Parameter interpretation method and system for oblique backscattering frequency sweep ionogram of ionized layer

Abstract

The invention discloses a parameter interpretation method and system for an oblique backscattering frequency sweep ionogram of an ionosphere. The method comprises the following steps: carrying out gray processing on the oblique backscattering frequency sweep ionogram; de-noising is carried out on the ionogram; image edges are extracted through a Canny edge detection operator; connecting frequency point signals submerged by noise through morphological closed operation to obtain a continuous signal edge; dividing an ionosphere vertical measurement signal and an oblique return scattering signal through connected domain analysis; delimiting an F-layer oblique return scattering signal starting area, and traversing the ionogram along the frequency direction to obtain a frequency-minimum group path curve of an oblique return scattering signal; fitting frequency-minimum group path curves of different frequency points by using LASSO regression to obtain an F-layer oblique return scattering front; the method has the advantages of being accurate in interpretation, small in noise interference and good in signal quality.

Description

Technical Field

[0001] The invention belongs to the technical field of ionospheric inversion, and in particular relates to a parameter interpretation method and system for an ionospheric oblique backscatter swept-frequency ionogram. Background Art

[0002] The principle of ionospheric oblique backscatter detection is that radio waves are transmitted by a transmitter and then incident obliquely on the ionosphere, and then reflected to the far ground or water surface of the earth. Due to the unevenness of the incident plane, some rays return along the original path and are received by the receiver. The detection equipment obtains the group delay information of oblique backscatter echoes at different frequencies through frequency sweep detection, and obtains the frequency-group delay two-dimensional ionogram. Due to spherical focusing and time focusing, the oblique backscatter echo has a very steep front edge, showing obvious signal enhancement on the ionogram, and its group path is the smallest among all oblique backscatter echoes at the detection frequency.

[0003] Since the oblique backscatter signal has experienced at least two ionospheric reflections and one ground or water surface scattering, the energy of the oblique backscatter signal is weak. In addition, as the detection frequency increases, the path of the front group of the oblique backscatter signal increases monotonically, which will bring greater path loss. Therefore, the signal-to-noise ratio of the oblique backscatter signal in the oblique backscatter ionogram is usually low, and due to the existence of a large number of interferences in the same frequency band in space, many frequency echoes are submerged in the interference, showing a front fracture in the oblique backscatter ionogram. Since the oblique backscatter detection power is large, the side lobes of the transmitting antenna can still radiate a large power signal. Part of the signal vertically enters the ionosphere and is vertically reflected and received by the receiver. Since the vertically reflected vertical detection signal has no scattering process, the energy of the vertical detection signal received by the receiver is even higher than that of the oblique backscatter signal. The existence of the above phenomenon makes it very difficult to automatically interpret the complete front of the oblique backscatter signal.

[0004] At present, the automatic interpretation method of the parameters of the oblique backscatter ionization map mainly relies on traversing the frequency axis and extracting the minimum group path of each frequency point as the front through the threshold method. However, this method has very high requirements on the oblique backscatter ionization map and is difficult to use under low-power detection conditions. It can only obtain a partially broken oblique backscatter front. At the same time, the interpretation result is greatly affected by the vertical measurement signal. Summary of the invention

[0005] The purpose of the present invention is to provide a parameter interpretation method and system for an ionospheric oblique backscatter swept-frequency ionogram, which utilizes the signal enhancement caused by the focusing effect at the front edge of the oblique backscatter signal in the ionospheric oblique backscatter ionogram, repairs the image submerged by interference through two-dimensional discrete wavelet reconstruction, uses the Canny edge detection operator to extract edges, and separates the oblique backscatter signal from the vertical reflection signal through morphological closing operations and prior knowledge of the path distribution of ionospheric signal groups, thereby realizing parameter interpretation, and has the advantages of accurate interpretation, low noise interference and good signal quality.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a parameter interpretation method of an ionospheric oblique backscatter swept frequency ionogram, comprising the following steps:

[0007] Grayscale processing and denoising of oblique backscattered swept frequency ionization images;

[0008] After the image edge is extracted by the Canny edge detection operator, the frequency point signals submerged by the noise are connected through the morphological closing operation to obtain a continuous signal edge;

[0009] The ionospheric vertical signal and the oblique backscatter signal are divided by connected domain analysis;

[0010] The starting area of ​​the F-layer oblique backscatter signal is delineated, and the ionization diagram is traversed along the frequency direction to obtain the frequency-minimum group path curve of the oblique backscatter signal. The frequency-minimum group path curve at different frequency points is fitted using LASSO regression to obtain the F-layer oblique backscatter front.

[0011] In the above scheme, the oblique backscatter swept frequency ionogram is firstly grayscale processed and denoised to improve the edge extraction effect. After the image edge is extracted by the Canny edge detection operator, the frequency point signals submerged by the noise are connected through the morphological closing operation to preprocess and repair the denoised image to obtain a continuous signal edge. The ionospheric vertical measurement signal and the oblique backscatter signal are divided by connected domain analysis to facilitate identification. After the ionospheric oblique backscatter signal group path area is determined, the LASSO regression is used to fit the minimum value of the ionospheric oblique backscatter signal group path at different frequency points to obtain the ionospheric oblique backscatter front.

[0012] Furthermore, the denoising includes performing two-dimensional discrete wavelet decomposition on the oblique backscattered swept frequency ionogram after grayscale processing, and then performing two-dimensional wavelet reconstruction to obtain the denoised oblique backscattered ionogram.

[0013] Furthermore, the Canny edge detection operator extracts the image edge including the following steps:

[0014] A Gaussian filter mask is defined, and each pixel of the image is scanned using the mask. The weighted average grayscale value of the pixels in the neighborhood determined by the mask is used to replace the value of the central pixel of the template. The image pixel gradient is calculated using the Sobel operator, and then a binary image containing the edge of the oblique backscattered signal and the vertical detection signal is obtained by setting two thresholds and traversing the edge pixels.

[0015] Furthermore, the morphological closing operation to connect the frequency point signals submerged by noise includes the following steps:

[0016] Construct a rectangular kernel , for the binary image containing the edge of the oblique backscattered signal and the vertical detection signal With nuclear The morphological closing operation is expressed as:

[0017] .

[0018] Furthermore, dividing the ionospheric vertical detection signal and the oblique backscatter signal by connected domain analysis includes the following steps:

[0019] After the connected domain analysis, the threshold is selected to remove the connected domains with too few connected pixels, and each connected domain is traversed and classified to obtain the ionization map background mask, vertical detection signal mask, and oblique backscatter signal mask.

[0020] Further, the method comprises the following steps: separately interpreting the ionospheric vertical measurement signal to obtain the local ionospheric critical frequency and the electron density peak height, and the interpreting steps are as follows:

[0021] Set the pixels at the oblique backscatter signal mask position of the ionogram to 0 to determine whether there is a vertical detection signal mask. If so, read the maximum frequency corresponding to the vertical detection signal mask as the ionospheric vertical detection signal x wave. Layer critical frequency, based on the ionospheric vertical signal x wave The magnetic rotation frequency above the critical frequency of the layer and the receiving equipment of the ionogram is Calculate the ionospheric vertical signal o wave Critical frequency and the ionosphere Layer peak height;

[0022] If the vertical signal mask does not exist, the ionosphere The critical frequency and peak height parameters of the layer are given by Model provided.

[0023] Further, the following steps are included:

[0024] Interpreting the vertical signal mask, extracting the group path of the minimum frequency point, and then calculating the minimum group path of the F layer oblique backscatter signal area;

[0025] Traverse along the frequency axis of the ionogram, set the pixels whose pixel values ​​of the corresponding positions of the oblique backscatter signal mask in the F layer oblique backscatter signal area are less than the set value to zero, and select the minimum group path of the non-zero pixel points at each frequency point to obtain the frequency-minimum group path curve of the oblique backscatter signal.

[0026] Furthermore, the F-layer oblique backscattering front is the oblique backscattering signal frequency point number, frequency-minimum group path curve, oblique backscattering signal frequency matrix, regularization parameter, polynomial coefficient matrix, polynomial coefficient matrix Norm of the fitted cubic polynomial function.

[0027] A parameter interpretation system for an ionospheric oblique backscatter swept-frequency ionogram, comprising:

[0028] The first main module is used for grayscale processing and denoising of the oblique backscattered swept frequency ionization image;

[0029] The second main module is used to extract the image edge through the Canny edge detection operator and then connect the frequency point signals submerged by noise through morphological closing operation to obtain continuous signal edges;

[0030] The third main module is used to divide the ionospheric vertical measurement signal and the oblique return scattering signal through connected domain analysis;

[0031] The fourth main module is used to define the starting area of ​​the oblique backscatter signal of the F layer, and traverse the ionization diagram along the frequency direction to obtain the frequency-minimum group path curve of the oblique backscatter signal; use LASSO regression to fit the frequency-minimum group path curve of different frequency points to obtain the oblique backscatter front of the F layer.

[0032] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer executes the steps of the above method.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. The present invention utilizes morphological closing operations to effectively improve the quality of the oblique backscattering signal of the ionization image, and can repair the broken oblique backscattering front under severe interference conditions;

[0035] 2. The present invention effectively separates the oblique backscatter signal of the ionogram from the vertical measurement signal, and while being able to obtain additional local ionospheric parameters, further reduces the interference of the vertical measurement signal and improves the accuracy of the front edge of the oblique backscatter signal;

[0036] 3. The present invention utilizes LASSO regression to fit the oblique backscattering front, effectively weakening the influence of noise on the front position, while the introduction of regularization terms reduces the risk of overfitting; and cooperates with two-dimensional discrete wavelet decomposition and two-dimensional wavelet reconstruction to improve the denoising effect;

[0037] 4. The present invention provides an image algorithm solution that does not require changing the hardware configuration, is easy to implement, and has good compatibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A schematic diagram of the principle of a method for interpreting parameters of an ionospheric oblique backscatter swept-frequency ionogram provided by an embodiment of the present invention;

[0039] Figure 2 A step diagram of a method for interpreting parameters of an ionospheric oblique backscatter swept-frequency ionogram provided by an embodiment of the present invention;

[0040] Figure 3 An example diagram of Canny edge extraction of an ionization image provided in an embodiment of the present invention;

[0041] Figure 4 An example diagram of performing a morphological closing operation on an edge binary image provided by an embodiment of the present invention;

[0042] Figure 5 An example diagram of a vertical reflection signal and an oblique backscattering signal marked by a connected domain analysis provided in an embodiment of the present invention;

[0043] Figure 6 An example diagram of vertical reflection signal extraction provided by an embodiment of the present invention;

[0044] Figure 7 An example diagram of oblique backscattering front extraction provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0045] In order to make the purpose, 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 in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention are arbitrarily combined with each other to form a new technical solution. This combination is not restricted by the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that this combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0046] The F layer is the ionosphere region with an altitude of generally 140 km to several thousand km. Layer and layer, The layer is located at Below the layer, the altitude is about 140~200 kilometers.

[0047] The present embodiment provides a method for interpreting parameters of an ionospheric oblique backscatter swept-frequency ionogram, which utilizes conventional oblique backscatter detection equipment and is beneficial to improving the quality of the oblique backscatter ionogram without increasing additional hardware costs. It can effectively separate the vertical reflection signal and the oblique backscatter signal on the ionogram, and additionally obtain local ionospheric parameters and a more continuous oblique backscatter front.

[0048] The basic principle of this embodiment is as follows Figure 1 As shown in the figure, an ionospheric oblique backscatter detector is used to transmit a coded modulated shortwave detection signal with a narrow bandwidth to the oblique upper ionosphere in a swept frequency observation mode. After being received by the receiver, it is cross-correlated with the code sequence for pulse compression to obtain a frequency~group path two-dimensional distribution ionogram. After constant false alarm, the oblique backscatter swept frequency ionogram is used for grayscale processing and wavelet denoising. After the Canny edge detection operator is used to extract the image edge, the frequency point signals submerged by the noise are connected through morphological closing operation, and the ionospheric vertical measurement signal and the oblique backscatter signal are divided by the centroid of the connected domain. The ionospheric vertical measurement signal is interpreted separately to obtain the local ionospheric critical frequency and electron density peak height, and then the ionospheric oblique backscatter signal group path area is determined. The minimum value of the ionospheric oblique backscatter signal group path at different frequency points is fitted by LASSO regression to obtain the ionospheric oblique backscatter front.

[0049] like Figure 2 As shown, the present invention implements the following steps:

[0050] Step 1, obtaining the oblique backscattered swept frequency ionization image after constant false alarm processing, and converting it into a grayscale image format;

[0051] It should be noted that the false alarm rate of the oblique backscattered swept frequency ionization image should be less than ;

[0052] Step 2, performing two-dimensional discrete wavelet decomposition on the oblique backscattered swept frequency ionogram obtained in the grayscale processing step 1, retaining the approximate coefficients, performing soft threshold quantization processing on the detail coefficients, and then performing two-dimensional wavelet reconstruction to obtain the denoised oblique backscattered ionogram;

[0053] It should be noted that if the x-axis scale of the oblique backscattered swept frequency ionization image is , the y-axis scale is , the pixel value function is , the two-dimensional scaling function is , the two-dimensional wavelet function transformed along the horizontal edge direction is , the two-dimensional wavelet function transformed along the vertical edge direction is , the two-dimensional wavelet function transformed along the diagonal edge direction is , the number of decomposition levels is , then the two-dimensional discrete scale and translation basis functions are expressed as formula (1) and formula (2) respectively:

[0054] Formula (1)

[0055] Formula (2)

[0056] If the number of pixels on the x-axis of the oblique backscattering swept frequency ionization image is , the number of pixels on the y-axis is , m represents the horizontal offset of the wavelet basis function, and n represents the vertical offset of the wavelet basis function. The pixel value function is , the low-frequency subband and high-frequency subband of discrete wavelet transform are expressed as formula (3) and formula (4) respectively:

[0057] Formula (3)

[0058] Formula (4)

[0059] in, Represents horizontal, vertical, and diagonal directions respectively.

[0060] The low-frequency subband is completely retained, and the high-frequency subband is denoised by soft threshold. If the matrix row sorting function is expressed as , then the matrix of high-frequency subband rows sorted from small to large It is expressed as:

[0061] Formula (5)

[0062] is the dimension. The preserved ratio value is expressed as , the value range is generally , threshold function It is expressed as:

[0063] Formula (6)

[0064] High frequency subband soft threshold denoising coefficients It is expressed as:

[0065] Formula (7)

[0066] is a symbolic function.

[0067] Two-dimensional wavelet reconstruction of oblique backscatter ionogram function It is expressed as:

[0068] Formula (8)

[0069] Step 3, using the Canny edge detection operator to perform edge detection on the oblique backscatter ionization image obtained in step 2, and obtain a binary image containing the edges of the oblique backscatter signal and the vertical detection signal;

[0070] It should be noted that the edge detection method is:

[0071] Define a 5*5 Gaussian filter mask, whose distribution is expressed as:

[0072] Formula (9)

[0073] Use the mask to scan each pixel of the image, and use the weighted average grayscale value of the pixels in the neighborhood determined by the mask to replace the value of the central pixel of the template, expressed as:

[0074] Formula (10)

[0075] The Sobel operator is used to calculate the image pixel gradient. The x-axis and y-axis gradients are expressed as formula (11) and formula (12) respectively:

[0076] Formula (11)

[0077] Formula (12)

[0078] The image pixel gradient amplitude and direction are expressed as equation (13) and equation (14) respectively:

[0079] Formula (13)

[0080] Formula (14)

[0081] The image gradient direction is classified into 8 directions: horizontal (left, right), vertical (up, down), diagonal (upper right, lower right, upper left, lower left), and traverses all pixels in the image. If the pixel gradient is the local maximum in the positive / negative gradient direction, the point is retained and defined as an edge pixel. Otherwise, the pixel value is set to zero. Set two thresholds and , and meet , traverse the edge pixels again, if the gradient amplitude of the edge pixels Greater than or equal to , then the edge pixel is marked as a strong edge point. If the gradient amplitude of the edge pixel is Less than But greater than or equal to , then the edge pixel is marked as a virtual edge point. If the gradient amplitude of the edge pixel is Less than , then the edge point is discarded and its pixel value is set to zero, and a binary image containing the edge of the oblique backscatter signal and the vertical detection signal is obtained; usually, and The values ​​of are and .

[0082] Step 4: Use morphological closing operation to fill in the frequency points that are submerged by the signal and interference, and the method to obtain continuous signal edges is:

[0083] Construct a rectangular kernel representation as:

[0084] Formula (15)

[0085] The binary image containing the oblique backscattered signal and the vertical detection signal edge obtained in step 3 With nuclear The morphological closing operation is expressed as:

[0086] Formula (16)

[0087] Step 5, performing a connected domain analysis on the signal edge obtained in step 4, removing the connected domains with too few connected pixels after selecting a threshold, and dividing the ionospheric vertical measurement signal, the oblique backscatter signal and the background noise according to the centroid height of the remaining connected domains;

[0088] It should be noted that the method of using connected domain analysis to remove the noise connected domain and separate the ionospheric vertical measurement signal, oblique backscatter signal and background noise is as follows:

[0089] If the signal edge The connected domain mask set obtained by connected domain analysis is , the background connected domain mask is , No. The signal or noise connected domain mask is , is the sum function, and the centroid of the connected domain is , refers to the group path corresponding to the centroid of the connected domain, is the frequency corresponding to the centroid of the connected domain; the total number of connected domains is , the ionization image background mask, vertical detection signal mask, and oblique backscatter signal mask obtained by traversing each connected domain and performing three classification are respectively expressed as equations (17), (18), and (19):

[0090] Formula (17)

[0091] Formula (18)

[0092] Formula (19)

[0093] In some embodiments, step 6 includes interpreting the ionospheric vertical measurement signal obtained in step 5 to obtain the local ionospheric critical frequency and the electron density peak height;

[0094] It should be noted that the method for interpreting the ionospheric vertical measurement signal and obtaining the local ionospheric critical frequency and electron density peak height is:

[0095] Set the pixels at the oblique backscatter signal mask position of the ionogram to 0 to determine whether there is a vertical detection signal mask. If so, read the maximum frequency corresponding to the vertical detection signal mask as the ionospheric vertical detection signal x wave. Layer critical frequency , if the magnetic rotation frequency above the receiving device where the ionogram is obtained is , then the ionospheric vertical signal o wave The layer critical frequency is expressed as:

[0096] Formula (20)

[0097] Ionosphere The layer peak height is expressed as:

[0098] Formula (21)

[0099] If the vertical signal mask does not exist, the ionosphere The critical frequency and peak height parameters of the layer are given by Model provided;

[0100] Step 6, using the group path of the minimum frequency point of the ionospheric vertical measurement signal obtained in step 5 to define the starting area of ​​the F layer oblique backscatter signal, and traversing the ionogram along the frequency direction to obtain a frequency-minimum group path curve of the oblique backscatter signal;

[0101] It should be noted that the method for defining the starting area of ​​the oblique backscatter signal and obtaining the frequency-minimum group path curve of the oblique backscatter signal is:

[0102] Interpret the ionospheric vertical signal mask obtained in step 5 and extract the group path of the minimum frequency point , then the minimum group path in the F layer oblique backscatter signal area satisfies:

[0103] Formula (22)

[0104] Traverse along the frequency axis of the ionogram, set the pixels with pixel values ​​less than 80 in the corresponding position of the oblique backscatter signal mask in the F layer oblique backscatter signal area to zero, and select the minimum group path of non-zero pixel points at each frequency point to obtain the frequency-minimum group path curve of the oblique backscatter signal;

[0105] Step 6, performing LASSO regression fitting on the obtained frequency-minimum group path curve of the oblique backscatter signal as the front edge of the F layer oblique backscatter signal;

[0106] It should be noted that the frequency-minimum group path curve of the oblique backscatter signal is subjected to LASSO regression fitting as the method for the frontier of the oblique backscatter signal:

[0107] The front edge of the oblique backscattered signal is fitted by a cubic polynomial function of frequency, expressed as:

[0108] Formula (23)

[0109] Among them, the polynomial coefficient matrix It is obtained by minimizing the cost function, which is expressed as:

[0110] Formula (24)

[0111] in, is the number of frequency points containing oblique backscattered signals, is the frequency-minimum group path curve obtained in step 6, is the frequency matrix of the oblique backscatter signal, is the regularization parameter, usually taken as , is the polynomial coefficient matrix of Norm.

[0112] In the above method, the edge detection effect can be adjusted by adjusting the size of the double threshold of the Canny edge detection operator as needed. By increasing the size of the regularization parameter of the LASSO regression fitting, the obtained oblique backscattering front can be made smoother, but if it is too large, the fitted front position will deviate from the true front position.

[0113] When implementing it, Figure 3As shown, in one embodiment, the present application performs Canny edge extraction on the oblique backscatter ionogram. Due to the implementation of the hysteresis threshold extraction algorithm, a large amount of discrete noise is suppressed, and the echo with higher frequency can be better retained even if the energy is weaker;

[0114] like Figure 4 As shown, this is an example diagram of the morphological closing operation performed on the edge binary image in this embodiment, a large number of isolated edges are connected, and the noise is further suppressed;

[0115] like Figure 5 As shown, it is an example diagram of marking vertical reflection signals and oblique backscatter signals after connected domain analysis in this embodiment. The green part is the oblique backscatter signal mask, and the red part is the vertical reflection signal mask. The classification is accurate, and more isolated noises are suppressed;

[0116] like Figure 6 As shown in the figure, it is an example diagram of vertical reflection signal extraction in this embodiment. The vertical reflection signal area is framed by using the oblique backscatter signal mask, and the pixel values ​​of the remaining area are set to zero to obtain an ionogram with only vertical reflection signals. The maximum frequency corresponding to the pixel can be obtained. , combined with the local magnetic rotation frequency, we can get Layer critical frequency With peak height ;

[0117] like Figure 7 As shown in the figure, it is an example diagram of the extraction of the oblique backscattering front in this embodiment. The oblique backscattering front is accurately extracted. Even if it is seriously interfered and many frequency signals are submerged, the oblique backscattering signal front can still be accurately extrapolated to Nearby, achieving the expected effect;

[0118] A parameter interpretation system for an ionospheric oblique backscatter swept-frequency ionogram, comprising:

[0119] The first main module is used for grayscale processing and denoising of the oblique backscattered swept frequency ionization image;

[0120] The second main module is used to extract the image edge through the Canny edge detection operator and then connect the frequency point signals submerged by noise through morphological closing operation to obtain continuous signal edges;

[0121] The third main module is used to divide the ionospheric vertical measurement signal and the oblique return scattering signal through connected domain analysis;

[0122] The fourth main module is used to define the starting area of ​​the oblique backscatter signal of the F layer, and traverse the ionization diagram along the frequency direction to obtain the frequency-minimum group path curve of the oblique backscatter signal; use LASSO regression to fit the frequency-minimum group path curve of different frequency points to obtain the oblique backscatter front of the F layer.

[0123] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer executes the steps of the above method.

[0124] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram, characterized in that: The steps include: Grayscale processing and denoising of oblique backscattered swept frequency ionization images; After the image edge is extracted by the Canny edge detection operator, the frequency point signals submerged by the noise are connected through the morphological closing operation to obtain a continuous signal edge; The ionospheric vertical signal and the oblique backscatter signal are divided by connected domain analysis; The starting area of ​​the F-layer oblique backscatter signal is delineated, and the ionization diagram is traversed along the frequency direction to obtain the frequency-minimum group path curve of the oblique backscatter signal. The frequency-minimum group path curve at different frequency points is fitted using LASSO regression to obtain the F-layer oblique backscatter front.

2. The method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram according to claim 1, characterized in that: The denoising includes performing two-dimensional discrete wavelet decomposition on the oblique backscattered swept frequency ionogram after grayscale processing, and then performing two-dimensional wavelet reconstruction to obtain the denoised oblique backscattered ionogram.

3. The parameter interpretation method of the ionospheric oblique backscatter swept frequency ionogram according to claim 1, characterized in that: The Canny edge detection operator extracts the image edge and includes the following steps: A Gaussian filter mask is defined, and each pixel of the image is scanned using the mask. The weighted average grayscale value of the pixels in the neighborhood determined by the mask is used to replace the value of the central pixel of the template. The image pixel gradient is calculated using the Sobel operator, and then a binary image containing the edge of the oblique backscattered signal and the vertical detection signal is obtained by setting two thresholds and traversing the edge pixels.

4. The method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram as claimed in claim 3, characterized in that: The morphological closing operation connects the frequency point signals submerged by noise and comprises the following steps: Construct a rectangular kernel , for the binary image containing the edge of the oblique backscattered signal and the vertical detection signal With nuclear The morphological closing operation is expressed as: 。 5. The method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram according to claim 1, characterized in that: The division of ionospheric vertical detection signals and oblique backscatter signals through connected domain analysis includes the following steps: After the connected domain analysis, the threshold is selected to remove the connected domains with too few connected pixels, and each connected domain is traversed and classified to obtain the ionization map background mask, vertical detection signal mask, and oblique backscatter signal mask.

6. The method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram according to claim 5, characterized in that: The method comprises the following steps: separately interpreting the ionospheric vertical measurement signal to obtain the local ionospheric critical frequency and the electron density peak height, and the interpretation steps are as follows: Set the pixels at the oblique backscatter signal mask position of the ionogram to 0 to determine whether there is a vertical detection signal mask. If so, read the maximum frequency corresponding to the vertical detection signal mask as the ionospheric vertical detection signal x wave. Layer critical frequency, based on the ionospheric vertical signal x wave The magnetic rotation frequency above the critical frequency of the layer and the receiving equipment of the ionogram is Calculate the ionospheric vertical signal o wave Critical frequency and the ionosphere Layer peak height; If the vertical signal mask does not exist, the ionosphere The critical frequency and peak height parameters of the layer are given by Model provided.

7. The method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram according to claim 5, characterized in that: The following steps are included: Interpreting the vertical signal mask, extracting the group path of the minimum frequency point, and then calculating the minimum group path of the F layer oblique backscatter signal area; Traverse along the frequency axis of the ionogram, set the pixels whose pixel values ​​of the corresponding positions of the oblique backscatter signal mask in the F layer oblique backscatter signal area are less than the set value to zero, and select the minimum group path of the non-zero pixel points at each frequency point to obtain the frequency-minimum group path curve of the oblique backscatter signal.

8. The method for interpreting parameters of an ionospheric oblique backscatter swept frequency ionogram according to claim 1, characterized in that: The F-layer oblique backscattering front is the oblique backscattering signal frequency point number, frequency-minimum group path curve, oblique backscattering signal frequency matrix, regularization parameter, polynomial coefficient matrix, polynomial coefficient matrix Norm of the fitted cubic polynomial function.

9. A parameter interpretation system for ionospheric oblique backscatter swept frequency ionogram, characterized in that: include: The first main module is used for grayscale processing and denoising of the oblique backscattered swept frequency ionization image; The second main module is used to extract the image edge through the Canny edge detection operator and then connect the frequency point signals submerged by noise through morphological closing operation to obtain continuous signal edges; The third main module is used to divide the ionospheric vertical measurement signal and the oblique return scattering signal through connected domain analysis; The fourth main module is used to define the starting area of ​​the oblique backscatter signal of the F layer, and traverse the ionization diagram along the frequency direction to obtain the frequency-minimum group path curve of the oblique backscatter signal; use LASSO regression to fit the frequency-minimum group path curve of different frequency points to obtain the oblique backscatter front of the F layer.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the computer is caused to perform the steps of the method according to any one of claims 1 to 8.

Images