A regional recognition method based on acoustic-magnetic composite intersection feature fusion

Abstract

This invention discloses a region identification method based on acoustic-magnetic composite intersection feature fusion, comprising the following steps: Step 1: Using electromagnetic and underwater acoustic sensors to collect n-dimensional electromagnetic and acoustic composite intersection feature quantities, forming n-dimensional intersection feature data S={S1,S2,…S n Step 2: Perform consistency preprocessing on the n-dimensional intersection feature data S, converting it into target region recognition support X = {X1, X2, ... X}. n Step 3: Based on the optimal confidence level C of each given feature quantity best ={C1,C2,…C n}, calculate the reliability coefficient λ of the feature combination; Step 4: Solve for all possible feature combinations {A} i The reliability M({A}) i Step 5: Identify the support X from the j-th (j=1,2,…m) target region. i,j (i = 1, 2, ..., n) Calculate all feature combinations {A} under this composite feature data. i The composite value F j (A i Step 6: Identify the support X from the j-th (j = 1, 2, ..., m) target region. i,j (i = 1, 2, ..., n) Calculate all feature combinations {A} under this composite feature data. i The credibility R of} j (A i Step 7: Solve for the fusion result W of the j-th (j=1,2,…m) n-dimensional intersection feature data of electromagnetic and acoustic composite detection. j (S); Step 8: Jump to step 5 until j = m, and obtain the final fusion result W(S).

Description

Technical Field

[0001] This invention relates to the field of target region autonomous identification algorithms, and in particular to a region identification method based on the fusion of acoustic-magnetic composite intersection features. Background Technology

[0002] Underwater target area identification technology utilizes electromagnetic, acoustic, and other physical field information to detect and identify objects within a certain distance range underwater. This technology can be applied in engineering fields such as shipwreck salvage, underwater target detection, and area localization. Due to the influence of the conductivity and magnetic permeability of seawater, current target area identification primarily employs electromagnetic or underwater acoustic physical field detection technologies. Traditional single-physical-field detection methods have relatively limited ability and fault tolerance in acquiring target information, often failing to detect or identify targets due to complex underwater environments or sensor malfunctions, making them unsuitable for the growing application demands. Summary of the Invention

[0003] To address the problems mentioned in the background section, the present invention provides the following technical solution:

[0004] A region identification method based on the fusion of acoustic-magnetic composite intersection features includes the following steps:

[0005] Step 1: Use electromagnetic and underwater acoustic sensors to collect data. The characteristic quantities of the electromagnetic and acoustic composite intersection form Dimensional intersection feature data S = {S1, S2, ... S n}, the data length of each feature is ;

[0006] Step 2: Based on the principle of local decision-making for feature quantities, perform... Cross-border feature data Perform consistency preprocessing, converting it into target region recognition support X = {X1, X2, ... X} n};

[0007] Step 3: Based on the optimal confidence level C of each given feature quantity best ={C1, C2, ... C n}, calculate the reliability coefficient of the characteristic combination. ;

[0008] Step 4: Solve for all possible combinations of features Reliability ;

[0009] Step 5: From the first Support for target region recognition Calculate all feature combinations under this composite feature data. composite value wherein j = 1,2, …m, wherein i = 1,2, …n;

[0010] Step six: the first bit target area recognition support Calculate the reliability of all feature combinations under the bit composite feature data ;

[0011] Step seven: according to the reliability and reliability of the feature combination , solve the fusion result of the first bit dimensional intersection feature data of electromagnetic and acoustic composite detection ;

[0012] Step eight: jump to step five until , that is, complete the fusion of all intersection feature data, get the final fusion result .

[0013] Further, in step one, the dimensional electromagnetic and acoustic composite intersection feature quantity includes electromagnetic field intersection feature quantity: electromagnetic field intensity and phase information of the axis accepted by the electromagnetic sensor, and axis electromagnetic field intensity ;

[0014] Also includes acoustic field intersection feature data: target noise sound pressure amplitude and target azimuth interval .

[0015] Further, in step one, the electromagnetic field intersection feature quantity is obtained by the following way: adopting a bar-shaped helical coil type electromagnetic transmitting antenna, whose axis is perpendicular to the longitudinal axis of the detection system, using , , three-axis electromagnetic receiving coil, wherein axis direction is the same as the longitudinal axis direction of the detection system, axis is vertical direction, respectively extracting electromagnetic field intensity and phase information of the axis, and axis electromagnetic field intensity ;

[0016] The acoustic field intersection feature quantity is obtained by the following way: adopting a low-frequency underwater acoustic receiving transducer, collecting the target radiation noise in the intersection process, and extracting the sound pressure amplitude​ , using mid-frequency active sonar, collecting target echo signals, and calculating target azimuth interval .

[0017] Further, in step two, when performing consistency preprocessing operation on dimensional feature data , the formula is as follows:

[0018]

[0019] wherein, is the consistency processing result of the th data point of the dimensional feature, represents the th data point of the dimensional feature, represents the th intersection point.

[0020] Further, in step three, when calculating the feature combination reliability coefficient , the formula is as follows:

[0021] .

[0022] Further, in step four, when solving the reliability , the formula is as follows:

[0023]

[0024] wherein, A i =(X1, X2, … X k ) , is the number of feature types of the feature combination .

[0025] Further, in step five, when calculating the composite value of all feature combinations , the formula is as follows:

[0026] .

[0027] Further, in step six, when calculating the reliability of all feature combinations , the formula is as follows:

[0028]

[0029]

[0030] wherein, and are feature combinations target region identification support degree of any two feature quantities, , is a preset parameter.

[0031] Further, in step seven, the fusion result of the multi-dimensional intersection feature data is solved When, the formula is as follows:

[0032]

[0033]

[0034] wherein, is the fusion confidence value of the feature combination is the serial number of the feature combination corresponding to the maximum fusion confidence value.

[0035] Compared with the prior art, the present application has the beneficial effects that:

[0036] The region identification method based on the fusion of the acoustic-magnetic composite intersection features provided by the present application can effectively fuse the multi-physical field features of the target, improve the detectability and identification reliability of the target, increase the fault tolerance and adaptability of the system, and thus obtain more reliable underwater target region identification effect. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1a and Figure 1b are setting diagrams for collecting intersection feature quantities in the specific embodiment of the region identification method based on the fusion of the acoustic-magnetic composite intersection features provided by the present application;

[0038] Figure 2a and Figure 2b are schematic diagrams of the intersection process in the specific embodiment;

[0039] Figure 3 is a schematic diagram of the change of the target region identification support degree of each dimension feature with the intersection distance in the specific embodiment;

[0040] Figure 4 is a schematic diagram of the change of the final fusion result of the visible intersection feature data with the intersection distance in the specific embodiment. DETAILED DESCRIPTION

[0041] In order to make the technical means, creative features, purposes and effects realized by the present application easy to understand, the following further describes how the present application is implemented in combination with the drawings and specific embodiments.

[0042] ​​​In one specific embodiment, the present invention provides a region identification method based on acoustic-magnetic composite intersection feature fusion, comprising the following steps:

[0043] Step 1: Use electromagnetic and underwater acoustic sensors to collect data. The characteristic quantities of the electromagnetic and acoustic composite intersection form Dimensional intersection feature data S = {S1, S2, ... S n}, the data length of each feature is .

[0044] In this embodiment, the... Electromagnetic and acoustic composite intersection characteristics, including electromagnetic field intersection characteristics: received by electromagnetic sensors. electromagnetic field strength of the shaft and phase information ,as well as electromagnetic field strength of the shaft It also includes sound field intersection characteristic data: the sound pressure amplitude of the target noise. and target azimuth range .

[0045] Reference Figure 1a As shown, Figure 1a This is a side view. The detection system can be installed on an underwater vehicle. The electromagnetic field intersection characteristics are obtained as follows: a rod-shaped helical coil electromagnetic transmitting antenna is used, with its axis perpendicular to the longitudinal axis of the detection system. , , A three-axis electromagnetic receiving coil, in which The axial direction is the same as the longitudinal axis of the detection system. With the axis in the vertical direction, extract respectively electromagnetic field strength of the shaft and phase information ,as well as electromagnetic field strength of the shaft .

[0046] Reference Figure 1b As shown, Figure 1b For the top view, the acoustic field intersection characteristics were obtained as follows: a low-frequency underwater acoustic receiver transducer was used to collect the target radiated noise during the intersection process, and the sound pressure amplitude was extracted. Using mid-frequency active sonar, target echo signals are acquired, and the target azimuth range is calculated and generated. .

[0047] Reference Figure 2a and Figure 2b As shown in this embodiment, the intersection distance between the detection system and a typical target is... The 5m intersection distance refers to the shortest distance between the detection system and the typical target during the intersection process. The intersection path considers the vertical projection point of the typical target on the intersection path as the 0 point, and the approaching process is negative and the moving away process is positive. The intersection angle is 60°.

[0048] Step two: according to the characteristic quantity local decision principle, the dimension intersection characteristic data is preprocessed for consistency and converted into target area recognition support X = {X1, X2, … X n}.

[0049] Specifically, when the dimension intersection characteristic data is preprocessed for consistency, the formula is as follows:

[0050]

[0051] wherein, is the consistency processing result of the i-th data point of the j-th feature, represents the i-th data point of the j-th feature, represents the i-th data point of the j-th feature, represents the i-th data point of the j-th feature, represents the i-th data point of the j-th feature, represents the i-th data point of the j-th feature, represents the i-th data point of the j-th feature, represents the i-th data point of the j-th feature. Referring to FIG. 1, it is a schematic diagram of the target area recognition support of each feature in the embodiment with the intersection distance. Figure 3

[0052] Step three: according to the given optimal confidence C best = {C1, C2, … C n} of each feature, the feature combination reliability coefficient is calculated. In the embodiment, C = 0.87, = 0.84, = 0.75, = 0.82, = 0.92.

[0053] Specifically, when the feature combination reliability coefficient is calculated, the formula is as follows:

[0054] .

[0055] Step four: the reliability of all possible feature combinations is solved.

[0056] Specifically, the reliability ​​When, the formula is as follows:

[0057]

[0058] Where, assume A i =(X1, X2, ... X k ) , This is a combination of features. The number of feature quantities.

[0059] Step 5: From the first Support for target region recognition Calculate all feature combinations under this composite feature data. composite value , where j=1,2,…m, and i=1,2,…n.

[0060] Specifically, calculate all feature combinations When dealing with composite values, the formula is as follows:

[0061] .

[0062] Step Six: From the first Support for target region recognition Calculate all feature combinations under this composite feature data. Credibility .

[0063] Specifically, calculate all feature combinations When determining the credibility, the formula is as follows:

[0064]

[0065]

[0066] in, and For feature combination Support for target region recognition for any two feature quantities. , As preset parameters, in this embodiment, 0.4 is acceptable. 0.7 is acceptable.

[0067] Step 7: Combining features Credibility and reliability Solving the electromagnetic-acoustic combined detection method Bit Fusion results of intersecting feature data .

[0068] Specifically, solving The fusion result of the intersection feature data At this time, the formula is as follows:

[0069]

[0070]

[0071] Wherein, is the fusion confidence value of the feature combination is the serial number of the feature combination corresponding to the maximum fusion confidence value

[0072] Step eight: jump to step five until , that is, all the fusion of intersection feature data is completed, and the final fusion result . Figure 4 is a schematic diagram of the change of the final fusion result of the visible intersection feature data in this embodiment with the intersection distance.

[0073] To sum up, the regional identification method based on the acoustic-magnetic composite intersection feature fusion provided by the present application can effectively fuse the multi-physical field features of the target based on the composite detection of electromagnetic and underwater acoustic, improve the detectability and identification reliability of the target, increase the fault tolerance and adaptability of the system, and thus obtain more reliable underwater target regional identification effect.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the purpose and scope of the present application, and all should be covered in the scope of the claims of the present application.​​

Claims

1. A method for regional recognition based on acoustic-magnetic composite intersection feature fusion, characterized in that, comprising the steps of: Step one: using electromagnetic, underwater acoustic sensor to collect electromagnetic, acoustic composite intersection characteristic quantity, forming intersection characteristic data S = {S1, S2, … S n}, the data length of each characteristic quantity is ; Step two: According to the local decision principle of characteristic quantity, the consistency preprocessing is carried out, and the target area recognition support X = {X1, X2, … Xn} is converted. The characteristic data of the World Exposition The consistency preprocessing is carried out, and the target area recognition support X = {X1, X2, … Xn} is converted. n}; Step three: Calculate the feature combination reliability coefficient best = {C1, C2, …C n} according to the given optimal confidence C of each feature quantity Step four: solving for all possible feature combinations reliability of the ; Step five: identifying the target region from the first bit target region identification support Calculate the composite value of all feature combinations under the bit composite feature data of the target region , where j = 1, 2, … m, where i = 1, 2, … n; Step six: identify the support degree of the target region from the first bit target region identification support Calculate the credibility of all feature combinations under the bit composite feature data ;​ Step seven: according to the feature combination of credibility and reliability , solve the fusion result of the first bit dimension intersection feature data of electromagnetic, acoustic composite detection ; Step eight: jump to step five until i.e. all the intersection feature data is fused to get the final fusion result ; In step one, the electromagnetic, acoustic composite intersection characteristic quantity, including electromagnetic field intersection characteristic quantity: electromagnetic field intensity accepted by electromagnetic sensor of the shaft and phase information , and electromagnetic field intensity of the shaft Also included are sound field intersection feature data: sound pressure amplitude of target noise and target azimuth interval ; In step two, the following is applied to Vicinity feature data When performing the consistency pre-processing operation, the following formula is applied: ； wherein, is the first dimensional characteristic of the first data point, is the first dimensional characteristic of the first data point, is the first intersection point; In step three, the feature combination reliability coefficient is calculated When, the formula is as follows: ； In step four, the reliability is solved When, the formula is as follows: ； wherein A i = (X1, X2,... X k ) , is the number of feature types of the feature combination . In step five, the composite value of all feature combinations is calculated using the following formula: In step five, the composite value of all feature combinations is calculated using the following formula: ； In step six, the confidence of all feature combinations is calculated using the following formula: ； ； wherein, and is a target region identification support degree of any 2 feature quantities of the feature combination , , is a preset parameter; In step seven, solving The fusion result of the characteristic data of the meeting When, the formula is as follows: ； ； wherein, is a combination of features is a fusion confidence value, is a sequence number of the combination of features corresponding to the maximum fusion confidence value.

2. The method of claim 1, wherein the method is based on a fusion of acoustic-magnetic composite intersection features. In step one, the electromagnetic field intersection characteristic quantities are obtained as follows: a rod-shaped spiral coil electromagnetic transmitting antenna is used, with its axis perpendicular to the longitudinal axis of the detection system. , , A three-axis electromagnetic receiving coil, in which The axial direction is the same as the longitudinal direction of the detection system. With the axis in the vertical direction, extract respectively electromagnetic field strength of the shaft and phase information ,as well as electromagnetic field strength of the shaft ; The sound field intersection characteristic quantity is obtained by using a low frequency underwater sound receiving transducer to collect the target radiation noise during the intersection, extracting the sound pressure amplitude , using a medium frequency active sonar to collect the target echo signal, and calculating to generate the target azimuth interval .

Images