A maritime audio recognition and analysis system

Through the maritime audio identification and analysis system, multiple audio acquisition units and marine life detection modules are used to monitor the seabed environment in real time and provide warnings, solving the threat of marine life attacks to the safety of maritime workers and achieving accurate judgment and positioning of marine life locations .

CN116013364BActive Publication Date: 2024-11-08HANGZHOU JIEFU ELECTRONIC TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211723733.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-30
Publication Date
2024-11-08
Estimated Expiration
2042-12-30

AI Technical Summary

Technical Problem

In maritime activities, attacks from marine organisms pose a threat to the safety of maritime workers, and it is difficult for existing technology to effectively judge and locate the location of marine organisms and provide real-time warnings.

Method used

A maritime audio recognition and analysis system is designed to monitor the seabed environment in real time through multiple audio acquisition units, combine it with the marine biological detection module to detect audio features and output positioning areas, which is convenient for providing warning and precise positioning.

Benefits of technology

It realizes accurate judgment and positioning of marine organisms, improves the safety of maritime staff, and provides real-time warning and precise positioning capabilities through the combination of the settings of multiple audio acquisition units and the combination of marine organism detection modules.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116013364B_ABST
    Figure CN116013364B_ABST
Patent Text Reader

Abstract

The present patent application relates to the field of sound recognition technology, and discloses a maritime audio recognition and analysis system, including an audio acquisition module, a marine life detection module and an output module, which has the following characteristics: by setting a plurality of audio acquisition units in the bottom of the ship, using the plurality of audio acquisition units to collect audio data, real-time monitoring of the seabed environment, and combining with the marine life detection module to detect whether the first voice information and the second voice information contain marine life audio features, if so, it indicates that there are marine life in the corresponding first area or the second area, and the output unit is used to output the positioning area, so as to provide a warning to underwater workers; 2. By using the setting of a plurality of audio acquisition units, it is convenient to preliminarily obtain the range of the marine life, and by using the positional relationship between adjacent audio acquisition units and the overlapping areas to reduce the range, it is convenient to locate the marine life and improve the accuracy of positioning.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of sound recognition, and in particular to a maritime audio recognition and analysis system. Background Art

[0002] In maritime activities, in addition to hull-related operations, operations on waterways, exploration-related operations, and rescue-related operations generally require diving into sea areas to carry out operations. However, there is a greater risk in sea area operations, namely, attacks by marine organisms. For highly aggressive marine organisms, if they appear during the operation, dangerous situations may occur, and the safety of maritime personnel cannot be guaranteed. Summary of the invention

[0003] The present invention is intended to provide a maritime audio recognition and analysis system, which has the characteristics of being easy to judge and locate the position of marine life and providing timely warnings to maritime workers.

[0004] To achieve the above object, the basic scheme of the present invention is as follows:

[0005] A maritime audio recognition and analysis system, characterized in that:

[0006] An audio acquisition module, wherein the audio acquisition module includes a plurality of audio acquisition units, wherein a positional relationship is preset between the respective audio acquisition units, and each audio acquisition unit is preset with a collection area, wherein the collection area reflects the range in which the corresponding audio acquisition unit can actually collect audio. When the operating equipment is in a stationary state, the audio acquisition module starts to work. When one of the audio acquisition units collects ambient sound, the audio acquisition unit is defined as a first acquisition unit, and the first acquisition unit collects ambient sound according to a preset duration and generates a first voice message. An audio acquisition unit that has an intersection with the collection area of ​​the first acquisition unit is defined as a second acquisition unit, and when the first acquisition unit collects ambient sound, the second acquisition unit synchronously collects ambient sound and generates a second voice message.

[0007] A marine life detection module, wherein the marine life detection module is configured with a marine life audio database, wherein the marine life audio database stores a number of marine life audio features, obtains the first voice information and determines whether the first voice information contains a marine life audio feature, wherein the marine life audio feature indicates the existence of a corresponding marine life, and if the marine life audio feature is contained, a first marine life signal is generated, and the marine life detection module obtains and determines whether each of the second voice information contains the marine life audio feature, and if so, a second marine life signal is generated;

[0008] A processing module, the processing module comprising

[0009] a first positioning unit, the first positioning unit receives the first marine biological signal and obtains a corresponding collection area to define as a first area, and generates a first positioning signal,

[0010] a second positioning unit, when the second positioning unit receives the second marine biological signal, acquiring a corresponding collection area to define as a second area, and generating a second positioning signal;

[0011] An output module, wherein the output module is used to receive the first positioning signal and the second positioning signal to output a positioning area. When the output module receives the first positioning signal and the second positioning signal, the positioning area represents the intersection of the first area and each of the second areas. When the output module only receives the first positioning signal, the positioning area represents the range of the first area excluding the second area.

[0012] Furthermore, the output module includes

[0013] an indicating unit, the indicating unit comprising

[0014] an illumination subunit, the illumination subunit being configured to emit light to indicate the first area,

[0015] a shielding subunit, the shielding subunit being capable of shielding the first area and adjusting a shielding range to obtain the positioning area;

[0016] A lateral driving unit is used to drive the indicating unit to move horizontally to a target area, where the target area represents an actual position of the first area in the ocean.

[0017] Further, the positioning area can be divided into a first output area and a second output area according to the distance from the first acquisition unit and the second acquisition unit, wherein the distance from the first output area to the first acquisition unit is smaller than the distance from the second output area to the first acquisition unit, and the distance from the first output area to the second acquisition unit is larger than the distance from the second output area to the second acquisition unit, and the maritime audio recognition and analysis system further includes a region adjustment module, and the region adjustment module includes

[0018] A volume acquisition unit, the volume acquisition unit corresponds to the audio acquisition unit one by one and is used to acquire volume data, and the volume data acquired by the volume acquisition units corresponding to the first acquisition unit and the second acquisition unit are defined in sequence as a first volume value and a second volume value, and the first volume value and the second volume value sequentially reflect the volume of each marine life audio feature in the first voice information and the second voice information at the same time.

[0019] A volume comparison unit, wherein the volume comparison unit compares the first volume value and the second volume value, and if the first volume value is greater than the second volume value, generates a first covering signal to the shielding subunit to shield the second output area; if the first volume value is less than the second volume value, generates a second covering signal to the shielding subunit to shield the first output area.

[0020] Furthermore, the output module also includes a longitudinal adjustment unit. When the first volume value and the second volume value are both smaller than a preset first volume threshold, an upward signal is generated to the longitudinal adjustment unit to expand the positioning area. When the first volume value and the second volume value are both larger than the preset first volume threshold, a downward signal is generated to the longitudinal adjustment unit to reduce the positioning area.

[0021] Furthermore, the lateral driving unit comprises an annular guide rail distributed on the bottom of the ship, and the indicating unit is driven by the annular guide rail.

[0022] Furthermore, the lighting subunit is configured as a lighting lamp and a lampshade mounted on the outside of the lighting lamp, the lampshade can enable the lighting lamp to form a circular lighting area, the shielding subunit includes a flexible shielding member and a plurality of driving members, the shielding member is installed in the lampshade, a light-transmitting hole is formed on the shielding member, and the driving member is arranged between the lampshade and the shielding member and is used to adjust the size and shape of the light-transmitting hole.

[0023] Furthermore, the longitudinal adjustment unit comprises a telescopic cylinder, and the telescopic cylinder adjusts the distance between the lampshade and the lighting lamp to change the size of the positioning area.

[0024] Furthermore, the marine life detection module is preset with an abnormal monitoring database, the abnormal monitoring database stores a number of abnormal sample models, and the marine life detection module includes an abnormal detection step, the abnormal detection step includes

[0025] Step 1: preprocessing the first voice information or the second voice information, wherein the preprocessing includes noise reduction processing to remove background noise;

[0026] Step 2: extract three features including multiple Mel frequency cepstral coefficients, resonance peaks and zero crossing rates;

[0027] Step 3: Match with each abnormal sample model to obtain the matching degree under each model;

[0028] Step 4: compare the matching degree with a threshold value. If the matching degree is greater than the threshold value, it indicates that there are marine life audio features.

[0029] Furthermore, the abnormal sample model is obtained by training the three features of Mel frequency cepstral coefficient, resonance peak and zero crossing rate as inputs of the neural network.

[0030] Furthermore, the audio acquisition module includes a random control unit, which controls each of the audio acquisition units to work synchronously within a preset fixed time period, and controls each of the audio acquisition units to work in a random order between two adjacent fixed time periods, and the random order conforms to the law of a random function.

[0031] Compared with the prior art, the beneficial effects of this solution are:

[0032] 1. By arranging multiple audio collection units in the bottom of the ship, the multiple audio collection units are used to collect audio data, and the seabed environment is monitored in real time. In combination with the marine life detection module, it is detected whether the first voice information and the second voice information contain marine life audio features. If they do, it indicates that there are marine life in the corresponding first area or second area, and the output unit is used to output the positioning area, so as to provide warnings to underwater operators;

[0033] 2. By setting up multiple audio acquisition units, it is convenient to preliminarily obtain the range of the marine life, and the positional relationship between adjacent audio acquisition units and the overlapping areas are used to reduce the range, which is convenient for locating the marine life and improving the accuracy of positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a structural schematic diagram of an embodiment of the present invention;

[0035] Figure 2 Schematic diagram of the distribution of acquisition units in an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of the structure of the output module in the embodiment of the present invention.

[0037] The reference numerals in the drawings of the specification include:

[0038] 1. Audio acquisition module; 11. Audio acquisition unit; 111. First acquisition unit; 112. Second acquisition unit; 113. First area; 114. Second area; 12. Random control unit; 2. Marine life detection module; 3. Processing module; 4. Output module; 41. Indication unit; 411. Illumination subunit; 412. Shielding subunit; 42. Lateral drive unit; 43. Longitudinal adjustment unit; 431. Telescopic cylinder; 5. Lampshade; 6. Shielding piece; 601. Light-transmitting hole; 7. Area adjustment module; 71. Volume acquisition unit; 72. Volume comparison unit. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0040] It should be noted that when a component is referred to as being "fixed to" another component, it may be directly on the other component or there may also be a component centered. When a component is considered to be "connected to" another component, it may be directly connected to the other component or there may also be a component centered. When a component is considered to be "set on" another component, it may be directly set on the other component or there may also be a component centered. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of processing modules, one or more of the related listed items.

[0042] Please also see Figure 1 to Figure 2 This embodiment provides a maritime audio recognition and analysis system, including an audio acquisition module 1, a marine life detection module 2, a processing module 3, an output module 4 and a regional adjustment module 7.

[0043] like Figure 1 and Figure 2 As shown, the audio collection module 1 includes a plurality of audio collection units 11, each of which has a preset positional relationship, and each of which has a preset collection area. The collection area reflects the range of audio that the corresponding audio collection unit 11 can actually collect. When the operating equipment is in a stationary state, that is, when the ship is stationary, the audio collection unit starts to work. Please refer to Figure 2In the figure, the circles represent the audio collection units 11, and the circular area represents the collection area. There is an intersection between the collection areas corresponding to adjacent audio collection units 11. When one of the audio collection units 11 collects the ambient sound, the audio collection unit 11 is defined as the first collection unit 111. The first collection unit 111 collects the ambient sound according to the preset time length and generates the first voice information. The audio collection unit 11 that has an intersection with the collection area of ​​the first collection unit 111 is defined as the second collection unit 112. When the first collection unit 111 collects the ambient sound, the second collection unit 112 synchronously collects the ambient sound and generates the second voice information.

[0044] The audio acquisition module 1 includes a random control unit 12, which controls each audio acquisition unit 11 to work synchronously within a preset fixed time period. The random control unit 12 controls each audio acquisition unit 11 to work in a random order between two adjacent fixed time periods, and the random order conforms to the law of random functions. This setting is mainly to reduce the workload of the system. Each audio acquisition unit 11 is synchronously controlled to perform acquisition work during a fixed time period, and a comprehensive inspection of the water body at the bottom of the hull is performed to ensure full coverage. The fixed time period can be set to the moment of maritime operation time, and then each audio acquisition unit 11 is controlled to work randomly, reducing the workload of the system. The use of random functions can ensure the randomness between each audio acquisition unit 11.

[0045] The marine life detection module 2 obtains the first voice information and determines whether the first voice information contains marine life audio features. The marine life audio features indicate that the animal has corresponding marine life. If the marine life audio features are contained, a first marine life signal is generated. The marine life detection module obtains and determines whether each second voice information contains marine life audio features. If so, a second marine life signal is generated.

[0046] like Figure 1 and Figure 2 As shown, the processing module 3 includes a first positioning unit and a second positioning unit. The first positioning unit receives the first marine biological signal and obtains the corresponding collection area to be defined as the first area 113, and generates a first positioning signal. When the second positioning unit receives the second marine biological signal, it obtains the corresponding collection area to be defined as the second area 114, and generates a second positioning signal.

[0047] The output module 4 is used to receive the first positioning signal and the second positioning signal to output the positioning area. When the output module 4 receives the first positioning signal and the second positioning signal, the positioning area represents the intersection of the first area 113 and each second area 114. When the output module 4 only receives the first positioning signal, the positioning area represents the range of the first area 113 excluding the second area 114.

[0048] like Figure 3 As shown, the output module 4 includes an indication unit 41 and a lateral drive unit 42. The indication unit 41 includes an illumination subunit 411 and a shielding subunit 412. The illumination subunit 411 is used to illuminate and indicate the first area 113, and the shielding subunit 412 can shield and adjust the shielding range on the first area 113 to obtain the positioning area. The lateral drive unit 42 is used to drive the indication unit 41 to move horizontally to the target area, and the target area represents the actual position of the first area 113 in the breeding circle.

[0049] In the embodiment, when the output module 4 receives the first positioning signal and the second positioning signal, in order to further narrow the scope of the positioning area, a region adjustment module 7 is provided to achieve the purpose of outputting a positioning area with a smaller area, so that a more accurate output result can be obtained. The positioning area is Figure 3 The intersection of the first area 113 and the second area 114, the positioning area can be divided into a first output area and a second output area according to the distance from the first acquisition unit 111 and the second acquisition unit 112, wherein the distance from the first output area to the first acquisition unit 111 is less than the distance from the second output area to the first acquisition unit 111, and the distance from the first output area to the second acquisition unit 112 is greater than the distance from the second output area to the second acquisition unit 112, which is equivalent to the left half of the intersection area being the first output area, and the right half of the intersection area being the second output area. This setting can reduce the area of ​​the output area by half. Specifically, the area adjustment module 7 includes a volume acquisition unit 71 and a volume comparison unit 72. The volume acquisition unit 71 corresponds to the audio acquisition unit 11 one by one and is used to acquire volume data. The volume data acquired by the volume acquisition unit 71 corresponding to the first acquisition unit 111 and the second acquisition unit 112 are defined as the first volume value and the second volume value in sequence. The first volume value and the second volume value reflect the volume of each marine biological audio feature in the first voice information and the second voice information at the same time in sequence. The volume can reflect the distance of the sound from the first acquisition unit 111 and the second acquisition unit 112. The volume comparison unit 72 compares the first volume value and the second volume value. If the first volume value is greater than the second volume value, a first covering signal is generated to the shielding subunit 412 to shield the second output area. If the first volume value is less than the second volume value, a second covering signal is generated to the shielding subunit 412 to shield the first output area.

[0050] like Figure 1 and Figure 3As shown, the output module 4 also includes a longitudinal adjustment unit 43. When the first volume value and the second volume value are both less than the preset first volume threshold, it means that the sound generation position is far from the first acquisition unit 111 and the second acquisition unit 112, and an up signal is generated to the longitudinal adjustment unit 43 to expand the positioning area, thereby improving the accuracy. When the first volume value and the second volume value are both greater than the preset first volume threshold, it means that the sound generation position is close to the first acquisition unit 111 and the second acquisition unit 112, and a down signal is generated to the longitudinal adjustment unit 43 to reduce the positioning area.

[0051] like Figure 3 As shown, the transverse driving unit 42 includes an annular guide rail (not shown) distributed on the bottom of the ship, and the indicating unit 41 is driven by a first linear slide rail 421. The lighting subunit 411 is configured as a lighting lamp and a lampshade 5 sleeved on the outside of the lighting lamp, and the lampshade 5 can make the lighting lamp form a circular lighting area. The shielding subunit 412 includes a flexible shielding member 6 and a plurality of driving members. The shielding member 6 is installed in the lampshade 5, and a light-transmitting hole 601 is formed on the shielding member 6. The driving member is arranged between the lampshade 5 and the shielding member 6 and is used to adjust the size and shape of the light-transmitting hole 601. The longitudinal adjustment unit 43 includes a telescopic cylinder 431, and the telescopic cylinder 431 adjusts the distance between the lampshade 5 and the lighting lamp to change the size of the positioning area.

[0052] The marine life detection module 2 is preset with an abnormal monitoring database, which stores a number of abnormal sample models. The marine life detection module 2 includes an abnormal detection step, which includes:

[0053] Step 1: preprocessing the first voice information or the second voice information, wherein the preprocessing includes noise reduction processing to remove background noise, and this setting can reduce the interference of background noise;

[0054] Step 2: extract three features including multiple Mel frequency cepstral coefficients, resonance peaks and zero crossing rates;

[0055] Step 3: Match with each abnormal sample model to obtain the matching degree under each model;

[0056] Step 4: compare the matching degree with the threshold. If the matching degree is greater than the threshold, it indicates that there are marine life audio features.

[0057] The abnormal sample model is trained by using three features, namely Mel-frequency cepstral coefficient, resonance peak and zero-crossing rate, as inputs of the neural network.

[0058] The above are only preferred embodiments of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A maritime audio recognition and analysis system, characterized in that: include An audio collection module, the audio collection module (1) comprising a plurality of audio collection units (11), each of the audio collection units (11) being preset with a positional relationship, and each of the audio collection units (11) being preset with a collection area, the collection area reflecting the range of the audio that the corresponding audio collection unit (11) can actually collect, when the operating equipment is in a stationary state, the audio collection module starts to work, when one of the audio collection units (11) collects ambient sound, the audio collection unit (11) is defined as a first collection unit (111), the first collection unit (111) collects ambient sound according to a preset time length and generates first voice information, the audio collection unit (11) having an intersection with the collection area of ​​the first collection unit (111) is defined as a second collection unit (112), when the first collection unit (111) collects ambient sound, the second collection unit (112) synchronously collects ambient sound and generates second voice information; A marine life detection module (2), wherein the marine life detection module (2) is configured with a marine life audio database, wherein the marine life audio database stores a plurality of marine life audio features, wherein the first voice information is obtained and it is determined whether the first voice information contains a marine life audio feature, wherein the marine life audio feature indicates the presence of a corresponding marine life, and if the marine life audio feature is contained, a first marine life signal is generated, wherein the marine life detection module obtains and determines whether each of the second voice information contains the marine life audio feature, and if so, a second marine life signal is generated; A processing module (3), wherein the processing module (3) comprises a first positioning unit, wherein the first positioning unit receives the first marine biological signal and obtains a corresponding collection area to define as a first area (113), and generates a first positioning signal, a second positioning unit, when the second positioning unit receives the second marine biological signal, obtains a corresponding collection area to define as a second area (114), and generates a second positioning signal; An output module (4), the output module (4) is used to receive the first positioning signal and the second positioning signal to output a positioning area. When the output module (4) receives the first positioning signal and the second positioning signal, the positioning area represents the intersection of the first area (113) and each of the second areas (114). When the output module (4) only receives the first positioning signal, the positioning area represents the range of the first area (113) excluding the second area (114).

2. The maritime audio recognition and analysis system according to claim 1, characterized in that: The output module (4) comprises An indication unit (41), the indication unit (41) comprising an illumination subunit (411), the illumination subunit (411) being used to illuminate and indicate the first area (113), a shielding subunit (412), the shielding subunit (412) being capable of shielding the first area (113) and adjusting the shielding range to obtain the positioning area; A lateral driving unit (42), the lateral driving unit (42) is used to drive the indicating unit (41) to move horizontally to a target area, the target area representing the actual position of the first area (113) in the ocean.

3. The maritime audio recognition and analysis system according to claim 2, characterized in that: The positioning area can be divided into a first output area and a second output area according to the distance from the first acquisition unit (111) and the second acquisition unit (112), wherein the distance from the first output area to the first acquisition unit (111) is smaller than the distance from the second output area to the first acquisition unit (111), and the distance from the first output area to the second acquisition unit (112) is larger than the distance from the second output area to the second acquisition unit (112), and the maritime audio recognition and analysis system further comprises a region adjustment module (7), wherein the region adjustment module (7) comprises A volume acquisition unit (71), the volume acquisition unit (71) corresponds to the audio acquisition unit (11) one by one and is used to acquire volume data, and the volume data acquired by the volume acquisition unit (71) corresponding to the first acquisition unit (111) and the second acquisition unit (112) are defined in sequence as a first volume value and a second volume value, the first volume value and the second volume value sequentially reflecting the volume of each marine organism audio feature in the first voice information and the second voice information at the same time. A volume comparison unit (72), wherein the volume comparison unit (72) compares the first volume value and the second volume value, and if the first volume value is greater than the second volume value, generates a first covering signal to the shielding subunit (412) to shield the second output area, and if the first volume value is less than the second volume value, generates a second covering signal to the shielding subunit (412) to shield the first output area.

4. The maritime audio recognition and analysis system according to claim 3, characterized in that: The output module (4) further comprises a longitudinal adjustment unit (43), which generates an up signal to the longitudinal adjustment unit (43) to expand the positioning area when the first volume value and the second volume value are both less than a preset first volume threshold, and generates a down signal to the longitudinal adjustment unit (43) to reduce the positioning area when the first volume value and the second volume value are both greater than the preset first volume threshold.

5. The maritime audio recognition and analysis system according to claim 4, characterized in that: The transverse driving unit (42) comprises an annular guide rail distributed on the bottom of the ship, and the indicating unit (41) is driven by the annular guide rail.

6. The maritime audio recognition and analysis system according to claim 4, characterized in that: The lighting subunit (411) is configured as a lighting lamp and a lampshade (5) mounted on the outside of the lighting lamp, wherein the lampshade (5) enables the lighting lamp to form a circular lighting area, and the shielding subunit (412) comprises a flexible shielding member (6) and a plurality of driving members, wherein the shielding member (6) is mounted in the lampshade (5), and a light-transmitting hole (601) is formed on the shielding member (6), and the driving member is arranged between the lampshade (5) and the shielding member (6) and is used to adjust the size and shape of the light-transmitting hole (601).

7. The maritime audio recognition and analysis system according to claim 6, characterized in that: The longitudinal adjustment unit (43) comprises a telescopic cylinder (431), and the telescopic cylinder (431) adjusts the distance between the lampshade (5) and the lighting lamp to change the size of the positioning area.

8. The maritime audio recognition and analysis system according to claim 1, characterized in that: The marine life detection module (2) is preset with an abnormal monitoring database, the abnormal monitoring database stores a number of abnormal sample models, and the marine life detection module (2) includes an abnormal detection step, the abnormal detection step includes Step 1: preprocessing the first voice information or the second voice information, wherein the preprocessing includes noise reduction processing to remove background noise; Step 2: extract three features including multiple Mel frequency cepstral coefficients, resonance peaks and zero crossing rates; Step 3: Match with each abnormal sample model to obtain the matching degree under each model; Step 4: compare the matching degree with a threshold value. If the matching degree is greater than the threshold value, it indicates that there are marine life audio features.

9. The maritime audio recognition and analysis system according to claim 8, characterized in that: The abnormal sample model is obtained by training the three features of Mel frequency cepstral coefficient, resonance peak and zero crossing rate as inputs of the neural network.

10. The maritime audio recognition and analysis system according to claim 1, characterized in that: The audio acquisition module (1) comprises a random control unit (12), wherein the random control unit (12) controls each of the audio acquisition units (11) to work synchronously within a preset fixed time period, and the random control unit (12) controls each of the audio acquisition units (11) to work in a random order between two adjacent fixed time periods, wherein the random order conforms to a random function rule.