Mean CFAR (Constant False Alarm Rate) detection method applying prior information
By combining Bayesian estimation and distribution information, the mean-class CFAR detection method is optimized, which solves the problems of false alarm and insufficient sea spike suppression ability of traditional methods under harsh sea conditions, and improves detection performance and small-objective detection accuracy.
Patent Information
- Application Number
- CN202411276534.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-12
AI Technical Summary
The traditional mean-class CFAR detection method generates false alarms under the influence of sea clutter above the third-level sea conditions, and lacks the ability to suppress sea spikes, resulting in a decrease in detection performance and insufficient utilization of historical frame information to improve the separability of the inspection statistics. .
Combining the distribution information of Bayesian estimation and test statistics, through the fusion of historical frame test statistics and current frame information, new test statistics are extracted using the continuously corrected prior distribution, and the Monte Carlo method and false alarm rate are used to calculate Get the threshold value and optimize the detection process through power transformation and maximum likelihood estimation.
The performance of the detector and the sea spike suppression ability are improved, the separability and accuracy of small-object detection are enhanced, and the performance of the detector is significantly improved, especially in harsh sea conditions.
Smart Images

Figure CN119044895A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a mean-type CFAR detection method using prior information, belonging to the field of radar target detection. Background Art
[0002] Detection of small targets at sea is one of the main tasks of sea detection radar. In sea conditions above level 3, the non-uniform and non-stationary characteristics of sea clutter can easily affect the test statistics of traditional mean-type CFAR, thus causing a large number of false alarms. When sea spikes appear, this type of test statistic has no ability to suppress them, which greatly affects the performance of the detector.
[0003] In traditional detection methods, the test statistic is often obtained from the echo of the current period. However, the echo information of the current period is relatively limited. Although the test statistic can contain more echo information by increasing the accumulation time, the detection efficiency of the radar will also be affected. In order to improve the separability of the test statistic, it is necessary to make full use of the historical echo information. It can be predicted that within a few seconds, there is a certain connection between the test statistics of the historical period and the current period. The current public literature has used this connection to provide prior information for the test statistic of the current period, thereby supporting the test statistic of the current period and making the test statistic more separable.
[0004] In the current processing method based on prior distribution information, the usual practice is to use the historical frame detection statistics to estimate the probability density function of the test statistic within a certain period of time, and select the test statistic with the largest probability density from the probability densities of multiple test statistics in the current frame as the test statistic of the current frame. Although the above method applies the prior distribution, it is not combined with the Bayesian estimation that is closely related to the prior distribution. As an existing parameter estimation method, Bayesian estimation can integrate historical frame information and current frame information to make a more reasonable estimate of the distribution parameters. When the Bayesian estimation method is not used and prior information is used to estimate the current frame features, the insufficient application of prior information will cause the feature estimate to be unable to efficiently obtain the prior information of the historical frame, which in turn leads to insufficient separability of the feature estimate.
[0005] In view of this feature, combined with the mean-like CFAR detection statistic, a target detection method with stronger separability and sea-spike suppression ability can be obtained, thereby improving the detection performance for small targets. Summary of the invention
[0006] To address the above needs, the present invention combines Bayesian estimation with the distribution information of the test statistic, fuses the information contained in the historical frame test statistic with the current frame, and uses the continuously revised prior distribution to extract new test statistics and form a new detection method.
[0007] The method for detecting mean-valued CFAR using prior information of the present invention is characterized in that it comprises the following steps: Step 1: Collect the clutter echo time series of multiple clutter units within a period of time, and extract the mean-class CFAR test statistic as training information; Step 2: Perform a power transformation on the test statistic in the training information to Gaussianize it and obtain the optimal power; Step 3: In the training phase, the clutter echo time series of multiple clutter units in the training information are subjected to test statistic extraction and power transformation, and then steps a to c are executed to obtain the test statistic output by the clutter unit, and the threshold value is obtained by using the Monte Carlo method and the false alarm rate; Step a: Take the time series of some clutter units, divide the sequence into multiple sequence segments of length N, count the mean of the sequence segments, and then use maximum likelihood to estimate the mean and variance of the mean sequence as the mean and variance of the prior distribution to complete the setting of prior information; Step b: Delimiting the unit to be detected and the reference unit by using the time series of another part of the clutter units; Step c: for the time series of the reference unit and the unit to be detected, iterate using the iterative perception algorithm of the Bayesian estimator; Step 4: In the detection phase, the echo time series of the unit to be detected and the reference unit are extracted by test statistics and power transformation, and then step c in step 3 is executed to obtain a new test statistic, which is compared with the threshold to obtain a decision result; Step 5: Use measured data to verify the proposed method.
[0008] Preferably, the specific steps of step 1 are as follows: The test statistic selection strategy of the SO-CFAR method is to first perform square-law detection on the IQ echo signal: (1); In the formula, represents the radar echo signal after square law detection, represents the real part of the radar IQ echo signal, Represents the imaginary part of the radar IQ echo signal; Secondly, traverse in the fast time dimension, select the unit to be detected and the corresponding reference unit, and calculate the test statistic of the unit to be detected according to the following formula: (2); In the formula represents the test statistic time series of the reference unit, It represents the echo time series of the unit to be detected after square law detection. , Represents the echo time series of the reference units on both sides of the unit to be detected after square-law detection.
[0009] Preferably, the specific steps of step 2 are as follows: Prioritize Gaussianization of the data to ensure that subsequent iterations of the prior distribution can be carried out stably. The power transformation can be performed according to the following formula: (3); In the formula, Refers to the test statistic sequence Through the power The time series after power transformation.
[0010] The method of seeking the optimal power uses maximum likelihood estimation and chi-square test. When traversing to find the optimal power, first use maximum likelihood estimation to estimate the mean and variance of the test statistic under the current power. Secondly, use the power-transformed data and the estimated mean and variance to perform a chi-square normality test, and output the power that can make the test statistic closest to the Gaussian distribution as the optimal power.
[0011] Preferably, the specific steps of step 3 are as follows: The test statistic time series of the unit to be tested after power transformation is recorded as , the test statistic time series of the reference unit is denoted by : Step a: Take the time series of some clutter units , the total number of sequences is K, the sequence is divided into K / N sequence segments of length N, the mean of the sequence segments is counted, and K / N mean sequences are obtained. The mean and variance of these K / N mean sequences are estimated by maximum likelihood as the mean of the prior distribution and variance , complete the setting of prior information; Step b: Delimiting the unit to be detected and the reference unit by using the time series of another part of the clutter units; Step c: Time series for reference cells and the unit to be tested , use the iterative perception algorithm of the Bayesian estimator to iterate and obtain the test statistic. The specific steps are as follows Ⅰ~Ⅵ: Ⅰ: Slide the window of the unit's time series, and the sliding window length is N , the sliding window step is step=1, and the time series length of each unit is M , the mean of the test statistic within the current time window and variance Make unbiased estimates; (4); Ⅱ: Secondly, using the mean and variance obtained by unbiased estimation, combined with the prior distribution, we can get the Bayesian estimate of the mean of the test statistic in the current window: , which is the mean of the posterior probability density function. Similarly, the variance of the posterior probability density function can also be calculated. : (5); In the formula, , refers to the mean and variance of the prior distribution, refers to the unbiased estimate of the variance of the current frame, Refers to the unbiased estimate of the mean of the current frame.
[0012] III: Based on the Bayesian estimate obtained by the current sliding window, the mean and variance of the Bayesian estimate are sequentially estimated, and the mean of the Bayesian estimate is updated and variance : (6); In the formula, k refers to the number of iterations, Refers to the Bayesian estimate obtained for the current frame.
[0013] IV: Based on the Bayesian estimate of the current sliding window, the mean and variance of the Bayesian estimate obtained by sequential estimation, and the variance of the current frame, the mean of the prior distribution and variance Make corrections, the correction method is as follows: (7); In the formula, Refers to the current frame ( k frame, and also the k Iterations) is an unbiased estimate of the sample variance.
[0014] V: If the current unit is the unit to be detected, execute this step, otherwise go to step VI; Collect the Bayesian estimates of multiple reference units at the same sliding window moment and calculate the mean of these Bayesian estimates The probability density under the posterior density function of the current unit , as the new test statistic output; when the iteration is stable, the Bayesian estimate is approximately regarded as being sampled from the prior distribution of the corresponding unit; since the reference unit is mostly a clutter unit, when the unit to be detected is also a clutter unit, the Bayesian estimates of the two can be almost regarded as being sampled from the same distribution, so the Bayesian estimate of the reference unit has a higher probability density under the probability density function of the unit to be detected; and when the unit to be detected is the target unit, there is a difference in the prior distribution of the two, resulting in the Bayesian estimate of the reference unit having a lower probability density under the probability density function of the unit to be detected. This phenomenon makes the new test statistic extracted by this method have a higher degree of discrimination than the original test statistic: (8); In the formula, , are the mean and variance of the prior distribution VI: Return to step I until the window reaches the end of the sequence; Step d: Sort the obtained test statistics from small to large and select The test statistic is used as the decision threshold , step 3 ends.
[0015] Preferably, the specific steps of step 4 are: In the detection phase, the unit to be detected is used and reference unit , The time echo sequence of is extracted according to formula (2) to obtain the test statistic of the unit to be tested. and the test statistic for the reference unit According to formula (10), the power transformation is performed to obtain the test statistic of the unit to be tested after the power transformation. and the test statistic after power transformation of the unit to be tested After step c, the probability density function of the unit to be detected can be obtained (See equation (8)) and the mean of the Bayesian estimates of the reference unit , and calculate Output as a new test statistic. With threshold For comparison, if This means that the unit to be detected is the target unit, otherwise it is clutter.
[0016] Preferably, the specific steps of step 5 are: to verify the effectiveness, use HH polarization and VV polarization radar staring data for verification, the target to be detected in the radar staring data is a channel buoy, select the unit where the channel buoy is located as the unit to be detected, and the distance unit around the unit where the channel buoy is located is used as the reference unit, and calculate the test statistic of the unit to be detected by formula (2): and the test statistic for the reference unit , refer to step 4 to get the new test statistic , and with the threshold For comparison, if This means that the unit to be detected is the target unit, otherwise it is clutter; the validity is verified based on the judgment result.
[0017] The beneficial effects of the present invention are as follows: First, the historical frame test statistics are fully utilized, which not only completes the correction of the prior distribution, but also completes the fusion of the historical frame test statistics and the current frame test statistics; Second, based on the prior distribution containing historical frame information, a more discriminative test statistic is extracted, which improves the performance of the detector; Third, after the power transformation, the mean-like CFAR statistics that were originally sensitive to sea spikes became more clustered, which had a certain inhibitory effect on sea spikes.
[0018] After verification by HH and VV measured data, the small target detection capability of the detector can be very effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 : Changes in the distribution histogram of the test statistic before and after the power transformation; Figure 2 : Prior probability density curves of the unit to be detected and the reference unit and the probability density diagram of the Bayesian estimate; Figure 3 : Binarization comparison diagram of the detection results of the present invention and the existing method in HH polarization and five-level sea state data sets; Figure 4 :Comparison of the detection results of the present invention and the original mean-class CFAR method on 9 data sets. Among them, sequence numbers 1 to 2 are HH polarization level 4 sea state data, sequence number 3 is HH polarization level 5 sea state data, and sequence numbers 4 to 9 are VV polarization level 5 sea state data. DETAILED DESCRIPTION
[0020] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and technical solutions.
[0021] Step 1: The test statistic selection strategy of the SO-CFAR method is to first perform square-law detection on the IQ echo signal: (1); In the formula, represents the radar echo signal after square law detection, represents the real part of the radar IQ echo signal, Represents the imaginary part of the radar IQ echo signal.
[0022] Secondly, traverse in the fast time dimension, select the unit to be detected and the corresponding reference unit, and calculate the test statistic of the unit to be detected according to the following formula: (2); In the formula represents the test statistic time series of the reference unit, It represents the echo time series of the unit to be detected after square law detection. , Represents the echo time series of the reference units on both sides of the unit to be detected after square-law detection.
[0023] Step 2: Prioritize Gaussianization of the data to ensure that subsequent iterations of the prior distribution can be carried out stably. The power transformation can be performed according to the following formula: (3); In the formula, Refers to the test statistic sequence Through the power The time series after power transformation.
[0024] The method of finding the best power is to use maximum likelihood estimation and chi-square test. When searching for the best power, first use maximum likelihood estimation to estimate the mean and variance of the test statistic under the current power. Secondly, use the power-transformed data and the estimated mean and variance to perform a chi-square normality test, and output the power that makes the test statistic closest to the Gaussian distribution as the best power. The distribution histogram of the test statistic without power transformation is as follows: Figure 1 As shown in (a), the distribution histogram of the test statistic after power transformation is as follows Figure 1 As shown in (b).
[0025] Step 3: Record the test statistic time series of the unit to be tested after power transformation as , the test statistic time series of the reference unit is denoted by : Step a: Take the time series of some clutter units , the total number of sequences is K, the sequence is divided into K / N sequence segments of length N, the mean of the sequence segments is counted, and K / N mean sequences are obtained. The mean and variance of these K / N mean sequences are estimated by maximum likelihood as the mean of the prior distribution and variance , complete the setting of prior information; Step b: Delimiting the unit to be detected and the reference unit by using the time series of another part of the clutter units; Step c: Time series for reference cells and the unit to be tested , use the iterative perception algorithm of the Bayesian estimator to iterate and obtain the test statistic. The specific steps are as follows Ⅰ~Ⅵ: Ⅰ: Slide the window of the unit's time series, and the sliding window length is N , the sliding window step is step=1, and the time series length of each unit is M , the mean of the test statistic within the current time window and variance Make unbiased estimates; (4); Ⅱ: Secondly, using the mean and variance obtained by unbiased estimation, combined with the prior distribution, we can get the Bayesian estimate of the mean of the test statistic in the current window: , which is the mean of the posterior probability density function. Similarly, the variance of the posterior probability density function can also be calculated. : (5); In the formula, , refers to the mean and variance of the prior distribution, refers to the unbiased estimate of the variance of the current frame, Refers to the unbiased estimate of the mean of the current frame.
[0026] III: Based on the Bayesian estimate obtained by the current sliding window, the mean and variance of the Bayesian estimate are sequentially estimated, and the mean of the Bayesian estimate is updated and variance : (6); In the formula, k refers to the number of iterations, Refers to the Bayesian estimate obtained for the current frame.
[0027] IV: Based on the Bayesian estimate of the current sliding window, the mean and variance of the Bayesian estimate obtained by sequential estimation, and the variance of the current frame, the mean of the prior distribution and variance Make corrections, the correction method is as follows: (7); In the formula, Refers to the current frame ( k frame, and also the k Iterations) is an unbiased estimate of the sample variance.
[0028] V: If the current unit is the unit to be detected, execute this step, otherwise go to step VI; Collect the Bayesian estimates of multiple reference units at the same sliding window moment and calculate the mean of these Bayesian estimates The probability density under the posterior density function of the current unit , as the new test statistic output; when the iteration is stable, the Bayesian estimate is approximately regarded as being sampled from the prior distribution of the corresponding unit; since the reference unit is mostly a clutter unit, when the unit to be detected is also a clutter unit, the Bayesian estimates of the two can be almost regarded as being sampled from the same distribution, so the Bayesian estimate of the reference unit has a higher probability density under the probability density function of the unit to be detected; and when the unit to be detected is the target unit, there is a difference in the prior distribution of the two, resulting in the Bayesian estimate of the reference unit having a lower probability density under the probability density function of the unit to be detected, such as Figure 2 As shown in Figure 2, this phenomenon makes the new test statistic extracted by this method have a higher degree of discrimination compared with the original test statistic: (8); In the formula, , are the mean and variance of the prior distribution VI: Return to step I until the window reaches the end of the sequence; Step d: Sort the obtained test statistics from small to large and select The test statistic is used as the decision threshold , step 3 ends.
[0029] Step 4: In the detection phase, use the unit to be detected and reference unit , The time echo sequence of is extracted according to formula (2) to obtain the test statistic of the unit to be tested. and the test statistic for the reference unit According to formula (10), the power transformation is performed to obtain the test statistic of the unit to be tested after the power transformation. and the test statistic after power transformation of the unit to be tested After step c, the probability density function of the unit to be detected can be obtained (See formula (8)) and the mean of the Bayesian estimates of the reference unit , and calculate Output as a new test statistic. With threshold For comparison, if This means that the unit to be detected is the target unit, otherwise it is clutter.
[0030] Step 5: To verify the effectiveness of the present invention, the radar staring data of HH polarization and VV polarization are used to verify the present invention. The target to be detected in the radar staring data is the channel buoy. The unit where the channel buoy is located is selected as the unit to be detected, and the distance unit around the unit where the channel buoy is located is selected as the reference unit. The test statistic of the unit to be detected is calculated by formula (2): and the test statistic for the reference unit . Referring to step 4, we can get the new test statistic , and with the threshold For comparison, if It means that the unit to be detected is the target unit, otherwise it is clutter. According to the judgment result, the effectiveness of the present invention is verified. The binary images of the detection results of the original method and the present invention are as follows Figure 3 As shown, the small target detection probability of the original method is only 11.59%, while the small target detection probability of the present invention is 93.87%, which can clearly prove the performance superiority of the present invention.
[0031] The detection results of the remaining HH polarization and VV polarization radar staring data sets are as follows: Figure 4 As shown, the superiority of the present invention over the traditional method can also be proved.
Claims
1. A mean-based CFAR detection method using prior information, characterized in that The following steps are involved: Step 1: Collect the clutter echo time series of multiple clutter units within a period of time, and extract the mean-class CFAR test statistic as training information; Step 2: Perform a power transformation on the test statistic in the training information to Gaussianize it and obtain the optimal power; Step 3: In the training phase, the clutter echo time series of multiple clutter units in the training information are subjected to test statistic extraction and power transformation, and then steps a to c are executed to obtain the test statistic output by the clutter unit, and the threshold value is obtained by using the Monte Carlo method and the false alarm rate; Step a: Take the time series of some clutter units, divide the sequence into multiple sequence segments of length N, count the mean of the sequence segments, and then use maximum likelihood to estimate the mean and variance of the mean sequence as the mean and variance of the prior distribution to complete the setting of prior information; Step b: Delimiting the unit to be detected and the reference unit by using the time series of another part of the clutter units; Step c: for the time series of the reference unit and the unit to be detected, iterate using the iterative perception algorithm of the Bayesian estimator; Step 4: In the detection phase, the echo time series of the unit to be detected and the reference unit are extracted by test statistics and power transformation, and then step c in step 3 is executed to obtain a new test statistic, which is compared with the threshold to obtain a decision result; Step 5: Use measured data to verify the proposed method.
2. A mean-based CFAR detection method using prior information according to claim 1, characterized in that: The specific steps of step 1 are as follows: The test statistic selection strategy of the SO-CFAR method is to first perform square-law detection on the IQ echo signal: (1); In the formula, represents the radar echo signal after square law detection, represents the real part of the radar IQ echo signal, Represents the imaginary part of the radar IQ echo signal; Secondly, traverse in the fast time dimension, select the unit to be detected and the corresponding reference unit, and calculate the test statistic of the unit to be detected according to the following formula: (2); In the formula represents the test statistic time series of the reference unit, It represents the echo time series of the unit to be detected after square law detection. , Represents the echo time series of the reference units on both sides of the unit to be detected after square-law detection.
3. A mean-based CFAR detection method using prior information according to claim 1, characterized in that: The specific steps of step 2 are as follows: Prioritize Gaussianization of the data to ensure that subsequent iterations of the prior distribution can be carried out stably. The power transformation can be performed according to the following formula: (3); In the formula, Refers to the test statistic sequence Through the power The time series after power transformation; The method of seeking the best power is to use the maximum likelihood estimation and chi-square test. When searching for the best power, the maximum likelihood estimation is first used to estimate the mean and variance of the test statistic under the current power. Secondly, the chi-square normality test is performed using the power-transformed data and the estimated mean and variance, and the power that makes the test statistic closest to the Gaussian distribution is output as the optimal power.
4. A mean-based CFAR detection method using prior information according to claim 3, characterized in that: The specific steps of step 3 are as follows: The test statistic time series of the unit to be tested after power transformation is recorded as , the test statistic time series of the reference unit is denoted by : Step a: Take the time series of some clutter units , the total number of sequences is K, the sequence is divided into K / N sequence segments of length N, the mean of the sequence segments is counted, and K / N mean sequences are obtained. The mean and variance of these K / N mean sequences are estimated by maximum likelihood as the mean of the prior distribution and variance , complete the setting of prior information; Step b: Delimiting the unit to be detected and the reference unit by using the time series of another part of the clutter units; Step c: Time series for reference cells and the unit to be tested , using the iterative perceptual algorithm of the Bayesian estimator to iterate and obtain the test statistic, Step d: Sort the obtained test statistics from small to large and select The test statistic is used as the decision threshold , step 3 ends.
5. A mean-based CFAR detection method using prior information according to claim 4, characterized in that: The specific steps of step c are as follows: Ⅰ: Slide the window of the unit's time series, and the sliding window length is N , the sliding window step is step=1, and the time series length of each unit is M , the mean of the test statistic within the current time window and variance Make unbiased estimates; (4); Ⅱ: Secondly, using the mean and variance obtained by unbiased estimation, combined with the prior distribution, we can get the Bayesian estimate of the mean of the test statistic in the current window: , which is the mean of the posterior probability density function. Similarly, the variance of the posterior probability density function can also be calculated. : (5); In the formula, , refers to the mean and variance of the prior distribution, refers to the unbiased estimate of the variance of the current frame, Refers to the unbiased estimate of the mean of the current frame; III: Based on the Bayesian estimate obtained by the current sliding window, the mean and variance of the Bayesian estimate are sequentially estimated, and the mean of the Bayesian estimate is updated and variance : (6); In the formula, k refers to the number of iterations, Refers to the Bayesian estimate obtained for the current frame; IV: Based on the Bayesian estimate of the current sliding window, the mean and variance of the Bayesian estimate obtained by sequential estimation, and the variance of the current frame, the mean of the prior distribution and variance Make corrections, the correction method is as follows: (7); In the formula, Refers to the unbiased estimate of the sample variance of the current frame; V: If the current unit is the unit to be detected, execute this step, otherwise go to step VI; Collect the Bayesian estimates of multiple reference units at the same sliding window moment and calculate the mean of these Bayesian estimates The probability density under the posterior density function of the current unit , as the new test statistic output; when the iteration is stable, the Bayesian estimate is approximately regarded as being sampled from the prior distribution of the corresponding unit; since the reference unit is mostly a clutter unit, when the unit to be detected is also a clutter unit, the Bayesian estimates of the two can be almost regarded as being sampled from the same distribution, so the Bayesian estimate of the reference unit has a higher probability density under the probability density function of the unit to be detected; and when the unit to be detected is the target unit, there is a difference in the prior distribution of the two, resulting in the Bayesian estimate of the reference unit having a lower probability density under the probability density function of the unit to be detected. This method uses the difference between the clutter unit and the unit to be detected to extract a new test statistic, as shown in formula (8), which has a higher discrimination than the original test statistic: (8); In the formula, , are the mean and variance of the prior distribution; VI: Return to step I until the window slides to the end of the sequence.
6. A mean-based CFAR detection method using prior information according to claim 4, characterized in that: The specific steps of step 4 are as follows: In the detection phase, the unit to be detected is used and reference unit , The time echo sequence of is extracted according to formula (2) to obtain the test statistic of the unit to be tested. and the test statistic for the reference unit , perform power transformation according to formula (3) to obtain the test statistic after power transformation of the unit to be tested and the test statistic after power transformation of the unit to be tested After step c, the probability density function of the unit to be detected is obtained , see formula (8) and the mean of the Bayesian estimates of the reference unit , and calculate As a new test statistic output, by With threshold For comparison, if This means that the unit to be detected is the target unit, otherwise it is clutter.
7. A mean-based CFAR detection method using prior information according to claim 3, characterized in that: The specific steps of step 5 are as follows: To verify the effectiveness, HH polarization and VV polarization radar staring data are used for verification. The target to be detected in the radar staring data is the channel buoy. The unit where the channel buoy is located is selected as the unit to be detected, and the distance unit around the unit where the channel buoy is located is selected as the reference unit. The test statistic of the unit to be detected is calculated by formula (2): and the test statistic for the reference unit , refer to step 4 to get the new test statistic , and with the threshold For comparison, if This means that the unit to be detected is the target unit, otherwise it is clutter; the validity is verified based on the judgment result.
Citation Information
Patent Citations
Radar range profile statistics and recognition method based on FA model in strong noise background
CN101598784A
Automated method for selecting training areas of sea clutter and detecting ship targets in polarimetric synthetic aperture radar imagery
WO2016097890A1
Cited By
Sea surface moving target detection method applying prior information
CN119247316A
A method for detecting moving targets on the sea surface using prior information
CN119247316B
Parameterized constant false alarm processing method suitable for any PolSAR detector
CN119511287A
