CN111505598B - FRFT domain-based three-feature joint detection device and method - Google Patents
FRFT domain-based three-feature joint detection device and method Download PDFInfo
- Publication number
- CN111505598B CN111505598B CN202010342976.2A CN202010342976A CN111505598B CN 111505598 B CN111505598 B CN 111505598B CN 202010342976 A CN202010342976 A CN 202010342976A CN 111505598 B CN111505598 B CN 111505598B
- Authority
- CN
- China
- Prior art keywords
- domain
- detection
- frft
- feature
- fractional
- 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
Links
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
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a three-feature joint detection device and method based on FRFT domain, wherein the method comprises the following steps: collecting radar echo data and performing FRFT (fractional Fourier transform) to obtain a fractional order domain data vector; extracting the characteristics of the fractional order domain data vector; and combining the extracted features, and realizing target detection contrast analysis by using a rapid convex hull detection algorithm. The invention discloses a sea surface target detection device based on fractional Fourier transform FRFT domain three-feature combination, which proves the feasibility of the FRFT domain three-feature combination detection algorithm by comparing the actual measurement sea clutter data with the detection performance curve experiment of single feature detection.
Description
Technical Field
The invention particularly relates to a three-feature combined detection device and method based on an FRFT domain, and belongs to the technical field of radar target detection.
Background
Sea surface environment is very complicated by sea condition, sea condition and wind speed, sea clutter presents non-Gaussian, non-uniform and non-stable three-not characteristics, a traditional sea surface target detection algorithm based on energy accumulation is a statistically optimal algorithm under the conditions of linear, stable and Gaussian sea clutter, radar target detection technology in uniform and stable clutter environment is researched to be mature day by day, but a practical complicated geographic position leads to clutter background to present non-uniform and non-stable characteristics, modeling clutter as single Gaussian distribution is not reasonable any more, a selected reference unit does not obey different distribution characteristics, and complexity of sea condition leads to detection performance reduction of classical radar target optimal algorithm.
Aiming at the detection problem in a complex environment, students at home and abroad have made a great deal of research, and aim to improve the detection performance and false inhibition capability of an algorithm. The conventional constant false alarm detector has the following problems: the traditional detector assumes that clutter background obeys Rayleigh distribution, and the square law is exponential distribution after detection, but the actual clutter environment is in non-Gaussian distribution, and the false alarm can be raised by adopting a detection strategy under Rayleigh distribution at the moment; because the environment is increasingly complex, the actual radar monitoring environment is not uniformly distributed any more, and non-uniform scenes such as interference targets, clutter edges and the like can exist, so that the performance of the optimal detector under a uniform background can be drastically reduced.
In the article "detection of the joint characteristic difference of targets under sea clutter background" of the university of Nanjing information engineering, the fractal parameters of sea clutter are extracted by adopting a trend-eliminating fluctuation analysis method, the power spectrum characteristics of the sea clutter are analyzed, and a convex hull training algorithm is utilized to obtain a sea clutter discrimination region. However, due to the dispersion of the energy distribution of the time domain signal, the effect of obtaining fractal characteristics by carrying out trend fluctuation elimination analysis in the time domain is poor. While there are a number of detector algorithms that improve performance under multi-target and clutter edges, there is no single detector that can solve the detection problem in all contexts.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a FRFT domain-based three-feature combined detection device and method, which aim at improving detection performance of different targets in actual sea clutter data experiments.
In a first aspect, the present invention provides a method for three-feature joint detection based on an FRFT domain, the method comprising the steps of:
collecting radar echo data and performing FRFT (fractional Fourier transform) to obtain a fractional order domain data vector;
carrying out three feature extraction of transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation on the fractional domain data vector;
and combining the extracted features, realizing target detection by using a rapid convex hull detection algorithm, and carrying out comparison analysis with a single feature detection algorithm.
With reference to the first aspect, further, the fractional order domain data vector is obtained by the following formula:
F(u)＝FrFt(R(t)，P)
wherein, frFt is fractional Fourier transform, F (u) is fractional domain data vector after FRFT of radar echo data R (t), u is scale of FRFT domain, t is time scale, P is the order of FRFT transform; f (u) = { F 1 ,F 2 ,F 3 ,…,F N The scale of the FRFT domain u ranges from u=1, 2,3, …, N being the total length of the fractional domain data vector F (u).
Further, the feature extraction includes:
step A: the amplitude values of the fractional order domain data vector F (u) are sorted,
F(u)＝Sort({F 1 ,F 2 ,F 3 ,…,F N })＝{F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) }
wherein Sort is the ascending order of vector magnitude values, { F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) The data vector is a fractional domain data vector with the amplitude values arranged in an ascending order, and n=1, 2,3, … and N;
and (B) step (B): fractional order domain data vector { F after ascending order of amplitude values (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) Dividing the amplitude value into K independent segments with equal length,
{F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) }＝{F 1 ,F 2 ,…,F k ,…,F K }
wherein F is k Representing a K independent segment fractional domain amplitude value short vector, the range of the segment scale K is k=1, 2,3, …, K is the fractional domain amplitude value short vector F k Is set to each F k The length of the vector is N k Fractional order domain amplitude value short vector F k Probability P falling within the kth segment scale k The calculation is as follows:
step C: extracting three groups of characteristics including transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation of the fractional domain data vector F (u) based on the step B,
the transform domain information entropy FIE is expressed as:
the transform domain peak-to-average ratio FPA is expressed as:
wherein max represents maximizing the vector;
the transform domain peak standard deviation FPS is expressed as:
where std denotes standard deviation of the vectors.
Further, the fast convex hull detection algorithm includes:
step a: calculating a convex hull CH (a) of a characteristic set a of the current pure sea clutter data, wherein the minimum set of the convex hull definition a is as follows:
wherein CH (·) represents a convex set composed of the feature set a, η is the element η i Is a combination of (a) i For the combined coefficients, I represents the number of elements in the set CH (a),is a real three-dimensional space;
L＝K*P fa k represents the total number of samples of the feature vector set a of the pure sea clutter data, and P fa The false alarm probability is given, and L is the number of samples to be deleted under the false alarm probability;
step b: removing L samples of convex hulls CH (a- { v) q }), where { v } q And is a set of points for L samples.
Step c: the overlapping area V of the convex set and the target area omega is obtained according to the comparison of the relative positions of the convex set and the target area omega in the step b,
V≡CH(a-{v q }))∩Ω；
step d: c, calculating the detection probability P under the current false alarm probability according to the step c d
Where # (. Cndot.) represents the number of elements in the collection,
the feature set extracted from the pure sea clutter signals is a= [ FIE; a FPA; FPS ].
Further, the method also comprises performing resolution analysis on the feature combination in a three-dimensional space, whereinOne convex hull in space is a convex polyhedron consisting of Q triangular faces, expressed as:
wherein SP represents a convex polyhedron composed of triangular faces, Q is the number of triangular faces,three vertices representing triangular faces constituting a convex hull, the symbol triangule (·) representing the convex hull is represented by +.>The triangular surface formed by three vertexes, S is a point set formed by all points; if the characteristic points of the data to be detected fall in the convex hull, judging as H 0 Indicating that no target exists, otherwise, judging as H 1 Indicating the presence of an object.
Furthermore, the combining the extracted three features, realizing target detection by using a rapid convex hull detection algorithm, and performing contrast analysis with single feature detection comprises the following steps: comparing the three-feature joint detection with the single-feature detection to obtain an optimal performance curve, wherein the single-feature detection is respectively as follows:
wherein eta 1 、η 2 And eta 3 The detection statistics determined based on clutter characteristics, respectively.
In a second aspect, the present invention provides a three-feature joint detection device based on FRFT domain, including
Training module: collecting pure clutter data, and performing FRFT (fast Fourier transform) to obtain a fractional order domain data vector; carrying out three feature extraction of transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation on the fractional domain data vector; combining the extracted three features, realizing target detection by using a rapid convex hull detection algorithm, and carrying out comparison analysis with single feature detection;
and a detection module: collecting data to be detected, and performing FRFT (fractional Fourier transform) to obtain a fractional order domain data vector; carrying out three feature extraction of transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation on the fractional domain data vector; obtaining an overlapping region through the relative position comparison of a convex set formed by the target region and the characteristic set of the current pure sea clutter data; according to the overlapping area and the target area, calculating the detection probability under the current false alarm probability; and combining the extracted three features, realizing target detection by using a rapid convex hull detection algorithm, and carrying out comparison analysis with single feature detection.
In a third aspect, the invention provides a FRFT domain-based three-feature joint detection device, which comprises a processor and a storage medium;
the storage medium is used for storing instructions;
the processor is operative to perform the steps of any one of the methods described in the preceding claims in accordance with the instructions.
In a fourth aspect, the invention also provides a computer readable storage medium having stored thereon a computer program, characterized in that the program when executed by a processor implements the steps of any of the methods described in the preceding claims.
Compared with the prior art, the invention has the following beneficial effects:
compared with other detectors, the FRFT domain three-feature combined detector provided by the invention can obtain better detection performance in the actual measurement of different targets in sea clutter data experiments. The rotation characteristic and the order continuity of FRFT transformation are utilized to find the transformation domain with the most concentrated energy, so that the optimality of the extracted features is ensured, and the method has the advantages of small calculated amount and stable performance; the fast convex hull detection algorithm utilizes clutter data characteristics to construct abnormal detection of a detection judgment area, and the size of the convex hull is controlled through the change of the false alarm rate, so that the detection performance is good.
Drawings
FIG. 1 is a flow chart of an FRFT domain three-feature object detector in an embodiment of the invention;
FIG. 2 is a three-dimensional distribution of three sets of features FIE, FPA and FPS extracted in accordance with an embodiment of the present invention;
FIG. 3 is a graph comparing detection performance curves of a single feature detector and a FRFT domain three-feature joint detector according to an embodiment of the invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
The FRFT domain-based three-feature combined target detector comprises the following technical measures: firstly, FRFT transformation is carried out on received echo signals by utilizing the property of fractional Fourier transformation, so that optimality of information entropy of three groups of extracted characteristic transformation domains, peak-to-average ratio of the transformation domains and standard deviation of peak values of the transformation domains is ensured; then, a rapid convex hull detection algorithm is applied to obtain a three-feature joint detector; and finally, comparing the single-feature detection statistic with the respective threshold to obtain a detection performance curve and comparing the detection performance curve with that of the three-feature combined detector.
As shown in fig. 1, the FRFT domain-based three-feature joint detector of the present invention specifically includes the following steps:
example 1: performing FRFT on the radar echo data, and performing feature extraction on the data in the FRFT domain;
FRFT transformation is performed on the radar echo data R (t),
F(u)＝FrFt(R(t),P) (1)
where FrFt is fractional fourier transform, F (u) is fractional domain data vector after FrFt transform of radar echo data R (t), u is scale of FrFt domain, t is time scale, and P is order of FrFt transform. F (u) = { F 1 ,F 2 ,F 3 ,…,F N }, wherein F u Is an element of the fractional-domain data vector F (u), the scale u of the FRFT domain ranges from u=1, 2,3, …, N being the total length of the fractional-domain data vector F (u).
The amplitude values of the fractional order domain data vector F (u) are sorted,
F(u)＝Sort({F 1 ,F 2 ,F 3 ,…,F N })＝{F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) } (2)
wherein Sort is the ascending order of vector magnitude values, { F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) And the data vector is a fractional domain data vector with the amplitude values arranged in an ascending order, wherein (N) represents the subscript of the elements arranged in the ascending order, and the range of (N) is n=1, 2,3, … and N.
Fractional order domain data vector { F after ascending order of amplitude values (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) Dividing into K independent segments with equal length according to magnitude of amplitude value
{F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) }＝{F 1 ,F 2 ,…,F k ,…,F K } (3)
Wherein F is k Representing a K independent segment fractional domain amplitude value short vector, the range of the segment scale K is k=1, 2,3, …, K is the fractional domain amplitude value short vector F k Is a total number of (a) in the number of (a). Set each F k The length of the vector is N k . Fractional order domain amplitude value short vector F k Probability P falling within the kth segment scale k Calculated as
The transform domain information entropy (FIE) of the fractional order domain data vector F (u) is expressed as
Clutter is potentially distinguishable from target amplitude for detection because the spectrum of the former is primarily distributed over wavy and rough surfaces, while the latter is more concentrated near a peak. To quantify this difference, the transform domain peak-to-average ratio (FPA) of the fractional domain data vector F (u) is expressed as
Where max represents maximizing the vector.
Considering that there is a difference in the degree of dispersion between FRFT domain clutter and each feature point of the target, the target will be more discrete than the clutter, and to quantify this difference, the transform domain peak standard deviation (FPS) of the fractional domain data vector F (u) is expressed as
Where std denotes standard deviation of the vectors.
Example 2: realizing target detection based on a rapid convex hull detection algorithm combined by three features;
by analyzing fig. 2, it is found that the characteristic points of the sea clutter are generally relatively concentrated, while the characteristic points of the target echo are relatively scattered, and by utilizing the phenomenon, the detection problem of the target is converted into a judgment of forming a convex hull by sea clutter characteristic data, if the characteristic points of the data to be detected fall in the convex hull, the judgment is that H0 indicates that the target is not present, otherwise, the judgment is that H1 indicates that the target is present.
Assuming a three-dimensional finiteAggregate asWherein eta i Refers to the elements in set S, I represents the number of elements in set S, +.>Is a real three-dimensional space. The minimum set of S is defined by the convex hull as:
wherein CH (·) represents a convex set composed of the set S, η is the element η i Is a combination of (a) i Is a combination coefficient.
At the position ofOne convex hull in space is a convex polyhedron consisting of Q triangular faces, expressed as:
wherein SP represents a convex polyhedron composed of triangular faces, Q is the number of triangular faces,three vertices (clockwise observation) representing triangular faces constituting a convex hull, the symbol triangle (·) representing the convex hull is represented by +.>And S is a point set formed by all points of a triangular surface formed by three vertexes.
Assuming that the feature vector extracted from the received pure sea clutter signal is a= [ FIE; a FPA; FPS (FPS)]The guiding area is omega, and the false alarm probability is P fa The specific process is as follows:
a: extracting a characteristic vector a= [ FIE ] of pure sea clutter data; a FPA; FPS ];
b: calculating a convex hull CH (a) of a characteristic set a of the current pure sea clutter data, wherein L=K×P fa K represents the total number of samples of the feature vector set a of the pure sea clutter data, and P fa The false alarm probability is given, and L is the number of samples to be deleted under the false alarm probability;
c: calculation of convex hulls CH (a- { v) after L samples were removed from set a q }), where { v } q Is a set of points of L samples, by CH (a- { v q And }) compared with the relative position of the guide region omega, the overlapping region of the two is V,
V≡CH(a-{v q }))∩Ω (10)
d: calculating the detection probability P under the current false alarm probability d ，
Where # (. Cndot.) represents the number of elements in the collection.
e: the false alarm probability is transformed to 1 from 0.001 according to the logarithmic scale, and the detection probability under each false alarm probability is calculated to obtain a detection performance curve.
Example 3: and comparing the performance curves of the single feature detection and the three feature combined detection.
The single feature detectors are respectively:
wherein eta 1 、η 2 And eta 3 The detection statistics determined based on clutter characteristics, respectively.
FIG. 3 is a comparison of performance curves based on single feature detection and the performance curves of the proposed FRFT domain three feature joint detection. The effectiveness of the three-feature joint detector based on the FRFT domain is verified.
The measured data is named 19931107_135603_starea17, the experimental use data is taught by s.haykin at the university of Memaser for HH polarization mode measured in an actual marine environment using IPIX radar. The data consists of 14 adjacent distance units, each distance unit has a sampling point number of 131072 (about 131 s), the 9 th distance unit where the target is located is called a unit to be detected, 8, 10, 11 are auxiliary units, and the rest is clutter units.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (9)
1. The FRFT domain-based three-feature joint detection method is characterized by comprising the following steps of:
collecting radar echo data and performing FRFT (fractional Fourier transform) to obtain a fractional order domain data vector;
carrying out three feature extraction of transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation on the fractional domain data vector;
and combining the extracted three features, realizing target detection by using a rapid convex hull detection algorithm, and carrying out comparison analysis with single feature detection.
2. The FRFT domain-based three-feature joint detection method of claim 1, characterized in that the fractional order domain data vector is obtained by the following formula:
F(u)＝FrFt(R(t),P)
wherein, frFt is fractional Fourier transform, F (u) is fractional domain data vector after FRFT of radar echo data R (t), u is scale of FRFT domain, t is time scale, P is the order of FRFT transform; f (F)(u)＝{F 1 ,F 2 ,F 3 ,…,F N The scale of the FRFT domain u ranges from u=1, 2,3, …, N being the total length of the fractional domain data vector F (u).
3. The FRFT domain-based three-feature joint detection method of claim 2, characterized in that the feature extraction includes:
step A: the amplitude values of the fractional order domain data vector F (u) are sorted,
F(u)＝Sort({F 1 ,F 2 ,F 3 ,…,F N })＝{F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) }，
wherein Sort is the ascending order of vector magnitude values, { F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) The data vector is a fractional domain data vector with the amplitude values arranged in an ascending order, and n=1, 2,3, … and N;
and (B) step (B): fractional order domain data vector { F after ascending order of amplitude values (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) Dividing the amplitude value into K independent segments with equal length,
{F (1) ,F (2) ,F (3) ,…,F (n) ,…F (N) }＝{F 1 ,F 2 ,…,F k ,…,F K }
wherein F is k Representing a K independent segment fractional domain amplitude value short vector, the range of the segment scale K is k=1, 2,3, …, K is the fractional domain amplitude value short vector F k Is set to each F k The length of the vector is N k Fractional order domain amplitude value short vector F k Probability P falling within the kth segment scale k The calculation is as follows:
step C: extracting three groups of characteristics including transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation of the fractional domain data vector F (u) based on the step B,
the transform domain information entropy FIE is expressed as:
the transform domain peak-to-average ratio FPA is expressed as:
wherein max represents maximizing the vector;
the transform domain peak standard deviation FPS is expressed as:
where std denotes standard deviation of the vectors.
4. The FRFT domain-based three-feature joint detection method of claim 3, wherein the fast convex hull detection algorithm includes:
step a: calculating a convex hull CH (a) of a characteristic set a of the current pure sea clutter data, wherein the minimum set of the convex hull definition a is as follows:
wherein CH (·) represents a convex set composed of the feature set a, η is the element η i Is a combination of (a) i For the combined coefficients, I represents the number of elements in the set CH (a),is a real three-dimensional space;
L＝K*P fa k represents the total number of samples of the feature vector set a of the pure sea clutter data, and P fa The false alarm probability is given, and L is the number of samples to be deleted under the false alarm probability;
step b: removing L samples of convex hulls CH (a- { v) q }), where { v } q -a set of points of L samples;
step c: the overlapping area V of the convex set and the target area omega is obtained according to the comparison of the relative positions of the convex set and the target area omega in the step b,
V≡CH(a-{v q }))∩Ω；
step d: c, calculating the detection probability P under the current false alarm probability according to the step c d
Wherein # (·) represents the number of elements in the collection;
the feature set extracted from the pure sea clutter signals is a= [ FIE; a FPA; FPS ].
5. The FRFT domain-based three-feature joint detection method of claim 4 further including performing a resolvable analysis of feature joint in three-dimensional space, atOne convex hull in space is a convex polyhedron consisting of Q triangular faces, expressed as:
wherein SP represents a convex polyhedron composed of triangular faces, Q is the number of triangular faces,representing three vertices of triangular faces constituting a convex hull, the symbol triangule (·) representing a convex hull composed of/>The triangular surface formed by three vertexes, S is a point set formed by all points; if the characteristic points of the data to be detected fall in the convex hull, judging as H 0 Indicating that no target exists, otherwise, judging as H 1 Indicating the presence of an object.
6. The FRFT domain-based three-feature joint detection method of claim 1, wherein the combining the extracted three features, implementing target detection by using a fast convex hull detection algorithm, and performing a comparison analysis with single feature detection includes: comparing the three-feature joint detection with the single-feature detection to obtain an optimal performance curve, wherein the single-feature detection is respectively as follows:
wherein eta 1 、η 2 And eta 3 The detection statistics are respectively determined based on clutter characteristics, FIE is transform domain information entropy, FPA is transform domain peak-to-average ratio, FPS is transform domain peak standard deviation, and H 0 Indicating the absence of target, H 1 Indicating the presence of an object.
7. Three-feature combined detection device based on FRFT domain is characterized by comprising
Training module: collecting pure clutter data, and performing FRFT (fast Fourier transform) to obtain a fractional order domain data vector; carrying out three feature extraction of transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation on the fractional domain data vector; combining the extracted three features, realizing target detection by using a rapid convex hull detection algorithm, and carrying out comparison analysis with single feature detection;
and a detection module: collecting data to be detected, and performing FRFT (fractional Fourier transform) to obtain a fractional order domain data vector; carrying out three feature extraction of transform domain information entropy, transform domain peak-to-average ratio and transform domain peak standard deviation on the fractional domain data vector; obtaining an overlapping region through the relative position comparison of a convex set formed by the target region and the characteristic set of the current pure sea clutter data; according to the overlapping area and the target area, calculating the detection probability under the current false alarm probability; and combining the extracted three features, realizing target detection by using a rapid convex hull detection algorithm, and carrying out comparison analysis with single feature detection.
8. The FRFT domain-based three-feature joint detection device is characterized by comprising a processor and a storage medium;
the storage medium is used for storing instructions;
the processor being operative according to the instructions to perform the steps of the method according to any one of claims 1 to 6.
9. Computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010342976.2A CN111505598B (en) | 2020-04-27 | 2020-04-27 | FRFT domain-based three-feature joint detection device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010342976.2A CN111505598B (en) | 2020-04-27 | 2020-04-27 | FRFT domain-based three-feature joint detection device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111505598A CN111505598A (en) | 2020-08-07 |
| CN111505598B true CN111505598B (en) | 2023-07-28 |
Family
ID=71864951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010342976.2A Active CN111505598B (en) | 2020-04-27 | 2020-04-27 | FRFT domain-based three-feature joint detection device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111505598B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113009443B (en) * | 2021-02-22 | 2023-09-12 | 南京邮电大学 | Sea surface target detection method and device based on graph communication density |
| CN113537062B (en) * | 2021-07-16 | 2024-01-09 | 浙江大学 | Anomaly detection method based on FrFT transformation and total variation regularization |
| CN117129958A (en) * | 2023-07-31 | 2023-11-28 | 湖南六九零六信息科技股份有限公司 | Sea surface weak target detection method based on multi-domain features |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102788969B (en) * | 2012-07-04 | 2015-01-28 | 中国人民解放军海军航空工程学院 | Sea surface micromotion target detection and feature extraction method based on short-time fractional Fourier transform |
| CN102914769A (en) * | 2012-10-19 | 2013-02-06 | 南京信息工程大学 | Joint fractal-based method for detecting small target under sea clutter background |
| CN105866758B (en) * | 2016-03-31 | 2018-06-05 | 西安电子科技大学 | Time-frequency bicharacteristic small targets detection in sea clutter method based on block albefaction clutter recognition |
| CN105738888B (en) * | 2016-03-31 | 2018-05-04 | 西安电子科技大学 | Bicharacteristic offshore floating small target detecting method based on ocean clutter cancellation |
-
2020
- 2020-04-27 CN CN202010342976.2A patent/CN111505598B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN111505598A (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111505598B (en) | FRFT domain-based three-feature joint detection device and method | |
| CN105319537B (en) | Marine radar co-channel interference suppression method based on spatial coherence | |
| CN102788969B (en) | Sea surface micromotion target detection and feature extraction method based on short-time fractional Fourier transform | |
| CN106778610B (en) | Intra-pulse modulation identification method based on time-frequency image characteristics | |
| CN109001708B (en) | Radar maneuvering target rapid fine processing method based on grading accumulation detection | |
| CN111190157B (en) | IPIX radar echo data time-frequency analysis method and system | |
| CN109471072B (en) | FPGA-based monopulse radar two-dimensional CFAR detection method and system | |
| CN112180354B (en) | High-frequency radar target joint detection method utilizing time-frequency analysis and constant false alarm technology | |
| CN110133632B (en) | Composite modulation signal identification method based on CWD time-frequency analysis | |
| CN112130142B (en) | Method and system for extracting micro Doppler features of complex moving target | |
| CN111985426A (en) | Sea clutter hybrid denoising algorithm based on variational modal decomposition | |
| CN113503871B (en) | Gravity matching method based on correlation filtering | |
| CN115359258B (en) | Weak and small target detection method and system for component uncertainty measurement | |
| CN110208767B (en) | Radar target rapid detection method based on fitting correlation coefficient | |
| CN113093119A (en) | Time-frequency constant false alarm high-frequency radar target detection method and system | |
| Fu et al. | A noise-resistant superpixel segmentation algorithm for hyperspectral images | |
| CN112083502B (en) | Magnetic different signal detection method based on parallel monostable stochastic resonance | |
| CN112329796B (en) | Infrared imaging cloud detection method and device based on visual saliency | |
| CN110490889B (en) | Radar target extraction method based on edge detection | |
| CN116224280B (en) | Radar target detection method, radar target detection device, radar equipment and storage medium | |
| CN116958189B (en) | Moving point target time-space domain track tracking method based on line segment correlation | |
| CN113269242B (en) | Target detection trace aggregation method based on peak clustering | |
| CN110412536B (en) | Target detection method based on variable-scale median ratio | |
| CN118011346A (en) | Time-frequency domain high-frequency ground wave radar target self-adaptive detection method | |
| Xu et al. | Effective Pulsar Dispersion Measure Estimation Based on Progressive Probabilistic Hough Transform |
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 |