An underwater target recognition method, device and computer readable storage medium

Abstract

The application belongs to the technical field of target identification, and relates to an underwater target identification method, device and computer readable storage medium. A plurality of intrinsic mode signals are obtained by decomposing a water acoustic signal mode. The plurality of intrinsic mode signals are input into a target identification model to output a plurality of underwater target prediction category probability distributions. The plurality of underwater target prediction category probability distributions are combined to obtain a new underwater target prediction category probability distribution. A weighted average probability distribution is obtained based on all underwater target prediction category probability distributions. An average similarity and a fusion evidence similarity are calculated based on each underwater target prediction category probability distribution and the weighted average probability distribution. If the fusion evidence similarity is less than or equal to the average similarity, the intrinsic mode signals of the water acoustic signal are screened, and underwater target identification is performed based on the screened target intrinsic mode signals. Through screening of signal components of the water acoustic signal, signal components containing error feature information are removed, and the underwater target identification precision is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of target recognition, in particular to an underwater target recognition method, device and computer readable storage medium. BACKGROUND

[0002] Underwater target detection is a technology that analyzes collected underwater acoustic signals, obtains target signal characteristics through processing, and then judges and identifies underwater target categories based on human or machine networks, which plays a very important role. With the development of science and technology, the further exploration and exploitation of marine resources, and the changes in the marine environment, the underwater environment has become increasingly complex, posing a potential threat to the marine ecological environment.

[0003] The prior art provides an underwater target detection method based on underwater acoustic signal decomposition and feature analysis recognition, which acquires underwater acoustic signals, decomposes the underwater acoustic signals, constructs a simple or complex convolutional neural network, extracts and fuses features of each signal component obtained by decomposition, and finally outputs the underwater target category corresponding to the underwater acoustic signals. However, due to the extreme complexity of underwater acoustic signals, for example, when the underwater target is moving, its signal will be superimposed with instantaneous interference (sound reflection of suddenly passing seabed organisms or seabed rocks), which results in that part of the signal components obtained by decomposition do not simply carry effective features about the target category, but are not simply environmental noise, but false superposition of interference signals and useful signals, i.e. pseudo-feature information carriers, which carry feature information unrelated or even conflicting with the true features of the underwater target category. Unlike noise signals, this pseudo-feature carrier does not simply mask effective feature information, but provides false feature information, which misleads the convolutional neural network and thus outputs an incorrect target recognition result, reducing the accuracy of underwater target recognition.

[0004] In summary, the existing underwater target recognition method based on signal decomposition ignores that part of the signal components obtained by decomposing the underwater acoustic signals act as pseudo-feature carriers, which provide feature information conflicting with that provided by other signal components, thus leading to an incorrect recognition result based on false feature information when recognizing the underwater target, affecting the accuracy of the underwater target recognition result. SUMMARY

[0005] To this end, the technical problem to be solved by the present application is to overcome the problem in the prior art that the underwater target recognition method based on signal decomposition ignores that part of the signal components obtained by decomposing the underwater acoustic signals act as pseudo-feature carriers, which provide feature information conflicting with that provided by other signal components, thus leading to an incorrect recognition result based on false feature information when recognizing the underwater target, affecting the accuracy of the underwater target recognition result.

[0006] To solve the above technical problems, the present application provides an underwater target recognition method, comprising:

[0007] S10: modal decomposition is performed on the underwater acoustic signal to obtain a plurality of intrinsic mode signals;

[0008] S20: the plurality of intrinsic mode signals are respectively input into a target recognition model to output a plurality of underwater target predicted class probability distributions, and a probability distribution set is obtained; the plurality of underwater target predicted class probability distributions are combined to obtain a new underwater target predicted class probability distribution, and the new underwater target predicted class probability distribution is added to the probability distribution set;

[0009] S30: the sum of the correlation coefficients of each underwater target predicted class probability distribution in the probability distribution set and the remaining underwater target predicted class probability distributions is taken as the support degree of each underwater target predicted class probability distribution; based on the underwater target predicted class probability distributions in the probability distribution set and their support degree weights, a weighted average probability distribution is obtained;

[0010] S40: the similarity between the underwater target predicted class probability distribution of each intrinsic mode signal and the weighted average probability distribution is taken as the evidence similarity of each intrinsic mode signal; based on the average of the evidence similarities of all intrinsic mode signals, an average similarity is obtained; the similarity between the new underwater target predicted class probability distribution and the weighted average probability distribution is taken as the fusion evidence similarity;

[0011] S50: the fusion evidence similarity and the average similarity are compared, if the fusion evidence similarity is less than or equal to the average similarity, the plurality of intrinsic mode signals of the underwater acoustic signal are screened, and step S20 is returned to be executed until the fusion evidence similarity is greater than the average similarity, and the target intrinsic mode signal obtained by screening is used for underwater target recognition.

[0012] Preferably, the screening of the plurality of intrinsic mode signals comprises:

[0013] A plurality of underwater acoustic signal samples are obtained in the underwater acoustic signal data set of each category of underwater target to obtain a subset of underwater acoustic signal samples of the category; a plurality of intrinsic mode signals of each underwater acoustic signal sample in all subsets of underwater acoustic signal samples are obtained;

[0014] The Fisher score method is used to calculate the discrimination performance score of each intrinsic mode signal for different categories of underwater targets;

[0015] The N intrinsic mode signals with the highest discrimination performance scores are taken as target intrinsic mode signals, so that the target intrinsic mode signals are screened from the plurality of intrinsic mode signals of the underwater acoustic signal.

[0016] Preferably, the Fisher score method is used to calculate the discrimination performance score of each intrinsic modal signal for different categories of underwater targets, including:

[0017] The mean value of the i-th intrinsic modal signal of all underwater acoustic signal samples in each subset of underwater acoustic signal samples is calculated to obtain the global mean value of the i-th intrinsic modal signal;

[0018] The mean value of the i-th intrinsic modal signal of each underwater acoustic signal sample in each subset of underwater acoustic signal samples is calculated to obtain the intra-class mean value of the i-th intrinsic modal signal in each subset of underwater acoustic signal samples;

[0019] Based on the sum of the squared differences between the intra-class mean value and the global mean value of the i-th intrinsic modal signal in each subset of underwater acoustic signal samples, the inter-class dispersion of the i-th intrinsic modal signal is obtained;

[0020] Based on the sum of the squared differences between the i-th intrinsic modal signal of each underwater acoustic signal sample in each subset of underwater acoustic signal samples and the intra-class mean value of the i-th intrinsic modal signal in each subset of underwater acoustic signal samples, the intra-class dispersion of the i-th intrinsic modal signal is obtained;

[0021] Based on the ratio of the inter-class dispersion and the intra-class dispersion of the i-th intrinsic modal signal, the discrimination performance score of the i-th intrinsic modal signal for different categories of underwater targets is obtained.

[0022] Preferably, the support weight of each underwater target predicted category probability distribution is the ratio of the support of the underwater target predicted category probability distribution to the sum of the supports of all underwater target predicted category probability distributions.

[0023] Preferably, the calculation formula of the correlation coefficient between underwater target predicted category probability distributions is:

[0024] ,

[0025] Wherein, represents the correlation coefficient between the i-th underwater target predicted category probability distribution and the j-th underwater target predicted category probability distribution; represents the probability value of the i-th predicted category in the i-th underwater target predicted category probability distribution; represents the number of categories of underwater targets;

[0026] The calculation formula of the support of the underwater target predicted category probability distribution is:​​​​​​​

[0027] ,

[0028] wherein, denotes the support of the i-th underwater target prediction class probability distribution; denotes the number of intrinsic modal signals; denotes the number of underwater target prediction class probability distributions in the probability distribution set; denotes the number of underwater target prediction class probability distributions in the probability distribution set;

[0029] The calculation formula of the weighted average probability distribution is:

[0030] ,

[0031] wherein, denotes the weighted average probability distribution; denotes the support weight of the i-th underwater target prediction class probability distribution; denotes the i-th underwater target prediction class probability distribution. Preferably, the calculation formula of the evidence similarity of the intrinsic modal signal is:

[0032]

[0033] ,

[0034] wherein, denotes the evidence similarity of the i-th intrinsic modal signal, , denotes the number of intrinsic modal signals; denotes the i-th underwater target prediction class probability distribution, ; denotes the weighted average probability distribution; denotes the transpose;

[0035] The calculation formula of the average similarity is:

[0036] ,

[0037] wherein, denotes the average similarity.

[0038] Preferably, the calculation formula of the global mean of the i-th intrinsic modal signal is:

[0039] ,

[0040] wherein, denotes the global mean of the i-th intrinsic modal signal; denotes the number of subsets of underwater acoustic signal samples;​​​ denotes the number of underwater acoustic signal samples in each subset of underwater acoustic signal samples; denotes the i-th intrinsic modal signal of the n-th underwater acoustic signal sample in the c-th subset of underwater acoustic signal samples;

[0041] The formula for calculating the intra-class mean of the i-th intrinsic modal signal in each subset of underwater acoustic signal samples is:

[0042] ,

[0043] wherein, denotes the intra-class mean of the i-th intrinsic modal signal in the c-th subset of underwater acoustic signal samples;

[0044] The formula for calculating the inter-class dispersion of the i-th intrinsic modal signal is:

[0045] ,

[0046] wherein, denotes the inter-class dispersion of the i-th intrinsic modal signal; denotes the weight of the squared difference between the intra-class mean of the i-th intrinsic modal signal in the c-th subset of underwater acoustic signal samples and the global mean;

[0047] The formula for calculating the intra-class dispersion of the i-th intrinsic modal signal is:

[0048] ,

[0049] wherein, denotes the intra-class dispersion of the i-th intrinsic modal signal;

[0050] The formula for calculating the performance score of the i-th intrinsic modal signal in distinguishing different classes of underwater targets is:

[0051] ,

[0052] wherein, denotes the performance score of the i-th intrinsic modal signal in distinguishing different classes of underwater targets.

[0053] Preferably, the underwater target recognition based on the target intrinsic modal signals screened out comprises:

[0054] inputting the target intrinsic modal signals screened out into a target recognition model respectively, and outputting the underwater target class prediction probability distribution of each target intrinsic modal signal;

[0055] The Dempster-Shafer fusion rule is used to combine the underwater target category prediction probability distribution of each target intrinsic modal signal, so that the underwater target category prediction comprehensive probability distribution is obtained, and thus the underwater target category corresponding to the underwater acoustic signal is obtained.

[0056] The application further provides an underwater target recognition device.

[0057] The signal decomposition module is configured to perform modal decomposition on the underwater acoustic signal to obtain a plurality of intrinsic modal signals.

[0058] The probability distribution acquisition module is configured to input the plurality of intrinsic modal signals into the target recognition model respectively, output a plurality of underwater target prediction category probability distributions, obtain a probability distribution set, combine the plurality of underwater target prediction category probability distributions, obtain a new underwater target prediction category probability distribution, and add the new underwater target prediction category probability distribution to the probability distribution set.

[0059] The probability distribution weighted fusion module is configured to take the sum of the correlation coefficients of each underwater target prediction category probability distribution and the remaining underwater target prediction category probability distributions in the probability distribution set as the support degree of each underwater target prediction category probability distribution, and obtain a weighted average probability distribution based on each underwater target prediction category probability distribution in the probability distribution set and the support degree weight thereof.

[0060] The similarity calculation module is configured to take the similarity between the underwater target prediction category probability distribution of each intrinsic modal signal and the weighted average probability distribution as the evidence similarity of each intrinsic modal signal, obtain an average similarity based on the average of the evidence similarities of all intrinsic modal signals, and take the similarity between the new underwater target prediction category probability distribution and the weighted average probability distribution as the fusion evidence similarity.

[0061] The signal screening and target recognition module is configured to compare the fusion evidence similarity and the average similarity, if the fusion evidence similarity is less than or equal to the average similarity, screen the plurality of intrinsic modal signals of the underwater acoustic signal, and return to the step of executing the probability distribution acquisition module until the fusion evidence similarity is greater than the average similarity, and perform underwater target recognition based on the target intrinsic modal signal obtained through screening.

[0062] The application further provides a computer readable storage medium, and the computer readable storage medium stores a computer program.

[0063] The underwater target recognition method provided by the application has the following beneficial effects:

[0064] Since the feature information of the pseudo feature signal conflicts with the feature information of other signals, resulting in different prediction results based on the feature information, the present application outputs a plurality of underwater target prediction class probability distributions based on the plurality of intrinsic mode signals obtained by decomposing the underwater acoustic signal, and takes each underwater target prediction class probability distribution as a raw evidence source, and then filters out the conflict evidence source and removes the corresponding intrinsic mode signal, so as to remove the pseudo feature signal in the plurality of intrinsic mode signals; specifically, the comprehensive evidence source (i.e. the new underwater target prediction class probability distribution) fused with all the original evidence source situations is obtained by merging each underwater target prediction class probability distribution, and the weighted average evidence is calculated by using the original evidence source and the comprehensive evidence source. Since the comprehensive evidence source can reflect the situation after the fusion of all original evidences from the overall perspective, the weighted average evidence can better reflect the potential synergy or conflict relationship between the original evidence sources. At the same time, the fusion evidence similarity is compared with the average similarity to determine whether there is a conflict evidence in the current original evidence source. Since the fusion evidence similarity represents the overall consistency of all original evidence sources with the weighted average evidence, it can more clearly reflect the credibility of the comprehensive evidence. When the fusion evidence similarity is low, it indicates that there is a conflict evidence source interfering with the overall evidence credibility. Then, the intrinsic mode signals are filtered until the fusion evidence similarity is greater than the average similarity. At this time, the intrinsic mode signals filtered out are signal components containing effective feature information, so that more accurate underwater target recognition results can be output based on these signal components. The present application introduces the evidence conflict theory to filter the signal components of the underwater acoustic signal, removes the signal components containing false feature information, so that the target recognition model is not disturbed by the false feature information, thereby improving the accuracy of underwater target recognition. BRIEF DESCRIPTION OF DRAWINGS

[0065] In order to make the content of the present application more easily understood, the present application will be further described in detail below according to specific embodiments of the present application and in conjunction with the drawings, in which:

[0066] Figure 1 The flow chart of the underwater target recognition method provided by the present application;

[0067] Figure 2 The structure schematic diagram of the underwater target recognition device provided by the present application. DETAILED DESCRIPTION

[0068] The present application will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the present application and implement it, but the embodiments are not limiting the present application.

[0069] Please refer to Figure 1 ,Figure 1 The flow chart of the underwater target recognition method provided in the present application is shown, and the method specifically comprises:

[0070] S10: Modal decomposition is performed on the underwater acoustic signal to obtain a plurality of intrinsic mode signals.

[0071] S20: The plurality of intrinsic mode signals are respectively input into a target recognition model to output a plurality of underwater target prediction class probability distributions, and a probability distribution set is obtained; the plurality of underwater target prediction class probability distributions are combined to obtain a new underwater target prediction class probability distribution, and the new underwater target prediction class probability distribution is added to the probability distribution set.

[0072] For example, the Dempster-Shafer fusion rule can be used to combine the plurality of underwater target prediction class probability distributions, and the formula is represented as:

[0073]

[0074] wherein, represents the new underwater target prediction class probability distribution; represents the underwater target prediction class probability distribution corresponding to the Ith intrinsic mode signal.

[0075] S30: The sum of the correlation coefficients of each underwater target prediction class probability distribution in the probability distribution set and the remaining underwater target prediction class probability distributions is taken as the support degree of each underwater target prediction class probability distribution; and a weighted average probability distribution is obtained based on the underwater target prediction class probability distributions in the probability distribution set and the support degree weights thereof.

[0076] S40: The similarity between the underwater target prediction class probability distribution of each intrinsic mode signal and the weighted average probability distribution is taken as the evidence similarity of each intrinsic mode signal; the average similarity is obtained based on the average of the evidence similarities of all intrinsic mode signals; and the similarity between the new underwater target prediction class probability distribution and the weighted average probability distribution is taken as the fusion evidence similarity.

[0077] S50: The fusion evidence similarity and the average similarity are compared, and if the fusion evidence similarity is less than or equal to the average similarity, the plurality of intrinsic mode signals of the underwater acoustic signal are screened, and step S20 is returned to be executed until the fusion evidence similarity is greater than the average similarity, and underwater target recognition is performed based on the target intrinsic mode signals obtained by screening.

[0078] ​Specifically, the present application takes the underwater target prediction category probability distribution of multiple intrinsic modal signals of the underwater acoustic signal as the evidence source. When one evidence source is different from other evidence sources, the evidence source is considered as conflict evidence. By eliminating the intrinsic modal signal corresponding to the conflict evidence source, the pseudo feature signal is eliminated, which can ensure the consistency of the remaining signals for the prediction result, thereby improving the accuracy of underwater target recognition.

[0079] Further, in the conflict evidence identification, the present application first takes all the underwater target prediction category probability distributions as original evidence sources, combines the original evidence sources to obtain a comprehensive evidence source (i.e. a new underwater target prediction category probability distribution), and calculates the weighted average evidence by using the original evidence sources and the comprehensive evidence source. Since the comprehensive evidence source can reflect the situation after the fusion of all original evidence sources from the overall perspective, the weighted average evidence can better reflect the potential synergy or conflict relationship between the original evidence sources. Meanwhile, the present application does not directly compare the evidence similarity of each original evidence source with the average similarity to determine the conflict evidence source, but compares the fusion evidence similarity with the average similarity to determine whether there is conflict evidence in the current original evidence source. Since the fusion evidence similarity represents the overall fit degree of all original evidence sources to the weighted average evidence, the reliability of the comprehensive evidence can be more clearly reflected. When the fusion evidence similarity is low, it indicates that there is a conflict evidence source interfering with the overall evidence reliability. This conflict evidence identification method takes the comprehensive evidence source as a relatively stable conflict identification benchmark, thereby improving the reliability of conflict evidence identification.

[0080] Specifically, the calculation formula of the correlation coefficient between the underwater target prediction category probability distributions in step S30 is:

[0081]

[0082] wherein, represents the correlation coefficient between the i-th underwater target prediction category probability distribution and the j-th underwater target prediction category probability distribution;

[0083] The calculation formula of the support degree of the underwater target prediction category probability distribution is:

[0084] ,​​​​​​​​​​

[0085] in, Indicates the first Support of the probability distribution of the predicted categories of underwater targets; Indicates the number of intrinsic mode signals; This represents the number of probability distributions for underwater target prediction categories in the probability distribution set;

[0086] Furthermore, the support weight of each underwater target prediction category probability distribution is the ratio of the support of that underwater target prediction category probability distribution to the sum of the support of all underwater target prediction category probability distributions.

[0087] Specifically, the formula for calculating the support of the probability distribution of underwater target prediction categories is as follows:

[0088] ,

[0089] ,

[0090] in, Indicates the first Support of the probability distribution of the predicted categories of underwater targets; Indicates the first Support weights of the probability distribution of underwater target prediction categories.

[0091] Furthermore, the formula for calculating the weighted average probability distribution is:

[0092] ,

[0093] in, This represents a weighted average probability distribution; Indicates the first Support weights of the probability distribution of underwater target prediction categories; Indicates the first Probability distribution of predicted categories for underwater targets.

[0094] Furthermore, the formula for calculating the evidence similarity of the intrinsic mode signals in step S40 is as follows:

[0095] ,

[0096] in, This represents the evidence similarity of the i-th intrinsic mode signal. , Indicates the number of intrinsic mode signals; Indicates the first Probability distribution of underwater target prediction categories ; This represents a weighted average probability distribution; This indicates transpose.

[0097] The formula for calculating the average similarity is:

[0098] ,

[0099] in, This represents the average similarity.

[0100] Specifically, when the similarity of the fused evidence is less than or equal to the average similarity, it indicates that there are conflicting evidence sources among the existing multiple evidence sources. Therefore, it is necessary to eliminate conflicting evidence sources. In some embodiments, in step S50, an intrinsic modal signal without conflicting evidence sources can be screened by random screening. However, random screening lacks theoretical guidance and is inefficient. Therefore, this application proposes a method for screening intrinsic modal signals.

[0101] Specifically, the filtering of multiple intrinsic mode signals in step S50 includes:

[0102] Step 1: Obtain multiple underwater acoustic signal samples from the underwater acoustic signal dataset for each category of underwater target to obtain a subset of underwater acoustic signal samples for that category; obtain multiple intrinsic mode signals for each underwater acoustic signal sample in all underwater acoustic signal sample subsets.

[0103] Step 2: Calculate the distinguishing performance score of each intrinsic mode signal for different categories of underwater targets using the Fisher scoring method.

[0104] Step 3: Select the N intrinsic mode signals with the highest performance scores as target intrinsic mode signals, thereby filtering out the target intrinsic mode signals from multiple intrinsic mode signals of the underwater acoustic signal.

[0105] Further, step 2 includes:

[0106] Step 2-1: Calculate the mean of the i-th intrinsic mode signal of all underwater acoustic signal samples in the subset of all underwater acoustic signal samples, and obtain the global mean of the i-th intrinsic mode signal.

[0107] Specifically, the formula for calculating the global mean of the i-th intrinsic mode signal is:

[0108] ,

[0109] in, This represents the global mean of the i-th intrinsic mode signal; Indicates the number of underwater acoustic signal sample subsets; This indicates the number of underwater acoustic signal samples in each subset of underwater acoustic signal samples; denotes the i-th intrinsic mode signal of the n-th underwater acoustic signal sample in the c-th underwater acoustic signal sample subset.

[0110] Step 2-2: Calculate the mean of the i-th intrinsic mode signal of each underwater acoustic signal sample in each underwater acoustic signal sample subset, to obtain the intra-class mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset.

[0111] Specifically, the calculation formula of the intra-class mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset is:

[0112] ,

[0113] wherein, denotes the intra-class mean of the i-th intrinsic mode signal in the c-th underwater acoustic signal sample subset.

[0114] Step 2-3: Based on the sum of the squared differences between the intra-class mean and the global mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset, obtain the inter-class dispersion of the i-th intrinsic mode signal.

[0115] Specifically, the calculation formula of the inter-class dispersion of the i-th intrinsic mode signal is:

[0116] ,

[0117] wherein, denotes the inter-class dispersion of the i-th intrinsic mode signal. denotes the weight of the squared difference between the intra-class mean and the global mean of the i-th intrinsic mode signal in the c-th underwater acoustic signal sample subset.

[0118] Step 2-4: Based on the sum of the squared differences between the i-th intrinsic mode signal of each underwater acoustic signal sample in each underwater acoustic signal sample subset and the intra-class mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset, obtain the intra-class dispersion of the i-th intrinsic mode signal.

[0119] Specifically, the calculation formula of the intra-class dispersion of the i-th intrinsic mode signal is:

[0120] ,

[0121] wherein, denotes the intra-class dispersion of the i-th intrinsic mode signal.

[0122] Step 2-5: Based on the ratio of the inter-class dispersion and the intra-class dispersion of the i-th intrinsic mode signal, obtain the performance score of the i-th intrinsic mode signal in distinguishing different types of underwater targets.

[0123] Specifically, the calculation formula of the discrimination performance score of the i-th intrinsic modal signal for different categories of underwater targets is:

[0124] ,

[0125] wherein, represents the discrimination performance score of the i-th intrinsic modal signal for different categories of underwater targets.

[0126] Since the inter-class dispersion is used to measure the dispersion degree of different categories of underwater acoustic signal samples on the intrinsic modal signal, and the intra-class dispersion is used to measure the compactness of the same category of underwater acoustic signal samples on the intrinsic modal signal, when the inter-class dispersion of a certain intrinsic modal signal is greater, the intra-class dispersion is smaller, that is, the discrimination performance score is greater, it means that the intrinsic modal signal can reflect the difference of different categories of underwater acoustic signals, and the ability to distinguish different categories of underwater acoustic signals is stronger. Therefore, the present application selects the N intrinsic modal signals with the maximum discrimination performance score as the target intrinsic modal signals, and selects the target intrinsic modal signals of the underwater acoustic signal, so as to improve the discrimination degree of the target recognition model for the underwater acoustic signal.

[0127] Further, the underwater target recognition based on the selected target intrinsic modal signal comprises:

[0128] The target recognition model is inputted into the selected target intrinsic modal signal, and the underwater target category prediction probability distribution of each target intrinsic modal signal is outputted.

[0129] The underwater target category prediction probability distribution of each target intrinsic modal signal is merged by using the Dempster-Shafer fusion rule, and the underwater target category prediction comprehensive probability distribution is obtained, so as to obtain the underwater target category corresponding to the underwater acoustic signal.

[0130] The effectiveness of the above-mentioned underwater target recognition method is illustrated by a specific example, and the marine underwater acoustic signal data set in the OUA data set is used in this example.

[0131] The marine underwater acoustic signal data set of the OUA data set is collected in a marine experiment, and the data set is obtained by a 128-unit horizontal line array (HLA) dragged by a 400-ton receiving ship. The device records the noise signals emitted by two source ships with a weight of 100 tons and 500 tons respectively, and the source ships run at different distances of 2 to 20 kilometers from the receiving ship. The total duration of the data set is 14.5 hours, and contains four different categories of classification data: target A: signal generated by ship A; target B: signal generated by ship B; target C: signal generated by A ship and B ship; non-target signal D.

[0132] The total signal duration is 38726 seconds, including 4567 seconds of background noise D, 4819 seconds of target A, 20220 seconds of target B, and 9120 seconds of target C. In order to maintain data balance, we selected 4500 sample experiments from each category. Given the relatively small size of the dataset, the length of each data sample is set to 1 second.

[0133] The target recognition model in this example is a back propagation (BP) neural network consisting of an input layer of 32 neurons, a hidden layer of 64 neurons, and an output layer. The specific steps for underwater target recognition include:

[0134] (1) Set up the data set: the test set is completely separated from the training set and the validation set, and the proportion is set to 7:1:2.

[0135] (2) Find the appropriate decomposition parameter K and use the multivariate variational mode decomposition method to decompose the underwater acoustic signal: use real data to analyze the signal energy ratio under different decomposition levels, hide the first component, and draw a picture to observe the decomposition energy ratio of the remaining components.

[0136] Specifically, after analysis, the number of decomposition modes K is set to 5 for the best effect, and the gray correlation degree between each decomposition component and the original signal is calculated. The results show that the decomposed signal shows a high degree of similarity to the original signal, and the decomposition process is effective.

[0137] (3) Preliminary prediction and merging: the multiple intrinsic mode signals obtained by decomposition are input into the target recognition model, and the corresponding underwater target prediction class probability distribution of each intrinsic mode signal is output; the multiple underwater target prediction class probability distributions are merged to obtain a new underwater target prediction class probability distribution.

[0138] (4) Conflict evidence acquisition and correction: based on all underwater target prediction class probability distributions, calculate the average similarity and fusion evidence similarity, compare the fusion evidence similarity and the average similarity, and determine whether there is conflict evidence in the multiple intrinsic mode signals. If there is, filter the multiple intrinsic mode signals until no conflict evidence exists in the filtered intrinsic mode signals.

[0139] (5) Target recognition: based on the filtered intrinsic mode signals without conflict evidence, perform underwater target recognition.

[0140] The accuracy of the validation set in this example increased by 0.48% to 4.89%, while the accuracy of the test set increased by 0.97% to 8.26%. As shown in Table 1, the data provided by this example before and after screening for different categories of underwater acoustic signal discrimination:

[0141] Table 1

[0142]

[0143] It is worth noting that through analysis of underwater target recognition results before and after the intrinsic modal signal screening, the present example finds that the underwater acoustic signals of categories B, C and D are more likely to be confused, and the discrimination degree of the three types of underwater acoustic signals is obviously improved after the intrinsic modal signal screening, indicating that the method provided by the present application can improve the accuracy of underwater target recognition.

[0144] Based on the underwater target recognition method provided in the above embodiments, the present embodiment further provides an underwater target recognition device, as shown in Figure 2 The device specifically comprises:

[0145] The signal decomposition module 10 is configured to perform modal decomposition on the underwater acoustic signal to obtain a plurality of intrinsic modal signals.

[0146] The probability distribution acquisition module 20 is configured to input the plurality of intrinsic modal signals into a target recognition model respectively, output a plurality of underwater target prediction category probability distributions, obtain a probability distribution set, combine the plurality of underwater target prediction category probability distributions, obtain a new underwater target prediction category probability distribution, and add the new underwater target prediction category probability distribution to the probability distribution set.

[0147] The probability distribution weighted fusion module 30 is configured to take the sum of the correlation coefficients of each underwater target prediction category probability distribution and the remaining underwater target prediction category probability distributions in the probability distribution set as the support degree of each underwater target prediction category probability distribution, and obtain a weighted average probability distribution based on the underwater target prediction category probability distributions in the probability distribution set and the support degree weights thereof.

[0148] The similarity calculation module 40 is configured to take the similarity between the underwater target prediction category probability distribution of each intrinsic modal signal and the weighted average probability distribution as the evidence similarity of each intrinsic modal signal, obtain an average similarity based on the average of the evidence similarities of all intrinsic modal signals, and take the similarity between the new underwater target prediction category probability distribution and the weighted average probability distribution as the fusion evidence similarity.

[0149] The signal screening and target recognition module 50 is configured to compare the fusion evidence similarity and the average similarity, if the fusion evidence similarity is less than or equal to the average similarity, screen the plurality of intrinsic modal signals of the underwater acoustic signal, and return to execute the step of the probability distribution acquisition module until the fusion evidence similarity is greater than the average similarity, and perform underwater target recognition based on the target intrinsic modal signal obtained through screening.

[0150] The present embodiment further provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the above-mentioned underwater target recognition method.

[0151] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, the methods can be tangibly embodied in a machine-readable storage medium having stored thereon instructions that can be used to program a computer to perform any of the operations described herein. The software implementation can be for example, in the form of a computer program product. The software implementation can be implemented in a centralized fashion in one computer system or processor, or in a distributed fashion where different elements are spread across several interconnected computer systems or processors.

[0152] The present application is described in reference to the drawings, which are as follows. Figure 1 Figure 1

[0153] Figure 1 Figure 1

[0154] Figure 1 Figure 1

[0155] It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.​​​​​​​​

Claims

1. A method for underwater target identification, characterized in that, include: S10: Perform mode decomposition on the underwater acoustic signal to obtain multiple intrinsic mode signals; S20: Input multiple intrinsic mode signals into the target recognition model respectively, output multiple underwater target prediction category probability distributions, and obtain a probability distribution set; merge the multiple underwater target prediction category probability distributions to obtain a new underwater target prediction category probability distribution, and add the new underwater target prediction category probability distribution to the probability distribution set; S30: The sum of the correlation coefficients between the probability distribution of each underwater target prediction category and the probability distribution of the remaining underwater target prediction categories in the probability distribution set is used as the support of each underwater target prediction category probability distribution; Based on the probability distribution of each underwater target prediction category and its support weight in the probability distribution set, a weighted average probability distribution is obtained. S40: The similarity between the underwater target prediction category probability distribution and the weighted average probability distribution of each intrinsic mode signal is used as the evidence similarity of each intrinsic mode signal; the average similarity is obtained based on the mean of the evidence similarities of all intrinsic mode signals; the similarity between the new underwater target prediction category probability distribution and the weighted average probability distribution is used as the fusion evidence similarity. S50: Compare the fused evidence similarity with the average similarity. If the fused evidence similarity is less than or equal to the average similarity, then filter the multiple intrinsic mode signals of the underwater acoustic signal and return to step S20 until the fused evidence similarity is greater than the average similarity. Underwater target identification is performed based on the filtered target intrinsic mode signals.

2. The underwater target identification method according to claim 1, characterized in that, Filtering multiple intrinsic mode signals includes: Multiple underwater acoustic signal samples are obtained from the underwater acoustic signal dataset for each category of underwater target to obtain a subset of underwater acoustic signal samples for that category; multiple intrinsic mode signals of each underwater acoustic signal sample in all underwater acoustic signal sample subsets are obtained. The Fisher scoring method was used to calculate the discrimination performance score of each intrinsic mode signal for different categories of underwater targets; The N intrinsic mode signals with the highest performance scores are selected as target intrinsic mode signals, thereby filtering out the target intrinsic mode signals from multiple intrinsic mode signals of the underwater acoustic signal.

3. The underwater target identification method according to claim 2, characterized in that, The Fisher scoring method was used to calculate the distinguishing performance score of each intrinsic mode signal for different categories of underwater targets, including: Calculate the mean of the i-th intrinsic mode signal of all underwater acoustic signal samples in the subset of all underwater acoustic signal samples, and obtain the global mean of the i-th intrinsic mode signal; Calculate the mean of the i-th intrinsic mode signal of each underwater acoustic signal sample in each underwater acoustic signal sample subset, and obtain the intra-class mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset; The inter-class dispersion of the i-th intrinsic mode signal is obtained by summing the squared differences between the intra-class mean and the global mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset. The intra-class dispersion of the i-th intrinsic mode signal is obtained by summing the squared differences between the i-th intrinsic mode signal and the intra-class mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset. Based on the ratio of inter-class dispersion to intra-class dispersion of the i-th intrinsic mode signal, the discrimination performance score of the i-th intrinsic mode signal for different categories of underwater targets is obtained.

4. The underwater target identification method according to claim 1, characterized in that, The support weight of each underwater target prediction category probability distribution is the ratio of the support of that underwater target prediction category probability distribution to the sum of the support of all underwater target prediction category probability distributions.

5. The underwater target identification method according to claim 1, characterized in that, The formula for calculating the correlation coefficient between the probability distributions of underwater target prediction categories is: ， in, Indicates the first The probability distribution of the predicted category of the first underwater target is related to the probability distribution of the second underwater target prediction category. The correlation coefficient between the probability distributions of the predicted categories of underwater targets; Indicates the first In the probability distribution of the underwater target prediction category, the th... The probability values ​​of each predicted category; Indicates the number of categories of underwater targets; Indicates the first In the probability distribution of the underwater target prediction category, the th... The probability values ​​of each predicted category; The formula for calculating the support of the probability distribution of underwater target prediction categories is: ， in, Indicates the first Support of the probability distribution of the predicted categories of underwater targets; Indicates the number of intrinsic mode signals; This represents the number of probability distributions for underwater target prediction categories in the probability distribution set; The formula for calculating the weighted average probability distribution is: ， in, This represents a weighted average probability distribution; Indicates the first Support weights of the probability distribution of underwater target prediction categories; Indicates the first Probability distribution of predicted categories for underwater targets.

6. The underwater target identification method according to claim 1, characterized in that, The formula for calculating the evidence similarity of intrinsic mode signals is: ， in, This represents the evidence similarity of the i-th intrinsic mode signal. , Indicates the number of intrinsic mode signals; Indicates the first Probability distribution of underwater target prediction categories ; This represents a weighted average probability distribution; Indicates transpose; The formula for calculating the average similarity is: ， in, This represents the average similarity.

7. The underwater target identification method according to claim 3, characterized in that, The formula for calculating the global mean of the i-th intrinsic mode signal is: ， in, This represents the global mean of the i-th intrinsic mode signal; Indicates the number of underwater acoustic signal sample subsets; This indicates the number of underwater acoustic signal samples in each subset of underwater acoustic signal samples; This represents the i-th intrinsic mode signal of the n-th underwater acoustic signal sample in the c-th underwater acoustic signal sample subset; The formula for calculating the intra-class mean of the i-th intrinsic mode signal in each underwater acoustic signal sample subset is: ， in, Let represent the intra-class mean of the i-th intrinsic mode signal in the c-th underwater acoustic signal sample subset; The formula for calculating the inter-class dispersion of the i-th intrinsic mode signal is: ， in, This represents the inter-class dispersion of the i-th intrinsic mode signal; This represents the weight of the squared difference between the intra-class mean and the global mean of the i-th intrinsic mode signal in the c-th underwater acoustic signal sample subset; The formula for calculating the intra-class dispersion of the i-th intrinsic mode signal is: ， in, This represents the intra-class dispersion of the i-th intrinsic mode signal; The formula for calculating the discrimination performance score of the i-th intrinsic mode signal for different categories of underwater targets is: ， in, This represents the performance score of the i-th intrinsic mode signal in distinguishing different categories of underwater targets.

8. The underwater target identification method according to claim 1, characterized in that, Underwater target identification based on the selected intrinsic mode signals includes: The selected target intrinsic mode signals are input into the target recognition model, and the underwater target category prediction probability distribution of each target intrinsic mode signal is output. The underwater target category prediction probability distribution of each target intrinsic mode signal is merged using the Dempster-Shafer fusion rule to obtain the comprehensive underwater target category prediction probability distribution, thereby obtaining the underwater target category corresponding to the underwater acoustic signal.

9. An underwater target identification device, characterized in that, include: The signal decomposition module is used to perform mode decomposition on the underwater acoustic signal to obtain multiple intrinsic mode signals; The probability distribution acquisition module is used to input multiple intrinsic mode signals into the target recognition model, output multiple underwater target prediction category probability distributions, and obtain a probability distribution set; the multiple underwater target prediction category probability distributions are merged to obtain a new underwater target prediction category probability distribution, and the new underwater target prediction category probability distribution is added to the probability distribution set; The probability distribution weighted fusion module is used to sum the correlation coefficients between the probability distribution of each underwater target prediction category and the probability distribution of the remaining underwater target prediction categories in the probability distribution set, and use this as the support of each underwater target prediction category probability distribution. Based on the probability distribution of each underwater target prediction category and its support weight in the probability distribution set, a weighted average probability distribution is obtained. The similarity calculation module is used to take the similarity between the underwater target prediction category probability distribution and the weighted average probability distribution of each intrinsic mode signal as the evidence similarity of each intrinsic mode signal; to obtain the average similarity based on the mean of the evidence similarity of all intrinsic mode signals; and to take the similarity between the new underwater target prediction category probability distribution and the weighted average probability distribution as the fusion evidence similarity. The signal filtering and target recognition module is used to compare the fused evidence similarity with the average similarity. If the fused evidence similarity is less than or equal to the average similarity, multiple intrinsic mode signals of the underwater acoustic signal are filtered, and the process returns to the step of the probability distribution acquisition module until the fused evidence similarity is greater than the average similarity. Underwater target recognition is then performed based on the filtered intrinsic mode signals of the target.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the underwater target identification method according to any one of claims 1 to 8.

Images