An adaptive target detection method in a non-uniform strong sea clutter scene

Abstract

An adaptive target detection method in a non-uniform strong sea clutter scene, a certain size of a protection area and a reference area are set around a to-be-detected scattering point on a range-Doppler image, the clutter area in the reference area is sliced out by using the area mean value of Doppler information, and the CACFAR threshold is calculated based on the clutter area, abnormal cells are filtered out, and clearer target echoes are obtained.

Description

Technical Field

[0001] This invention relates to the field of microwave radar target detection technology, and in particular to a target detection method suitable for non-uniform strong sea clutter scenarios. Background Technology

[0002] When high-resolution radar detects targets against a non-uniform, strong sea clutter background, the equivalent backscattering coefficient increases, and most of the sea clutter energy is projected onto a few range cells, resulting in uneven energy distribution and the appearance of clutter "abnormal cells" with suddenly increased power. This leads to a complex and variable background environment for the detector reference cell. If a traditional cell averaging (CA CFAR) detector is used, the false alarm rate and mis-alarm rate increase, resulting in reduced radar detection performance. Target detection against a non-uniform, strong sea clutter background is one of the hot and challenging research topics in the field of signal detection. Overcoming the influence of strong sea clutter or suppressing it, and removing the influence of strong sea clutter "abnormal cells" with suddenly increased energy, are the main approaches to solving this problem. Summary of the Invention

[0003] This disclosure provides a method for filtering sea clutter by setting an adaptive reference cell threshold, which improves the accuracy of target detection in complex backgrounds with non-uniform strong sea clutter.

[0004] For environments with uniform noise and clutter, the in-phase I-channel and quadrature Q-channel echo signals received from the matched filter receiver serve as the input signals for the entire algorithm. These signals then pass through a square-law detector and are sent to the first-level threshold detection module and the CFAR module, outputting the target detection result. (See attached image) Figure 1 As shown. The main function of the first-level threshold detection module is to set a fixed threshold and remove uniform noise from the preprocessed data. The main function of the CFAR module is to set an adaptive threshold through a sliding window, filter out "abnormal cells", and output the target detection results.

[0005] Against a uniform sea clutter background, target echoes can be considered relatively clear, and target detection can be performed using the classic and reliable two-dimensional CACFAR. However, for echoes against a non-uniform, strong sea clutter background, the edge detection of the target points obtained by using two-dimensional CACFAR is inaccurate (an extension compared to the first case). This disclosure mainly addresses the problem of how to use a CFAR module to filter out "abnormal cells" and obtain relatively clear target echoes against a non-uniform, strong sea clutter background.

[0006] The adaptive target detection method for non-uniform strong sea clutter scenarios disclosed herein comprises the following steps:

[0007] S101: Select the scattering point to be detected on the range-Doppler image, set a protective area of ​​a certain size around the scattering point to be detected, and divide the surrounding area into several reference areas, including the main and side lobe areas;

[0008] S102: Select the reference regions S1 and S2 with the largest and second largest mean values, respectively;

[0009] S103: Slice out the clutter region in the reference regions S1 and S2 to obtain new reference regions N1 and N2;

[0010] S104: Calculate the mean values ​​of signals N1 and N2, select the region with the larger mean value as the final reference region LastRefArea, and calculate the CA CFAR threshold of the scattering region to be detected according to the following formula:

[0011]

[0012] in: It is the signal mean within the reference region LastRefArea, and α is the threshold amplitude factor, obtained by the following formula:

[0013]

[0014] Where: N is the number of reference units, P fa This refers to the false alarm rate.

[0015] Furthermore, in step S103, the method for slicing out the clutter region in the reference region includes: selecting a peak point in the reference region and defining the area near the peak point as the clutter region, i.e., the new reference region.

[0016] Furthermore, the method for obtaining the new reference region specifically includes:

[0017] For the initially defined reference region, the Doppler mean is calculated to obtain the Doppler index midposD of the peak value;

[0018] On a single distance channel of the distance dimension, calculate the mean of the Doppler dimensions (midposD-2: midposD+2) to obtain the index of the maximum value, which is the distance dimension index of the peak value, midposR.

[0019] Centered on (midposD, midposR), a region with Doppler dimensions ±Dref / 3 and distance dimensions ±3*interval is defined as the new reference region, where Dref is the Doppler length of the initially defined reference region and interval is the interval between each distance channel.

[0020] Furthermore, in step S103, if the reference region S1 or S2 is a sidelobe region, then no slicing is required, and the reference region is the new reference region.

[0021] Furthermore, the Doppler dimension span of the protected area or reference area is 10-20 Doppler units, the distance dimension span is the target size * 1.2, and the span of the sidelobe region is 1-3 Doppler units or 1-3 distance units.

[0022] Furthermore, a sliding window is used to perform pixel-by-pixel detection on scattering points that have exceeded the first threshold.

[0023] Compared with the prior art, the beneficial effects of this disclosure are: (1) It makes full use of Doppler information and sets an adaptive reference cell threshold to filter out sea clutter, which can effectively "filter out" anomalous cells; (2) It improves the accuracy of target detection under complex backgrounds of non-uniform strong sea clutter; (3) The calculation method is simple and highly practical. Attached Figure Description

[0024] The above and other objects, features and advantages of this disclosure will become more apparent from the more detailed description of exemplary embodiments of this disclosure taken in conjunction with the accompanying drawings, in which the same reference numerals generally represent the same components.

[0025] Figure 1 This demonstrates the general workflow of target detection;

[0026] Figure 2 This is a flowchart of the CFAR module processing in this publication;

[0027] Figure 3 This is a schematic diagram of distance-Doppler two-dimensional CA-CFAR detection;

[0028] Figure 4 This is a diagram showing the results of traditional CA-CFAR testing.

[0029] Figure 5 A schematic diagram illustrating the application of the sliced ​​CFAR reference region method of this disclosure;

[0030] Figure 6 This is a diagram showing the detection results of CA-CFAR on the slice. Detailed Implementation

[0031] Preferred embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While preferred embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0032] This disclosure primarily addresses the problem of how to use a CFAR module to filter out "abnormal cells" and obtain relatively clear target echoes against a background of non-uniform strong sea clutter.

[0033] According to the exemplary CFAR module processing flowchart of this disclosure, as follows: Figure 2 As shown, a sliding window is used to detect scattering points that have passed the first-level threshold pixel by pixel. The most critical step is how to slice N1 and N2 from S1 and S2.

[0034] The exemplary steps and related methods are detailed below:

[0035] (1) Select the scattering point to be detected on the range-Doppler image, and design a protection area and a reference area of ​​a certain size, as shown in the attached figure. Figure 3 As shown, the dots represent the scattering points of the target to be detected, the central rectangular area represents the target protection area, Rpro is the length of the range protection unit, Dpro is the length of the Doppler protection unit, Rref is the length of the range reference unit, and Dref is the length of the Doppler reference unit. The surrounding rectangular area is divided into 10 reference areas, of which reference areas 7, 8, 9, and 10 are sidelobe areas.

[0036] Preferably, the Doppler dimension span of the selected protected area or reference area is 10-20 Doppler units, the range dimension span is the target size * 1.2, and the span of the sidelobe region is 1-3 Doppler units or 1-3 range units.

[0037] (2) Select the regions S1 and S2 with the largest and second largest mean values ​​from the 10 regions.

[0038] (3) Slice out the clutter region in the reference regions S1 and S2 to obtain new reference regions N1 and N2 (if S1 and S2 are sidelobe regions, then there is no need to slice out the clutter region, that is, S1 and S2 are N1 and N2). Calculate the mean values ​​of the signals N1 and N2, select the region with the larger mean value as the final reference region LastRefArea, and calculate the CA CFAR threshold for that scattering point. The CA CFAR threshold is calculated by the following formula:

[0039]

[0040] in: It is the signal mean within the reference region LastRefArea, and α is the threshold amplitude factor, obtained by the following formula:

[0041]

[0042] Where: N is the number of reference units, P fa This refers to the false alarm rate.

[0043] The specific method for slicing the clutter region N1 within the S1 reference region is as follows: Select the peak point of the S1 region and delineate the region near the peak point as N1. In the specific operation:

[0044] 1. Obtain the Doppler index midposD of the peak value by averaging the Doppler values ​​(row by row);

[0045] 2. On a single distance channel divided into 10 equal parts (interval value), calculate the mean of the Doppler dimensions (midposD-2:midposD+2) (column-wise) to obtain the index of the maximum value, which is the distance dimension index midposR of the peak value;

[0046] With (midposD, midposR) as the center point, a region with Doppler dimensions ±Dref / 3 and distance dimensions ±3*interval is defined as the new reference region N1.

[0047] Application Examples

[0048] In a radar test, the scenario was a non-uniform, strong sea clutter situation. The result of the radar echo signal exceeding the first threshold was as follows: Figure 4 As shown in (a), when using traditional two-dimensional CFAR, the effect is as follows: Figure 4 As shown in (b), it can be seen that many sea clutter "abnormal cells" were not filtered out.

[0049] When using the slice CFAR detection proposed in this disclosure, a sliding window is used to perform pixel-by-pixel detection for scattering points that have passed the first-level threshold. The specific operation is as follows:

[0050] 1. Select the scattering point to be detected on the range-Doppler image, and design a protection area and a reference area of ​​a certain size, such as... Figure 2 As shown, the dots represent the scattering points of the target to be detected, the central rectangular area represents the target protection area, and the surrounding rectangular areas are divided into 10 reference areas.

[0051] 2. Select the regions with the largest and second largest mean values ​​from the 10 regions, namely S1 and S2. In this example, these are reference region 1 and reference region 6, respectively.

[0052] 3. Slice out the clutter region in the S1 and S2 reference areas. The slicing result is as follows: Figure 5 As shown in N1 and N2, the new reference regions N1 and N2 obtained from the two slices are selected, and the one with the larger mean is used as the final reference region LastRefArea to calculate the CA CFAR threshold of the detection points.

[0053] The effects of using this disclosure are as follows: Figure 6 As shown, this disclosure can effectively "filter out" abnormal units.

[0054] The above technical solutions are merely exemplary embodiments of the present invention. For those skilled in the art, based on the application methods and principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the methods described in the specific embodiments of the present invention. Therefore, the methods described above are merely preferred and not restrictive.

Claims

1. An adaptive target detection method for non-uniform strong sea clutter scenarios, comprising the following steps: S101: Select the scattering point to be detected on the range-Doppler image, set a protective area of ​​a certain size around the scattering point to be detected, and divide the surrounding area into several reference areas, including the main and side lobe areas; S102: Select the reference regions S1 and S2 with the largest and second largest mean values; S103: Slice out the clutter region in the reference regions S1 and S2 to obtain new reference regions N1 and N2; S104: Calculate the mean values ​​of signals N1 and N2, select the region with the larger mean value as the final reference region LastRefArea, and calculate the CA CFAR threshold of the scattering point to be detected according to the following formula: (1) in: It is the signal mean within the reference area LastRefArea. The threshold amplitude factor is obtained from the following formula: (2) in: For the reference unit number, False alarm rate; In step S103, the method for slicing the clutter region in the reference region includes: selecting a peak point in the reference region and defining the area near the peak point as the clutter region, i.e., the new reference region; specifically including: For the initially defined reference region, the Doppler mean is calculated to obtain the Doppler index midposD of the peak value; On a single distance channel of the distance dimension, calculate the mean of the Doppler dimensions (midposD-2: midposD+2) to obtain the index of the maximum value, which is the distance dimension index of the peak value, midposR. Centered on (midposD, midposR), a region with Doppler dimensions ±Dref / 3 and distance dimensions ±3*interval is defined as the new reference region, where Dref is the Doppler length of the initially defined reference region and interval is the interval between each distance channel.

2. The detection method according to claim 1, characterized in that, In step S103, if the reference region S1 or S2 is a sidelobe region, then no slicing is required, and the reference region is the new reference region.

3. The detection method according to claim 1 or 2, characterized in that, The Doppler dimension span of the protected area or reference area is 10-20 Doppler units, the range dimension span is the target size * 1.2, and the span of the sidelobe region is 1-3 Doppler units or 1-3 range units.

4. The detection method according to claim 1, characterized in that, For scattering points that exceed the first threshold, a sliding window is used for pixel-by-pixel detection.

Images