An offshore flow detection method, system, medium, device and terminal

Abstract

The present application belongs to the field of computer vision and security technology field, disclose a kind of offshore flow detection method, system, medium, equipment and terminal, utilize Sentienl-1 dual polarization SAR image to detect crack flow, crack flow characteristics are enhanced processing in combination with VH and VV polarization data, crack flow in SAR image is detected by edge detection operator and threshold segmentation method.The present application detects offshore flow that occurs near the coast of Portugal from Sentienl-1 dual polarization SAR image, and the polarization difference of VH and VV polarization data is used to enhance the characteristics of offshore flow in the image, and then the offshore flow in the SAR image is extracted by edge detection operator and threshold segmentation method, this method has achieved good detection effect, can effectively enhance crack flow characteristics, realize the area, extension length and flow direction of offshore flow extracted from SAR image and detect the distribution of crack flow from SAR image.

Description

Technical Field

[0001] This invention belongs to the fields of computer vision and security technology, and particularly relates to an offshore current detection method, an offshore current detection system, a computer-readable storage medium, a computer device, and an information data processing terminal. Background Technology

[0002] Rip currents, also known as rip currents, are small but strong and concentrated surface currents caused by the combined effects of breaking waves and uneven shallow water topography. Rip currents are prone to occur near sandbars, headlands, and protruding structures. Continuous breaking waves cause water to accumulate on the coastline, flowing back into the sea through concentrated pathways. The flow direction is almost perpendicular to the coastline, and the current is short-lived and fast. The rip current head, neck, coastal current, and compensating rip current constitute the entire rip current system. The calm surface of rip currents can be deceptive, quickly sweeping inattentive swimmers into deeper water. According to the U.S. National Oceanic and Atmospheric Administration (NOAA), rip currents are a leading cause of beach drowning accidents, yet they have not received widespread public awareness. In recent years, rip currents have received increasing attention from scholars worldwide, leading to extensive research.

[0003] In the early days, people observed rip currents through field observations, digital images, and videos. In recent years, ultra-high frequency shore-based radar has been widely used to monitor rip currents. This method can provide timely information about the sea area and obtain information that is difficult to obtain with traditional equipment.

[0004] Ocean radar can provide radial velocity characteristics of ocean currents and enhance the radar echo intensity of sea surface roughness, thereby imaging ocean currents. Haller et al. demonstrated the ability of X-band and horizontally polarized ocean radar to detect crack current features. Trizna used ocean coherent radar to collect four months of data and conducted detection and analysis of crack current features.

[0005] However, marine radar can only be deployed on specific beaches for observation, limiting its detection range, and the data is not publicly available and difficult to obtain. Microwave remote sensing technology is a feasible method for monitoring ocean currents. Among them, Synthetic Aperture Radar (SAR) has advantages such as all-weather, all-time capability, high resolution, and multiple polarization modes. Radar echoes from the sea surface are highly sensitive to micro-scale waves and their changes. By transmitting and receiving radar echoes from the sea surface, radar can capture ocean surface currents; therefore, SAR can be used for the identification and detection of offshore currents.

[0006] Da Silva identified offshore current features from two satellite-borne SAR datasets (ENVISAT ASAR and ERS-2 SAR) of the west coast of Portugal and performed SAR image diagnostics using a simple wind comparison model. Arry Retnowati identified and detected the breaking wave zone and surf zone of the rift current on ALOS PALSAR Fine Beam Single (FBS) HH polarization images of Paragtritis Beach.

[0007] Detecting rip currents using UHF shore-based radar offers high measurement accuracy in both time and space, but its measurement range is limited, requiring stations to be established near designated coastlines, making it difficult to set up stations in harsh coastal environments, and resulting in high instrument maintenance costs. In contrast, detecting rip currents from satellite remote sensing imagery achieves high spatial measurement accuracy and can detect rip currents within the satellite-covered sea area, broadening the geographical scope of research. This method is also relatively inexpensive, but it requires sifting through massive amounts of imagery to identify images containing rip currents and cannot track the rip current's occurrence and termination over time. Research on rip current detection using spaceborne SAR data is relatively limited, and existing studies have not fully utilized the multi-polarization capabilities of spaceborne SAR, only detecting rip currents from single-polarization imagery. Sentinel-1A satellite's high-resolution SAR data provides wide swath width and dual-polarization capabilities; therefore, there is an urgent need to design a rip current detection method based on Sentinel-1 dual-polarization SAR data.

[0008] Based on the above analysis, the problems and shortcomings of the existing technology are as follows:

[0009] (1) Currently, the spatial range of offshore current detection by ultra-high frequency shore-based radar is limited. Stations can only be built near designated coastlines, and offshore current detection cannot be carried out in sea areas where no stations are built.

[0010] (2) There are relatively few studies on detecting crack flow using spaceborne SAR data, and existing studies have not made full use of the multi-polarization capability of spaceborne SAR, only detecting crack flow from single-polarization images.

[0011] The difficulty in solving the above problems and defects is as follows:

[0012] (1) The spatial range of ocean current measurement by shore-based radar is limited, which is an inherent limitation of this method. SAR data can be used to detect offshore currents in a wider area, but preliminary screening from massive amounts of data is required.

[0013] (2) The number of satellites with multipolarization capability is limited, and work can only be carried out from the currently available multipolar satellite data.

[0014] The significance of solving the above problems and defects is as follows:

[0015] (1) Using SAR to detect rip currents is a supplement to observation methods such as shore-based radar, and can identify rip currents in a wider area.

[0016] (2) Currently, crack currents are only identified from single-polarization data of Envisat ASAR and ERS-2SAR satellites. Detecting rip currents using Sentinel-1SAR dual-polarization data fully utilizes the dual-polarization capability of the satellites to achieve more effective extraction of rip currents. Summary of the Invention

[0017] To overcome the problems existing in related technologies, this invention discloses an embodiment of an offshore current detection method, system, medium, device, and terminal based on Sentienl-1 dual-polarization SAR data. The technical solution is as follows:

[0018] This invention is implemented as follows: the rip current detection method includes:

[0019] Crack flows were detected using Sentienl-1 dual-polarization SAR images. Crack flow features were enhanced by combining VH and VV polarization data. Crack flows in SAR images were detected by edge detection operators and threshold segmentation methods.

[0020] In one embodiment, the offshore current detection method further includes:

[0021] Polarization combination parameters are extracted using VV and VH polarization data; the polarization difference (PD) is used to enhance the crack current features and improve the contrast between the offshore current and the background; then, the crack current is extracted from the SAR image using the edge detection operator and threshold segmentation method, and the relevant parameters of the crack current unit are estimated.

[0022] In one embodiment, the polarization combination parameters include polarization difference (PD), polarization ratio (PR), and polarization conversion ratio (PCR); wherein, the polarization difference (PD) represents the difference between the target and the background, as shown in the following formula:

[0023]

[0024] The polarization ratio PR is the ratio of the backscattering coefficients of VV polarization and VH polarization, as shown in the following formula:

[0025]

[0026] The polarization conversion ratio is used as an indicator to evaluate the performance of polarization conversion capability. The polarization conversion ratio (PCR) of incident b-polarization is defined as the ratio of depolarization to total polarization backscattering, as shown in the following formula:

[0027]

[0028] Another object of the present invention is to provide an offshore current detection system applying the aforementioned offshore current detection method, the offshore current detection system comprising:

[0029] The data preprocessing module is used to acquire Sentinel 1 data and preprocess the Sentinel 1 data.

[0030] The image cropping module is used to crop the preprocessed data;

[0031] The PD calculation module is used to obtain images containing cracked flows using VV polarization data and VH polarization data;

[0032] The gradient calculation module is used to calculate the threshold using the Sobel operator and then use the edge to obtain the binary gradient mask of the crack flow.

[0033] The closing and opening operation modules are used to perform closing and opening operations on images using the disk structure.

[0034] The connected component extraction module is used to extract the maximum connected component of the fractured flow and visualize the fractured flow extraction results.

[0035] Another object of the present invention is to provide a computer device comprising a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the following steps:

[0036] Crack flows were detected using Sentienl-1 dual-polarization SAR images. Crack flow features were enhanced by combining VH and VV polarization data. Crack flows in SAR images were detected by edge detection operators and threshold segmentation methods.

[0037] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps:

[0038] Crack flows were detected using Sentienl-1 dual-polarization SAR images. Crack flow features were enhanced by combining VH and VV polarization data. Crack flows in SAR images were detected by edge detection operators and threshold segmentation methods.

[0039] Another object of the present invention is to provide an information data processing terminal for implementing the offshore current detection system.

[0040] Combining all the above technical solutions, the advantages and positive effects of this invention are as follows:

[0041] First, this invention detects rip currents near the coast of Portugal using Sentienl-1 dual-polarization SAR images. It enhances the rip current features in the images by utilizing the polarization difference between VH and VV polarization data, and then extracts the rip currents from the SAR images using edge detection operators and threshold segmentation. This method achieves good detection results and can extract the area, extension length, and direction of the rip currents from the SAR images.

[0042] Secondly, this invention, targeting Sentinel-1 dual-polarization SAR data, selected two images containing rip currents. One image showed a rip current with high backscattering characteristics against a low backscattering background, while the other showed a rip current with low backscattering characteristics, appearing as a "sawtooth" shape in the image. Polarization operations (including PD, PR, and PCR) from VV and VH polarization data were used to enhance the crack current features. Table 3 clearly shows that in bright conditions, the crack current to background ratio in the PD image is as high as 3.6691, while that in the VV polarization NRCS image is only 1.1943. This indicates that the crack current to background ratio in the PD image is more than three times that of the VV polarization NRCS image. PR's enhancement effect on crack currents is slightly better than that in the NRCS image, but PCR's enhancement effect is not significant. In dark conditions, the background to crack current ratio in the PD image is the highest among the enhancement parameters, representing the best enhancement effect. By comparing and analyzing these three parameters, it can be found that polarization difference (PD) can effectively enhance the characteristics of crack flow, and can be used to enhance the detection of crack flow targets. Furthermore, by using edge detection operators and threshold segmentation, crack flow can be successfully extracted from SAR images. Finally, the morphology and distribution of crack flow are detected, and the parameters of crack flow units in the image, including area, extension length, and flow direction, can be estimated. Compared with manually outlining crack flow contours in SAR images, the number of pixels occupied by crack flow units detected by this method is almost the same, and the relative error between the two methods is approximately 2.91%. Therefore, it can be proven that this invention can effectively enhance crack flow characteristics, and the distribution of crack flow can be further detected from SAR images using edge detection operators and threshold segmentation.

[0043] Third, the offshore current detection method provided by this invention makes full use of Sentienl-1 dual-polarization SAR imagery to detect crack currents, combines VH and VV polarization data to enhance crack current features, and then uses edge detection operators and threshold segmentation methods to detect crack currents in SAR images.

[0044] Fourth, the technical solution of this invention, after being transformed, will increase research experience for the method of detecting rip currents through SAR images, and can provide relevant departments with another feasible way to detect rip currents, especially for situations where it is necessary to study rip currents that occurred at specific times and locations in the past.

[0045] Fifth, the technical solution of this invention has achieved the first detection of rip currents using Sentinel-1 dual-polarization spaceborne SAR data both domestically and internationally. This method fills the gap in the industry both domestically and internationally in the use of spaceborne SAR data and the dual-polarization capability of SAR to detect rip currents.

[0046] Sixth, this invention is the first to detect rip currents from massive Sentinel-1 SAR data, proving the feasibility of detecting rip currents from spaceborne SAR data.

[0047] Seventh, this invention is a new attempt to detect rip currents using the dual polarization characteristics of spaceborne SAR data, proving the feasibility of detecting rip currents from spaceborne SAR data. Attached Figure Description

[0048] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0049] Figure 1 This is a flowchart of the offshore current detection method provided in an embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram of the offshore current detection method provided in this embodiment of the invention;

[0051] Figure 3 This is a structural block diagram of the offshore current detection system provided in an embodiment of the present invention;

[0052] Figure 4 This is a schematic diagram of the fracture flow in the Sentinel-1 data amplitude image provided in an embodiment of the present invention, wherein, Figure 4 (a) is a schematic diagram of the fracture flow in the Sentinel-1 data amplitude image acquired on June 23, 2015. Figure 4 (b) is a schematic diagram of the rift flow in the Sentinel-1 data amplitude image acquired on September 21, 2016;

[0053] Figure 5 This is a schematic diagram illustrating the preprocessing of Sentinel-1 SAR data provided in an embodiment of the present invention, wherein... Figure 5 (a) is the VV polarization amplitude image before processing. Figure 5 (b) is the image after radiometric calibration. Figure 5 (c) is the topographically corrected image. Figure 5 (d) is the image after the land masking;

[0054] Figure 6 This is the VV polarization σ0 image of Sentinel-1 provided in this embodiment of the invention (September 21, 2016);

[0055] Figure 7This is a comparative schematic diagram of two types of crack flow characteristics provided in this embodiment of the invention in VV and VH polarized SAR amplitude data, wherein, Figure 7 (a) is a VH image acquired on June 23, 2015. Figure 7 (b) is a VV image acquired on June 23, 2015. Figure 7 (c) is a VH image acquired on September 21, 2016. Figure 7 (d) is a VV image acquired on September 21, 2016;

[0056] Figure 8 These are comparison images of two crack flow characteristics provided in this embodiment of the invention before and after PD, PR, and PCR enhancement, wherein... Figure 8 (a) is the NRCS image of the crack flow under bright feature conditions. Figure 8 (b) is the PD image of the crack flow under bright feature conditions. Figure 8 (c) is the PR image of the fractured flow under bright feature conditions. Figure 8 (d) is a PCR image of split flow under bright characteristic conditions. Figure 8 (e) is the NRCS image of the fractured flow under dark feature conditions. Figure 8 (f) is the PD image of the fracture flow under dark feature conditions. Figure 8 (g) is the PR image of the fractured flow under dark feature conditions. Figure 8 (h) is a PCR image of the split flow under dark characteristic conditions;

[0057] Figure 9 This is a schematic diagram of an artificial mask for the cracked flow and background area provided in an embodiment of the present invention, wherein, Figure 9 (a) is a schematic diagram of the fractured flow region with bright characteristics. Figure 9 (b) is a schematic diagram of the background area showing the highlighted features. Figure 9 (c) is a schematic diagram of the fractured flow region with dark features. Figure 9 (d) is a schematic diagram of the background area showing the dark features.

[0058] Figure 10 These are diagrams showing the results of each step in the complete offshore current detection process provided in this embodiment of the invention. Figure 10 (a) is the original SAR image. Figure 10 (b) is the image after PD enhancement. Figure 10 (c) is the binarized image. Figure 10 (d) is the image after the opening operation. Figure 10 (e) is the image after the closing operation. Figure 10 (f) is a schematic diagram of the reduced study area. Figure 10 (g) is the image after extracting the largest connected component. Figure 10(h) involves overlaying the extracted crack flow onto the original SAR image. Figure 10 (i) is the extracted crack flow contour line superimposed on the original image;

[0059] Figure 11 This is a schematic diagram of offshore current parameter calculation provided in an embodiment of the present invention;

[0060] Figure 12 These are experimental effect diagrams provided in the embodiments of the present invention, wherein, Figure 12 (a) is a schematic diagram of the detected rip current. Figure 12 (b) is a schematic diagram of the rip current extracted by human drawing;

[0061] In the diagram: 1. Data preprocessing module; 2. Image cropping module; 3. PD calculation module; 4. Grayscale stretching module; 5. Gradient calculation module; 6. Closing and opening operation module; 7. Connected component extraction module. Detailed Implementation

[0062] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0063] like Figure 1 As shown, the offshore current detection method provided in this embodiment of the invention includes the following steps:

[0064] S101, Data preprocessing: Obtain Sentinel 1 data and preprocess the Sentinel 1 data;

[0065] S102, Image cropping and PD calculation: The preprocessed data is cropped, and a local image containing the crack flow is obtained using VV polarization data and VH polarization data;

[0066] S103, Gray-scale stretching: Stretches the gray-scale of the PD image to enhance contrast;

[0067] S104, Calculate the gradient: Calculate the threshold using the Sobel operator, and then use the edge again to obtain the binary gradient mask of the crack flow;

[0068] S105, Closing and Opening Operations: Performing closing and opening operations on images using disk structures;

[0069] S106, Extracting the maximum connected component and visualizing offshore currents: Extract the maximum connected component of the rift current and visualize the rift current extraction results.

[0070] The schematic diagram of the rip current detection method provided in this embodiment of the invention is as follows: Figure 2 As shown.

[0071] like Figure 3 As shown, the offshore current detection system provided in this embodiment of the invention includes:

[0072] Data preprocessing module 1 is used to acquire Sentinel 1 data and preprocess the Sentinel 1 data;

[0073] Image cropping module 2 is used to crop the preprocessed data;

[0074] PD calculation module 3 is used to obtain a local image containing crack flow using VV polarization data and VH polarization data;

[0075] Gray-scale stretching module 4 is used to stretch the gray-scale of the PD image, thereby enhancing the contrast.

[0076] Gradient calculation module 5 is used to calculate the threshold using the Sobel operator and then use the edge to obtain the binary gradient mask of the crack flow.

[0077] Module 6, which performs closing and opening operations on images using a disk structure;

[0078] The connected component extraction module 7 is used to extract the maximum connected component of the fractured flow and visualize the fractured flow extraction results.

[0079] This invention discloses an offshore current detection method based on satellite remote sensing SAR data, belonging to the fields of computer vision and security technology. The invention aims to detect offshore currents occurring near the coast of Portugal using Sentienl-1 dual-polarization SAR imagery. It enhances the offshore current features in the imagery by utilizing the polarization difference between VH and VV polarization data, and then extracts the offshore current from the SAR imagery using edge detection operators and threshold segmentation. This method achieves good detection results, enabling the extraction of the area, length, and direction of the offshore current from SAR imagery.

[0080] Study Area and Data: This invention utilizes Sentinel-1 bipolar interferometric wide-strip (IW) data collected along the Portuguese coast on June 23, 2015, and September 21, 2016, with a spatial resolution of approximately 20 meters. Figure 4As shown in Table 1, the relevant parameters of the data include geographical location and wind field data at the time of data collection. Wind speed and direction were provided by the Figueira da Foz weather station in Portugal (https: / / www.worldweatheronline.com / figueira-da-foz-weather-historry / coimbra / pt.aspx).

[0081] The Portuguese coast is an exposed Atlantic coastline, with most beaches facing the eastern boundary of the North Atlantic. Satellite data shows that rip currents are present in numerous locations along the nearshore beaches of Portugal. The topography of the study area is characterized by quasi-continuous, underwater-wavering nearshore sandbars approximately 500 meters offshore, with well-defined boundaries and varying heights from 2 to 6 meters, characteristic of rip current channels. The average tidal range in this area is approximately 2.2 meters. According to wind field information provided by the Figueira da Foz weather station, the study area experienced low wind speeds during the SAR data acquisition period used in this invention.

[0082] Table 1 Research Data

[0083]

[0084]

[0085] Data preprocessing: First, preprocessing is performed on the two Sentinel-SAR images, mainly including the following steps: orbit correction, radiometric calibration, terrain correction, and land masking. Specifically, to reduce the impact of radiometric bias and obtain the target's backscattering coefficient, radiometric calibration is performed on the images. The calibrated images are shown below. Figure 5 As shown in (b), the SAR image at this time has geometric distortion, so terrain correction is required. This invention uses a method based on the Range-Doppler (RD) model for terrain correction. The processed image is as follows. Figure 5 As shown in (c). The image behind the land cover is as follows. Figure 5 As shown in (d).

[0086] Data analysis reveals two main characteristics of rip currents: one is a "bright" characteristic, corresponding to situations where rip current units exhibit high backscattering against a low backscattering background, such as... Figure 4 As shown in (a); another is the "dark" feature, corresponding to the case where the offshore current unit has low backscattering, such as... Figure 4 As shown in (b).

[0087] Brightness feature analysis of offshore currents:

[0088] In cases of bright features, rip currents appear as enhanced backscattering patches because breaking waves appear as bright bands against a sea surface background with low backscattering. The white dashed box marks the location of the rip current, and this invention shows the rip current neck and head. The rip current erupts from the shore, forming a relatively narrow rip current neck, which then spreads outwards to form a "rounded" rip current head with a larger area.

[0089] A wind contrast model based on a simple first-order Bragg scattering theory can be used to explain the enhanced backscattering characteristics of the rift unit. In this model, the modulation effect of the relative changes in rift and wind speed on the Bragg wave is considered. The variation in the wind-wave spectrum is described by the equilibrium equation of the wave action spectral density N(k, x, t).

[0090]

[0091] Where β and γ represent the wave growth rate and wave damping coefficient, respectively; δ is a phenomenological coefficient describing the nonlinear limit of the wave spectrum; K w It's a contrast of wind.

[0092]

[0093] Where k is the wave number of the surface wave, and β and β0 are the wave growth rates under and without turbulence influence, respectively. The expression for β0 assumes that wave generation is isotropic, i.e.:

[0094] β0=0.04(U0 *2 k 2 / ω) (3)

[0095]

[0096] In the formula, U0 * It is the frictional velocity unaffected by the ocean current in (3), and ω is the natural frequency of the surface wave, given by the following formula:

[0097]

[0098] Where g is the acceleration due to gravity, σ is the surface tension, and ρ is the density of water. In (4), U* is the wind friction velocity, θ is the angle between the surface wave propagation and the wind direction, and c p It is the phase velocity of the wave.

[0099] When V w When the speed is less than 7 m / s, the empirical formula U* = 0.034V can be used. w The wind speed V at a standard height of 10m is obtained. w Relationship with U*. Typical offshore current velocity U c The corresponding effective sea surface wind speed V e :

[0100] V e =V w +U c (6)

[0101]

[0102] Where θ v For V w with U c The included angle.

[0103] For the Sentinel-1 SAR image from June 23, 2015, V calculated based on the CMOD5 model e Approximately 2.4 m / s. (Using V) w =2.4 m / s (direction 260.5°) is the background wind speed, U c = 0.75 m / s (typical fracture velocity), θ v At 39°, V is obtained in the turbulent flow. e =2.8 m / s, far exceeding the threshold for generating Bragg waves (V≈2.0 m / s). This explains why the rift flow exhibits enhanced backscattering characteristics against a dark background. Wind near the threshold compared to K w It should be large enough to produce a strong positive contrast signal. At very low wind speeds, the wind speed is below the threshold that produces any measurable radar backscattering. However, once the “effective” wind speed increases, it may rise above the wind threshold (above the radar noise lower limit) due to the presence of strong rifts, thus producing a measurable SAR backscattering level.

[0104] Dark feature analysis of rip currents:

[0105] For the dark features, the images were acquired under low wind speeds, and some oil slicks are visible near the shore, giving the surf area a "jagged" appearance. This is manifested as a seemingly periodic alternation between the nearshore oil slick and the wave breaking field. This is caused by a non-uniform wave breaking field, which produces higher backscattering compared to areas where breaking has not occurred. The oil slicks in the image correspond to deeper waters (rift channels) where the waves are less steep and less prone to breaking.

[0106] On a uniformly open beach along the coast, obliquely incident waves drive the currents in the nearshore surf area. It is well known that coastal currents can be unstable because they may undergo shearing across the coast. Shear-unsteady rifts are not common and only occur on uniformly open nearshore beaches when exposed to highly obliquely incident waves. On the beaches along the Portuguese coast, waves are incident at a large angle to the shore, consistent with the basic characteristics of shear-unsteady rifts. Therefore, it can be preliminarily determined that... Figure 6 The fractured flow in the middle is a shear-unsteady fractured flow.

[0107] Furthermore, the fracture characteristics in VV polarization images and VH polarization images are also different. For example... Figure 7 As shown, two types of split flow features can be clearly seen in the VV polarization image, while they are completely invisible in the VH polarization image.

[0108] For bright features, the ratio of crack currents to sea surface background in the VV image is approximately 1.19, while in the VH image this ratio is approximately 0.96. For dark features, the ratio of crack currents to sea surface background in the VV image is approximately 0.78, while in the VH image this ratio is approximately 0.98. Table 2 shows the ratio of crack currents to background in the VV and VH images for bright and dark features.

[0109] Table 2. Ratios of backscattering coefficients between fractured flow and ocean background under different polarizations

[0110]

[0111] Polarization combination enhancement:

[0112] Radar acquires target information by emitting electromagnetic waves and receiving scattered echoes. Polarization is an inherent characteristic of electromagnetic waves, representing the trajectory of an electric field oscillating in a plane perpendicular to its propagation direction. This invention proposes three parameters for polarization combination to enhance the crack flow characteristics.

[0113] Polarization difference:

[0114] Polarized Difference (PD) is the difference between the backscattering coefficients of VV polarization and VH polarization. In this invention, its physical meaning represents the difference between the target and the background, as shown in Equation (8). Because in VH images, the crack flow is almost mixed with the background and is difficult to distinguish, while in VV images, the crack flow and the background show different characteristics.

[0115]

[0116] Polarization ratio (PR):

[0117] The polarization ratio (PR) is the ratio of the backscattering coefficients of VV polarization to VH polarization, as shown in equation (9).

[0118]

[0119] Polarization conversion ratio (PCR):

[0120] The polarization conversion ratio (PCR) is a metric used to evaluate the performance of polarization conversion capability. The polarization conversion ratio (PCR) of incident beta polarization is defined as the ratio of depolarization to total polarization backscattering, as shown in the following expression:

[0121]

[0122] Where σ0 represents the scattering coefficients of polarizations a and b; a and b can be linearly polarized or circularly polarized, and a≠b. PCR is used to characterize the dielectric inhomogeneity (dielectric profile) and spatial anisotropy (roughness) of rough surfaces, and can serve as a new and reliable index for surface parameter inversion. In this invention, the PCR parameter can be used to enhance the characteristics of offshore currents.

[0123] The ability to identify fractured flows can be enhanced by combining parameters through polarization operations. Figure 8 The images show comparisons before and after PD, PR, and PCR enhancement.

[0124] To observe the enhancement effect of the three parameters on the fractured flow, masks of the fractured flow region and the background region were manually drawn, such as... Figure 9 As shown.

[0125] The average values ​​of VV polarization NRCS, PD, PR, and PCR under the mask are calculated below. Table 3 shows the ratio of crack flow to background for VV polarization NRCS, PD, PR, and PCR under bright and dark feature conditions. The former is the target value divided by the background value, and the latter is the background value divided by the target value.

[0126] Table 3. Ratio of fracture flow to background flow after enhancement with different parameters

[0127]

[0128] In bright features, the ratio of crack flow to background in PD images is more than three times that of VV polarized NRCS images. PR enhances crack flow slightly better than NRCS images, but PCR does not significantly enhance crack flow.

[0129] In the case of dark features, the background-to-fracture ratio of the PD image is the highest among the enhancement parameters, representing the best enhancement effect. The enhancement effects of PR and PCR are negligible because their background-to-fracture ratios are lower than those of NRCS.

[0130] Therefore, it is clear that PD more effectively displays the characteristics of fractured flow compared to the other parameters in Table 3. Thus, PD will be used to enhance the detection of fractured flow targets.

[0131] Detection methods and analysis

[0132] This invention presents a method for detecting rip currents near the coast of Portugal by identifying and extracting them from Sentinel-1 SAR imagery. The method utilizes VV and VH polarization data to extract polarization combination parameters, such as polarization difference (PD), polarization ratio (PR), and polarization conversion ratio (PCR). In the Sentinel-1 SAR image example used in this invention, the enhancement effect of polarization difference (PD) is optimal. Therefore, by enhancing the crack current features through polarization difference (PD), the contrast between the rip current and the background is improved. Furthermore, by using edge detection operators and threshold segmentation, crack currents can be successfully extracted from SAR imagery. Moreover, relevant parameters of crack current units, including crack current area, extension length, and direction, can be estimated, achieving good detection results.

[0133] Offshore current detection principle: This invention primarily targets the detection of offshore currents exhibiting high backscattering against a dark background. First, the PCR (Potential Chronology of Offshore Currents) is calculated using VV and VH polarization data to highlight offshore current features and enhance the contrast between the offshore current and the background. Then, a binary gradient mask is obtained using the Sobel operator. Next, closing and opening operations are performed to extract the contour information of the offshore current units. Based on the connected component characteristics of local regions, only the offshore current is retained, while other background interference elements are removed. Finally, the offshore current units are visualized. The complete offshore current detection process includes: data preprocessing, image cropping, PD (Potential Chronology of Offshore Currents) calculation, grayscale stretching, gradient calculation, closing and opening operations, extraction of the largest connected component, and offshore current visualization, etc. Figure 10 As shown.

[0134] Offshore current detection experiment: Based on the principle of offshore current detection, the preprocessed data is first cropped, and local images containing crack currents are obtained using VV and VH polarization data. Then, grayscale stretching is performed on the PD image to further enhance contrast. Next, the Sobel operator is used to calculate the threshold, and the binary gradient mask of the crack currents is obtained again using the edges. Then, a "disk" structure is used to perform closure and opening operations on the image. The gradient mask after erosion and dilation clearly shows the distribution of crack currents. Finally, the maximum connected component of the crack currents is extracted, and the crack current extraction results are visualized. The results of each step in the complete crack current detection process are as follows: Figure 11 As shown.

[0135] Offshore current parameter calculation: Total area of ​​an offshore current unit: By counting the number of pixels with a value of 1 in the binarized image and multiplying it by the actual area represented by each pixel, the area of ​​a single offshore current can be estimated. In this invention, the area of ​​the offshore current is approximately 99531 m². 2 .

[0136] The maximum extension length of the rip current unit can be calculated by multiplying the length of the longest side of the lowest circumscribed rectangle by the actual length represented by a unit pixel. In this invention, the maximum extension length of the rip current is approximately 552m.

[0137] The angle between the rip current unit's direction and the coastline: The coastline of the west coast of Portugal is nearly straight. The exit angle of the rip current can be calculated using the coastline angle and the angle of the lowest bounding rectangle of the individual rip current. In this invention, the angle between the rip current direction and the coastline is calculated as follows: the angle between the rip current direction and the coastline is approximately 55°. Figure 12 The angle indicated by the arc is shown.

[0138] Conclusion: This invention selected two images containing rip currents from Sentinel-1 dual-polarization SAR data. One image showed a rip current with high backscattering characteristics against a low backscattering background, while the other showed a low backscattering characteristic, appearing as a "sawtooth" pattern in the image. Polarization operations (PD, PR, and PCR) from VV and VH polarization data were used to enhance the crack current features. Table 3 clearly shows that in bright conditions, the crack current to background ratio in the PD image is as high as 3.6691, while that in the VV polarization NRCS image is only 1.1943. This indicates that the crack current to background ratio in the PD image is more than three times that of the VV polarization NRCS image. PR has a slightly better enhancement effect on crack currents than NRCS images, but PCR has no significant enhancement effect. In dark conditions, the background to crack current ratio in the PD image is the highest among the enhancement parameters, representing the best enhancement effect. By comparing and analyzing these three parameters, it can be found that polarization difference (PD) can effectively enhance the characteristics of crack flow, and can be used to enhance the detection of crack flow targets. Furthermore, by using edge detection operators and threshold segmentation, crack flow can be successfully extracted from SAR images. Finally, the morphology and distribution of crack flow are detected, and the parameters of crack flow units in the image, including area, extension length, and flow direction, can be estimated. Compared with manually outlining crack flow contours in SAR images, the number of pixels occupied by crack flow units detected by this method is almost the same, and the relative error between the two methods is approximately 2.91%. Therefore, it can be proven that this invention can effectively enhance crack flow characteristics, and the distribution of crack flow can be further detected from SAR images using edge detection operators and threshold segmentation.

[0139] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented, in whole or in part, as a computer program product, the computer program product includes one or more computer instructions. When the computer program instructions are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0140] Evidence of the effects of the embodiments: such as Figure 12 As shown in (a) to (b), to verify the experimental results, the outline of the fracture flow was manually drawn from the original SAR image, and the number of internal pixels was obtained. To reduce subjective error, the experiment was conducted 20 times, with an average value of approximately 619, which was taken as the true value. It is known that the number of pixels inside the fracture flow cells extracted by the computer is 637, thus the absolute error of the experiment is 18 (pixels), and the relative error is approximately 2.91%. Therefore, it can be proven that the method proposed in this invention can extract fracture flows effectively.

[0141] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for detecting rip currents, characterized in that, The offshore current detection method is as follows: crack currents are detected using Sentienl-1 dual-polarization SAR images, crack current features are enhanced by combining VH and VV polarization data, and crack currents in SAR images are detected by edge detection operators and threshold segmentation methods. The rip current detection method further includes: Polarization combination parameters were extracted using VV polarization and VH polarization data. These polarization combination parameters included polarization difference (PD), polarization ratio (PR), and polarization conversion ratio (PCR). The polarization ratio PR is the ratio of the backscattering coefficients of VV polarization and VH polarization, as shown in the following formula: ； The polarization difference (PD) represents the difference between the target and the background, as shown in the following formula: ； The polarization conversion ratio (PCR) is an indicator used to evaluate the performance of polarization conversion capability. The PCR of incident b-polarization is defined as the ratio of depolarization to total polarization backscattering, as shown in the following formula: ； The polarization difference (PD) is used to enhance the characteristics of the fractured current and improve the contrast between the offshore current and the background. The offshore current includes bright offshore currents with high backscattering against a low backscattering background and dark offshore currents with low backscattering and a sawtooth appearance. The PD-enhanced image is then subjected to grayscale stretching to further enhance contrast. Then, the crack currents are extracted from the SAR image using an edge detection operator and a threshold segmentation method, and the relevant parameters of the crack current units are estimated. The extraction of crack currents from the SAR image using the edge detection operator and threshold segmentation method includes: calculating the threshold using the Sobel operator and obtaining the binary gradient mask of the crack currents; performing opening and closing operations on the image sequentially using a disk structure; and extracting the maximum connected component of the crack currents as the detection result. The relevant parameters include the total area of ​​the rip current, the maximum extension length, and the angle between the flow direction and the coastline.

2. An offshore current detection system applying the offshore current detection method as described in claim 1, characterized in that, The rip current detection system includes: The data preprocessing module (1) is used to acquire Sentinel 1 data and preprocess the Sentinel 1 data; Image cropping module (2) is used to crop the preprocessed data; The PD calculation module (3) is used to obtain a local image containing crack flow using VV polarization data and VH polarization data, thereby enhancing the contrast. The gradient calculation module (5) is used to calculate the threshold using the Sobel operator and then use the edge to obtain the binary gradient mask of the crack flow. The closing and opening operation module (6) is used to perform closing and opening operations on images using graph structures; The connected component extraction module (7) is used to extract the maximum connected component of the fractured flow and visualize the fractured flow extraction results.

3. A computer device for performing the offshore current detection method according to claim 1, characterized in that, The computer device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the following steps: Crack flows were detected using Sentienl-1 dual-polarization SAR images. Crack flow features were enhanced by combining VH and VV polarization data. Crack flows in SAR images were detected by edge detection operators and threshold segmentation methods.

4. A computer-readable storage medium for storing data of the offshore current detection method of claim 1, comprising a computer program that, when executed by a processor, causes the processor to perform the following steps: Crack flows were detected using Sentienl-1 dual-polarization SAR images. Crack flow features were enhanced by combining VH and VV polarization data. Crack flows in SAR images were detected by edge detection operators and threshold segmentation methods.

5. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the offshore current detection system as described in claim 2.

Images