequency domain spectrum.

[0152] Optionally, the generating module 1004 is specifically configured to detect the oblique stripe noise region in the target SAR image and output the detection result; if the detection result indicates that there is no oblique stripe noise in the oblique stripe noise region, the target SAR image is output.

[0153] Optionally, the processing module 1003 is also used to obtain an initial SAR image; judge the initial SAR image based on a preset judgment method and output a judgment result; when the judgment result indicates that there are null value areas in the initial SAR image, fill the null value areas in the initial SAR image; and perform boundary extension processing on the filled initial SAR image to generate a SAR image to be processed.

[0154] An embodiment of the present invention further provides an electronic device, comprising: at least one processor; and a memory in communication with the at least one processor. The memory stores a computer program executable by the at least one processor, and when the computer program is executed by the at least one processor, the electronic device executes the method of the embodiment of the present invention.

[0155] An embodiment of the present invention further provides a non-transitory machine-readable medium storing a computer program, wherein the computer program, when executed by a processor of a computer, is used to cause the computer to execute the method of the embodiment of the present invention.

[0156] An embodiment of the present invention further provides a computer program product, including a computer program, wherein the computer program, when executed by a processor of a computer, is used to enable the computer to execute the method of the embodiment of the present invention.

[0157] refer to Fig.11 , a block diagram of an electronic device that can be used as a server or client of an embodiment of the present invention will now be described, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic devices are intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0158] like Fig.11 As shown, the electronic device includes a computing unit 1101, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the electronic device can also be stored. The computing unit 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0159] Multiple components in the electronic device are connected to the I / O interface 1105, including: an input unit 1106, an output unit 1107, a storage unit 1108, and a communication unit 1109. The input unit 1106 can be any type of device that can input information to the electronic device, and the input unit 1106 can receive input digital or character information, and generate key signal input related to user settings and / or function control of the electronic device. The output unit 1107 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 1108 can include but is not limited to a disk, an optical disk. The communication unit 1109 allows the electronic device to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0160] The computing unit 1101 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a CPU, a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1101 performs the various methods and processes described above. For example, in some embodiments, the method embodiments of the present invention may be implemented as a computer program, which is tangibly contained in a machine-readable medium, such as a storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on an electronic device via a ROM 1102 and / or a communication unit 1109. In some embodiments, the computing unit 1101 may be configured to perform the above-described method in any other appropriate manner (e.g., by means of firmware).

[0161] The computer programs for implementing the methods of the embodiments of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer programs are executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer programs may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0162] In the context of an embodiment of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable signal medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or 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 foregoing.

[0163] It should be noted that the term "including" and its variations used in the embodiments of the present invention are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The modifications of "one" and "multiple" mentioned in the embodiments of the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless the context clearly indicates otherwise, it should be understood as "one or more".

[0164] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0165] The various steps described in the method implementation methods provided in the embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method implementation methods may include additional steps and / or omit the steps shown. The scope of protection of the present invention is not limited in this respect.

[0166] The term "embodiment" in this specification refers to specific features, structures or characteristics described in conjunction with the embodiment that can be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. The various embodiments in this specification are described in a related manner, and the same or similar parts between the various embodiments refer to each other. In particular, for the device, equipment, and system embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts refer to the partial description of the method embodiment.

[0167] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the attached claims.

Claims

1. A method for removing oblique stripe noise from SAR images, characterized in that: include: Performing two-dimensional discrete Fourier transform on the SAR image to be processed and the oblique stripe noise region in the SAR image respectively to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region, wherein the oblique stripe noise region includes oblique stripe noise; Determining a characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and position information of the characteristic domain relative to the SAR image frequency domain spectrum; Performing consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information to generate a processed target SAR image frequency domain spectrum; Performing a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image; The performing of consistent diffusion interpolation processing on the feature domain to repair the SAR image frequency domain spectrum based on the position information to generate a processed target SAR image frequency domain spectrum includes: Based on the image size of the SAR image and the position information of the feature domain relative to the frequency domain spectrum of the SAR image, a target binary image is generated, wherein pixels corresponding to the position information in the binary image are marked as non-zero values; Determine a target repair area in the SAR image frequency domain spectrum based on the value of each pixel in the target binary image; Determine a Euclidean distance between a region boundary of the target repair region and a plurality of pixels in the target repair region, so as to determine a filling order corresponding to each of the plurality of pixels based on the Euclidean distance; Based on the filling order, the multiple pixels in the target repair area are processed in sequence to generate the target SAR image frequency domain spectrum.

2. The method according to claim 1, characterized in that The determining of the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and the position information of the characteristic domain relative to the SAR image frequency domain spectrum, includes: Morphological processing is performed on the frequency domain spectrum of the noise region, and the characteristic domain corresponding to the oblique stripe noise is determined in the frequency domain spectrum of the noise region.

3. The method according to claim 2, characterized in that The determining of the characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and the position information of the characteristic domain relative to the SAR image frequency domain spectrum, includes: Based on the pixel ratio of the SAR image frequency domain spectrum, the feature domain is interpolated by a cubic spline interpolation method to determine the position information of the feature domain relative to the SAR image frequency domain spectrum.

4. The method according to claim 1, characterized in that: The step of sequentially processing the plurality of pixels in the target repair area based on the filling order to generate the target SAR image frequency domain spectrum includes: For a target pixel among the multiple pixels, taking the coordinates of the target pixel as the center, determining a weighted average value of all pixel values ​​within a preset radius in the SAR image frequency domain spectrum; The target pixels are mapped based on the weighted average value to generate the target SAR image frequency domain spectrum.

5. The method according to claim 1, characterized in that After performing a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image, the method further includes: Detecting the oblique stripe noise area in the target SAR image and outputting the detection result; When the detection result indicates that the oblique stripe noise does not exist in the oblique stripe noise area, the target SAR image is output.

6. The method according to claim 1, characterized in that Also includes: Acquire initial SAR image; Judging the initial SAR image based on a preset judgment method, and outputting a judgment result; If the judgment result indicates that there is a null value area in the initial SAR image, filling processing is performed on the null value area in the initial SAR image; The initial SAR image after the filling process is subjected to boundary extension process to generate the SAR image to be processed.

7. A device for removing oblique stripe noise from SAR images, characterized in that: include: a transform module, configured to perform two-dimensional discrete Fourier transform on a to-be-processed SAR image and an oblique stripe noise region in the SAR image, respectively, to generate a SAR image frequency domain spectrum corresponding to the SAR image and a noise region frequency domain spectrum corresponding to the oblique stripe noise region, wherein the oblique stripe noise region includes oblique stripe noise; A determination module, configured to determine a characteristic domain corresponding to the oblique stripe noise in the noise region frequency domain spectrum, and position information of the characteristic domain relative to the SAR image frequency domain spectrum; A processing module, configured to perform a consistent diffusion interpolation process on the feature domain, so as to repair the SAR image frequency domain spectrum based on the position information, and generate a processed target SAR image frequency domain spectrum; The generating module is used to perform a two-dimensional discrete Fourier inverse transform on the frequency domain spectrum of the target SAR image to generate a target SAR image corresponding to the SAR image.

8. An electronic device comprising: A processor, and a memory storing a program, wherein the program comprises instructions, which, when executed by the processor, cause the processor to perform the method according to any one of claims 1 to 6.

9. A non-transitory machine-readable medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

Images