CN116106835A

CN116106835A - Self-adaptive target detection method in heterogeneous strong sea clutter scene

Info

Publication number: CN116106835A
Application number: CN202211140604.7A
Authority: CN
Inventors: 张洪纲; 王丹; 张哲�; 苏亚哲; 周敬亮
Original assignee: Bit Raco Electronic Information Technology Co ltd
Current assignee: Bit Raco Electronic Information Technology Co ltd
Priority date: 2022-09-20
Filing date: 2022-09-20
Publication date: 2023-05-12

Abstract

A self-adaptive target detection method under a non-uniform strong sea clutter scene is characterized in that a protection area and a reference area with a certain size are arranged around a scattering point to be detected on a distance-Doppler image, the clutter area in the reference area is sliced by utilizing the area mean value of Doppler information, a CACFAR threshold is calculated based on the clutter area, and abnormal units are filtered to obtain a clearer target echo.

Description

Self-adaptive target detection method in heterogeneous strong sea clutter scene

Technical Field

The invention relates to the technical field of microwave radar target detection, in particular to a target detection method suitable for a heterogeneous strong sea clutter situation.

Background

When the high-resolution radar detects under the heterogeneous strong sea clutter background, the equivalent backscattering coefficient is increased, most of sea clutter energy is projected to a few distance units, so that energy distribution is uneven, clutter 'abnormal units' with suddenly increased power appear, and the background environment where the detector reference units are located is complex and changeable. If a conventional unit average constant false alarm rate (CA CFAR) detector is used for detection, the false alarm rate and false alarm rate increase, and the radar detection performance decreases. The problem of target detection under the heterogeneous strong sea clutter background is one of hot spot difficult problems studied in the current signal detection field, and the main way for solving the problem is to overcome the influence of the strong sea clutter or inhibit the strong sea clutter and remove the influence of a strong sea clutter abnormal unit with suddenly increased energy.

Disclosure of Invention

The method for filtering the sea clutter by setting the self-adaptive reference unit threshold improves the accuracy of target detection under the heterogeneous strong sea clutter complex background.

For the uniform noise and clutter adding environment, echo in-phase I-path signals and quadrature Q-path signals received from a matched filter receiver are used as input signals of the whole algorithm, and then the input signals are sent to a first-level threshold detection module and a CFAR module through a square law detector, and target detection results are output. As shown in fig. 1. The primary threshold detection module is used for setting a fixed threshold and removing uniform noise of the preprocessed data. The CFAR module has the main function of setting a self-adaptive threshold through a sliding window, filtering abnormal units and outputting a target detection result.

Under the uniform sea clutter background, the target echo can be considered to be clearer, and the classical reliable two-dimensional CA CFAR is used for target detection; however, for the echo under the heterogeneous strong sea clutter background, the detection of the trace point edge of the target obtained by the detection by using the two-dimensional CA CFAR is still inaccurate (compared with the first case, the expansion exists). The method mainly solves the problems of filtering abnormal units and obtaining a clearer target echo by using the CFAR module under the heterogeneous strong sea clutter background.

The self-adaptive target detection method under the heterogeneous strong sea clutter scene provided by the disclosure comprises the following steps:

s101: selecting a scattering point to be detected on a distance-Doppler image, setting a protection area with a certain size on the periphery of the scattering point to be detected, and dividing the peripheral area into a plurality of reference areas, including a main sidelobe area;

s102: selecting reference areas S1 and S2 with the largest average value and the largest average value;

s103: slicing clutter areas in the S1 and S2 reference areas to obtain new reference areas N1 and N2;

s104: solving the mean value of N1 and N2 signals, selecting a region with a larger mean value as a final reference region LastRefArea, and calculating a CA CFAR threshold of the scattering region to be detected according to the following formula:

wherein:

is the signal mean value in the reference area LastRefArea, and alpha is a threshold amplitude factor, and is obtained by the following formula: />

Wherein: n is the reference unit number, P _fa Is the false alarm rate.

Further, in the step S103, the method for slicing out the clutter region in the reference region includes: and selecting peak points in the reference area, and defining an area near the peak points as a clutter area, namely a new reference area.

Further, the method for acquiring the new reference area specifically includes:

for the initially set reference area, calculating the average value according to the Doppler dimension to obtain the Doppler dimension index midcosD of the peak value;

on a single distance channel of the distance dimension, calculating the average value of Doppler dimensions (midassosd-2: midassosd+2), and obtaining the index of the maximum value, namely the distance dimension index midassosr of the peak value;

and taking (midcosD, midcosR) as a central point, and defining a region with Doppler dimension +/-Dref/3 and distance dimension +/-3 x interval as the new reference region, wherein Dref is the Doppler dimension length of the initially set reference region, and interval is the interval of each distance channel.

Further, in the step S103, if the reference area S1 or S2 is a side lobe area, no slicing is needed, and the reference area is a new reference area.

Further, the Doppler dimension span of the protection area or the reference area is 10-20 Doppler units, the range dimension span is 1.2 of the target size, and the span of the sidelobe area is 1-3 Doppler units or 1-3 range units.

Further, a sliding window is adopted for detecting the scattering points passing through the first-level threshold pixel by pixel.

Compared with the prior art, the beneficial effects of the present disclosure are: (1) Doppler information is fully utilized, and a self-adaptive reference unit threshold is set to filter sea clutter, so that abnormal units can be well filtered; (2) The accuracy of target detection under the heterogeneous strong sea clutter complex background is improved; and (3) the calculation method is simple and convenient, and the practicability is high.

Drawings

The foregoing and other objects, features and advantages of the disclosure will be apparent from the following more particular descriptions of exemplary embodiments of the disclosure as illustrated in the accompanying drawings wherein like reference numbers generally represent like parts throughout the exemplary embodiments of the disclosure.

FIG. 1 shows a general flow of target detection;

FIG. 2 is a flow chart of the CFAR module process of the present disclosure;

FIG. 3 is a schematic diagram of a range-Doppler two-dimensional CA-CFAR detection;

FIG. 4 is a diagram showing the effect of conventional CA-CFAR detection;

FIG. 5 is a schematic illustration of a sliced CFAR reference area method employing the present disclosure;

FIG. 6 is a graph showing the detection effect of CA-CFAR on a slice.

Detailed Description

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

The method mainly solves the problems of filtering abnormal units and obtaining a clearer target echo by using the CFAR module under the heterogeneous strong sea clutter background.

An exemplary CFAR module process flow diagram according to the present disclosure is shown in fig. 2, employing a sliding window for pixel-by-pixel detection of scattering points that are past a level one threshold. The most critical step is how to slice N1, N2 from S1, S2.

Exemplary steps and associated methods are detailed below:

(1) Selecting a scattering point to be detected on a distance-Doppler image, and designing a protection area and a reference area with a certain size, as shown in figure 3, wherein the round point is the scattering point of the target to be detected, the middle rectangular area is the target protection area, rpro is the length of a distance-dimension protection unit, dpro is the length of a Doppler-dimension protection unit, rref is the length of a distance-dimension reference unit, dref is the length of the Doppler-dimension reference unit, and the peripheral rectangular area is divided into 10 reference areas, wherein the reference areas 7, 8, 9 and 10 are side lobe areas.

Preferably, the range of the Doppler dimension span of the protection area or the reference area is selected to be 10-20 Doppler units, the range dimension span is 1.2 of the target size, and the range of the sidelobe area is 1-3 Doppler units or 1-3 range units.

(2) And selecting areas S1 and S2 with the largest average value and the largest average value from the 10 areas.

(3) The clutter regions in the S1 and S2 reference regions are sliced out to obtain new reference regions N1 and N2 (if S1 and S2 are side lobe regions, the clutter regions do not need to be sliced out, namely S1 and S2 are N1 and N2). Solving the mean value of N1 and N2 signals, selecting a region with larger mean value as a final reference region LastRefArea, and calculating a CA CFAR threshold of the scattering point. The CA CFAR threshold is calculated from the following equation:

wherein:

is the signal mean value in the reference area LastRefArea, and alpha is a threshold amplitude factor, and is obtained by the following formula:

wherein: n is the reference unit number, P _fa Is the false alarm rate.

The specific way to slice the clutter region N1 in the S1 reference region is: and selecting a peak point of the S1 region, and defining a region near the peak point as N1. In a specific operation:

1. obtaining a Doppler index midcosD of a peak value by averaging Doppler dimensions (according to rows);

2. on a single distance channel of the equal point (interval value is interval) of the distance dimension 10, calculating the average value of Doppler dimensions (MidposD-2: midposD+2) in columns, and obtaining the index of the maximum value, namely the distance dimension index MidposR of the peak value;

the region of the Doppler dimension + -Dref/3 and the distance dimension + -3 x interval is defined as a new reference region N1 with (midcosD, midcosR) as a center point.

Application example

In a certain radar test, the scene is a non-uniform strong sea clutter scene, the result of the first-level threshold crossing of the radar echo signal is shown in fig. 4 (a), and when the conventional two-dimensional CFAR is adopted, the effect is shown in fig. 4 (b). It can be seen that there are many sea clutter "outliers" that are not filtered out.

When the slice CFAR detection provided by the disclosure is adopted, the sliding window is adopted for pixel-by-pixel detection on the scattering points passing through the first-level threshold, and the specific operation is as follows:

1. selecting a scattering point to be detected on a distance-Doppler image, and designing a protection area and a reference area with certain sizes, as shown in fig. 2, wherein the dots are scattering points of an object to be detected, the middle rectangular area is an object protection area, and the peripheral rectangular area is divided into 10 reference areas.

2. Among 10 regions, regions S1 and S2 having the largest average value and the largest average value are selected, and in this example, reference region 1 and reference region 6 are selected.

3. The clutter regions in the S1, S2 reference regions are sliced out, and the slicing results are shown as N1, N2 in fig. 5. And selecting the new reference areas N1 and N2 with larger average value as a final reference area LastRefArea to calculate the CA CFAR threshold of the detection point.

Using the effects of the present disclosure as shown in fig. 6, it can be seen that the present disclosure can "filter out" abnormal cells well.

The foregoing technical solutions are merely exemplary embodiments of the present invention, and various modifications and variations can be easily made by those skilled in the art based on the application methods and principles disclosed in the present invention, not limited to the methods described in the foregoing specific embodiments of the present invention, so that the foregoing description is only preferred and not in a limiting sense.

Claims

1. A self-adaptive target detection method under a non-uniform strong sea clutter scene comprises the following steps:

wherein:

wherein: n is the reference unit number, P _fa Is the false alarm rate.

2. The method according to claim 1, wherein in the step S103, the method for slicing out the clutter region in the reference region includes: and selecting peak points in the reference area, and defining an area near the peak points as a clutter area, namely a new reference area.

3. The detection method according to claim 2, wherein the method for acquiring the new reference area specifically comprises:

4. The detection method according to claim 2, wherein in the step S103, if the reference area S1 or S2 is a side lobe area, slicing is not required, and the reference area is a new reference area.

5. The method of any one of claims 1-4, wherein the guard or reference region has a doppler dimension span of 10-20 doppler cells, a range dimension span of 1.2 target dimensions, and a side lobe region span of 1-3 doppler cells or 1-3 range cells.

6. The method of claim 1, wherein the scatter points that are past the first threshold are detected pixel by pixel using a sliding window.