A comprehensive identification method and system for aerial targets based on multi-attribute decision making

Abstract

The embodiment of the present application provides a method and system for comprehensive identification of aerial targets with multi-attribute decision-making, which relates to the field of target identification technology. The method includes: obtaining multiple identification results of aerial targets; generating intuitive fuzzy matrix data according to multiple identification results; obtaining identification weight set data corresponding to multiple identification means according to the intuitive fuzzy matrix data; obtaining basic probability assignment data according to evidence theory and intuitive fuzzy matrix data; correcting the basic probability assignment data according to the identification weight set data to obtain comprehensive identification result matrix data corresponding to multiple detection cycles; obtaining period identification weight set data corresponding to multiple detection cycles according to the conflict degree data of multiple detection cycles; performing multi-period information fusion on the comprehensive identification result matrix data to obtain multi-period comprehensive identification results. The method can achieve the technical effect of improving the reliability and effectiveness of aerial target identification.

Description

Technical Field

[0001] The present application relates to the field of target recognition technology, and in particular to a method, system, electronic device and computer-readable storage medium for comprehensive recognition of aerial targets with multi-attribute decision-making. Background Art

[0002] At present, as an important combat operation in air defense operations, the identification of friend or foe of aerial targets can provide important information support for commanders to make air defense decisions. Accurate and efficient identification of friend or foe of aerial targets will help promote the smooth implementation of air defense operations, reduce the risk of accidental attacks and injuries, and improve the effectiveness of joint operations. In order to ensure the reliability of identification of friend or foe of aerial targets, it is necessary to use a variety of technical means and procedural means to identify the friend or foe attributes of aerial targets, and then integrate the information obtained by these means to obtain reliable comprehensive identification results.

[0003] In the existing technology, in order to achieve the smooth implementation of aerial target friend-or-foe identification, most studies mainly use DS evidence theory, intuitionistic fuzzy sets, cloud models, three-way decision-making and other methods, and combine these mathematical tools according to actual conditions to complete the task of aerial target friend-or-foe identification; however, the above methods are greatly affected by the uncertainty of various factors in aerial target friend-or-foe identification, and their effectiveness and reliability are not high. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a method, system, electronic device and computer-readable storage medium for comprehensive identification of aerial targets with multi-attribute decision-making, which can achieve the technical effect of improving the reliability and effectiveness of aerial target identification.

[0005] The present application embodiment provides a method for comprehensive identification of aerial targets based on multi-attribute decision making, including:

[0006] Acquire multiple recognition results of the aerial target, wherein the multiple recognition results are obtained by multiple recognition means in multiple detection cycles;

[0007] Generate intuitive fuzzy matrix data according to the multiple recognition results;

[0008] Obtaining recognition weight set data corresponding to the plurality of recognition means according to the intuitive fuzzy matrix data;

[0009] Obtain basic probability assignment data corresponding to the multiple detection cycles and the multiple identification means respectively according to the evidence theory and the intuitive fuzzy matrix data;

[0010] Modifying the basic probability assignment data according to the recognition weight set data to obtain comprehensive recognition result matrix data corresponding to the multiple detection cycles;

[0011] Obtaining cycle identification weight set data corresponding to the plurality of detection cycles according to the conflict degree data of the plurality of detection cycles;

[0012] The multi-cycle information fusion is performed on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result.

[0013] In the above implementation process, the application of multi-attribute decision-making is combined with the recognition results, recognition means and detection cycles, and intuitive fuzzy sets and evidence theory are used to model multi-attribute decision-making. Then, the basic probability assignment data is corrected using the recognition weight set data. Then, the conflict degree data is used to determine the detection cycle weight and multi-cycle fusion is performed to obtain a reasonable multi-cycle comprehensive recognition result of the aerial target. Therefore, the comprehensive recognition method of aerial targets based on multi-attribute decision-making can achieve the technical effect of improving the reliability and effectiveness of aerial target recognition.

[0014] Furthermore, the step of obtaining recognition weight set data corresponding to the plurality of recognition means according to the intuitive fuzzy matrix data comprises:

[0015] Obtaining cross entropy data corresponding to the plurality of recognition means according to the intuitionistic fuzzy matrix data;

[0016] The recognition weight set data is obtained according to the cross entropy data.

[0017] Furthermore, the step of obtaining basic probability assignment data corresponding to the multiple detection cycles and the multiple identification means respectively according to the evidence theory and the intuitive fuzzy matrix data includes:

[0018] Generate the basic probability assignment formula corresponding to the Mass function based on evidence theory;

[0019] The basic probability assignment data is obtained according to the basic probability assignment formula and the intuitive fuzzy matrix data.

[0020] Furthermore, the step of modifying the basic probability assignment data according to the recognition weight set data to obtain comprehensive recognition result matrix data corresponding to the multiple detection cycles includes:

[0021] Modify the basic probability assignment data according to the recognition weight set data and the evidence discount method to obtain recognition modified basic probability assignment matrix data;

[0022] The recognition-corrected basic probability assignment matrix data is fused to obtain the comprehensive recognition result matrix data.

[0023] Furthermore, before the step of obtaining cycle identification weight set data corresponding to the plurality of detection cycles according to the conflict degree data of the plurality of detection cycles, the step further includes:

[0024] Obtaining the Jousselme evidence distance of the Mass function corresponding to the plurality of detection cycles;

[0025] The conflict degree data of the plurality of detection cycles are generated according to the Jousselme evidence distance.

[0026] Furthermore, the step of performing multi-cycle information fusion on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result includes:

[0027] Correcting the comprehensive identification result matrix data according to the period identification weight set data to obtain period correction basic probability assignment matrix data;

[0028] The multi-cycle comprehensive recognition result is generated according to the cycle-corrected basic probability assignment matrix data and the DS combination rule.

[0029] In a second aspect, an embodiment of the present application provides a comprehensive aerial target identification system for multi-attribute decision-making, including:

[0030] An acquisition module, used to acquire multiple recognition results of aerial targets, wherein the multiple recognition results are obtained by multiple recognition means within multiple detection cycles;

[0031] An intuitionistic fuzzy matrix module, used to generate intuitionistic fuzzy matrix data according to the multiple recognition results;

[0032] A recognition weight module, used for obtaining recognition weight set data corresponding to the plurality of recognition means according to the intuitive fuzzy matrix data;

[0033] A basic probability assignment module, used for obtaining basic probability assignment data corresponding to the multiple detection cycles and the multiple identification means respectively according to the evidence theory and the intuitive fuzzy matrix data;

[0034] An identification correction module, used to correct the basic probability assignment data according to the identification weight set data to obtain comprehensive identification result matrix data corresponding to the multiple detection cycles;

[0035] A cycle weight module, used to obtain cycle identification weight set data corresponding to the multiple detection cycles according to the conflict degree data of the multiple detection cycles;

[0036] The multi-cycle fusion module is used to perform multi-cycle information fusion on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result.

[0037] Furthermore, the recognition weight module includes:

[0038] A cross entropy unit, used to obtain cross entropy data corresponding to the multiple recognition means according to the intuitionistic fuzzy matrix data;

[0039] An identification weight unit is used to obtain the identification weight set data according to the cross entropy data.

[0040] Furthermore, the basic probability assignment module includes:

[0041] Mass function unit, used to generate the basic probability assignment formula corresponding to the Mass function according to the evidence theory;

[0042] A basic probability assignment unit is used to obtain the basic probability assignment data according to the basic probability assignment formula and the intuitive fuzzy matrix data.

[0043] Furthermore, the identification and correction module includes:

[0044] An identification correction unit, used to correct the basic probability assignment data according to the identification weight set data and the evidence discount method to obtain identification correction basic probability assignment matrix data;

[0045] The recognition fusion unit is used to fuse the recognition-corrected basic probability assignment matrix data to obtain the comprehensive recognition result matrix data.

[0046] Furthermore, the system further comprises:

[0047] An evidence distance module, used to obtain the Jousselme evidence distance of the Mass function corresponding to the multiple detection cycles;

[0048] A conflict degree module is used to generate conflict degree data of the multiple detection cycles according to the Jousselme evidence distance.

[0049] Furthermore, the multi-cycle fusion module includes:

[0050] A period correction unit, used to correct the comprehensive recognition result matrix data according to the period recognition weight set data to obtain period correction basic probability assignment matrix data;

[0051] The multi-cycle fusion unit is used to generate the multi-cycle comprehensive recognition result according to the cycle-corrected basic probability assignment matrix data and the DS combination rule.

[0052] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in any one of the first aspects when executing the computer program.

[0053] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which instructions are stored. When the instructions are executed on a computer, the computer executes the method as described in any one of the first aspects.

[0054] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when running on a computer, enables the computer to execute the method as described in any one of the first aspects.

[0055] Other features and advantages disclosed in the present application will be described in the following description, or some features and advantages can be inferred or determined without doubt from the description, or can be learned by implementing the above-mentioned technology disclosed in the present application.

[0056] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A schematic diagram of the structure of a method and means for comprehensive identification of aerial targets provided in an embodiment of the present application;

[0059] Figure 2 A flowchart of a method for comprehensive identification of aerial targets using multi-attribute decision making provided in an embodiment of the present application;

[0060] Figure 3 A flowchart of another method for comprehensive identification of aerial targets using multi-attribute decision making provided in an embodiment of the present application;

[0061] Figure 4 A structural block diagram of the comprehensive identification system for aerial targets with multi-attribute decision-making provided in an embodiment of the present application;

[0062] Figure 5 A structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0064] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0065] The embodiments of the present application provide a method, system, electronic device and computer-readable storage medium for comprehensive identification of aerial targets based on multi-attribute decision-making, which can be applied to the process of friend-or-foe identification of aerial targets; the method for comprehensive identification of aerial targets based on multi-attribute decision-making combines the application of multi-attribute decision-making based on recognition results, recognition means and detection cycles, uses intuitive fuzzy sets and evidence theory to model multi-attribute decision-making, then uses recognition weight set data to correct basic probability assignment data, and then uses conflict degree data to determine detection cycle weights and perform multi-cycle fusion, thereby obtaining a reasonable multi-cycle comprehensive identification result of aerial targets; thus, the method for comprehensive identification of aerial targets based on multi-attribute decision-making can achieve the technical effect of improving the reliability and effectiveness of aerial target identification.

[0066] See also Figure 1 , Figure 1 A schematic diagram of the structure of the method and means for comprehensive identification of aerial targets provided in an embodiment of the present application.

[0067] For example, based on the idea of ​​air situation identification under the framework of "airspace control", the concept of comprehensive identification of air targets is given, that is, comprehensive identification of air targets refers to the comprehensive use of various technical means such as radar and procedural means such as airspace coordination measures represented by minimum risk routes, low-altitude crossing corridors, and air channels to identify the enemy and friendly attributes of air targets. Figure 1It can be seen that when conducting comprehensive identification of air targets, a variety of appropriate active and program identification means can be selected according to the actual combat situation to avoid misjudgment caused by suppression of a single means, and ensure the accuracy of the enemy identification of air targets. It should be noted that the above means can be used for both enemy identification and "airspace control", which is a concrete manifestation of the unified action under the framework of "airspace control". In addition to the above identification means, attribute tags of various types of enemy identification of air targets can also be introduced to identify the threat level of air targets, and the attribute tags of air targets can be dynamically updated according to the identification situation (such as "our side", "may be our side", "may be the enemy", "enemy" and other attribute tags), realizing the organic combination of enemy attribute identification and threat assessment of air targets, which is conducive to the air defense department to take corresponding air defense countermeasures in time, thereby reducing the risk of accidental attack and accidental injury and ensuring air defense safety.

[0068] In order to better model the multi-attribute decision-making for the comprehensive identification of aerial targets, some basic concepts of intuitionistic fuzzy sets and DS evidence theory in the algorithm tools are first introduced to provide theoretical support for the subsequent comprehensive identification of aerial targets based on intuitionistic fuzzy sets and evidence theory.

[0069] For example, intuitionistic fuzzy sets are based on "fuzzy sets" and provide a fuzzy description of information as "neither this nor that", that is, expressing the degree of "support, neutrality, or opposition" of the information, which can better describe the fuzzy relationship between information.

[0070] For example, let X be a given domain, for If there is a mapping: μ A (x):X→[0,1] and γ A (x):X→[0,1], such that 0≤μ A (x)+γ A (x)≤1 always holds true, then an intuitionistic fuzzy set A on the domain X is called:

[0071] A={ <x,μ A (x),γ A (x)>|x∈U} (1);

[0072] Among them, μ A (x) and γ A (x) represent the membership and non-membership of element x to the intuitionistic fuzzy set A respectively.

[0073] For example, let X be a given domain, and the hesitation degree of each intuitionistic fuzzy set A on the domain X is called:

[0074] π A (x) = 1-μ A (x)-γ A (x) (2);

[0075] Among them, the hesitation degree π A (x) represents the hesitation degree of element x about intuitionistic fuzzy set A.

[0076] For example, let X be a given domain, and the element x in the domain X is said to belong to the tuple consisting of the membership function and non-membership of the intuitionistic fuzzy set A <μ A (x),γ A (x)> is an intuitionistic fuzzy number. From all the intuitionistic fuzzy numbers z ij = <μ ij ,γ ij >(i=1,2,…,m;j=1,2,…,n)composed of the matrix Z=(z ij ) m×n is the intuitionistic fuzzy matrix.

[0077] For example, according to the definition of Mass function in DS evidence theory, in order to facilitate subsequent information fusion, the Mass function on the identification framework Θ can be represented by intuitionistic fuzzy numbers, as shown in the following formula:

[0078]

[0079] For example, the DS evidence theory is a theory proposed by Dempster and Shafter for processing uncertain information in multi-source information fusion. Its good mathematical properties and combination rules effectively improve the processing capability of uncertain information.

[0080] For example, let the pairwise mutually exclusive finite complete set θ = {θ1, θ2, …, θ n} is the identification frame. For any subset A and power set 2 of the identification frame Θ Θ , if there is a mapping: m:2 Θ →[0,1] satisfies the following conditions:

[0081]

[0082] Where m is called the Mass function on the identification frame Θ, and m(A) is called the basic probability assignment of A. If m(A)>0, A is called the focal element.

[0083] For example, let m1 and m2 be independent Mass functions under the identification framework Θ, and their focal elements are B1, B2, ..., B i and C1,C2,…,C j , then the DS combination rules are as follows:

[0084]

[0085]

[0086] Among them, m represents the new Mass function after fusion; k is the conflict coefficient (conflict degree), which indicates the degree of conflict between evidences.

[0087] See also Figure 2 , Figure 2 A flowchart of a method for comprehensive identification of aerial targets based on multiple attribute decision making is provided in an embodiment of the present application. The method for comprehensive identification of aerial targets based on multiple attribute decision making includes the following steps:

[0088] S100: Acquire multiple recognition results of the aerial target, where the multiple recognition results are obtained through multiple recognition means within multiple detection cycles.

[0089] Exemplarily, the identification means may be radar, data link and airspace coordination measures, etc.; the identification results may be "our side", "may be our side", "may be the enemy", "enemy", etc.; each identification result is directly determined by the commander according to operational needs after corresponding demonstration and research. Once determined, it will not change with changes in time and methods in subsequent modeling applications.

[0090] S200: Generate intuitive fuzzy matrix data according to multiple recognition results.

[0091] For example, let the set X consisting of n recognition results of comprehensive recognition of aerial targets = {x1, x2, ..., x n} is the “identity set” in the model, which can correspond to the “solution set” in multi-attribute decision making; the set O composed of m identification means used in comprehensive identification is {o1, o2, …, o m} is the “means set”, which can correspond to the “attribute set” in multi-attribute decision making; the set T consisting of k cycles of detection of aerial targets k (k=1,2,…,K) is a “period set”, which can correspond to the “expert set” in multi-attribute decision making. Therefore, the detection period T is determined by using the multi-attribute decision making language to describe it. k For the recognition result x j In identifying means o i The evaluation value under this condition. Assume that the above evaluation value can be obtained by intuitionistic fuzzy number So we can get the corresponding intuitionistic fuzzy matrix That is, intuitive fuzzy matrix data is generated.

[0092] S300: Obtaining recognition weight set data corresponding to multiple recognition methods according to intuitive fuzzy matrix data.

[0093] For example, the effectiveness of the identification results of the aerial target obtained by different identification means is different, so the identification means can be assigned a weight value; for example, the detection period can be obtained as T by identifying the weight set data. k When the identification means o i The weight of .

[0094] S400: Obtain basic probability assignment data corresponding to multiple detection cycles and multiple identification methods according to evidence theory and intuitive fuzzy matrix data.

[0095] For example, the detection period can be determined as T according to formula (3): k When the recognition result x j And the corresponding identification means o i Assign basic probability to obtain basic probability assignment data.

[0096] S500: Modify the basic probability assignment data according to the recognition weight set data to obtain comprehensive recognition result matrix data corresponding to multiple detection cycles.

[0097] Exemplarily, the above-mentioned basic probability assignment data is corrected because when fusing the recognition results, it is necessary to ensure that the importance of the recognition results is the same. Since these recognition results are derived from data detected by different recognition means, the importance of these recognition results is determined by the capabilities of the recognition means themselves. Different recognition means have different abilities to obtain information, which results in different importance of the recognition results. This destroys the prerequisite for data fusion and will have a certain impact on the fusion results. Therefore, it is necessary to correct the basic probability assignment data to reduce these impacts.

[0098] S600: Obtaining cycle identification weight set data corresponding to the plurality of detection cycles according to the conflict degree data of the plurality of detection cycles.

[0099] Exemplarily, in order to improve the accuracy of multi-cycle fusion results, it is necessary to process the uncertain information in the multi-cycle fusion process to obtain the cycle identification weight set data corresponding to multiple detection cycles; optionally, the uncertainty information in the multi-cycle fusion is measured by combining the conflict coefficient k and the Jousselme evidence distance, and the evidence discount method is used to correct the Mass function of each cycle.

[0100] S700: Perform multi-cycle information fusion on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result.

[0101] For example, in order to achieve effective aggregation of detection cycle information, it is necessary to first correct the recognition results of each detection cycle, and then perform multi-cycle information fusion. This is because the different information acquisition capabilities of each recognition method make the recognition results of each detection cycle different, resulting in certain conflicts in the constructed evidence. Therefore, these recognition results should be corrected to ensure the smooth implementation of multi-cycle information fusion. Similarly, the weight of the detection cycle is used as an indicator to measure the importance of these recognition results and used as a discount coefficient to correct the comprehensive recognition result matrix data.

[0102] In some embodiments, the comprehensive identification method for aerial targets based on multi-attribute decision-making combines the application of multi-attribute decision-making by starting from the recognition results, recognition means and detection cycle, uses intuitive fuzzy sets and evidence theory to model multi-attribute decision-making, then uses the recognition weight set data to correct the basic probability assignment data, and then uses the conflict degree data to determine the detection cycle weight and perform multi-cycle fusion, thereby obtaining a reasonable multi-cycle comprehensive identification result for aerial targets; thus, the comprehensive identification method for aerial targets based on multi-attribute decision-making can achieve the technical effect of improving the reliability and effectiveness of aerial target identification.

[0103] Qingjian Figure 3 , Figure 3 A flowchart of another method for comprehensive identification of aerial targets based on multi-attribute decision-making provided in an embodiment of the present application.

[0104] Exemplarily, S300: the step of obtaining recognition weight set data corresponding to a plurality of recognition means according to the intuitive fuzzy matrix data includes:

[0105] S310: Obtain cross entropy data corresponding to multiple recognition methods according to the intuitive fuzzy matrix data;

[0106] S320: Obtain and identify weight set data based on cross entropy data.

[0107] For example, in order to improve the accuracy of measuring the uncertainty information contained in the intuitive fuzzy number, a divergence-based AIFS cross entropy can be used to measure the uncertainty information in the intuitive fuzzy number, and it can be used as a method for determining the weight of the "attribute set" in the multi-attribute decision-making problem. Here, it is to solve the problem of determining the weight of the identification means in the comprehensive identification of aerial targets.

[0108] For example, the uncertainty in the intuitionistic fuzzy set is mainly composed of intuitionistic fuzziness and fuzziness. Among them, intuitionistic fuzziness is mainly composed of the hesitation degree π in the intuitionistic fuzzy number A (x), the fuzziness is mainly determined by the membership degree μ in the intuitive fuzzy number A (x) and non-membership γ A The degree of difference of (x) ΔA (x i ), that is, Δ A (x i )=|μ A (x i )-γ A (x i )|.

[0109] Exemplarily, the axiomatic definition of divergence-based AIFS cross entropy is as follows:

[0110] Let X be a given domain, A be an intuitionistic fuzzy set on the domain X, then the intuitionistic fuzzy entropy of the intuitionistic fuzzy set A is a real-valued function of the hesitation degree, that is, E I (A) = g(π A ):AIFS→[0,1], and satisfies the following properties:

[0111] 1) For x∈X, if and only if π A When (x) = 0, E I (A)=0.

[0112] 2) For x∈X, if and only if π A When (x) = 1, E I (A)=1.

[0113] 3) E I (A) With π A (x i ) increases with the increase of, where i = 1, 2,…, n.

[0114] E I (A C )=E I (A).

[0115] For example, let X be a given domain, A be an intuitionistic fuzzy set on the domain X, then the fuzzy entropy of the intuitionistic fuzzy set A is a function of the membership μ A (x) and non-membership degree γ A (x), that is, E F (A) = h(Δ A ):AIFS→[0,1], and satisfies the following properties:

[0116] 1) For x∈X, if and only if Δ A When (x) = 1, E F (A) = 0, that is, A is an exact set.

[0117] 2) For x∈X, if and only if Δ A When (x) = 0, E F (A)=1.

[0118] 3) E F (A) With Δ A (x) increases, where i = 1, 2,…, n.

[0119] 4) E F (A C )=E F (A).

[0120] For example, the divergence-based AIFS cross entropy can be expressed as:

[0121]

[0122] The corresponding intuitive fuzzy entropy and fuzzy entropy are:

[0123]

[0124]

[0125] Therefore, formula (7) can also be expressed as:

[0126]

[0127] That is, the AIFS cross entropy E based on divergence S (A) is actually the intuitionistic fuzzy entropy E I (A) and fuzzy entropy E F (A) The arithmetic mean.

[0128] Then, we can get the detection period as T k When the identification means o i The weight is:

[0129]

[0130] The identification means o i The corresponding weight set ω (identification weight set data) is:

[0131]

[0132] Exemplarily, S400: the step of obtaining basic probability assignment data corresponding to a plurality of detection cycles and a plurality of identification means respectively according to evidence theory and intuitionistic fuzzy matrix data includes:

[0133] S410: Generate a basic probability assignment formula corresponding to the Mass function according to the evidence theory;

[0134] S420: Obtain basic probability assignment data according to the basic probability assignment formula and the intuitive fuzzy matrix data.

[0135] For example, to aggregate the identification means (attribute) information, the detection period should first be determined as T according to formula (3): k When the recognition result x j And the corresponding identification means o i The basic probability assignment formula is as follows:

[0136]

[0137] Exemplarily, S500: the step of modifying the basic probability assignment data according to the recognition weight set data to obtain comprehensive recognition result matrix data corresponding to a plurality of detection cycles includes:

[0138] S510: Correcting the basic probability assignment data according to the recognition weight set data and the evidence discount method to obtain recognition corrected basic probability assignment matrix data;

[0139] S520: Fusing the recognition-corrected basic probability assignment matrix data to obtain comprehensive recognition result matrix data.

[0140] Exemplarily, the weight of the identification means is used as an indicator to measure the importance of the basic probability assignment data and is used as a discount coefficient to correct the basic probability assignment data.

[0141] According to the identification means weight obtained by formula (12), the basic probability assignment obtained by formula (13) is corrected using the evidence discount method. The formula is as follows:

[0142]

[0143] The discount factor

[0144] The detection period is T k When the recognition result x j And the corresponding identification means o i The revised basic probability assignment matrix is ​​as follows:

[0145]

[0146] Using the DS combination rule, the modified basic probability assignments are fused to obtain the comprehensive recognition result matrix (comprehensive recognition result matrix data) of each detection cycle as follows:

[0147]

[0148] Exemplarily, before the step of S600: obtaining cycle identification weight set data corresponding to the plurality of detection cycles according to the conflict degree data of the plurality of detection cycles, the method further includes:

[0149] S601: Obtaining the Jousselme evidence distance of the Mass function corresponding to multiple detection cycles;

[0150] S602: Generate conflict degree data of multiple detection cycles according to the Jousselme evidence distance.

[0151] Exemplarily, in order to improve the accuracy of the multi-cycle fusion result, it is necessary to process the uncertain information in the multi-cycle fusion process, that is, to correct the Mass function obtained by formula (16). In the embodiment of the present application, the uncertainty information in the multi-cycle fusion is measured by combining the conflict coefficient k and the Jousselme evidence distance, and the evidence discount method is used to correct the Mass function of each cycle.

[0152] The Jousselme evidence distance of the Mass function of each detection period is calculated using the following formula:

[0153]

[0154] Among them, m p and m q is the corresponding Mass function m p and m q The vector form of Is a 2 n ×2 n The similarity matrix of elements can be expressed as:

[0155]

[0156] Assume the detection period T k With T g The obtained Mass function m k and m g The degree of conflict between them is cf kg (Conflict degree data, conflict coefficient), conflict degree cf kg It can be expressed as:

[0157]

[0158] Among them, k kg Indicates m k With m g Conflict of evidence between kg Indicates m k With m g The Jousselme evidence distance between them; θ represents any hypothesis on the identification framework Θ; argmax θ∈Θ (BetP m (θ)) represents the maximum supported hypothesis on the identification framework Θ.

[0159] Therefore, the detection period T k The corresponding evidence is detected period T g The degree of support provided by the evidence kg Can be expressed as sup kg =1-cf kg , then the detection period T k Weight w k It can be expressed as:

[0160]

[0161] The detection period T k The corresponding weight set W is:

[0162] W=(w1,w2,…,w k ) T (twenty one).

[0163] Exemplarily, S700: the step of performing multi-period information fusion on the comprehensive recognition result matrix data according to the period recognition weight set data to obtain the multi-period comprehensive recognition result includes:

[0164] S710: Correcting the comprehensive identification result matrix data according to the period identification weight set data to obtain period correction basic probability assignment matrix data;

[0165] S720: Generate a multi-period comprehensive recognition result according to the period-corrected basic probability assignment matrix data and the DS combination rule.

[0166] Exemplarily, the embodiment of the present application uses the weight of the detection cycle as an indicator to measure the importance of these comprehensive recognition result matrix data and uses it as a discount coefficient to correct the comprehensive recognition result matrix data.

[0167] Then, the detection period T obtained by equation (16) is calculated using the evidence discount method. k The basic probability assignment is corrected, that is:

[0168]

[0169] Where, the discount factor β = w k / w max , w max =max(w k ), k=1,2,…,K.

[0170] Corrected detection period T k The basic probability assignment matrix is ​​as follows:

[0171]

[0172] Using the DS combination rule, the final multi-cycle comprehensive recognition result can be obtained as follows:

[0173]

[0174] In some implementation scenarios, suppose that in a certain air defense operation, our side intends to use "radar (o1)", "data link (o2)" and "airspace coordination measures (o3)" as the identification means for the comprehensive identification of air targets; in order to facilitate making reasonable air defense decisions, the comprehensive identification results are divided into "our side (x1)", "may be our side (x2)", "may be the enemy (x3)", and "enemy (x4)"; to ensure the accuracy of identification, the air targets are detected for a total of four cycles, namely T1, T2, T3, and T4. The data detected by the above three means are converted into intuitionistic fuzzy language, and the intuitionistic fuzzy numbers corresponding to each detection cycle can be obtained as follows:

[0175] Table 1 Intuitive fuzzy numbers of each period

[0176]

[0177] The corresponding intuitive fuzzy matrix can be obtained from Table 1. According to the data in Table 1, the detection period T can be obtained from formula (7): k Identification means o i The AIFS cross entropy is as follows:

[0178]

[0179]

[0180] According to equations (11) and (12), the detection period T can be obtained k Identification means o i The weight matrix is ​​as follows:

[0181]

[0182] According to formula (13), the detection period T can be calculated k Identification means o i The basic probability assignments obtained are shown in Table 2:

[0183] Table 2 BPA obtained by each identification method in each detection cycle

[0184]

[0185] According to the data in Table 2, the data in the above table can be corrected and integrated by combining equations (14)-(16) to obtain the detection period T k The comprehensive recognition results are:

[0186] Table 3 Comprehensive recognition results of each detection cycle

[0187]

[0188] The detection period T can be calculated from equations (17)-(19): k The degree of conflict between g for:

[0189]

[0190] According to equations (20) and (21), the detection period T can be obtained k The weights are as follows:

[0191] W = [0.253 0.254 0.246 0.248] T (29);

[0192] By performing multi-cycle information fusion on the data in Table 3, the multi-cycle fusion recognition results are as follows:

[0193]

[0194] Therefore, the comprehensive identification result of the aerial target is "our side". It can be preliminarily seen that the use of multi-attribute decision-making ideas to identify aerial targets from a distance can achieve the desired purpose.

[0195] For example, compared with the DS evidence theory, the method provided in the embodiment of the present application can identify the correct friend-or-foe attributes of aerial targets, avoid misjudgment, and reduce the risk of accidental hits and injuries. This is because the method in this paper corrects the evidence when aggregating the identification means information and the detection cycle information, thereby minimizing the adverse effects of high-conflict evidence on the comprehensive identification results. Through the above comparison, it can be seen that even in the presence of high-conflict evidence, the use of multi-attribute decision-making ideas can still better solve the problem of comprehensive identification of aerial targets.

[0196] See also Figure 4 , Figure 4 The structural block diagram of the comprehensive identification system of aerial targets for multi-attribute decision-making provided in the embodiment of the present application includes:

[0197] An acquisition module 100 is used to acquire multiple recognition results of an aerial target, where the multiple recognition results are obtained through multiple recognition means in multiple detection cycles;

[0198] An intuitionistic fuzzy matrix module 200, used to generate intuitionistic fuzzy matrix data according to a plurality of recognition results;

[0199] The recognition weight module 300 is used to obtain recognition weight set data corresponding to multiple recognition means according to the intuitive fuzzy matrix data;

[0200] A basic probability assignment module 400 is used to obtain basic probability assignment data corresponding to multiple detection cycles and multiple identification methods according to evidence theory and intuitive fuzzy matrix data;

[0201] The recognition correction module 500 is used to correct the basic probability assignment data according to the recognition weight set data to obtain the comprehensive recognition result matrix data corresponding to the multiple detection cycles;

[0202] The cycle weight module 600 is used to obtain cycle identification weight set data corresponding to the multiple detection cycles according to the conflict degree data of the multiple detection cycles;

[0203] The multi-cycle fusion module 700 is used to perform multi-cycle information fusion on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result.

[0204] Exemplarily, the recognition weight module 300 includes:

[0205] A cross entropy unit, used to obtain cross entropy data corresponding to multiple recognition methods according to the intuitive fuzzy matrix data;

[0206] The identification weight unit is used to obtain and identify weight set data based on the cross entropy data.

[0207] Exemplarily, the basic probability assignment module 400 includes:

[0208] Mass function unit, used to generate the basic probability assignment formula corresponding to the Mass function according to the evidence theory;

[0209] The basic probability assignment unit is used to obtain basic probability assignment data according to the basic probability assignment formula and the intuitive fuzzy matrix data.

[0210] Exemplarily, the identification and correction module 500 includes:

[0211] An identification correction unit, used to correct the basic probability assignment data according to the identification weight set data and the evidence discount method to obtain identification correction basic probability assignment matrix data;

[0212] The recognition fusion unit is used to fuse the recognition-corrected basic probability assignment matrix data to obtain comprehensive recognition result matrix data.

[0213] Exemplarily, the multi-attribute decision-making air target integrated identification system also includes:

[0214] The evidence distance module is used to obtain the Jousselme evidence distance of the Mass function corresponding to multiple detection cycles;

[0215] The conflict degree module is used to generate conflict degree data for multiple detection cycles based on the Jousselme evidence distance.

[0216] Exemplarily, the multi-cycle fusion module 700 includes:

[0217] A period correction unit, used to correct the comprehensive recognition result matrix data according to the period recognition weight set data to obtain period correction basic probability assignment matrix data;

[0218] The multi-period fusion unit is used to generate multi-period comprehensive recognition results according to the period-corrected basic probability assignment matrix data and the DS combination rule.

[0219] This application also provides an electronic device, see Figure 5 , Figure 5 A block diagram of an electronic device provided in an embodiment of the present application. The electronic device may include a processor 510, a communication interface 520, a memory 530, and at least one communication bus 540. The communication bus 540 is used to realize direct connection and communication between these components. The communication interface 520 of the electronic device in the embodiment of the present application is used to communicate signaling or data with other node devices. The processor 510 may be an integrated circuit chip with signal processing capabilities.

[0220] The processor 510 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor 510 can also be any conventional processor, etc.

[0221] The memory 530 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electric erasable programmable read-only memory (EEPROM), etc. The memory 530 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 510, the electronic device can execute the above-mentioned Figures 1 to 3 The method embodiment involves various steps.

[0222] Optionally, the electronic device may further include a storage controller and an input / output unit.

[0223] The memory 530, storage controller, processor 510, peripheral interface, input and output unit components are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses 540. The processor 510 is used to execute executable modules stored in the memory 530, such as software function modules or computer programs included in the electronic device.

[0224] The input and output unit is used to provide users with the task creation and to create a start optional time period or preset execution time for the task to realize the interaction between the user and the server. The input and output unit can be, but is not limited to, a mouse and a keyboard.

[0225] Understandably, Figure 5 The structure shown is for illustration only, and the electronic device may also include Figure 5 More or fewer components as shown, or with Figure 5 Different configurations are shown. Figure 5 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0226] An embodiment of the present application further provides a storage medium having instructions stored thereon. When the instructions are run on a computer, the computer program is executed by a processor to implement the method described in the method embodiment. To avoid repetition, the method will not be described here.

[0227] The present application also provides a computer program product, which, when executed on a computer, enables the computer to execute the method described in the method embodiment.

[0228] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0229] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0230] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0231] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0232] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0233] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

Claims

1. A comprehensive identification method for aerial targets based on multi-attribute decision making, characterized in that: include: Acquire multiple recognition results of the aerial target, wherein the multiple recognition results are obtained by multiple recognition means in multiple detection cycles; Generate intuitive fuzzy matrix data according to the multiple recognition results; Obtaining recognition weight set data corresponding to the plurality of recognition means according to the intuitive fuzzy matrix data; Obtain basic probability assignment data corresponding to the multiple detection cycles and the multiple identification means respectively according to the evidence theory and the intuitive fuzzy matrix data; Modifying the basic probability assignment data according to the recognition weight set data to obtain comprehensive recognition result matrix data corresponding to the multiple detection cycles; Obtaining cycle identification weight set data corresponding to the plurality of detection cycles according to the conflict degree data of the plurality of detection cycles; The multi-cycle information fusion is performed on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result.

2. The method for comprehensive identification of aerial targets based on multiple attribute decision making according to claim 1 is characterized in that: The step of obtaining recognition weight set data corresponding to the plurality of recognition means according to the intuitive fuzzy matrix data comprises: Obtaining cross entropy data corresponding to the plurality of recognition means according to the intuitionistic fuzzy matrix data; The recognition weight set data is obtained according to the cross entropy data.

3. The method for comprehensive identification of aerial targets based on multi-attribute decision making according to claim 2 is characterized in that: The step of obtaining basic probability assignment data corresponding to the multiple detection cycles and the multiple identification means respectively according to the evidence theory and the intuitive fuzzy matrix data comprises: Generate the basic probability assignment formula corresponding to the Mass function based on evidence theory; The basic probability assignment data is obtained according to the basic probability assignment formula and the intuitive fuzzy matrix data.

4. The method for comprehensive identification of aerial targets based on multi-attribute decision making according to claim 2 is characterized in that: The step of modifying the basic probability assignment data according to the recognition weight set data to obtain comprehensive recognition result matrix data corresponding to the multiple detection cycles includes: Modify the basic probability assignment data according to the recognition weight set data and the evidence discount method to obtain recognition modified basic probability assignment matrix data; The recognition-corrected basic probability assignment matrix data is fused to obtain the comprehensive recognition result matrix data.

5. The method for comprehensive identification of aerial targets based on multiple attribute decision making according to claim 1 is characterized in that: Before the step of obtaining cycle identification weight set data corresponding to the plurality of detection cycles according to the conflict degree data of the plurality of detection cycles, the method further includes: Obtaining the Jousselme evidence distance of the Mass function corresponding to the plurality of detection cycles; The conflict degree data of the plurality of detection cycles are generated according to the Jousselme evidence distance.

6. The method for comprehensive identification of aerial targets based on multiple attribute decision making according to claim 1 is characterized in that: The step of performing multi-cycle information fusion on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result comprises: Correcting the comprehensive identification result matrix data according to the period identification weight set data to obtain period correction basic probability assignment matrix data; The multi-cycle comprehensive recognition result is generated according to the cycle-corrected basic probability assignment matrix data and the DS combination rule.

7. A multi-attribute decision-making aerial target comprehensive identification system, characterized in that: include: An acquisition module, used to acquire multiple recognition results of aerial targets, wherein the multiple recognition results are obtained by multiple recognition means within multiple detection cycles; An intuitionistic fuzzy matrix module, used to generate intuitionistic fuzzy matrix data according to the multiple recognition results; A recognition weight module, used for obtaining recognition weight set data corresponding to the plurality of recognition means according to the intuitive fuzzy matrix data; A basic probability assignment module, used for obtaining basic probability assignment data corresponding to the multiple detection cycles and the multiple identification means respectively according to the evidence theory and the intuitive fuzzy matrix data; An identification correction module, used to correct the basic probability assignment data according to the identification weight set data to obtain comprehensive identification result matrix data corresponding to the multiple detection cycles; A cycle weight module, used to obtain cycle identification weight set data corresponding to the multiple detection cycles according to the conflict degree data of the multiple detection cycles; The multi-cycle fusion module is used to perform multi-cycle information fusion on the comprehensive recognition result matrix data according to the cycle recognition weight set data to obtain a multi-cycle comprehensive recognition result.

8. The multi-attribute decision-making aerial target comprehensive identification system according to claim 7 is characterized in that: The recognition weight module comprises: A cross entropy unit, used to obtain cross entropy data corresponding to the multiple recognition means according to the intuitionistic fuzzy matrix data; An identification weight unit is used to obtain the identification weight set data according to the cross entropy data.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for an integrated aerial target identification system based on multi-attribute decision making as described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the computer executes the multi-attribute decision-making aerial target integrated identification system method according to any one of claims 1 to 6.

Images