CN113689449B - Inversion method and system for characteristic parameters of mesoscale vortices - Google Patents
Inversion method and system for characteristic parameters of mesoscale vortices Download PDFInfo
- Publication number
- CN113689449B CN113689449B CN202110977814.0A CN202110977814A CN113689449B CN 113689449 B CN113689449 B CN 113689449B CN 202110977814 A CN202110977814 A CN 202110977814A CN 113689449 B CN113689449 B CN 113689449B
- Authority
- CN
- China
- Prior art keywords
- image
- filtering processing
- mesoscale
- sar
- vortex
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10044—Radar image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an inversion method and system of mesoscale vortex characteristic parameters, wherein the inversion method comprises the steps of collecting an ocean SAR image, preprocessing the ocean SAR image in a radiation correction and shading trend processing mode, and obtaining a first image; carrying out primary filtering processing on the first image to obtain a second image with enhanced mesoscale vortex characteristics, and obtaining the SAR image gradient of the second image; based on morphological filtering processing, after secondary filtering processing is carried out on the second image, a coordinate set of each edge line of the second image is obtained through an edge detection method according to the gradient of the SAR image, and a mesoscale vortex characteristic parameter inversion model is constructed through a Hough transformation detection operator; the inversion system comprises an image acquisition module, a first filtering processing module, a second filtering processing module and an inversion module; the invention realizes the effectiveness of the ocean mesoscale vortex inversion method of simulation data, aviation SAR observation data and other similar satellite data, and provides a new technical idea for the technical scheme in the field.
Description
Technical Field
The application relates to the technical field of ocean parameter inversion, in particular to an inversion method and system of medium-scale vortex characteristic parameters.
Background
The ocean mesoscale vortex has a large coverage area in the ocean, and the scale is hundreds of kilometers. The space morphology is one of the main apparent characteristics of the mesoscale vortexes, and the mesoscale vortex space morphologies with different types and different mechanisms show respective laws. If the shape of the ocean cold vortex is in a fixed round cake shape, the shape of the warm vortex is in a shape with different strengths, such as a circle, a strip, a silk and a tongue. The mesoscale vortices in the sea present a circular, ribbon, tongue-like structure with weak edges on the SAR image.
The determination of the mesoscale vortex ROI (region of interest) is a difficulty of mesoscale vortex parameter inversion, and comprises the steps of judging whether the ROI region exists or not and determining the ROI region so as to reduce the target range, eliminate the interference of other ocean phenomena and invert mesoscale vortex characteristic parameters. Because the structure of the mesoscale vortex image is complex, judging whether the mesoscale vortex region exists or not and judging the approximate range of the mesoscale vortex region by using an image processing method are difficult to realize.
Disclosure of Invention
The invention aims to determine an ROI (region of interest) by adopting a manual intervention method, and in order to realize the aim, the invention provides an inversion method of a mesoscale vortex characteristic parameter, which comprises the following steps:
acquiring an ocean SAR image, and preprocessing the ocean SAR image in a radiation correction and shading trend processing mode to obtain a first image;
carrying out primary filtering processing on the first image to obtain a second image with enhanced mesoscale vortex characteristics, and obtaining the SAR image gradient of the second image;
based on morphological filtering processing, after secondary filtering processing is carried out on the second image, a coordinate set of each edge line of the second image is obtained through an edge detection method according to the SAR image gradient;
and constructing a mesoscale vortex characteristic parameter inversion model through a Hough transformation detection operator based on the coordinate set, wherein the mesoscale vortex characteristic parameter inversion model is used for judging the center of an ellipse by fitting the ellipse position presented by the mesoscale vortex phenomenon, acquiring the center of a vortex, and inverting the mesoscale vortex characteristic parameter according to the coordinate set.
Preferably, in the process of performing the first filtering processing on the first image, the first filtering processing is performed on the first image by using a laplacian pyramid method, so as to obtain a first low-pass image and a multi-level high-pass image of the first image.
Preferably, in the process of performing the first filtering processing on the first image, performing the second filtering processing on the first image based on the discrete Contourlet transform to obtain a second low-pass image of the first image and high-frequency components distributed in multiple scales and multiple directions;
and acquiring a second image according to the first low-pass image, the second low-pass image, the multi-level high-pass image and the high-frequency component.
Preferably, during the second filtering process of the first image, a double-layer filter PDFB is constructed through LP and DFB, wherein the double-layer filter PDFB is used to approximate the original image in a contour segment manner.
Preferably, in the second filtering process performed on the second image, the equation of the morphological filtering process is:
wherein Δ G represents a difference between a maximum value and a minimum value of the gray scale of the image in the structural element size region; Δ D represents the distance defined when calculating the gradient.
Preferably, in the process of acquiring the coordinate set, the method for edge detection includes the following steps:
s101, smoothing the second image through a Gaussian filter;
s102, calculating the amplitude and the direction of the gradient of the SAR image based on the finite difference of the first-order partial derivative;
s103, carrying out non-maximum suppression on the amplitude of the gradient of the SAR image;
and S104, edge determination is carried out by adopting a thresholding method through reducing the number of the false edges of the smoothed second image, and a coordinate set is obtained.
Preferably, in the process of acquiring the coordinate set, vectorization processing is performed on the second image subjected to edge detection, a characteristic curve is separated, and the coordinate set is constructed by extracting curve coordinate data.
Preferably, in the process of constructing the coordinate set, the curve coordinate data is acquired according to an edge tracking method.
An inversion system of mesoscale vortex characteristic parameters, comprising:
the image acquisition module is used for acquiring the marine SAR image, and preprocessing the marine SAR image in a radiation correction and shading trend processing mode to obtain a first image;
the first filtering processing module is used for carrying out primary filtering processing on the first image to obtain a second image with enhanced mesoscale vortex characteristics and obtaining the SAR image gradient of the second image;
the second filtering processing module is used for obtaining a coordinate set of each edge line of the second image by an edge detection method according to the SAR image gradient after performing secondary filtering processing on the second image based on morphological filtering processing;
and the inversion module is used for constructing a medium-scale vortex characteristic parameter inversion model through a Hough transformation detection operator based on the coordinate set, wherein the medium-scale vortex characteristic parameter inversion model is used for judging the center of an ellipse by fitting the ellipse position presented by the medium-scale vortex phenomenon, acquiring the center of the vortex, and inverting the medium-scale vortex characteristic parameter according to the coordinate set.
Preferably, the inversion system further comprises,
the data storage module is used for storing system data of the inversion system;
the communication module is used for data interaction of the inversion system;
the system control module is used for realizing logic control among all modules of the inversion system;
and the display module is used for displaying the SAR image and the mesoscale vortex characteristic parameters of the SAR image.
The invention discloses the following technical effects:
the invention realizes the effectiveness of the ocean mesoscale vortex inversion method of simulation data, aviation SAR observation data and other similar satellite data, and provides a new technical idea for the technical scheme in the field.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow of inversion of characteristic parameters of mesoscale vortices according to the present invention;
FIG. 2 is a Contourlet transform filter bank structure according to an embodiment of the present invention;
FIG. 3 is a software interface for inverting the characteristic parameters of meso-scale vortices according to an embodiment of the present invention;
FIG. 4 is a strong mesoscale eddy current field distribution according to an embodiment of the present invention;
FIG. 5 is a strong mesoscale vortex simulated SAR image according to an embodiment of the present invention;
FIG. 6 is a mesoscale vortex inversion of strong vortex L-band three-level sea state simulation data according to an embodiment of the present invention;
FIG. 7 is a medium intensity meso-scale eddy current field distribution according to an embodiment of the present invention;
FIG. 8 is a medium intensity mesoscale vortex simulated SAR image in accordance with an embodiment of the present invention;
fig. 9 is a mesoscale vortex inversion of the L-band secondary sea state simulation data of the mesoscale vortices in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
As shown in fig. 1 to 9, the present invention provides an inversion method of a mesoscale vortex characteristic parameter, including the following steps:
acquiring an ocean SAR image, and preprocessing the ocean SAR image in a radiation correction and shading trend processing mode to obtain a first image;
carrying out primary filtering processing on the first image to obtain a second image with enhanced mesoscale vortex characteristics, and obtaining the SAR image gradient of the second image;
based on morphological filtering processing, after secondary filtering processing is carried out on the second image, a coordinate set of each edge line of the second image is obtained through an edge detection method according to the SAR image gradient;
and constructing a mesoscale vortex characteristic parameter inversion model through a Hough transformation detection operator based on the coordinate set, wherein the mesoscale vortex characteristic parameter inversion model is used for judging the center of an ellipse by fitting the ellipse position presented by the mesoscale vortex phenomenon, acquiring the center of a vortex, and inverting the mesoscale vortex characteristic parameter according to the coordinate set.
Further, in the process of performing primary filtering processing on the first image, performing primary filtering processing on the first image by using a laplacian pyramid method to obtain a first low-pass image and a multi-level high-pass image of the first image.
Further, in the process of performing primary filtering processing on the first image, performing secondary filtering processing on the first image based on discrete Contourlet transformation to obtain a second low-pass image of the first image and high-frequency components distributed in multiple dimensions and multiple directions;
and acquiring a second image according to the first low-pass image, the second low-pass image, the multi-level high-pass image and the high-frequency component.
Further, in the second filtering process of the first image, a double-layer filter PDFB is constructed by LP and DFB, wherein the double-layer filter PDFB is used to approximate the original image in a contour segment manner.
Further, in the second filtering process of the second image, the equation of the morphological filtering process is:
wherein, Δ G represents the difference between the maximum value and the minimum value of the gray scale of the image in the structural element size area; Δ D represents the distance defined when calculating the gradient.
Further, in the process of acquiring the coordinate set, the method for edge detection includes the following steps:
s101, smoothing the second image through a Gaussian filter;
s102, calculating the amplitude and the direction of the gradient of the SAR image based on the finite difference of the first-order partial derivative;
s103, carrying out non-maximum suppression on the amplitude of the gradient of the SAR image;
and S104, edge determination is carried out by adopting a thresholding method through reducing the number of the false edges of the smoothed second image, and a coordinate set is obtained.
Furthermore, in the process of obtaining the coordinate set, vectorization processing is performed on the second image subjected to edge detection, a characteristic curve is separated, and the coordinate set is constructed by extracting curve coordinate data.
Further, in the process of constructing the coordinate set, curve coordinate data are obtained according to an edge tracking method.
An inversion system for mesoscale vortex characteristic parameters, comprising:
the image acquisition module is used for acquiring the marine SAR image, and preprocessing the marine SAR image in a radiation correction and shading trend processing mode to obtain a first image;
the first filtering processing module is used for carrying out primary filtering processing on the first image to obtain a second image with enhanced mesoscale vortex characteristics and obtaining the SAR image gradient of the second image;
the second filtering processing module is used for performing secondary filtering processing on the second image based on morphological filtering processing, and then acquiring a coordinate set of each edge line of the second image through an edge detection method according to the gradient of the SAR image;
and the inversion module is used for constructing a mesoscale vortex characteristic parameter inversion model through Hough transformation detection operators based on the coordinate set, wherein the mesoscale vortex characteristic parameter inversion model is used for judging the center of an ellipse by fitting the elliptic position presented by the mesoscale vortex phenomenon, acquiring the center of a vortex, and inverting the mesoscale vortex characteristic parameter according to the coordinate set.
Further, the inversion system may further include,
the data storage module is used for storing system data of the inversion system;
the communication module is used for data interaction of the inversion system;
the system control module is used for realizing logic control among all modules of the inversion system;
and the display module is used for displaying the SAR image and the mesoscale eddy characteristic parameters of the SAR image.
Example 1: according to the characteristics of the mesoscale vortexes on the SAR image, the mesoscale vortexes are inverted by using a digital image processing method, and the specific flow is as follows:
(1) SAR image preprocessing
When the SAR image is imaged, due to different incident angles, the image can be dark on the bright side, so that the radiation correction can be performed during imaging processing, and the SAR image can be processed according to the brightness trend of the image in the post-processing process, so that the brightness distribution of the image is uniform on the whole.
(2) Mesoscale vortex ROI region determination
The determination of the mesoscale vortex ROI (region of interest) is a difficulty of mesoscale vortex parameter inversion, and comprises the steps of judging whether the ROI region exists or not and determining the ROI region so as to reduce the target range, eliminate the interference of other ocean phenomena and invert mesoscale vortex characteristic parameters. Because the structure of the mesoscale vortex image is complex, judging whether the mesoscale vortex region exists or not and judging the approximate range of the mesoscale vortex region by an image processing method are difficult to achieve, and therefore a manual intervention method is adopted to determine the ROI region at the stage.
(3) Mesoscale vortex feature enhancement
The interference of speckle noise and sea wave clutter exists in the SAR mesoscale vortex image, and the interference is caused to the extraction of mesoscale vortex parameters, so that the SAR image must be filtered, the speckle noise and the sea wave clutter are filtered as far as possible, and the mesoscale vortex characteristic is highlighted. In the past decades, many SAR image speckle filtering methods have been proposed, mainly classified into two major categories, the first category is speckle elimination during imaging, and the other category is speckle elimination by filtering after imaging. Speckle removal during imaging is most classically multi-view processing, which is done at the cost of reduced image resolution, and therefore most speckle filtering methods are done after imaging. Early methods were primarily based on fourier transforms, and further work was to extend Kalman and Bayesian filtering to two dimensions, both of which required statistical models of the signal, and filters based on local statistical parameters were gradually developed as natural artifacts could not be described by a single statistical model.
Contourlet is a new analysis tool developed by Curvelet, the transformation satisfies anisotropic scale relation, has good directionality, and can accurately capture the edge contour information in the image into sub-bands with different scales and different directions. From domestic and foreign documents in recent years, Contourlet has been well applied to the field of image processing, Contourlet transformation is introduced into the field of image enhancement, and due to the fact that Contourlet transformation is lack of translation invariance, a pseudo Gibbs phenomenon can be generated during threshold processing, and therefore a novel method for enhancing the image based on the translation invariance Contourlet transformation by the aid of the project is provided.
The discrete Contourlet transform generally consists of two steps. Firstly, the traditional Laplacian pyramid method is applied to multi-level decomposition, the Laplacian pyramid decomposition is similar to multi-resolution analysis of wavelet transformation, and the space V is divided into a plurality of spaces0Step-by-step two-division into a series of subspaces Vj0And { Wj}，Vj0Is of the scale
Approximate subspace of, WjFor its quadrature complement, a resolution of 2 is includedj-1Details of the time signal. The two-dimensional squared integrable space can then be expressed as:
wherein L is2(R2) Two-dimensional square integrable space, and ^ space quadrature. j is a function of0At the low frequency scale, j is the scale level of the multi-scale decomposition.
In the first step of decomposition, a residual image between a low-pass sampling approximation of an original image and a low-pass prediction image is generated, the obtained low-pass image is continuously decomposed to obtain a low-pass approximation image and a difference image of the next layer, and a low-pass image and a multi-level high-pass image are finally obtained through gradual filtering. The Laplacian pyramid filter does not sample the high-pass image, and only one band-pass image is generated in each pyramid layer, so that the frequency aliasing phenomenon can be effectively avoided.
The LP and DFB combine to form a two-layer filter bank structure, also called a pyramid-wise direction filter bank PDFB, also called discrete Contourlet transform, since the PDFB essentially approximates the original image in contour segments. A schematic diagram of the discrete Contourlet transform is shown in fig. 2. The original image is decomposed by a PDFB structure to obtain a low-pass image and high-frequency components distributed on multiple scales and multiple directions. The number of directions increases exponentially with increasing l.
The invention obtains the main frequency of the wave clutter and the energy distribution characteristic in the Contourlet transform domain through the simulation and analysis of the ocean background, reduces the influence of spot noise and the wave clutter on the mesoscale phenomenon signal through the filtering in the Contourlet domain, and finally achieves the purpose of image enhancement.
(4) Image gradient calculation
And calculating the SAR image gradient by using a two-dimensional gradient calculation method.
(5) Morphological filtering process
Due to the weak edge characteristic of the mesoscale vortexes, the mesoscale vortex region obtained by image segmentation has great uncertainty, and needs morphological filtering for smoothing. Morphological processing is generally defined as the result of the original image dilation minus the original image erosion, algebraically in the form:
wherein, f is the original image data; g is a structural element and g is a structural element,
for the dilation operation, Θ is the erosion operation.
If the morphological gradient operator is given in the form of a numerical difference, then this is simply understood as the maximum difference per unit distance,
the delta G represents the difference value between the maximum value and the minimum value of the gray scale of the image in the size area of the structural element; Δ D represents the distance defined when calculating the gradient. Therefore, the operator not only can be used as an edge detection operator to extract the morphological characteristics of the mesoscale vortex, but also has certain advantages in the robustness, continuity and contrast of noise resistance.
(6) Edge detection
And (4) carrying out edge detection on the image after being subjected to mathematical morphology filtering by adopting a Canny operator. The Canny operator is an optimal detection algorithm of the step-shaped edge, and the algorithm is based on the following principle: two conditions must be satisfied, one is that noise can be effectively suppressed, and the other is that the boundary must be determined as accurately as possible; measuring according to the signal-to-noise ratio and the positioning product to obtain an optimized approximation operator; belongs to a method of smoothing first and then derivation. The Canny edge detection algorithm comprises four steps: firstly, smoothing an image by using a Gaussian filter; secondly, calculating the amplitude and the direction of the gradient by adopting the finite difference of the first-order partial derivatives; thirdly, carrying out non-maximum suppression on the gradient amplitude; and finally, reducing the number of false edges, and determining the edges by adopting a thresholding method.
After feature extraction, the original image is converted into a binary image which implicitly contains vortex boundary information, and vectorization is also needed to be carried out on the images, a specific feature curve is separated, and curve coordinate data are obtained. Specifically, an edge tracking method is adopted, the algorithm firstly detects all nodes in a binary image, then starts from each node, searches neighborhood pixels according to a certain principle, carries out line segment tracking on different directions with different priority levels under a certain boundary limit, and finally obtains a coordinate set of each edge line in the image.
(7) Vortex center detection
The method for detecting and extracting the mesoscale vortex features has a plurality of methods, and because of the complexity of the imaging process of the marine remote sensing image and the complexity of the image, the detection features of the vortex are difficult to extract based on the gray value connected region such as the image, and the edge detection of the remote sensing image can be carried out by directly utilizing a Hough transform circle detection operator. Due to the complexity of the vortex formation, the detected edge curve is not generally a regular circle, but only approximates a circle or ellipse in a rough sense. According to the position of the vortex edge, fitting the ellipse position presented by the mesoscale vortex phenomenon through a Hough transform detection operator, judging the center of the ellipse, and taking the center as the center of the vortex. The premise of vortex center detection is that the SAR image can completely reflect the whole mesoscale vortex, and if the SAR only images a vortex partial area, the vortex center is difficult to invert.
FIG. 3 shows a part of the interface of the mesoscale vortex feature parameter inversion software module.
Simulation data inversion test
Because the data of the high-resolution actually measured marine mesoscale eddy current field is lacked, the marine mesoscale eddy current field is simulated according to the flow velocity gradient range found by the literature by taking satellite fusion data as the reference. The flow velocity data of the ocean mesoscale vortex field is based on multi-satellite fusion data of a sea surface flow field published by a NASA (national analysis of health) website, the spatial resolution of the data is 25 kilometers, and mesoscale vortices in an extended area of east black tide of Japan sea are selected, wherein the extended area is a typical ocean mesoscale vortex area. The foreign literature data shows that the gradient of the mesoscale eddy current field is 10-5(1/s) to 10-4And (1/s) magnitude, simulating strong, medium and weak medium-scale eddy current fields according to satellite data flow velocity data and medium-scale eddy current field gradient data reported by literature, performing subsequent simulation on the basis to obtain an SAR simulation image, and inverting medium-scale eddy parameters by using a medium-scale eddy inversion model.
FIG. 5 is a simulation SAR of scale-in-one vortexThe image of the ocean surface current field distribution is shown in fig. 4. The maximum value of the mesoscale eddy velocity gradient is about 1 × 10-4(1/s) belongs to strong mesoscale vortices, the simulated SAR parameters are L wave band, VV polarization, central incidence angle of about 48 degrees, geometric resolution azimuth direction of 40m, distance direction of 20m, radiation resolution of about 1.12dB and sea surface wind speed of 5 m/s.
And performing mesoscale vortex parameter inversion on the three-level sea state L wave band VV polarization SAR simulation image according to the mesoscale vortex inversion model, as shown in FIG. 6.
FIG. 8 is a simulation SAR image of mesoscale eddy current with a maximum mesoscale eddy current gradient of about 5 × 10-5(1/s), the sea surface current field distribution is shown in FIG. 7. The simulated SAR parameters are L wave band, VV polarization, central incidence angle of about 48 degrees, geometric resolution of 100m, radiation resolution of 1.12dB and sea surface wind speed of 5 m/s.
And performing mesoscale vortex parameter inversion on the mesoscale vortex simulation image of the mesoscale intensity and secondary sea state L wave band VV polarization SAR according to the mesoscale vortex inversion model, as shown in FIG. 9.
TABLE 1
The mesoscale vortex inversion result is compared with the input flow field data, and the detection position deviation and the central deviation of the mesoscale vortex in the two examples are obtained and are shown in table 1.
The method disclosed by the invention is used for carrying out inversion test on the aviation SAR observation data and carrying out inversion test on other similar satellite data to prove the applicability of the method disclosed by the invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus once an item is defined in one figure, it need not be further defined and explained in subsequent figures, and moreover, the terms "first", "second", "third", etc. are used merely to distinguish one description from another and are not to be construed as indicating or implying relative importance.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present invention, which are used for illustrating the technical solutions of the present invention and not for limiting the same, and the protection scope of the present invention is not limited thereto, although the present invention is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the embodiments of the present invention. Are intended to be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (6)
1. An inversion method of a mesoscale vortex characteristic parameter is characterized by comprising the following steps:
acquiring an ocean SAR image, and preprocessing the ocean SAR image in a radiation correction and shading trend processing mode to obtain a first image;
carrying out primary filtering processing on the first image to obtain a second image with enhanced mesoscale vortex characteristics, and acquiring the SAR image gradient of the second image;
based on morphological filtering processing, after secondary filtering processing is carried out on the second image, a coordinate set of each edge line of the second image is obtained through an edge detection method according to the SAR image gradient;
constructing a mesoscale vortex characteristic parameter inversion model through a Hough transform detection operator based on the coordinate set, wherein the mesoscale vortex characteristic parameter inversion model is used for judging an ellipse center by fitting an ellipse position presented by a mesoscale vortex phenomenon, acquiring the center of a vortex, and inverting the mesoscale vortex characteristic parameter according to the coordinate set;
in the process of carrying out primary filtering processing on the first image, carrying out primary filtering processing on the first image by using a Laplacian pyramid method to obtain a first low-pass image and a multi-level high-pass image of the first image;
in the process of carrying out primary filtering processing on the first image, carrying out secondary filtering processing on the first image based on discrete Contourlet transformation to obtain a second low-pass image of the first image and high-frequency components distributed on multiple scales and multiple directions;
acquiring the second image according to the first low-pass image, the second low-pass image, the multi-level high-pass image and the high-frequency component;
in the process of carrying out second filtering processing on the first image, constructing a double-layer filter PDFB through LP and DFB, wherein the double-layer filter PDFB is used for approximating an original image in a contour segment mode;
in the process of performing the second filtering process on the second image, the equation of the morphological filtering process is as follows:
▽f＝ΔG/ΔD
wherein, Δ G represents the difference between the maximum value and the minimum value of the gray scale of the image in the structural element size area; Δ D represents the distance defined when calculating the gradient.
2. The method for inverting the characteristic parameters of the mesoscale vortices as claimed in claim 1, wherein:
in the process of acquiring the coordinate set, the method for edge detection comprises the following steps:
s101, smoothing the second image through a Gaussian filter;
s102, calculating the amplitude and the direction of the SAR image gradient based on the finite difference of the first-order partial derivative;
s103, carrying out non-maximum suppression on the amplitude of the SAR image gradient;
and S104, edge determination is carried out by adopting a thresholding method through reducing the number of the smoothed false edges of the second image, and the coordinate set is obtained.
3. The method for inverting the characteristic parameters of the mesoscale vortices as claimed in claim 2, wherein:
and in the process of acquiring the coordinate set, vectorizing the second image subjected to the edge detection, separating a characteristic curve, and extracting curve coordinate data to construct the coordinate set.
4. The method for inverting the characteristic parameters of the mesoscale vortices as claimed in claim 3, wherein:
and in the process of constructing the coordinate set, acquiring the curve coordinate data according to an edge tracking method.
5. An inversion system of mesoscale vortex characteristic parameters, comprising:
the image acquisition module is used for acquiring an ocean SAR image, and preprocessing the ocean SAR image in a radiation correction and shading trend processing mode to obtain a first image;
a first filtering processing module, configured to perform primary filtering processing on the first image to obtain a second image with enhanced mesoscale eddy features, and obtain an SAR image gradient of the second image, where,
in the process of carrying out primary filtering processing on the first image, carrying out primary filtering processing on the first image by using a Laplacian pyramid method to obtain a first low-pass image and a multi-level high-pass image of the first image;
in the process of carrying out primary filtering processing on the first image, carrying out secondary filtering processing on the first image based on discrete Contourlet transformation to obtain a second low-pass image of the first image and high-frequency components distributed on multiple scales and multiple directions;
acquiring the second image according to the first low-pass image, the second low-pass image, the multi-level high-pass image and the high-frequency component;
in the process of carrying out second filtering processing on the first image, constructing a double-layer filter PDFB through LP and DFB, wherein the double-layer filter PDFB is used for approximating an original image in a contour segment mode;
a second filtering processing module, configured to perform secondary filtering processing on the second image based on morphological filtering processing, and then obtain a coordinate set of each edge line of the second image by an edge detection method according to the gradient of the SAR image, where an equation of the morphological filtering processing is:
▽f＝ΔG/ΔD
wherein Δ G represents a difference between a maximum value and a minimum value of the gray scale of the image in the structural element size region; Δ D represents the distance defined when calculating the gradient;
and the inversion module is used for constructing a mesoscale vortex characteristic parameter inversion model through a Hough transformation detection operator based on the coordinate set, wherein the mesoscale vortex characteristic parameter inversion model is used for judging the center of an ellipse by fitting the elliptic position presented by the mesoscale vortex phenomenon, acquiring the center of a vortex, and inverting the mesoscale vortex characteristic parameter according to the coordinate set.
6. The system for inverting the characteristic parameters of the mesoscale vortices as claimed in claim 5, wherein:
the inversion system may further comprise a data storage system,
the data storage module is used for storing system data of the inversion system;
the communication module is used for carrying out data interaction on the inversion system;
the system control module is used for realizing logic control among all modules of the inversion system;
a display module for displaying the SAR image and the meso-scale vortex characteristic parameters of the SAR image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110977814.0A CN113689449B (en) | 2021-08-25 | 2021-08-25 | Inversion method and system for characteristic parameters of mesoscale vortices |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110977814.0A CN113689449B (en) | 2021-08-25 | 2021-08-25 | Inversion method and system for characteristic parameters of mesoscale vortices |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113689449A CN113689449A (en) | 2021-11-23 |
| CN113689449B true CN113689449B (en) | 2022-06-07 |
Family
ID=78582109
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110977814.0A Active CN113689449B (en) | 2021-08-25 | 2021-08-25 | Inversion method and system for characteristic parameters of mesoscale vortices |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113689449B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114283128B (en) * | 2021-12-16 | 2024-04-26 | 中国人民解放军海军潜艇学院 | Ocean mesoscale vortex edge detection method and system based on multi-parameter threshold |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105844104A (en) * | 2016-03-29 | 2016-08-10 | 中国海洋大学 | Mesoscale vortex visualization algorithm based on Morphing technique |
| CN106951854B (en) * | 2017-03-16 | 2019-06-04 | 中国人民解放军国防科学技术大学 | A kind of SAR image internal wave of ocean parameter extracting method |
| US10983250B1 (en) * | 2020-09-29 | 2021-04-20 | James L. Regens | Component-based system for computer implemented multi-dimensional gridded mesoscale meteorological projection |
| CN112766194A (en) * | 2021-01-26 | 2021-05-07 | 上海海洋大学 | Detection method for mesoscale ocean eddy |
| CN112991309A (en) * | 2021-03-29 | 2021-06-18 | 中国人民解放军海军潜艇学院 | Automatic detection method for internal wave region of satellite SAR image |
-
2021
- 2021-08-25 CN CN202110977814.0A patent/CN113689449B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113689449A (en) | 2021-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Modava et al. | Integration of spectral histogram and level set for coastline detection in SAR images | |
| CN109272489B (en) | Infrared weak and small target detection method based on background suppression and multi-scale local entropy | |
| Brandtberg et al. | Automated delineation of individual tree crowns in high spatial resolution aerial images by multiple-scale analysis | |
| Nasiri et al. | Infrared small target enhancement based on variance difference | |
| Bovik | On detecting edges in speckle imagery | |
| CN101727662B (en) | SAR image nonlocal mean value speckle filtering method | |
| CN107220988B (en) | Part image edge extraction method based on improved canny operator | |
| CN107808383B (en) | Rapid detection method for SAR image target under strong sea clutter | |
| CN103761731A (en) | Small infrared aerial target detection method based on non-downsampling contourlet transformation | |
| CN105303526A (en) | Ship target detection method based on coastline data and spectral analysis | |
| CN108229342B (en) | Automatic sea surface ship target detection method | |
| CN110660065B (en) | Infrared fault detection and identification algorithm | |
| Lestriandoko et al. | Circle detection based on hough transform and Mexican Hat filter | |
| CN113077514B (en) | SAR image ship weak trail enhancement and detection method | |
| CN116152115B (en) | Garbage image denoising processing method based on computer vision | |
| CN113689449B (en) | Inversion method and system for characteristic parameters of mesoscale vortices | |
| CN109872358B (en) | Marine oil film identification method for shipborne radar image based on active contour model | |
| CN112150474A (en) | Underwater bubble image feature segmentation and extraction method | |
| CN105869156B (en) | A kind of infrared small target detection method based on fuzzy distance | |
| CN112669332A (en) | Method for judging sea and sky conditions and detecting infrared target based on bidirectional local maximum and peak local singularity | |
| CN112164079A (en) | Sonar image segmentation method | |
| CN116953674A (en) | Rapid target detection algorithm in sonar imaging | |
| CN110097569B (en) | Oil tank target detection method based on color Markov chain significance model | |
| Kim et al. | Oil spill detection from RADARSAT-2 SAR image using non-local means filter | |
| CN109427042B (en) | Method for extracting layered structure and spatial distribution of local sea area sedimentary layer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |