Adaptive statistical sea clutter suppression method based on x-band marine radar

Abstract

The application discloses an adaptive statistical sea clutter suppression method based on an X-band marine radar, which comprises the following steps: 1, obtaining a radar sample image; 2, intercepting regional sample data; 3, calculating the root mean square error S of the regional sample; 4, setting a sliding window; 5, obtaining a radar real-time image and caching; 6, sliding window processing, which specifically comprises: determining the sliding range of the sliding window; calculating the intensity Vc of the central pixel point and the root mean square error Sn of the intensity of all the pixel points except Vc; by comparing Sn with S, the threshold threshold Thr of the current sliding window is adaptively determined; then Vc is compared with Thr to obtain the intensity value of the central pixel point after sea clutter suppression in the current sliding window. c c The application adopts the method of environment detection to select a suitable threshold to suppress sea clutter and improve the signal-to-noise ratio, and simultaneously adopts the method of region to adaptively calculate the threshold, which conforms to the regional characteristic of sea clutter, can improve the signal-to-noise ratio and reduce the false alarm and missed alarm probability.​

Description

Technical Field

[0001] This invention relates to the fields of harbor environmental monitoring and ship management, and in particular to an adaptive statistical sea clutter suppression method based on X-band marine radar. Background Technology

[0002] In various scenarios of marine radar applications, the rapid and reliable detection of maritime targets is becoming increasingly important. However, the operating environment of marine radar is complex and variable, and the received echo signals include not only targets but also noise, interference, and sea clutter. Among these signal types, sea clutter is the most complex. Due to its high power level, significant non-Gaussian and non-stationary characteristics, and its susceptibility to environmental influences, sea clutter is a significant factor affecting target detection capabilities. Based on practical needs, research on sea clutter characteristics is needed to understand its variation patterns, thereby achieving effective sea clutter suppression and further improving the radar signal-to-noise ratio.

[0003] Extensive research has been conducted both domestically and internationally on the characteristics of sea clutter. Generally, this research can be categorized into two levels: first, research based on physical mechanisms, which assumes the sea surface follows a theoretical model and, within a certain parameter range, uses electromagnetic scattering theory and the sea surface model to calculate backscattered echoes and study sea clutter characteristics; second, research based on data, which involves conducting actual experiments based on research needs, using sea clutter experimental data to analyze sea clutter characteristics and their variation trends with radar parameters, marine environmental parameters, and other factors, and establishing sea clutter models with or without considering certain mechanisms. These two parts of research are complementary.

[0004] Currently, widely used sea clutter suppression methods include STC, CFAR, and slow threshold. However, existing methods generally do not consider the regional characteristics of sea clutter, because sea clutter is usually a regional signal in patches, and the echo signal of sea clutter changes frequently, which can easily cause false alarms or missed alarms, thus affecting the detection effect of target signals. Based on a systematic review of the more representative sea clutter suppression methods, this paper proposes an adaptive statistical sea clutter suppression method based on X-band marine radar, focusing on the research content of sea clutter amplitude distribution characteristics, spectral characteristics, correlation, and non-stationary and nonlinear characteristics. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing an adaptive statistical sea clutter suppression method based on X-band marine radar. This adaptive statistical sea clutter suppression method based on X-band marine radar can sense changes in the sea clutter environment around the target in real time and generate a set of sea clutter suppression thresholds that change in real time with the intensity of sea clutter, thereby improving the signal-to-noise ratio and increasing the probability of target detection in strong sea clutter backgrounds.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] An adaptive statistical sea clutter suppression method based on X-band marine radar includes the following steps.

[0008] Step 1: Acquire radar sample images: Read several radar images of the ocean to be monitored from the X-band marine radar within the set monitoring time, and select the radar images that meet the set sample indicators from all the radar images read as sample images; among them, the sample indicators include target sample indicators and sea clutter sample indicators.

[0009] Step 2: Extract regional sample data: From the radar sample image obtained in Step 1, extract m target areas and m sea clutter areas; where m ≥ 5; the m target areas must include all the set target sample indicators; the m sea clutter areas must include all the set sea clutter sample indicators; each target area and each sea clutter area is a square with an area of ​​TSize.

[0010] Step 3: Calculate the root mean square error S of the regional samples: For the m target regions and m sea clutter regions extracted in Step 2, calculate the root mean square error S of the pixel intensity.

[0011] Step 4, Set up the sliding window: Let the size of the sliding window be n×n, then the formula for calculating n is:

[0012]

[0013] in:

[0014]

[0015] In the formula, n t This is for intermediate computational costs.

[0016] Step 5: Acquire and cache real-time radar images: Acquire real-time radar images and cache n consecutive radial data points; the n radial data points are ordered in chronological order as the 1st, 2nd, ..., n-1st and nth data points.

[0017] Step 6, sliding window processing, specifically includes the following steps.

[0018] Step 6-1: Determine the sliding window range: Take the (n+1) / 2th radial data cached in Step 5 as the radar data to be processed; for the radar data to be processed, the sliding window starts from the radar center origin and slides pixel by pixel according to the principle and order of azimuth 0-360° and distance from 0m to the maximum detection range of the radar.

[0019] Step 6-2, Calculate V C : Calculate the intensity V of the center pixel of the current sliding window C .

[0020] Step 6-3, Calculate S n :S n This is the root mean square error of the intensity of all pixels within the current sliding window, excluding the intensity of the central pixel.

[0021] Step 6-4: Calculate Thr C : S in step 6-3 n Compare the value of S calculated in step 3 with the pixel intensity value whose probability is at the critical point between the target data and sea clutter data, and use it as the threshold value Thr of the current sliding window. C .

[0022] Step 6-5, Sliding Window Processing: Set the current sliding window's V... C Thr with the current sliding window threshold C By comparison, the intensity value Vf of clutter suppression at the central pixel within the current sliding window is obtained, specifically:

[0023]

[0024] Step 6-6, continuous sliding window processing: Slide the window sequentially to the next pixel for clutter suppression processing. Repeat steps 6-2 to 6-6 until the (n+1) / 2th radial data is processed. Output the clutter suppression intensity value Vf of the central pixel after all sliding window processing.

[0025] Step 7, Cache Update: Delete the first radial data that was cached in Step 5, and update the second, ..., n-1 and nth radial data that were cached in Step 5 to the first, second, ... and n-1th radial data that were cached in the new cache, respectively; at the same time, read a new radial data as the nth radial data that was cached in the new cache; repeat Steps 6 to 7 to complete the sea clutter suppression processing of the radar real-time image.

[0026] In step 1, the target sample index includes the ship length L, which includes at least the following five cases: (1) L≤50m; (2) 50m<L≤100m; (3) 100m<L≤150m; (4) 150m<L≤200m; (5) L>200m.

[0027] In step 1, the sea clutter sample indicators include wave height H, which includes at least the following four cases: (1) H≤1m; (2) 1m<H≤1.5m; (3) 1.5m<H≤2m; (4) H≥2m.

[0028] In step 2, Tsize > T0, where T0 is the total number of pixels occupied by the largest target in the captured target area.

[0029] In step 2, Tsize = 1.5T0.

[0030] In step 6-4, Thr C The specific calculation method is as follows:

[0031] (1) When S n <S, at this time the dispersion of the data within the sliding window is small, indicating that sea clutter data is predominant;

[0032] a) When V C ≥V mean At this time, the probability that the central pixel is the target data is high;

[0033] If V C <TV, then Thr C =V mean ;

[0034] If V C ≥TV, then Thr C =V max ;

[0035] In the formula, V mean and V max These are the average and maximum intensity values ​​of all pixels within the current sliding window, excluding the central pixel; TV is the average intensity value of the m target regions captured in step 2.

[0036] b) When V C <V mean At this time, the central pixel is likely to be sea clutter data;

[0037] If V C <SV, then Thr C =V max ;

[0038] If V C If ≥SV, then Thr C =V mean ;

[0039] In the formula, SV is the average intensity value of the m sea clutter regions extracted in step 2;

[0040] (2) When S n≥S, at this point the dispersion of the data within the sliding window is relatively large, indicating that the target data is predominant;

[0041] a) When V C ≥V mean At this time, the probability that the central pixel is the target data is high;

[0042] If V C <V max And V C If ≥SV, then Thr C =V mean ;

[0043] If V C <V max And V C <SV, then Thr C =V max ;

[0044] If V C ≥V max And V C ≥TV, then Thr C =V min ;

[0045] If V C ≥V max And V C <TV, then Thr C =V mean ;

[0046] In the formula, V min It is the minimum intensity of all pixels within the current sliding window, excluding the central pixel.

[0047] b) When V C <V mean At this time, the central pixel is likely to be sea clutter data;

[0048] If V C <V min Then Thr C =V max ;

[0049] If V C ≥V min And V C If ≥SV, then Thr C =V mean ;

[0050] If V C ≥V min And V C <SV, then Thr C =V max .

[0051] In step 6-4, V mean and S n The calculation formulas are as follows:

[0052]

[0053]

[0054] In the formula, V1, V2, ..., V n*n-1 These are the intensity values ​​of the n*n-1 pixels within the current sliding window, excluding the central pixel.

[0055] In step 2, m = 5.

[0056] The present invention has the following beneficial effects:

[0057] This invention employs an environmental detection method to select an appropriate threshold to suppress sea clutter and improve the signal-to-noise ratio. Furthermore, this method uses a regional approach to calculate the threshold, which better reflects the regional occurrence of sea clutter, effectively selecting the threshold, improving the signal-to-noise ratio, and reducing the probability of false alarms and missed alarms. Attached Figure Description

[0058] Figure 1 The flowchart of an adaptive statistical sea clutter suppression method based on X-band marine radar according to the present invention is shown.

[0059] Figure 2 A schematic diagram illustrating the processing principle of sliding windows in this invention is shown. Detailed Implementation

[0060] The present invention will now be described in further detail with reference to the accompanying drawings and specific preferred embodiments.

[0061] In the description of this invention, it should be understood that the terms "azimuth" and "distance" refer to the coordinate system relationship centered on the radar, and "radial data" represents the set of all pixels in a certain azimuth, also known as "main trigger".

[0062] like Figure 1 As shown, an adaptive statistical sea clutter suppression method based on X-band marine radar includes the following steps.

[0063] Step 1: Acquire radar sample images: Read several radar images of the ocean to be monitored from the X-band marine radar within the set monitoring time, and select the radar images that meet the set sample indicators from all the radar images read as sample images; among them, the sample indicators include target sample indicators and sea clutter sample indicators.

[0064] The target sample indicators mentioned above include ship length L, which must include at least the following five cases:

[0065] (1) L≤50m;

[0066] (2) 50m < L ≤ 100m;

[0067] (3) 100m < L ≤ 150m;

[0068] (4) 150m < L ≤ 200m;

[0069] (5) L>200m.

[0070] The above sea clutter sample indicators, including wave height H, must include at least the following four cases:

[0071] (1) H≤1m;

[0072] (2) 1m < H ≤ 1.5m;

[0073] (3) 1.5m < H ≤ 2m;

[0074] (4) H≥2m.

[0075] In order for the acquired radar sample images to have the above-mentioned sea clutter sample indicators, the above-mentioned monitoring time should generally be no less than one month, preferably one month.

[0076] Step 2: Extract regional sample data: For the radar sample image obtained in Step 1, preferably use the B display local magnification method to observe the intensity and outline of the target, as well as the intensity and distribution of sea clutter, and extract m target areas and m sea clutter areas; where m≥5, preferably m=5.

[0077] The m-block target area must include all the set target sample indicators; the m-block sea clutter area must include all the set sea clutter sample indicators.

[0078] Each target region and each sea clutter region is a square with an area of ​​TSsize. Here, Tsize > T0, where T0 is the total number of pixels occupied by the largest target (e.g., a ship with a length L > 200m) in the captured target region. In this embodiment, Tsize is preferably 1.5T0.

[0079] Step 3: Calculate the root mean square error S of the regional samples: For the m target regions and m sea clutter regions extracted in Step 2, calculate the root mean square error S of the pixel intensity.

[0080] The preferred method for calculating the root mean square error S includes the following steps.

[0081] Step 3-1: Calculate the average intensity TV of the target sample data. The specific calculation formula is as follows:

[0082]

[0083] In the formula, TV1, TV2, ..., TV i TV m are the average values ​​of pixels in the target regions of the 1st, 2nd, i-th, and m-th blocks, respectively; where 1≤i≤m.

[0084] Step 3-2: Calculate the average intensity SV of the sea clutter sample data. The specific calculation formula is as follows:

[0085]

[0086] In the formula, SV1, SV2, ..., SV i 、…、SV m These are the average values ​​of pixels within the sea clutter regions of the 1st, 2nd, i-th, and m-th blocks, respectively.

[0087] Step 3-3: Calculate the root mean square error S. The specific calculation formula is as follows:

[0088]

[0089] in:

[0090]

[0091] In the formula, Let be the average intensity of pixels in the m target regions and the m sea clutter regions.

[0092] Step 4, Set up the sliding window: Let the size of the sliding window be n×n, then the formula for calculating n is:

[0093]

[0094] in:

[0095]

[0096] In the formula, n t This is for intermediate computational costs.

[0097] Step 5: Acquire and cache real-time radar images: Acquire real-time radar images and cache n consecutive radial data points; the n radial data points are ordered in chronological order as the 1st, 2nd, ..., n-1st and nth data points.

[0098] Step 6, sliding window processing, specifically includes the following steps.

[0099] Step 6-1: Determine the sliding window range: Take the (n+1) / 2th radial data cached in Step 5 as the radar data to be processed; for the radar data to be processed, the sliding window starts from the radar center origin and slides pixel by pixel according to the principle and order of azimuth 0-360° and distance from 0m to the maximum detection range of the radar. The specific window processing principle is illustrated in the diagram below. Figure 2 As shown.

[0100] Since the pixels at 0°, start distance, and end distance cannot be combined to form an n×n sliding window, no sliding window processing is performed, and the processing result closest to it is used instead.

[0101] Step 6-2, Calculate V C : Calculate the intensity V of the center pixel of the current sliding window C .

[0102] Step 6-3, Calculate S n :S n This is the root mean square error of the intensity of all pixels within the current sliding window, excluding the central pixel. The specific calculation formula is as follows:

[0103] In step 6-4, V mean and S n The calculation formulas are as follows:

[0104]

[0105] in:

[0106]

[0107] In the formula, V1, V2, ..., V n*n-1 These are the intensity values ​​of the n*n-1 pixels within the current sliding window, excluding the central pixel.

[0108] V mean It is the average intensity of all pixels within the current sliding window, excluding the central pixel.

[0109] Step 6-4: Calculate Thr C : S in step 6-3 n Compare the value of S calculated in step 3 with the pixel intensity value whose probability is at the critical point between the target data and sea clutter data, and use it as the threshold value Thr of the current sliding window. C .

[0110] The above Thr C The preferred specific calculation method is:

[0111] (1) When S n <S, at this time the dispersion of the data within the sliding window is small, representing that sea clutter data is predominant.

[0112] a) When V C ≥V mean At this time, the probability that the central pixel is the target data is high.

[0113] If V C <TV, then Thr C =V mean .

[0114] If V C ≥TV, then Thr C =V max .

[0115] In the formula, V max This represents the maximum intensity of all pixels within the current sliding window, excluding the central pixel.

[0116] b) When V C <V mean At this time, the probability that the central pixel is sea clutter data is high.

[0117] If V C <SV, then Thr C =V max .

[0118] If V C If ≥SV, then Thr C =V mean .

[0119] (2) When S n ≥S indicates that the data dispersion within the sliding window is relatively large, representing that the target data is predominant.

[0120] a) When V C ≥V mean At this time, the probability that the central pixel is the target data is high.

[0121] If V C <V max And V C If ≥SV, then Thr C =V mean .

[0122] If V C <V max And V C <SV, then Thr C =V max .

[0123] If V C ≥V max And V C ≥TV, then Thr C =V min .

[0124] If V C ≥V max And V C <TV, then Thr C =V mean .

[0125] In the formula, V min It is the minimum intensity of all pixels within the current sliding window, excluding the central pixel.

[0126] b) When V C <V mean At this time, the probability that the central pixel is sea clutter data is high.

[0127] If V C <V min Then Thr C =V max .

[0128] If V C ≥V min And V C If ≥SV, then Thr C =V mean .

[0129] If V C ≥V min And V C <SV, then Thr α =V max .

[0130] Step 6-5, Sliding Window Processing: Set the current sliding window's V... C Thr with the current sliding window threshold C By comparison, the intensity value Vf of clutter suppression at the central pixel within the current sliding window is obtained, specifically:

[0131]

[0132] Step 6-6, continuous sliding window processing: Slide the window sequentially to the next pixel for clutter suppression processing. Repeat steps 6-2 to 6-6 until the (n+1) / 2th radial data is processed. Output the clutter suppression intensity value Vf of the central pixel after all sliding window processing.

[0133] Step 7, Cache Update: Delete the first radial data that was cached in Step 5, and update the second, ..., n-1 and nth radial data that were cached in Step 5 to the first, second, ... and n-1th radial data that were cached in the new cache, respectively; at the same time, read a new radial data as the nth radial data that was cached in the new cache; repeat Steps 6 to 7 to complete the sea clutter suppression processing of the radar real-time image.

[0134] This invention uses X-band marine radar to detect targets and employs a sliding window threshold method to adaptively suppress sea clutter, thereby effectively improving the radar's signal-to-noise ratio and enhancing its target detection capability.

[0135] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. An adaptive statistical sea clutter suppression method based on X-band marine radar, characterized in that: The steps include the following: Step 1: Acquire radar sample images: Read several radar images of the ocean to be monitored from the X-band marine radar within the set monitoring time, and select the radar images that meet the set sample indicators from all the radar images read as sample images; wherein, the sample indicators include target sample indicators and sea clutter sample indicators. Step 2: Extract regional sample data: From the radar sample image obtained in Step 1, extract m target regions and m sea clutter regions; where m ≥ 5; the m target regions must include all the set target sample indicators; the m sea clutter regions must include all the set sea clutter sample indicators; each target region and each sea clutter region is a square with an area of ​​TSize. Step 3: Calculate the root mean square error S of the region samples: For the m target regions and m sea clutter regions extracted in Step 2, calculate the root mean square error S of the pixel intensity. Step 4, Set up the sliding window: Let the size of the sliding window be n×n, then the formula for calculating n is: in: In the formula, n t This is for intermediate calculations; Step 5: Acquire and cache real-time radar images: Acquire real-time radar images and cache n consecutive radial data points; the n radial data points are ordered in chronological order as the 1st, 2nd, ..., (n-1)th, and nth data points. Step 6, sliding window processing, specifically includes the following steps: Step 6-1: Determine the sliding window range: Take the (n+1) / 2th radial data cached in Step 5 as the radar data to be processed; for the radar data to be processed, the sliding window starts from the radar center origin and slides pixel by pixel according to the principle and order of azimuth 0-360° and distance from 0m to the maximum detection range of the radar. Step 6-2, Calculate V C : Calculate the intensity V of the center pixel of the current sliding window C ; Step 6-3, Calculate S n :S n The root mean square error of the intensity of all pixels within the current sliding window, excluding the intensity of the central pixel. Step 6-4: Calculate Thr C : S in step 6-3 n Compare the value of S calculated in step 3 with the pixel intensity value whose probability is at the critical point between the target data and sea clutter data, and use it as the threshold value Thr of the current sliding window. C ; Step 6-5, Sliding Window Processing: Set the current sliding window's V... C Thr with the current sliding window threshold C By comparison, the intensity value Vf of clutter suppression at the central pixel within the current sliding window is obtained, specifically: Step 6-6, continuous sliding window processing: slide the sliding window to the next pixel in sequence to perform clutter suppression processing, repeat steps 6-2 to 6-6 until the (n+1) / 2th radial data is processed, and output the clutter suppression intensity value Vf of the central pixel after all sliding window processing. Step 7, Cache Update: Delete the first radial data that was cached in Step 5, and update the second, ..., n-1 and nth radial data that were cached in Step 5 to the first, second, ... and n-1th radial data that were cached in the new cache, respectively; at the same time, read a new radial data as the nth radial data that was cached in the new cache; repeat Steps 6 to 7 to complete the sea clutter suppression processing of the radar real-time image.

2. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 1, characterized in that: In step 1, the target sample index includes the ship length L, which includes at least the following five cases: (1) L≤50m; (2) 50m<L≤100m; (3) 100m<L≤150m; (4) 150m<L≤200m; (5) L>200m.

3. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 1, characterized in that: In step 1, the sea clutter sample indicators include wave height H, which includes at least the following four cases: (1) H≤1m; (2) 1m<H≤1.5m; (3) 1.5m<H≤2m; (4) H≥2m.

4. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 1, characterized in that: In step 2, Tsize > T0, where T0 is the total number of pixels occupied by the largest target in the captured target area.

5. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 4, characterized in that: In step 2, Tsize = 1.5T0.

6. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 1, characterized in that: In step 6-4, Thr C The specific calculation method is as follows: (1) When S n <S, at this time the dispersion of the data within the sliding window is small, indicating that sea clutter data is predominant; a) When V C ≥V mean At this time, the probability that the central pixel is the target data is high; If V C <TV, then Thr C =V mean ; If V C ≥TV, then Thr C =V max ; In the formula, V mean and V max These are the average and maximum intensity values ​​of all pixels within the current sliding window, excluding the central pixel; TV is the average intensity value of the m target regions captured in step 2. b) When V C <V mean At this time, the central pixel is likely to be sea clutter data; If V C <SV, then Thr C =V max ; If V C If ≥SV, then Thr C =V mean ; In the formula, SV is the average intensity value of the m sea clutter regions extracted in step 2; (2) When S n ≥S, at this point the dispersion of the data within the sliding window is relatively large, indicating that the target data is predominant; a) When V C ≥V mean At this time, the probability that the central pixel is the target data is high; If V C <V max And V C If ≥SV, then Thr C =V mean ; If V C <V max And V C <SV, then Thr C =V max ; If V C ≥V max And V C ≥TV, then Thr C =V min ; If V C ≥V max And V C <TV, then Thr C =V mean ; In the formula, V min It is the minimum intensity of all pixels within the current sliding window, excluding the central pixel. b) When V C <V mean At this time, the central pixel is likely to be sea clutter data; If V C <V min Then Thr C =V max ; If V C ≥V min And V C If ≥SV, then Thr C =V mean ; If V C ≥V min And V C <SV, then Thr C =V max .

7. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 6, characterized in that: In step 6-4, V mean and S n The calculation formulas are as follows: In the formula, V1, V2, ..., V n*n-1 These are the intensity values ​​of the n*n-1 pixels within the current sliding window, excluding the central pixel.

8. The adaptive statistical sea clutter suppression method based on X-band marine radar according to claim 1, characterized in that: In step 2, m = 5.

Images