Underwater organism intelligent monitoring method and system and electronic equipment

Abstract

The invention relates to the technical field of underwater organism intelligent monitoring scheme design, in particular to an underwater organism intelligent monitoring method and system and electronic equipment. The invention provides an efficient, accurate and intelligent underwater organism monitoring method and system and electronic equipment. According to the scheme, underwater biological species identification and quantity statistics are completed only through innovative processing of underwater sound signals, the precision of underwater biological species identification and quantity statistics is extremely high, dynamic changes of the underwater ecological environment can be monitored in real time, and timely decision support is provided for ecological protection and fishery management through an alarm mechanism. Compared with the prior art, the system has higher intelligent degree, practicability, reliability and expansibility, has a wide application prospect, and reduces the application cost of an underwater biological monitoring system to a great extent.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater organism intelligent monitoring scheme design, and in particular to an underwater organism intelligent monitoring method and system, and electronic equipment. Background Art

[0002] Underwater ecosystems are one of the most important ecosystems on Earth. The diversity and quantity of underwater organisms are of great significance for maintaining ecological balance. However, with the intensification of human activities (such as overfishing, pollution and climate change), the habitats and populations of underwater organisms are under serious threat. In order to protect underwater biological resources, timely monitoring of the species and quantity of underwater organisms has become an important task in ecological research, fishery management and environmental protection.

[0003] Traditional biological monitoring methods mainly rely on image processing. Image processing methods often require high-definition cameras, which are often expensive. In addition, a camera has a very limited field of view and is easily affected by light and floating sand underwater, resulting in unreliable results. In addition, due to the complex underwater terrain, organisms will block each other. It is difficult to accurately monitor the number of underwater organisms using image processing solutions, and the error is very large. In addition, existing equipment usually only provides monitoring data output and lacks intelligent alarm mechanisms for abnormal situations (such as abnormal increase or decrease in the number of certain organisms).

[0004] Therefore, there is an urgent need for an intelligent underwater biological monitoring system based on acoustic signals that can simultaneously identify and count the species of multiple underwater organisms and promptly alarm abnormal situations to meet the actual needs of ecological research, fishery management and environmental protection.

[0005] Therefore, the prior art needs to be further developed. Summary of the invention

[0006] The purpose of the present invention is to overcome the above technical deficiencies and provide an underwater biological intelligent monitoring method and system, and electronic equipment to solve the problems existing in the prior art.

[0007] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides an underwater organism intelligent monitoring method, comprising: S100, collecting underwater acoustic signals of a target water area through an acoustic sensor, wherein the acoustic signals include sound signals emitted by underwater organisms and background environmental noise signals; S200, filtering and noise reduction processing are performed on the collected underwater acoustic signal to remove the background environmental noise signal; S300, feature extraction step: extracting time domain and frequency domain features from the acoustic signal after filtering and noise reduction processing, the time domain features include the period, frequency distribution, duration and energy distribution of the signal, and the frequency domain features include spectrum characteristics, spectrum energy distribution and frequency change pattern; S400, acoustic feature classification step: inputting the extracted acoustic features into a pre-trained classification model, and determining the type of underwater organisms based on the classification results of the acoustic features; S500, acoustic pattern recognition step: based on the periodic characteristics and time series pattern of the underwater creatures' sounds, counting the sound frequency of each underwater creature in a unit time, and calculating the number of underwater creatures; S600, result output step: generate and output the monitoring results of the types and quantity of underwater organisms, determine whether the types and quantities meet the alarm conditions, if so, output the relevant alarm signal, if not, do not output the relevant alarm signal.

[0008] Specifically, the determining whether the type and quantity meet the alarm condition, if so, outputting the relevant alarm signal, if not, not outputting the relevant alarm signal, includes: If the number of the preset type of organisms is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, output an alarm signal to prompt the user to pay attention; If the number of the preset type of organisms is greater than or equal to a second preset threshold, outputting an alarm signal to prompt the user to take preliminary measures; If the total number of all organisms exceeds a third preset threshold, an alarm signal is output to prompt the user to take immediate measures; If the number of the preset type of organisms is less than the first preset threshold, no alarm signal is output.

[0009] Specifically, in step S200, the filtering and noise reduction processing includes using an adaptive filtering algorithm to suppress the background noise signal.

[0010] Specifically, in step S300, the feature extraction step includes extracting frequency domain features of the acoustic signal based on short-time Fourier transform or wavelet transform.

[0011] Specifically, in step S400, the classification model is a sound classification model based on deep learning, and the model is trained by a labeled underwater bioacoustic dataset.

[0012] Specifically, in step S500, the periodic characteristics include repetition interval time, pulse width and amplitude change of the underwater biological sound signal.

[0013] Specifically, in step S500, the time series pattern analysis includes modeling and identifying the sound patterns of underwater organisms based on a hidden Markov model.

[0014] Specifically, in step S600, the result output step includes transmitting the monitoring result to a remote terminal via a wireless communication module.

[0015] According to a second aspect of the present invention, there is provided an underwater organism intelligent monitoring system, comprising: Acoustic sensor module: used to collect underwater acoustic signals of the target water area through acoustic sensors, wherein the acoustic signals include sound signals emitted by underwater organisms and background environmental noise signals; Signal processing module: used to filter and reduce noise on the collected underwater acoustic signals to remove background environmental noise signals; A feature extraction module, used to extract time domain and frequency domain features from the acoustic signal after filtering and noise reduction processing, wherein the time domain features include the period, frequency distribution, duration and energy distribution of the signal, and the frequency domain features include spectrum characteristics, spectrum energy distribution and frequency change pattern; Classification module: used to input the extracted acoustic features into the pre-trained classification model and determine the type of underwater organisms based on the classification results of the acoustic features; Quantity statistics module: used to count the frequency of sound produced by each underwater creature per unit time and calculate the number of underwater creatures based on the periodic characteristics and time series patterns of the sounds produced by underwater creatures; Result output module: used to generate and output the monitoring results of the types and quantity of underwater organisms, and determine whether the types and quantities meet the alarm conditions. If so, the relevant alarm signal is output; if not, the relevant alarm signal is not output.

[0016] According to a third aspect of the present invention, there is provided an electronic device, comprising: a memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned underwater biological intelligent monitoring method is implemented.

[0017] Beneficial effects: The present invention proposes an efficient, accurate and intelligent underwater biological monitoring method, system and electronic equipment. This solution realizes the identification and quantity statistics of underwater biological species only through innovative processing of underwater acoustic signals, and the accuracy of underwater biological species identification and quantity statistics is extremely high. It can monitor the dynamic changes of the underwater ecological environment in real time, and provide timely decision-making support for ecological protection and fishery management through an alarm mechanism. Compared with the existing technology, the present invention has a higher degree of intelligence, practicality, reliability and scalability, has broad application prospects, and greatly reduces the application cost of underwater biological monitoring systems. It has the following significant beneficial effects: 1. Efficient underwater species identification: The present invention can efficiently and accurately identify a variety of underwater organisms by extracting time and frequency domain features of underwater acoustic signals and using a deep learning model for classification. Experimental results show that the accuracy of the present invention in identifying underwater organism species can reach more than 90%, which is significantly higher than the recognition accuracy of traditional methods.

[0018] 2. Accurate biological population statistics: By analyzing the periodic characteristics and time series patterns of underwater biological sound signals, the present invention can accurately distinguish the sound pulses of a single organism from overlapping signals, thereby achieving accurate statistics of the number of different types of underwater organisms. Compared with traditional acoustic equipment, the statistical error of the present invention is reduced to less than 5%.

[0019] 3. Real-time alarm mechanism: The present invention can generate real-time alarm signals according to preset alarm conditions (such as the number of a certain organism exceeding a threshold or the total number being abnormal), and output alarm information in local and remote ways. The introduction of the alarm mechanism can help users to timely discover ecological anomalies (such as the spread of invasive species or changes in the number of endangered species), so as to take corresponding protective measures.

[0020] 4. Non-invasive and eco-friendly: The present invention performs monitoring based on acoustic signals, does not need to contact or capture underwater organisms, avoids damage to underwater organisms and habitats, and meets the requirements of ecological protection.

[0021] 5. Low cost and high reliability: The present invention optimizes the signal processing algorithm and data transmission method, and completes the underwater biological species identification and quantity statistics only through innovative processing of underwater acoustic signals, which significantly reduces the hardware cost and operating energy consumption of the system. There is no need to monitor underwater organisms through image processing, which greatly improves the reliability of the system and greatly reduces the application cost of the underwater biological monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of the underwater organism intelligent monitoring method provided in a specific embodiment of the present invention; Figure 2 It is a schematic diagram of the system composition of the underwater biological intelligent monitoring system provided in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0024] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0025] See also Figure 1 The present invention provides an underwater biological intelligent monitoring method, comprising: S100: Collect underwater acoustic signals of a target water area through an acoustic sensor, where the acoustic signals include sound signals emitted by underwater organisms and background environmental noise signals.

[0026] Specifically, the determining whether the type and quantity meet the alarm condition, if so, outputting the relevant alarm signal, if not, not outputting the relevant alarm signal, includes: If the number of the preset type of organisms is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, output an alarm signal to prompt the user to pay attention; If the number of the preset type of organisms is greater than or equal to a second preset threshold, outputting an alarm signal to prompt the user to take preliminary measures; If the total number of all organisms exceeds a third preset threshold, an alarm signal is output to prompt the user to take immediate measures; If the number of the preset type of organisms is less than the first preset threshold, no alarm signal is output.

[0027] In a preferred embodiment of the present invention, the following alarm triggering conditions are defined to determine whether the monitoring results need to trigger an alarm signal: Species alarm: If the number of a certain underwater organism detected exceeds the set threshold , then the type alarm is triggered.

[0028] The formula is as follows: Example: Set the alarm threshold for fish A to , and the monitoring result shows that the number of fish A is 15, then the species alarm is triggered. It can be understood that the alarm threshold of fish A can be specifically set according to the actual needs of the user of the present invention, as long as it is applicable to the underwater biological intelligent monitoring method proposed by the present invention. The alarm threshold of fish A is set to , which is obtained by the technicians of the present invention through a large number of tests and can well implement the type of alarm scheme described in the present invention.

[0029] Reason: Abnormal numbers of certain underwater organisms (such as invasive species or endangered species) may pose a threat to the ecosystem and require timely alarm.

[0030] Quantity alarm: If the total number of all underwater organisms detected Exceeding the set threshold , then the quantity alarm is triggered.

[0031] The formula is as follows: Example: Set the total quantity threshold to , and the monitoring result shows that the total number is 60, then the number alarm is triggered. It can be understood that the total number threshold can be specifically set according to the actual needs of the user of the present invention, as long as it is applicable to the underwater biological intelligent monitoring method proposed by the present invention. The total number threshold is set to , which is obtained by the technicians of the present invention through a large number of tests and can well implement the quantity alarm solution described in the present invention.

[0032] Rationale: Unusual fluctuations in the abundance of underwater organisms may indicate changes in water quality, environmental stress, or other ecological problems.

[0033] Time window alarm: If the number of a certain organism exceeds the threshold for multiple consecutive time windows , a continuous alarm is triggered.

[0034] Method: Record the number in each time window , if continuous In the time window , then trigger the alarm.

[0035] Example: Setting , and the number of fish A exceeds 10 in three consecutive 10-second windows, a continuous alarm is triggered.

[0036] Reason: Continuous alarm can avoid false alarm of single abnormal data. It is understandable that The size of the time window can be specifically set according to the actual needs of the user according to the present invention, as long as it is applicable to the underwater biological intelligent monitoring method proposed by the present invention. , setting the time window to 10 seconds is obtained by the technicians of the present invention through a large number of tests, and can well implement the continuous alarm solution described in the present invention.

[0037] It should be noted here that the acoustic sensor uses a directional microphone array, and the preferred model is HTI-96-MIN, which is characterized by high sensitivity and wide-band response (10 Hz to 150 kHz), and can effectively collect low-frequency to high-frequency acoustic signals underwater.

[0038] It should be noted here that the sensor selection: the high sensitivity and wide frequency response of the HTI-96-MIN microphone array can capture weak signals from underwater organisms.

[0039] Understandably, regarding high frequency signals (>1 kHz): High-frequency signals attenuate quickly. The typical detection radius of the HTI-96-MIN microphone array for high-frequency signals is usually in the range of 10 to 50 meters, which is suitable for close-range monitoring of sound signals of small organisms (such as fish).

[0040] It is understandable that regarding intermediate frequency signals (100 Hz to 1 kHz): The propagation distance of intermediate frequency signals is relatively long. The typical detection radius of the HTI-96-MIN microphone array for intermediate frequency signals is usually in the range of 50 meters to 200 meters, which is suitable for monitoring the sound signals of medium-sized creatures (such as dolphins).

[0041] Understandably, regarding low frequency signals (<100 Hz): Low-frequency signals have the longest propagation distance. The typical detection radius of the HTI-96-MIN microphone array for low-frequency signals can reach one kilometer, which is suitable for monitoring the sound signals of large organisms (such as whales).

[0042] Sampling rate: A sampling rate of 200 kHz is able to cover the biological sound range of 10 Hz to 20 kHz while avoiding aliasing effects.

[0043] Model parameters: The architecture and hyperparameters of CNN have been verified through multiple experiments and can achieve high classification accuracy (>90%) with a limited number of samples.

[0044] Environmental adaptation: The automatic parameter adjustment mechanism can adapt to different water depths and environmental noise conditions, improving the robustness of the system.

[0045] Array configuration: An array consisting of 4 microphones with a microphone spacing of 10 cm, an array center frequency of 50 kHz, and a directional gain of 8 dB, which can enhance the directionality of the target sound source and suppress the surrounding environmental noise.

[0046] Deployment location: The acoustic sensor is deployed at a depth of 10 to 20 m underwater to avoid interference from near-surface noise (such as waves and water flow) while ensuring that the signal covers the target waters.

[0047] It should be noted here that the sampling rate is set to 200 kHz, which can capture the high-frequency sound signals of underwater creatures (for example, the frequency range of fish calls is 20 Hz to 150 Hz, while the frequency of whale calls can reach 10 kHz to 50 kHz).

[0048] Sampling time: Each acquisition time is 10 seconds to ensure the capture of the complete biological sound signal cycle.

[0049] Sampling interval: The interval between two adjacent acquisitions is 2 seconds to avoid signal overlap.

[0050] S200: Filter and perform noise reduction processing on the collected underwater acoustic signal to remove background environmental noise signal.

[0051] Specifically, in step S200, the filtering and noise reduction processing includes using an adaptive filtering algorithm to suppress the background noise signal.

[0052] It should be noted here that an adaptive filtering algorithm (such as LMS algorithm, least mean square algorithm) is used to reduce the noise of the signal. The specific parameters are as follows: Filter order: set to 16; Step size parameter (μ): set to 0.01, which strikes a balance between reducing noise and avoiding divergence; Noise reference signal: The background noise signal is obtained through additionally deployed environmental noise sensors (such as pressure sensors or accelerometers) as a reference signal for adaptive filtering.

[0053] S300, feature extraction step: extracting time domain and frequency domain features from the acoustic signal after filtering and noise reduction processing, the time domain features include the period, frequency distribution, duration and energy distribution of the signal, and the frequency domain features include spectrum characteristics, spectrum energy distribution and frequency change pattern.

[0054] Specifically, in step S300, the feature extraction step includes extracting frequency domain features of the acoustic signal based on short-time Fourier transform or wavelet transform.

[0055] It should be noted here that: Perform short-time Fourier transform (STFT) on the filtered signal to extract time-frequency domain features: Window function: Use the Hamming Window with a window length of 256 points; Overlap rate: Set to 50% to ensure a balance between time resolution and frequency resolution; Frequency Range: The frequency range of 10 Hz to 20 kHz is screened to remove invalid high-frequency noise and low-frequency water flow interference.

[0056] Calculate the signal-to-noise ratio (SNR) of the signal using the following formula: in, is the signal power, If the SNR is lower than 10 dB, the signal will be marked as a low-quality signal and re-acquired.

[0057] Furthermore, the following time domain features are extracted: Period: Calculates the repetition period of the signal pulse, which is used to identify organisms that make periodic sounds (such as whales); Duration: The duration of a single pulse, used to distinguish short pulses (such as fish calls) from long pulses (such as whale songs); Energy distribution: The strength of the signal is measured by the root mean square (RMS) value, using the formula: in, For the The amplitude value of the sampling points, is the total number of sampling points.

[0058] Furthermore, the following frequency domain features are extracted: Spectral peak: The frequency component with the largest energy in the spectrum, used to identify the main vocal frequency of the organism; Spectral bandwidth: The frequency range of the spectrum peaks, used to distinguish between narrowband and broadband vocalization organisms; Spectral entropy: measures the complexity of spectrum distribution, the formula is: in, For frequency The normalized power spectral density at .

[0059] Furthermore, principal component analysis (PCA) is used to reduce the dimension of the extracted features, retaining the first 90% of the variance information and reducing the computational complexity.

[0060] S400, acoustic feature classification step: input the extracted acoustic features into a pre-trained classification model, and determine the type of underwater organisms based on the classification results of the acoustic features.

[0061] Specifically, in step S400, the classification model is a sound classification model based on deep learning, and the model is trained by a labeled underwater bioacoustic dataset.

[0062] Specifically, the method includes: A deep convolutional neural network (CNN) is used to classify the extracted acoustic features. The model architecture is as follows: Input layer: input 128-dimensional features (obtained after PCA dimensionality reduction); Convolutional layer 1: 32 3×3 convolution kernels, using ReLU activation function; Pooling layer 1: 2×2 maximum pooling layer; Convolutional layer 2: 64 3×3 convolution kernels, ReLU activation function; Pooling layer 2: 2×2 maximum pooling layer; Fully connected layer 1: 128 neurons, ReLU activation function; Output layer: N neurons (corresponding to N types of underwater creatures), using Softmax activation function.

[0063] Specifically, the training process of the classification model of the present invention includes: Dataset: Use an annotated underwater biological acoustic dataset (such as Fish Call DB or Marine MammalSound Library). The dataset includes acoustic samples of 10 types of underwater organisms, with no less than 500 samples of each type. Data enhancement: Enhance data diversity by adding white noise and spectrum shifting; Hyperparameters: Batch Size: 32 Learning rate: 0.001; Optimizer: Adam; Epochs: 50 Loss function: cross entropy loss function, the formula is: in, is the true label, is the predicted probability.

[0064] Furthermore, the method of species statistics includes: based on the output of the classification module, counting the identification results of each underwater organism. Threshold screening is performed on the predicted probability output by the classification model. If the predicted probability of a certain type of organism is Greater than the set threshold (like ), then this type of organism will be included in the statistical results.

[0065] The formula is as follows: Category List Reason: Setting a threshold can prevent low-confidence predictions from interfering with the results. The present invention is obtained by the technicians through a large number of tests, and can well implement the underwater biological intelligent monitoring method of the present invention and improve the reliability of the monitoring results.

[0066] S500, acoustic pattern recognition step: based on the periodic characteristics and time series patterns of the sounds made by underwater creatures, the sound frequency of each underwater creature in a unit time is counted, and the number of underwater creatures is calculated.

[0067] Specifically, in step S500, the periodic characteristics include repetition interval time, pulse width and amplitude change of the underwater biological sound signal.

[0068] Specifically, in step S500, the time series pattern analysis includes modeling and identifying the sound patterns of underwater organisms based on a hidden Markov model.

[0069] Further, the method of periodic characteristic analysis includes: Model selection: Hidden Markov model (HMM) is used to model the periodic characteristics of underwater organisms' sounds.

[0070] Parameter settings: State number: set to 3, representing the states of voice, intermittent and silent respectively; Observation vector: energy and spectral characteristics of the acoustic signal; Model training: Parameter estimation is performed using the Baum-Welch algorithm.

[0071] Furthermore, the method of time series pattern recognition includes: Model selection: Long short-term memory network (LSTM) is used to model the time series of vocalization patterns.

[0072] Parameter settings: Number of hidden layer units: 64; Time window length: 10 seconds; Objective: To predict the period and frequency variation patterns of vocalizations.

[0073] Specifically, the present invention also includes an environment adaptation step: Depth adaptation: The frequency response range of the acoustic sensor is adjusted according to the water depth. For every 10 m increase in water depth, the frequency response range is reduced by 10% (e.g. 100 kHz to 90 kHz) to compensate for the attenuation of high-frequency signals in water.

[0074] Noise adaptation: Dynamically adjust filter parameters (such as step size μ and filter order) according to the ambient noise level (SNR) to ensure noise suppression effect.

[0075] Furthermore, the present invention uses the least squares method (LSE) to optimize the parameters of the acoustic sensor, with the goal of minimizing the prediction error: in, is the predicted value, is the true value.

[0076] Specifically, the method of quantitative statistics includes: based on the output result of the acoustic pattern recognition module, counting the sound frequency of each underwater creature in unit time.

[0077] Method: Perform time series analysis on the vocalization pulses of each species, and calculate the number of vocalizations Fi per unit time (such as 10 seconds) as the estimated number of this type of organism.

[0078] The formula is as follows: Fi = pulse count / time interval Reason: The periodic characteristics and time series patterns of acoustic signals can accurately estimate the number of organisms and avoid the errors of direct counting.

[0079] S600, result output step: generate and output the monitoring results of the types and quantity of underwater organisms, determine whether the types and quantities meet the alarm conditions, if so, output the relevant alarm signal, if not, do not output the relevant alarm signal.

[0080] Specifically, in step S600, the result output step includes transmitting the monitoring result to a remote terminal via a wireless communication module.

[0081] Furthermore, the present invention outputs monitoring results every 10 seconds, including the species and quantity of detected organisms; Furthermore, the present invention generates reports on a daily basis, including the changing trends of biological distribution and quantity.

[0082] The present invention adopts ZigBee wireless communication (2.4 GHz frequency band) to transmit data to the remote terminal, with a transmission rate of 250 kbps and a transmission range of 100 m.

[0083] See also Figure 2 The present invention provides another embodiment, which provides an underwater organism intelligent monitoring system, the underwater organism intelligent monitoring system comprising: Acoustic sensor module 100: used to collect underwater acoustic signals of the target water area through an acoustic sensor, wherein the acoustic signals include sound signals emitted by underwater organisms and background environmental noise signals; Signal processing module 200: used to filter and reduce noise on the collected underwater acoustic signals to remove background environmental noise signals; A feature extraction module 300 is used to extract time domain and frequency domain features from the acoustic signal after filtering and noise reduction processing, wherein the time domain features include the period, frequency distribution, duration and energy distribution of the signal, and the frequency domain features include spectrum characteristics, spectrum energy distribution and frequency change pattern; Classification module 400: used to input the extracted acoustic features into a pre-trained classification model, and determine the type of underwater organisms based on the classification results of the acoustic features; The number counting module 500 is used to count the frequency of sound production of each underwater organism in a unit time and calculate the number of underwater organisms based on the periodic characteristics and time series patterns of the underwater organisms' sound production; Result output module 600: used to generate and output monitoring results of the types and quantities of underwater organisms, determine whether the types and quantities meet the alarm conditions, and if so, output relevant alarm signals; if not, do not output relevant alarm signals.

[0084] In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the underwater biological intelligent monitoring method is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with the necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide the necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0085] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0086] It will be appreciated by a person skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0087] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0088] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for intelligent monitoring of underwater organisms, characterized in that: The method comprises: S100, collecting underwater acoustic signals of a target water area through an acoustic sensor, wherein the acoustic signals include sound signals emitted by underwater organisms and background environmental noise signals; S200, filtering and noise reduction processing are performed on the collected underwater acoustic signal to remove the background environmental noise signal; S300, feature extraction step: extracting time domain and frequency domain features from the acoustic signal after filtering and noise reduction processing, the time domain features include the period, frequency distribution, duration and energy distribution of the signal, and the frequency domain features include spectrum characteristics, spectrum energy distribution and frequency change pattern; S400, acoustic feature classification step: inputting the extracted acoustic features into a pre-trained classification model, and determining the type of underwater organisms based on the classification results of the acoustic features; S500, acoustic pattern recognition step: based on the periodic characteristics and time series pattern of the underwater creatures' sounds, counting the sound frequency of each underwater creature in a unit time, and calculating the number of underwater creatures; S600, result output step: generate and output the monitoring results of the types and quantity of underwater organisms, determine whether the types and quantities meet the alarm conditions, if so, output the relevant alarm signal, if not, do not output the relevant alarm signal.

2. The underwater organism intelligent monitoring method according to claim 1, characterized in that: The determining whether the type and quantity meet the alarm condition, if so, outputting the relevant alarm signal, if not, not outputting the relevant alarm signal, includes: If the number of the preset type of organisms is greater than or equal to the first preset threshold and less than or equal to the second preset threshold, output an alarm signal to prompt the user to pay attention; If the number of the preset type of organisms is greater than or equal to a second preset threshold, outputting an alarm signal to prompt the user to take preliminary measures; If the total number of all organisms exceeds a third preset threshold, an alarm signal is output to prompt the user to take immediate measures; If the number of the preset type of organisms is less than the first preset threshold, no alarm signal is output.

3. The underwater organism intelligent monitoring method according to claim 1, characterized in that: In step S200, the filtering and noise reduction processing includes using an adaptive filtering algorithm to suppress the background noise signal.

4. The underwater organism intelligent monitoring method according to claim 1, characterized in that: In step S300, the feature extraction step includes extracting frequency domain features of the acoustic signal based on short-time Fourier transform or wavelet transform.

5. The underwater organism intelligent monitoring method according to claim 1, characterized in that: In step S400, the classification model is a sound classification model based on deep learning, and the model is trained by a labeled underwater bioacoustic dataset.

6. The underwater organism intelligent monitoring method according to claim 1, characterized in that: In step S500, the periodic characteristics include the repetition interval time, pulse width and amplitude change of the underwater biological sound signal.

7. The underwater organism intelligent monitoring method according to claim 6, characterized in that: In step S500, the time series pattern analysis includes modeling and identifying the sound patterns of underwater organisms based on a hidden Markov model.

8. The underwater organism intelligent monitoring method according to claim 1, characterized in that: In step S600, the result output step includes transmitting the monitoring result to a remote terminal via a wireless communication module.

9. An underwater biological intelligent monitoring system, characterized in that: include: Acoustic sensor module: used to collect underwater acoustic signals of the target water area through acoustic sensors, wherein the acoustic signals include sound signals emitted by underwater organisms and background environmental noise signals; Signal processing module: used to filter and reduce noise on the collected underwater acoustic signals to remove background environmental noise signals; A feature extraction module, used to extract time domain and frequency domain features from the acoustic signal after filtering and noise reduction processing, wherein the time domain features include the period, frequency distribution, duration and energy distribution of the signal, and the frequency domain features include spectrum characteristics, spectrum energy distribution and frequency change pattern; Classification module: used to input the extracted acoustic features into the pre-trained classification model and determine the type of underwater organisms based on the classification results of the acoustic features; Quantity statistics module: used to count the frequency of sound produced by each underwater creature per unit time and calculate the number of underwater creatures based on the periodic characteristics and time series patterns of the sounds produced by underwater creatures; Result output module: used to generate and output the monitoring results of the types and quantity of underwater organisms, and determine whether the types and quantities meet the alarm conditions. If so, the relevant alarm signal is output; if not, the relevant alarm signal is not output.

10. An electronic device, characterized in that: include: Memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the underwater biological intelligent monitoring method according to any one of claims 1 to 8 is implemented.

Images