CN111624605B - Marine radar target detection method based on angle dimension echo characteristics - Google Patents

Marine radar target detection method based on angle dimension echo characteristics Download PDF

Info

Publication number
CN111624605B
CN111624605B CN202010461180.9A CN202010461180A CN111624605B CN 111624605 B CN111624605 B CN 111624605B CN 202010461180 A CN202010461180 A CN 202010461180A CN 111624605 B CN111624605 B CN 111624605B
Authority
CN
China
Prior art keywords
echo
radar
samples
ring
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010461180.9A
Other languages
Chinese (zh)
Other versions
CN111624605A (en
Inventor
卢志忠
文保天
胡佳幸
张润博
郭树渊
张玉莹
孙雷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Harbin Engineering University
Original Assignee
Harbin Engineering University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN202010461180.9A priority Critical patent/CN111624605B/en
Publication of CN111624605A publication Critical patent/CN111624605A/en
Application granted granted Critical
Publication of CN111624605B publication Critical patent/CN111624605B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/937—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of marine craft
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G01S7/414—Discriminating targets with respect to background clutter

Abstract

The invention provides a marine radar target detection method based on angle dimension echo characteristics. In an off-line state, concentric rings are taken from radar images and defined as one-dimensional arrays with the same distance and different directions, two types of angle dimensional echo samples on the n concentric rings are selected, one type of angle dimensional echo samples is samples containing target radar echoes, the other type of angle dimensional echo samples is pure sea clutter radar echo samples without targets, near-far attenuation compensation is completed, one echo characteristic parameter R is selected, probability density curves of the two types of samples are calculated, and a detection threshold T capable of distinguishing the two types of samples is obtained; collecting data on a single concentric ring of the X-band navigation radar, and calculating to obtain a radar echo characteristic value K on each ring; and comparing the obtained echo characteristic value K with a detection threshold value T, and judging whether a target exists on the ring. The method can more accurately and quickly judge the target in the navigation process, thereby reducing the influence of false alarm and false alarm omission on navigation.

Description

Marine radar target detection method based on angle dimension echo characteristics
Technical Field
The invention relates to a radar target detection method.
Background
With the continuous development of the field of ship navigation, the target detection problem under the background of sea clutter becomes a hotspot for research. The accuracy of target detection has important significance in the aspects of route planning, effective obstacle avoidance and the like in the navigation process of the ship.
The traditional target detection method is mainly based on a Constant False Alarm Rate (CFAR) technology, and the technology can realize a constant false alarm rate by adaptively changing a threshold value, thereby ensuring the detection probability. At present, according to different detection types, the detection can be generally divided into an average value type CFAR detection and an ordered statistic type CFAR detection. In 1968, Cell Averaging, CA, CFAR detector was first proposed by H M Finn et al, which performs best in all respects under uniform clutter conditions, but has limited overall performance under multi-target and non-uniform ambient conditions. To solve this problem, Hansen and Trunk in 1973 and 1978, respectively, proposed a constant false alarm rate for cell averaging (GO-CFAR) [1] and a constant false alarm rate for cell averaging (SO-CFAR) algorithm [2 ]. Subsequently, Rohling proposed an order-statistic constant false alarm rate (OS-CFAR) detector in 1983, which maintained good results in a multi-objective background [3 ]. In 2015, people such as Tanjie put forward a double-threshold CFAR algorithm according to characteristics and processing modes of civil navigation radar (JRC) video images, so that detection accuracy is improved while detection of marine targets under different backgrounds is realized [4 ]. In 2016, Zhao Li Ka studied deeply the performance of several classical CFAR detectors applicable to radar target detection in homogeneous clutter, clutter edges, and multi-target environments [5 ]. See references [1-5] (Amoozegar F, surrounding frequency algorithm Target detection in university: a Neural processing algorithm [ C ] Applications of organic Networks V.International Society for Optics and photometics, 1994.Trunk, G.V.Range Resolution of Targets Using Automatic Detectors [ J ] IEEE Transactions on an aqueous and Electronic Systems,1978, AES-14(5): sword 750-755.Rohling H.R. radar CFhrcing in Clutt and Multi Target detection [ J ] IEEE Transactions and optoelectronic Systems [ J ] 19, Australian analysis and noise system [ J ] 19, Australian and CFR [ 19 ] 19, Beijing K ] 99. R.R. K.R. K. constant radar detection in High, K.R. K. 5. detection of Targets and R.R.R. K. R.R. K. 5. R.R. K. detection of Targets and R.R.R.S. K. A. R. K. 5. R. K. detection of Targets and R.R.R.R. K. detection of Targets and R. A. K. A. 3, K. A. detection of Targets
In terms of detection efficiency, all CFAR detectors disclosed in the literature at present adopt a point-by-point scanning detection mode to detect targets, and with the increase of complexity of CFAR algorithms and the refinement of radar resolution, the influence of target detection speed on a radar system cannot be ignored. In order to realize the rapidity of detection, in 2019, a method applied to the rapid detection method of the target of the X-band navigation radar is discussed, and a detection algorithm is optimized into 2 steps: firstly, line-by-line (radial dimension) detection is carried out, and scanning lines with targets are screened out (rough detection); and the second step is to perform point-by-point detection (fine detection) only on the scanning lines with the targets screened out [6 ]. See reference [6] (parallel root. research on SVM-based navigation radar target detection technology [ D ]. Harbin engineering university, 2019.)
When the electromagnetic wave transmitted by the navigation radar antenna propagates in the air, the electromagnetic wave attenuates along with the increase of the propagation distance, so that the average intensity of the echo of the electromagnetic wave at a near place received by the antenna is higher than that of the echo at a far place, namely, a near-far effect exists. The fast detectors disclosed in the prior document [6] all adopt a mode of radially selecting scanning lines, are influenced by the near-far effect, and have lower target detection accuracy.
Disclosure of Invention
The invention aims to provide a navigation radar target detection method based on angle dimension echo characteristics, which can improve the accuracy of extracted radar echo characteristic data and improve the accuracy of target detection.
The purpose of the invention is realized by the following steps:
step 1, taking concentric rings from a radar image in an off-line state, defining the concentric rings as one-dimensional arrays with the same distance and different directions, selecting two types of angle dimensional echo samples on the n concentric rings, wherein one type of angle dimensional echo samples is samples containing target radar echoes, and the other type of angle dimensional echo samples is pure sea clutter radar echo samples without targets, completing near-far attenuation compensation, selecting an echo characteristic parameter R, calculating probability density curves of the two types of samples, and obtaining a detection threshold T capable of distinguishing the two types of samples;
step 2, collecting data on a single concentric ring of the X-band navigation radar, and calculating to obtain a radar echo characteristic value K on each ring;
and 3, comparing the obtained echo characteristic value K with a detection threshold value T, and judging whether a target exists on the ring.
The present invention further comprises:
1. the step 1 specifically comprises the following steps:
step 1.1, carrying out an observation test off line, reading a radar original file and drawing an image by using software, and carrying out same-frequency interference suppression processing on a radar image by using a selected filtering algorithm;
step 1.2, concentric rings are taken from the radar image with the same frequency removed, the concentric rings are defined as one-dimensional arrays with the same distance and different directions, two types of angle dimensional echo samples on the n concentric rings are selected, one type of angle dimensional echo samples are samples containing target radar echoes, and the other type of angle dimensional echo samples are pure sea clutter radar echo samples without targets;
step 1.3, compensating each concentric ring data of the two types of samples by using a selected attenuation compensation mode;
step 1.4, selecting an echo characteristic parameter R, and acquiring a group of given angle dimension radar echo data sampling points (x) on a single ringi，yi) 1, 2, …, m, where xiRepresenting the ith point, y, on a single ringiPresentation sheetThe radar echo intensity of the ith point on each ring, m represents the total number of angle dimension sampling points on each ring, and the values of echo characteristic parameters R of all samples are calculated;
step 1.5, respectively drawing probability density curves of the characteristic R of the radar echo containing the target and the pure sea clutter radar echo according to the value of the characteristic R of the two types of echoes with a certain amount obtained in the step 1.4;
and 1.6, solving the detection threshold value T of the characteristic R by using two probability density curves and adopting a selection strategy.
2. The step 2 specifically comprises the following steps:
step 2.1, extracting the angle direction, the radial distance and the echo intensity of a single ring of the radar to be detected;
and 2.2, calculating a value K of the characteristic R of the echo intensity on the single radar ring according to the step 1.4 in the step 1.
3. The step 3 specifically comprises the following steps:
step 3.1, when the value K of the characteristic R of the echo intensity on the radar single ring is larger than a detection threshold T, if the value R of the echo containing the target radar is smaller than the threshold T during off-line testing, judging that no target exists; if the R value containing the target radar echo is larger than the threshold T in the off-line test, determining that a target exists;
step 3.2, when the value K of the characteristic R of the echo intensity on the radar single ring is smaller than a detection threshold T, if the R value of the echo containing the target radar is smaller than the threshold T during off-line testing, determining that a target exists; and if the R value containing the target radar echo is larger than the threshold T in the offline test, judging that no target exists.
4. The step 1.3 specifically comprises:
step 1.3.1, replacing the echo intensity of each ring of data by a value obtained by subtracting the average value of the echo intensity of each ring of data from the echo intensity of each ring of data;
and step 1.3.2, removing sample points of the part, with the amplitude being obviously lower, of the echo signal, influenced by the stern trace, in each ring of data after attenuation compensation.
5. Step 1.4 specifically includes:
step 1.4.1, a calculation formula of fitting angle dimension radar echo data is as follows:
Figure BDA0002510977310000031
in the formula: a iskK is 0, 1, 2, …, n represents the coefficient of the multivariate function,
Figure BDA0002510977310000032
for functions formed by polynomials of degree not exceeding n, xiRepresenting the ith point, y, on a single ringiRepresenting the radar echo intensity at the ith point on a single ring, m representing the total number of angular dimension sample points on a single ring, n representing the highest order of x in the polynomial
Find akAnd k is 0, 1, 2, …, n is as follows:
Figure BDA0002510977310000041
then there are:
Figure BDA0002510977310000042
in the formula: y isiRepresenting the radar echo intensity at the ith point on a single ring,
Figure BDA0002510977310000043
angle dimensional radar echo data representing a fit of the ith ring, m represents the total number of sample points on a single ring, n represents the highest order of x in the polynomial, akK is 0, 1, 2, …, n represents the coefficient of the multivariate function;
step 1.4.2, if the fitting correlation coefficient is selected as the echo characteristic parameter, calculating the fitting correlation coefficient according to the fitting angle dimension radar echo data obtained in the step 1.4.1,
fitting a calculation formula of the correlation coefficient:
Figure BDA0002510977310000044
in the formula, N represents the data length,
Figure BDA0002510977310000045
representing the value of the radar echo for the ith ring fit,
Figure BDA0002510977310000046
means for indicating the intensity average of the i-th ring radar echo, yiRepresenting the radar echo intensity at the ith point on a single ring.
Aiming at the problems in the prior art, the invention provides a marine radar target rapid detection technology based on angle dimension echo characteristics. The navigation radar type applicable to the method is as follows: in order to realize the rapidity of detection, a detection algorithm is optimized into 2 steps: the first step is that the angle dimension data set (same distance and different directions) of each concentric ring in the radar echo is integrally detected, the rings with targets are screened out, and coarse detection is completed; and in the second step, point-by-point detection is carried out only on the rings with the targets screened out, so that fine detection is completed. The algorithm involved in the patent belongs to the coarse detection stage and is used for screening out concentric rings with targets.
When the method is specifically implemented, after the detection threshold is determined through an offline field test, the characteristic value of the radar single ring which is actually measured can be compared with the detection threshold, a large amount of radar single ring data which do not contain a target can be rejected in advance, the influence of the near-far effect on the target detection performance is solved, the detection precision is improved, the detection efficiency is improved, the requirements of the radar system on the real-time performance and the accuracy of target detection are met, and the method is used as the first step of target rapid detection.
Compared with the prior art, the marine radar target rapid detection technology provided by the invention has the advantages that:
(1) according to the method, the target is quickly detected by using the angle dimension radar echo characteristics in the X-band marine radar image, attenuation compensation is performed on data of each concentric ring, the influence of the near-far effect on the radar echo characteristics can be effectively avoided, and the accuracy of the target detector is improved.
(2) The method utilizes the angle dimension radar echo characteristics in the X-band marine radar image to carry out target rapid detection, determines the detection threshold value by counting a large amount of actually-measured navigation radar data, and enhances the engineering practicability of the method.
Drawings
FIG. 1A single radar image drawn by the MATLAB program.
FIG. 2 shows the angle dimension data selection.
Fig. 3 contains radar echo samples of target echoes.
FIG. 4 a radar echo sample of pure sea clutter.
FIG. 5 attenuates radar echo samples containing target echoes after compensation.
FIG. 6 attenuates radar echo samples of compensated pure sea clutter.
Figure 7 contains a fitted correlation coefficient curve for the target echo.
FIG. 8 is a curve of fitted correlation coefficients for pure sea clutter.
FIG. 9 is a probability density curve of two types of samples under the characteristic of fitting correlation coefficients.
FIG. 10 is a flow chart of an embodiment of the present invention.
Detailed Description
The invention is further illustrated below by way of example.
The first embodiment is as follows:
the invention discloses a marine radar target rapid detection technology based on angle dimension echo characteristics, and aims to solve the problem that the distance effect of radial dimension radar echoes affects radar echo characteristics, improve the accuracy of extracted radar echo characteristic data and improve the accuracy of target detection. The implementation steps are as follows:
step 1, selecting a wavelet sample in an angle dimension, and determining a detection threshold value T. And (3) carrying out an observation test off line, and taking concentric rings from the radar image, wherein the concentric rings are defined as one-dimensional arrays with the same distance and different directions. Two types of angle dimension echo samples on the n concentric rings are selected, one type is a sample containing target radar echoes, and the other type is a pure sea clutter radar echo sample without targets. And completing the distance attenuation compensation. And selecting a proper echo characteristic parameter R, and calculating the probability density curve of the two types of samples according to the characteristic parameter R to obtain a threshold T capable of distinguishing the two types of samples.
And 2, extracting the echo characteristic value K of the original image of the radar to be detected. And collecting data (including angle direction, radial distance, echo intensity and the like) on a single concentric ring of the X-waveband navigation radar, and calculating to obtain a radar echo characteristic value K on each ring.
And 3, judging whether the target exists or not. And comparing the obtained echo characteristic value K with a detection threshold value T, and judging whether a target exists on the ring.
Example two:
embodiment two is based on embodiment one, step 1 further includes the following steps:
step 1.1, carrying out an observation test off line, and reading a radar original file and drawing an image by using software. And performing same-frequency interference suppression processing on the radar image by using the selected filtering algorithm.
Step 1.2, concentric rings are taken from the radar image with the same frequency, the concentric rings are defined as one-dimensional arrays with the same distance and different directions, two types of angle-dimensional echo samples on the n concentric rings are selected, one type is a sample containing target radar echoes, and the other type is a pure sea clutter radar echo sample without targets.
And 1.3, compensating the data of each concentric ring for the two types of samples by using a selected attenuation compensation mode.
Step 1.4, selecting a proper echo characteristic parameter R (such as a fitting correlation coefficient, a variance, a mean value and the like), and acquiring a set of given angle dimension radar echo data sampling points (x) on a single ringi，yi) (i-1, 2, …, m) where xiRepresenting the ith point, y, on a single ringiIndicating the radar echo intensity at the ith point on a single ring and m indicating the total number of angular dimension sampling points on a single ring. The values of R were calculated for all samples.
And step 1.5, respectively drawing probability density curves of the characteristic R of the radar echo containing the target and the characteristic R of the pure sea clutter radar echo according to the characteristic R values of the two types of echoes with a certain amount obtained in the step 1.4.
And 1.6, solving the detection threshold value T of the characteristic R by using two probability density curves and adopting a selection strategy.
Example three:
in the third embodiment, the step 2 further comprises the following steps on the basis of the second embodiment:
and 2.1, extracting the angle direction, the radial distance and the echo intensity on the single ring of the radar to be detected.
And 2.2, calculating a value K of the characteristic R of the echo intensity on the single radar ring according to the step 1.4 in the step 1.
Example four:
in the fourth embodiment, the step 3 further comprises the following steps on the basis of the third embodiment:
step 3.1, when the value K of the characteristic R of the echo intensity on the radar single ring is larger than the detection threshold T, judging that no target exists (the R value containing the target radar echo is smaller than the threshold T in the offline test) or a target exists (the R value containing the target radar echo is larger than the threshold T in the offline test)
And 3.2, when the value K of the characteristic R of the echo intensity on the radar single ring is smaller than a detection threshold T, judging that a target exists (the R value containing the target radar echo is smaller than the threshold T in the offline test) or no target exists (the R value containing the target radar echo is larger than the threshold T in the offline test).
Example five:
example five step 1.3 is based on example four and further comprises the following steps:
and 1.3.1, selecting an attenuation compensation mode of subtracting a single ring mean value to compensate the data of each ring of the two types of samples. The specific method comprises the following steps: and replacing the echo intensity of each ring of data by a value obtained by subtracting the average value of the echo intensities of the ring of data from the echo intensity of each ring of data.
And step 1.3.2, removing sample points of the part, with the amplitude being obviously lower, of the echo signal, influenced by the stern trace, in each ring of data after attenuation compensation.
Example six:
example six is that step 1.4 further comprises the following steps on the basis of example five:
step 1.4.1, if the fitting correlation coefficient is selected as the echo characteristic parameter, a group of given angle dimension radar echo data sampling points (x) acquired on a single ringi，yi) (i ═ 1, 2, …, m), fitting angle dimensional radar echo data derived with a least squares fitting polynomial are given.
The calculation formula of the fitting angle dimension radar echo data is as follows:
Figure BDA0002510977310000071
in the formula: a is ak(k-0, 1, 2, …, n) represents the coefficient of a multivariate function,
Figure BDA0002510977310000072
for functions formed by polynomials of degree not exceeding n, xiRepresenting the ith point, y, on a single ringiRepresenting the radar echo intensity at the ith point on a single ring, m representing the total number of angular dimension sample points on a single ring, n representing the highest order of x in the polynomial
Find ak(k is 0, 1, 2, …, n) is such that:
Figure BDA0002510977310000073
then there are:
Figure BDA0002510977310000074
in the formula: y isiRepresenting the radar echo intensity at the ith point on a single ring,
Figure BDA0002510977310000075
represents the ithAngle-dimensional radar echo data fitted to a ring, m representing the total number of sample points on a single ring, n representing the highest order of x in the polynomial, ak(k-0, 1, 2, …, n) represents the coefficients of a multivariate function.
And step 1.4.2, if the fitting correlation coefficient is selected as the echo characteristic parameter, calculating the fitting correlation coefficient according to the fitting angle dimensional radar echo data obtained in the step 1.4.1.
Fitting a calculation formula of the correlation coefficient:
Figure BDA0002510977310000081
in the formula, N represents the data length,
Figure BDA0002510977310000082
representing the value of the radar echo for the ith ring fit,
Figure BDA0002510977310000083
means for indicating the intensity average of the i-th ring radar echo, yiRepresenting the radar echo intensity at the ith point on a single ring.
The implementation process of the invention is shown in figure 10, and can be specifically divided into the following steps, wherein the first step is to select a echo sample in an angle dimension, determine a detection threshold value T, the second step is to extract an echo characteristic value K of an original image of a radar to be detected, and the third step is to judge whether a target exists or not.
The marine radar used in the following embodiments is an X-band navigation radar, and operates in a short pulse mode, echo data is stored in a polar coordinate form in lines after being digitized, the time interval between two adjacent storage lines is less than 1ms, the time for a radar antenna to scan for one circle is about 2.5s, the number of the radar image bus is 2048, each line has 2048 pixel points, the azimuth resolution is about 0.1 °, and the radial resolution is about 2.5 m.
The main technical parameters of the marine radar are shown in the first table:
TABLE technical parameters of a marine radar
Figure BDA0002510977310000084
With reference to the attached drawings 1-10, the specific experimental steps of the invention are as follows:
in a first step, a wavelet sample is selected in the angular dimension and a detection threshold T is determined. The method comprises the following steps:
step 1.1, carrying out an off-line observation test, and reading and drawing an image of a radar original file by using MATLAB software, wherein the attached drawing 1 is a drawn single radar image. The method for restraining the same frequency interference of the radar image by using the spatial domain correlation method comprises the following specific steps: in the area close to the radar antenna, the echo strength value of each image element point is replaced by the median of the echo strengths of two non-noise points with the same radial distance on the left and right adjacent storage lines of the point. In the area far away from the radar antenna, the echo intensity value of each pixel point is replaced by the median of the echo intensities of 7 non-noise points in the 3 × 3 neighborhood window of the point.
Step 1.2, taking the angle in the radar image as a basic unit in the radar image after the same frequency is removed, taking concentric rings in the image, wherein the attached drawing 2 is a schematic diagram of an angle dimension data selection mode, a black inner ring in the image is a starting ring for selecting target echoes, a black outer ring in the image is a termination ring for selecting target echoes, and rings between two concentric rings can be selected as radar echo samples containing targets. By the angle dimension selection mode, two types of angle dimension echo samples on 3455 rings are selected, wherein 1084 rings contain target radar echo samples, and 2371 rings contain pure sea clutter radar echo samples without targets. Fig. 3 and 4 are result graphs after two types of sample angle dimensions are selected.
And 1.3, selecting an attenuation compensation mode of subtracting a single ring mean value to compensate the data of each ring of the two types of samples. The specific method comprises the following steps: and replacing the echo intensity of each ring of data by a value obtained by subtracting the average value of the echo intensities of the ring of data from the echo intensity of each ring of data. In each ring of data after attenuation compensation, sample points of the part of the echo signal with obviously low amplitude caused by the influence of the stern trace are removed, and the removal range is from the 900 th point to the 1300 th point in the example. Fig. 5 and 6 are graphs of the results of two types of sample attenuation compensation.
Step 1.4, for a set of given angle dimension radar echo data sampling points (x) acquired on a single ringi，yi) (i ═ 1, 2, …, m), fitting angle dimensional radar echo data and fitting correlation coefficients obtained by least squares fitting polynomial are given, and fig. 7 and 8 show fitting correlation coefficients of two types of echo signals.
The calculation formula of the fitting angle dimension radar echo data is as follows:
Figure BDA0002510977310000091
in the formula: a is ak(k-0, 1, 2, …, n) represents coefficients of a multivariate function,
Figure BDA0002510977310000092
for functions formed by polynomials of degree not exceeding n, xiRepresenting the ith point, y, on a single ringiThe radar echo intensity of the ith point on the single ring is represented, m represents the total number of angle dimensional sampling points on the single ring, the value of the example is 1648 (the sampling point influenced by a trail is removed), n represents the highest order of x in the polynomial, and the value is 2
Find ak(k is 0, 1, 2, …, n) and:
Figure BDA0002510977310000093
then there are:
Figure BDA0002510977310000101
in the formula: y isiRepresenting the radar echo intensity at the ith point on a single ring,
Figure BDA0002510977310000102
representThe fitted angle dimensional radar echo data for the ith ring, m represents the total number of samples on the single ring, which is 1648 in this case (the samples affected by the trail are removed), n represents the highest order of x in the polynomial as 2, ak(k-0, 1, 2, …, n) represents the coefficient of a multivariate function
Fitting a calculation formula of the correlation coefficient:
Figure BDA0002510977310000103
wherein, N represents the data length, the value of the present example is 1648 (the sampling point influenced by the trail is removed),
Figure BDA0002510977310000104
representing the value of the radar echo for the i-th ring fit,
Figure BDA0002510977310000105
means for indicating the intensity average of the i-th ring radar echo, yiRepresenting the intensity of the radar echo at the ith point on a single ring
And step 1.5, respectively drawing probability density curves of the fitting correlation coefficients of the radar echo containing the target and the pure sea clutter radar echo through the fitting correlation coefficient values of the 3455 two types of echoes obtained in the step 1.4. FIG. 9 is a graph of probability density for two types of samples.
Step 1.6, the current selection strategy is to use the fitting correlation coefficient 0.7755 corresponding to the intersection point of the two probability density curves as the detection threshold T. (in addition to this selection strategy, other selection strategies can be adopted, such as constant false alarm rate, etc.)
And secondly, extracting an echo characteristic value K of the original image of the radar to be detected. The method comprises the following steps:
and 2.1, extracting the angle direction, the radial distance and the echo intensity on the single ring of the radar to be detected.
And 2.2, calculating a value K of a fitting correlation coefficient of the echo intensity on the radar single ring according to the step 1.4 in the step 1.
And thirdly, judging whether the target exists or not. The method comprises the following steps:
and 3.1, judging that a target exists when the fitting correlation coefficient value K of the echo intensity on the radar single ring is greater than a detection threshold value 0.7755.
And 3.2, judging that no target exists when the fitting correlation coefficient value K of the echo intensity on the single radar ring is smaller than a detection threshold value 0.7755.
The rapid detection technology for the marine radar target based on the angle dimension echo characteristics provided by the invention is used for carrying out experimental analysis on a large amount of radar data and sea condition information of relevant time periods, which are acquired by a test ship in the course of navigation in the east sea area in 2018. In the experiment, data of 2018, 6 and 8 days, 10 and 20 days and 10 and 23 days are selected, wherein the data comprise a target sample 1172 group, a pure sea clutter sample 2456 group and a total 3628 group. A comparative experiment was performed using the target detection method and the method of the present invention, respectively, which extract echo features in the radial dimension.
Chi-square test is a hypothesis test method which is currently applied in a plurality of fields, and has more outstanding advantages and effects in the aspects of feature selection and the like. In the feature selection process of the current target and sea clutter classification problem, chi-square test is introduced to judge the relative correlation between the selected features and two types of samples, and the bigger chi-square value indicates the larger degree of characterization of the features on the sample characteristics.
In the CHI-square test method, the CHI value is calculated as follows:
Figure BDA0002510977310000111
in the formula: t represents a feature, CtRepresenting sample categories, and for the convenience of explanation, the meanings of the rest variables in the formula are shown in a second table:
checking formula meaning by using table two chi square
Figure BDA0002510977310000112
The experimental data were counted according to the determined detection threshold, and the results are shown in table three, table four, and table five:
TABLE III sample representation of target detection method based on radial dimension extraction fitting correlation coefficient
Figure BDA0002510977310000113
TABLE IV sample Performance of target detection method based on Angle dimension extraction fitting correlation coefficient (inventive method)
Figure BDA0002510977310000114
CHI-value comparison of the five methods in Table
Figure BDA0002510977310000115
Experimental results show that the CHI value of the target rapid detection method using the angle dimension echo characteristics is far larger than that of the target rapid detection method using the radial dimension echo characteristics, and radar echoes containing targets and radar echoes of pure sea clutter can be better distinguished.
The marine radar target rapid detection technology based on the angle dimension echo characteristics improves the accuracy and efficiency of target detection. The technology solves the problem of influence of the radial dimension radar echo far and near effect on radar echo characteristics, so that the accuracy of extracted radar echo characteristic data is improved, and the accuracy of target detection is improved.

Claims (5)

1. A marine radar target detection method based on angle dimension echo characteristics is characterized by comprising the following steps:
step 1, taking concentric rings from a radar image in an off-line state, defining the concentric rings as one-dimensional arrays with the same distance and different directions, selecting two types of angle dimensional echo samples on the n concentric rings, wherein one type of angle dimensional echo samples is samples containing target radar echoes, and the other type of angle dimensional echo samples is pure sea clutter radar echo samples without targets, completing near-far attenuation compensation, selecting an echo characteristic parameter R, calculating probability density curves of the two types of samples, and obtaining a detection threshold T capable of distinguishing the two types of samples;
step 1.1, carrying out an observation test off line, reading a radar original file and drawing an image by using software, and carrying out same-frequency interference suppression processing on a radar image by using a selected filtering algorithm;
step 1.2, taking concentric rings from the radar image with the same frequency, defining the concentric rings as one-dimensional arrays with the same distance and different directions, and selecting two types of angle-dimensional echo samples on the n concentric rings, wherein one type of angle-dimensional echo samples is samples containing target radar echoes, and the other type of angle-dimensional echo samples is pure sea clutter radar echo samples without targets;
step 1.3, compensating each concentric ring data of the two types of samples by using a selected attenuation compensation mode;
step 1.4, selecting an echo characteristic parameter R, and obtaining a group of given angle dimension radar echo data sampling points (x) on a single ringi，yi) 1, 2, …, m, wherein xiRepresenting the ith point, y, on a single ringiThe radar echo intensity of the ith point on the single ring is represented, m represents the total number of angle dimension sampling points on the single ring, and the values of echo characteristic parameters R of all samples are calculated;
step 1.5, respectively drawing probability density curves of the characteristic R of the radar echo containing the target and the characteristic R of the pure sea clutter radar echo according to the value of the characteristic R of the two types of echoes with a certain amount obtained in the step 1.4;
step 1.6, solving a detection threshold T of the characteristic R by using two probability density curves and adopting a selection strategy;
step 2, collecting data on a single concentric ring of the X-waveband navigation radar, and calculating to obtain a radar echo characteristic value K on each ring;
and 3, comparing the obtained echo characteristic value K with a detection threshold value T, and judging whether a target exists on the ring.
2. The method for detecting marine radar target based on angle dimension echo characteristics according to claim 1, wherein the step 2 specifically comprises:
step 2.1, extracting the angle direction, the radial distance and the echo intensity on the single ring of the radar to be detected;
and 2.2, calculating a value K of the characteristic R of the echo intensity on the single radar ring according to the step 1.4 in the step 1.
3. The method for detecting marine radar target based on angular dimension echo characteristics according to claim 2, wherein the step 3 specifically comprises:
step 3.1, when the value K of the characteristic R of the echo intensity on the radar single ring is larger than a detection threshold T, if the value R of the echo containing the target radar is smaller than the threshold T during off-line testing, judging that no target exists; if the R value containing the target radar echo is larger than the threshold T in the off-line test, determining that a target exists;
step 3.2, when the value K of the characteristic R of the echo intensity on the radar single ring is smaller than a detection threshold T, if the R value of the echo containing the target radar is smaller than the threshold T during off-line testing, determining that a target exists; and if the R value containing the target radar echo is larger than the threshold T in the offline test, judging that no target exists.
4. The method for detecting marine radar target based on angular dimension echo characteristics according to claim 3, wherein the step 1.3 specifically comprises:
step 1.3.1, replacing the echo intensity of each ring of data by a value obtained by subtracting the average value of the echo intensity of each ring of data from the echo intensity of each ring of data;
and step 1.3.2, removing sample points of the part, with the amplitude being obviously lower, of the echo signal, influenced by the stern trace, in each ring of data after attenuation compensation.
5. The method for detecting marine radar target based on angle dimension echo characteristics according to claim 4, wherein the step 1.4 specifically comprises:
step 1.4.1, a calculation formula of fitting angle dimension radar echo data is as follows:
Figure FDA0003626285280000021
in the formula: a is akK is 0, 1, 2, …, n represents the coefficient of the multivariate function,
Figure FDA0003626285280000022
a function formed of all polynomials of degree not exceeding n, xiRepresenting the ith point, y, on a single ringiRepresenting the radar echo intensity of the ith point on a single ring, m representing the total number of angle dimension sampling points on the single ring, and n representing the highest order of x in the polynomial;
find akAnd k is 0, 1, 2, …, n is such that:
Figure FDA0003626285280000023
then there are:
Figure FDA0003626285280000024
in the formula:
Figure FDA0003626285280000025
angular dimensional radar echo data representing a fit of the ith ring;
step 1.4.2, if the fitting correlation coefficient is selected as the echo characteristic parameter, calculating the fitting correlation coefficient according to the fitting angle dimension radar echo data obtained in the step 1.4.1,
fitting a calculation formula of the correlation coefficient:
Figure FDA0003626285280000026
in the formula, N represents the data length,
Figure FDA0003626285280000031
and representing the intensity average value of the ith ring radar echo.
CN202010461180.9A 2020-05-27 2020-05-27 Marine radar target detection method based on angle dimension echo characteristics Active CN111624605B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010461180.9A CN111624605B (en) 2020-05-27 2020-05-27 Marine radar target detection method based on angle dimension echo characteristics

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010461180.9A CN111624605B (en) 2020-05-27 2020-05-27 Marine radar target detection method based on angle dimension echo characteristics

Publications (2)

Publication Number Publication Date
CN111624605A CN111624605A (en) 2020-09-04
CN111624605B true CN111624605B (en) 2022-07-15

Family

ID=72271395

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010461180.9A Active CN111624605B (en) 2020-05-27 2020-05-27 Marine radar target detection method based on angle dimension echo characteristics

Country Status (1)

Country Link
CN (1) CN111624605B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117724048A (en) * 2024-02-07 2024-03-19 中国人民解放军海军航空大学 RDPH feature-based improved OS-CFAR detection algorithm, device and medium

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6750805B1 (en) * 2002-12-20 2004-06-15 The Boeing Company Full polarization synthetic aperture radar automatic target detection algorithm
JP2013246022A (en) * 2012-05-25 2013-12-09 Furuno Electric Co Ltd Target detection device, radar apparatus, target detection method, and target detection program
CN105137424A (en) * 2015-07-17 2015-12-09 电子科技大学 Real-beam scanning radar angular super-resolution method under clutter background
CN105319537A (en) * 2015-10-16 2016-02-10 哈尔滨工程大学 Navigation radar co-frequency interference inhibition method based on spatial correlation
CN108872982A (en) * 2018-05-10 2018-11-23 北京航空航天大学 The extraction of Multiple Scattering feature and correction processing method in radar target RCS far to field transformation
CN109683160A (en) * 2019-03-19 2019-04-26 长沙深之瞳信息科技有限公司 A kind of object detection method suitable for through-wall radar
CN110208767A (en) * 2019-06-05 2019-09-06 哈尔滨工程大学 A kind of radar target rapid detection method based on fitting correlation coefficient
CN110221266A (en) * 2019-06-11 2019-09-10 哈尔滨工程大学 A kind of marine radar target rapid detection method based on support vector machines
CN110412550A (en) * 2019-07-20 2019-11-05 中国船舶重工集团公司第七二四研究所 A kind of region adaptivity sea object detection method based on context-aware
CN110940953A (en) * 2019-11-27 2020-03-31 自然资源部第一海洋研究所 Three-dimensional detection method for targets in ground wave radar sea clutter
CN111123212A (en) * 2019-12-11 2020-05-08 安徽四创电子股份有限公司 Signal processing method of scene surveillance radar based on complex clutter background

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2297835A (en) * 1995-02-08 1996-08-14 Secr Defence Three dimensional detection of contraband using x rays
CN103116153B (en) * 2013-01-31 2014-11-12 西安电子科技大学 Eight-unit uniform circular array antenna steering vector construction method in case of mast shielding
CN103475388B (en) * 2013-09-13 2016-08-03 山东科技大学 Communication system multi-user test method based on S order parameter reproducing kernel ICA

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6750805B1 (en) * 2002-12-20 2004-06-15 The Boeing Company Full polarization synthetic aperture radar automatic target detection algorithm
JP2013246022A (en) * 2012-05-25 2013-12-09 Furuno Electric Co Ltd Target detection device, radar apparatus, target detection method, and target detection program
CN105137424A (en) * 2015-07-17 2015-12-09 电子科技大学 Real-beam scanning radar angular super-resolution method under clutter background
CN105319537A (en) * 2015-10-16 2016-02-10 哈尔滨工程大学 Navigation radar co-frequency interference inhibition method based on spatial correlation
CN108872982A (en) * 2018-05-10 2018-11-23 北京航空航天大学 The extraction of Multiple Scattering feature and correction processing method in radar target RCS far to field transformation
CN109683160A (en) * 2019-03-19 2019-04-26 长沙深之瞳信息科技有限公司 A kind of object detection method suitable for through-wall radar
CN110208767A (en) * 2019-06-05 2019-09-06 哈尔滨工程大学 A kind of radar target rapid detection method based on fitting correlation coefficient
CN110221266A (en) * 2019-06-11 2019-09-10 哈尔滨工程大学 A kind of marine radar target rapid detection method based on support vector machines
CN110412550A (en) * 2019-07-20 2019-11-05 中国船舶重工集团公司第七二四研究所 A kind of region adaptivity sea object detection method based on context-aware
CN110940953A (en) * 2019-11-27 2020-03-31 自然资源部第一海洋研究所 Three-dimensional detection method for targets in ground wave radar sea clutter
CN111123212A (en) * 2019-12-11 2020-05-08 安徽四创电子股份有限公司 Signal processing method of scene surveillance radar based on complex clutter background

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
Radar CFARThresholding in Clutter and Multiple Target Situations;Rohling H.;《IEEE Transactions on Aerospace and Electronic Systems》;19831231;608-621 *
Radar HRRP Group-Target Recognition Based on Combined Methods in the Backgroud of Sea Clutter;Liyang Zhu等;《2018 International Conference on Radar (RADAR)》;20181231;6pp *
Underwater 3D target positioning by inhomogeneous illumination based on binocular stereo vision;Bing Zheng等;《OCEANS 2012 MTS/IEEE - YEOSU》;20111231;4pp *
X波段导航雷达图像反演海面高程技术研究;卫延波;《中国优秀博硕士学位论文全文数据库(博士)信息科技辑》;20180615;I136-90 *
基于SVM的导航雷达目标检测技术研究;骈根;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》;20190815;C032-41 *
机动平台SAR场景成像和动目标成像算法研究;董祺;《中国优秀博硕士学位论文全文数据库(博士)信息科技辑》;20190215;I136-60 *
海洋信息获取、传输、处理及融合前沿研究评述;徐文等;《中国科学:信息科学》;20160820(第08期);全文 *

Also Published As

Publication number Publication date
CN111624605A (en) 2020-09-04

Similar Documents

Publication Publication Date Title
US8094060B2 (en) Method of detecting a target
CN112731307B (en) RATM-CFAR detector based on distance-angle joint estimation and detection method
CN109597065B (en) False alarm suppression method and device for through-wall radar detection
CN110208806B (en) Marine radar image rainfall identification method
CN110208767B (en) Radar target rapid detection method based on fitting correlation coefficient
CN111157953A (en) Two-stage threshold constant false alarm detection algorithm under strong ground clutter
CN111323756B (en) Marine radar target detection method and device based on deep learning
CN111856411A (en) Double-threshold target detection method and device
CN111323757B (en) Target detection method and device for marine radar
CN111624605B (en) Marine radar target detection method based on angle dimension echo characteristics
CN108318881A (en) Marine radar image rainfall recognition methods based on K parameter
CN111680537A (en) Target detection method and system based on laser infrared compounding
CN109581366B (en) Discrete sidelobe clutter identification method based on target steering vector mismatch
CN108008374B (en) Sea surface large target detection method based on energy median
CN108387879B (en) Clutter map unit median detection method based on adaptive normalized matched filtering
EP1812807B1 (en) Waveform management method using a doppler mapping by statistical segmentation
Arivazhagan et al. Significance based ship detection from SAR imagery
CN106249241B (en) A kind of self-adapting clutter power statistic algorithm
KR101813357B1 (en) Clutter elimination Method and Apparatus for applying variable data map and variable statistic characteristics information for identifying underwater fixed target and clutter
CN114265015A (en) Clutter characteristic perception-based refined radar detection method
Faramarzi et al. Infrared dim and small targets detection based on multi-scale local contrast measure utilizing efficient spatial filters
CN109358324B (en) Target detection method based on ground monitoring radar spectrum data
CN111695087A (en) Method and system for detecting composite mine detection target
Zhou et al. Research on autonomous detection method of underwater unmanned vehicle
Madjidi et al. Cfar ship detection in sar images based on the generalized rayleigh mixture models

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant