A fish density monitoring method based on big data and acoustic signals

Abstract

The present invention discloses a fish density monitoring method based on big data and acoustic wave signals, comprising: constructing an acoustic wave reflection source type identification model to determine the acoustic wave reflection source type and location of the acoustic wave reflection signal; identifying and eliminating multipath reflection signals from the same acoustic wave reflection source based on the fish school acoustic wave reflection signal; constructing a fish school density prediction model to predict the fish school density and determine the fish school activity index; judging the fish school health status based on the baseline fish school activity index threshold and the fish school activity index; judging the fish school disease category based on the appearance image of fish with abnormal health status; formulating fish school health intervention measures based on the fish school disease category and judging the effectiveness of the fish school health intervention measures. The present invention improves the real-time and accuracy of fish school density and fish school health monitoring, reduces risks in aquaculture, and significantly improves the efficiency and scientific nature of aquaculture management through automated water quality control and drug administration equipment.

Description

Technical Field

[0001] The present invention relates to the field of aquaculture monitoring, and in particular to a fish density monitoring method based on big data and acoustic wave signals. Background Art

[0002] In the aquaculture industry, real-time monitoring of fish density, health, and the environment in which they thrive is crucial for improving profitability and ensuring fish health. However, current monitoring methods have numerous limitations, resulting in inefficient and costly aquaculture management. In particular, fish health issues are often detected late, missing the optimal time for intervention. These management deficiencies not only impact aquaculture profitability but can also lead to risks such as large-scale fish die-offs and environmental degradation. Therefore, improving the accuracy and real-time nature of monitoring has become a pressing challenge for the industry. To address these issues, acoustic detection technology has become an increasingly important tool for fish monitoring. This technology relies on the propagation and reflection of sound waves in water, analyzing the reflected signals after they hit fish to determine their location and density. Compared to traditional manual detection methods, acoustic detection technology can provide more real-time information on fish distribution and reduce the delays associated with manual detection. However, despite significant progress in improving monitoring efficiency, practical applications still face numerous technical challenges. For example, the complex and dynamic underwater environment, influenced by factors such as water flow, temperature, and salinity, makes acoustic signals susceptible to interference. Multipath reflection of sound waves is particularly common in larger bodies of water, causing duplication and confusion in reflected signals, impacting detection accuracy. Multipath reflection signals refer to the multiple reflections generated when sound waves encounter obstacles or different layers of water. These signals, when superimposed, can easily lead to misjudgments of fish location and density, compromising data reliability. In addition to the multipath reflection issue, existing acoustic detection equipment can mostly only provide basic information on fish distribution and struggles to capture more complex dynamic characteristics of fish schools, such as their health, swimming speed, and behavioral changes. Existing technologies can mostly only monitor static fish density and are unable to effectively track real-time movement trends and activity indices. This results in a lack of timely warnings of abnormal fish behavior or health issues in the early stages. Furthermore, the processing accuracy and resolution of acoustic detection signals are limited by the equipment and algorithms, making it difficult to accurately predict changes in fish density. For large-scale, intensive aquaculture, this technological limitation leads to unscientific density management, which in turn affects fish health and breeding efficiency. Furthermore, fish health management relies primarily on manual inspections and periodic sampling, which are not only time-consuming and labor-intensive, but also suffer from significant latency. When fish develop health problems, external symptoms are often difficult to detect early. By the time the disease spreads, the optimal time for intervention is often too late. As aquaculture scales up, the limitations of manual detection methods become increasingly apparent, making fish health monitoring more difficult and hindering farmers from taking timely and effective measures to prevent the spread of disease. Although acoustic detection technology has been widely used in fish monitoring, it still faces issues such as multipath reflection signal interference, low monitoring data accuracy, insufficient real-time performance, and delayed health management.These problems seriously affect the decision-making efficiency and risk control in the aquaculture process. Therefore, a new technical means is urgently needed to improve the accuracy, real-time and automation level of fish monitoring to better support scientific aquaculture and environmental management. Summary of the Invention

[0003] The present invention addresses the problems existing in the above-mentioned prior art and provides a fish density monitoring method based on big data and acoustic wave signals, which mainly includes:

[0004] According to the sound wave reflection signal of the water area to be detected, a sound wave reflection source type identification model is constructed to determine the type of the sound wave reflection source of the sound wave reflection signal and the location of the sound wave reflection source;

[0005] Based on the acoustic wave reflection signals of the school of fish, the similarity of the spectral characteristics of the acoustic wave reflection signals of the school of fish is calculated, and the multipath reflection signals from the same acoustic wave reflection source are identified and eliminated;

[0006] Based on the echo width, spectral characteristics, phase difference and phase offset of the fish school acoustic reflection signal, a fish school density prediction model is constructed to predict the fish school density. The fish school activity index is determined based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish school in the water area.

[0007] The health status of the fish school is determined based on the baseline fish school activity index threshold and the fish school activity index. The drone-mounted camera is used to obtain images of the fish body appearance of fish with abnormal health status.

[0008] Based on images of fish with abnormal health status, identify fish behavioral characteristics and external symptoms, and determine the type of fish disease;

[0009] Based on the fish disease categories, fish health intervention measures are formulated and sent to automated water quality control equipment and automatic drug dosing equipment for fish health management. The fish activity index is monitored in real time to determine the effectiveness of the fish health intervention measures.

[0010] Furthermore, the method of constructing a sound wave reflection source type identification model based on the sound wave reflection signal of the water area to be detected, determining the sound wave reflection source type of the sound wave reflection signal, and determining the position of the sound wave reflection source includes:

[0011] The acoustic detection equipment emits acoustic waves to the water area to be detected according to the set acoustic wave emission parameters, and the directional array sensor obtains the acoustic wave reflection signal and the corresponding timestamp information; the acoustic wave reflection signal is separated according to the acoustic wave reflection signal of the water area to be detected by the independent component analysis algorithm to obtain the acoustic wave reflection signals of different acoustic wave reflection sources, and save them to the acoustic wave reflection monitoring database; the short-time Fourier transform algorithm is used to convert the time domain acoustic wave reflection signal into frequency domain data, and its spectral characteristics are extracted, which include frequency, frequency peak and amplitude; the historical acoustic wave emission frequency, amplitude, and spectral characteristics of the acoustic wave reflection signal are obtained through the acoustic wave reflection monitoring database, and the type of acoustic wave reflection source is marked. The support vector machine algorithm is used for model training to construct an acoustic wave reflection source type recognition model to determine the acoustic wave reflection source type of the acoustic wave reflection signal, including fish school acoustic wave reflection sources and non-fish school acoustic wave reflection sources; the echo time of the acoustic wave reflection signal of the fish school is obtained, and the position of the fish school relative to the sensor array is calculated through time difference estimation to determine the position of the fish school.

[0012] Furthermore, the method of calculating the similarity of the frequency spectrum characteristics of the fish school acoustic wave reflection signals based on the fish school acoustic wave reflection signals, and identifying and eliminating multipath reflection signals from the same acoustic wave reflection source, includes:

[0013] The acoustic wave reflection signals of fish schools are obtained through the acoustic wave reflection monitoring database. The similarity of the spectral characteristics of the acoustic wave reflection signals of fish schools is calculated using the method of calculating the Pearson correlation coefficient. If the similarity is greater than a preset similarity threshold and the echo time difference is less than a preset difference threshold, the acoustic wave reflection signals of the fish school are determined to be multipath reflection signals of the same acoustic wave reflection source; the multipath reflection signals in the same acoustic wave reflection source are eliminated, and the unique acoustic wave reflection signal of the acoustic wave reflection source is retained.

[0014] Furthermore, the fish density prediction model is constructed based on the echo width, spectral characteristics, phase difference and phase offset of the fish acoustic reflection signal to predict the fish density, and the fish activity index is determined based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish in the water area, including:

[0015] Based on the unique sound wave reflection signal from the sound wave reflection source, the phase difference and phase offset of the sound wave transmission signal are determined; based on the echo width, spectral characteristics, phase difference and phase offset of the sound wave reflection signal of the fish school, a recurrent neural network algorithm is used for model training to construct a fish school density prediction model to predict the fish school density; the fish school density is continuously monitored. If the fish school density change rate is higher than the preset density threshold, the drone-mounted camera is used to obtain the appearance image of the fish body and save it to the fish school appearance image database; if the fish school density change rate is lower than the preset density threshold, the echo signals at different time points are compared to judge the position change of the fish school and determine the swimming speed of the fish school; based on the swimming speed, density, swimming speed volatility and speed change rate of the fish school in the water area, the fish school activity index is determined.

[0016] It also includes determining a fish activity index based on the swimming speed, density, swimming speed fluctuation and speed change rate of fish schools in the water area, specifically including:

[0017] According to the swimming speed of the fish school in the water area, the fish school swimming speed fluctuation formula is used Calculate the fluctuation of fish swimming speed σ V (t), where V(t) is the swimming speed of the fish school at time t, is the average swimming speed in the time period T; based on the swimming speed, density, swimming speed volatility and speed change rate of the fish school in the water area, the fish school activity index formula is used Calculate the fish activity index FAI(t), where V(t) is the average swimming speed of the fish school, D(t) is the fish school density, and σ V (t) is the fluctuation of the swimming speed of the fish school, is the rate of change of the swimming speed of the fish school.

[0018] Furthermore, judging the health status of the fish school based on the benchmark fish school activity index threshold and the fish school activity index, and using the drone-mounted camera to obtain the appearance image of the fish school with abnormal health status, includes:

[0019] Based on the calculated fish activity index, the water environment parameters corresponding to the detection time of the fish activity index are obtained in real time through the water environment sensor and saved in the fish monitoring database. The water environment parameters include flow rate, dissolved oxygen content, water temperature, and turbidity; based on the water environment parameters obtained in real time, the baseline fish activity index threshold is adjusted; based on the baseline fish activity index threshold and the fish activity index, the health status of the fish is judged, and the fish health status is normal or abnormal; if the difference between the fish activity index and the baseline fish activity index threshold is greater than the preset difference threshold, the health status of the fish is judged to be abnormal, and the camera mounted on the drone is used to obtain the appearance image of the fish with abnormal health status, and save it to the fish appearance image database.

[0020] It also includes adjusting the baseline fish activity index threshold based on real-time water environment parameters, including:

[0021] Through the fish monitoring database, historical water environment parameters are obtained, and the baseline fish activity index thresholds corresponding to the historical water environment parameters are marked. The random forest regression algorithm is used for model training to construct a fish activity index threshold adjustment model. According to the real-time water environment parameters, the fish activity index threshold adjustment model is used to adjust the baseline fish activity index threshold. The health status of the fish population is judged based on the adjusted baseline fish activity index threshold and the real-time fish activity index.

[0022] Furthermore, the identification of fish behavior characteristics and external symptoms based on the fish appearance images with abnormal health status, and determination of the fish disease category, includes:

[0023] Through the fish appearance image database, historical fish appearance images with abnormal health status are obtained, and behavioral characteristics and external symptoms are marked. A convolutional neural network is used for model training to build a fish feature recognition model; based on the real-time acquired fish appearance images with abnormal health status and fish appearance images of fish with a density change rate higher than a preset density threshold, the fish behavioral characteristics and fish external symptoms are identified. The fish behavioral characteristics include but are not limited to abnormal swimming and floating heads, and the fish external symptoms include but are not limited to surface lesions, discoloration, scale abnormalities, ulcers, and parasites; the fish disease category is determined based on the fish behavioral characteristics and fish external symptoms; and the fish health status and fish disease category are sent to the fish health management center in real time through fish health monitoring equipment.

[0024] Furthermore, according to the fish disease category, fish health intervention measures are formulated and sent to the automated water quality control equipment and automatic drug delivery equipment for fish health management, real-time monitoring of fish activity index, and judgment of the effectiveness of fish health intervention measures, including:

[0025] Through the Fish Health Management Center's dispatching system, fish health intervention measures are formulated based on the disease type. These intervention measures include water quality adjustment requirements, the type and dosage of medication required, and are recorded and sent to automated water quality control and automatic drug delivery equipment. Water quality parameters are adjusted using oxygen enrichment and pH adjustment equipment. Automatic drug delivery equipment injects drugs into the water at a preset delivery rate and cycle based on the disease prevention or treatment requirements specified in the fish health adjustment work order, and this delivery information is transmitted to the Fish Health Management Center. The fish activity index is monitored in real time, and the current fish activity index is compared with the baseline fish activity index threshold to determine the effectiveness of the fish health intervention measures. Feedback is then sent to the Fish Health Management Center. If the fish health intervention measures are effective and the fish health status returns to normal, the Fish Health Management Center's dispatching system shuts down the water quality control and automatic drug delivery equipment, halting water quality adjustment and drug delivery. If the fish health intervention measures are ineffective, new fish health intervention measures are generated, and the water quality control and automatic drug delivery equipment are readjusted until the fish health status returns to normal.

[0026] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0027] The present invention provides a fish density monitoring method based on big data and acoustic signals. By constructing an acoustic reflection source identification model and a fish density prediction model, the present invention can effectively identify and eliminate multipath reflection signals, thereby improving the accuracy of fish location and density monitoring. By analyzing the swimming speed, density, swimming speed volatility, and speed change rate of fish, the present invention calculates the fish activity index, enabling real-time monitoring of fish behavior and health status. The present invention can also dynamically adjust the baseline activity index threshold based on water environmental parameters to ensure assessment accuracy under different water quality conditions. When abnormal fish health is detected, drones automatically capture and analyze the appearance of the abnormal fish, quickly determine the type of fish disease, and provide health intervention measures. This present invention not only improves the real-time and accuracy of fish density and fish health monitoring, reducing risks in aquaculture, but also significantly improves the efficiency and scientific nature of aquaculture management through automated water quality control and drug administration equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of a fish density monitoring method based on big data and acoustic wave signals according to the present invention;

[0029] Figure 2 Schematic diagram of a fish density monitoring method based on big data and acoustic wave signals according to the present invention;

[0030] Figure 3This is another schematic diagram of a fish density monitoring method based on big data and acoustic wave signals according to the present invention. DETAILED DESCRIPTION

[0031] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] like Figure 1-3 In this embodiment, a fish density monitoring method based on big data and acoustic wave signals may specifically include:

[0033] Step S101 : constructing a sound wave reflection source type identification model based on the sound wave reflection signal of the water area to be detected, determining the type of the sound wave reflection source of the sound wave reflection signal, and determining the position of the sound wave reflection source.

[0034] The acoustic detection equipment transmits acoustic waves according to the set acoustic emission parameters into the water area to be monitored. The directional array sensor acquires the acoustic reflection signals and corresponding timestamp information. Based on the acoustic reflection signals from the water area to be monitored, the independent component analysis algorithm is used to separate the acoustic reflection signals, obtaining the acoustic reflection signals from different sources and storing them in the acoustic reflection monitoring database. The short-time Fourier transform algorithm is used to convert the time-domain acoustic reflection signals into frequency-domain data, extracting their spectral features, including frequency, frequency peak, and amplitude. The acoustic reflection monitoring database is used to obtain historical acoustic emission frequency, amplitude, and spectral features of the acoustic reflection signals. The acoustic reflection source type is annotated and trained using a support vector machine algorithm to construct an acoustic reflection source identification model. This model identifies the acoustic reflection source type of the acoustic reflection signals, including those from fish schools and those from non-fish schools. The echo time of the acoustic reflection signals from the fish schools is obtained, and the position of the fish schools relative to the sensor array is calculated by time difference estimation to determine the fish locations.

[0035] For example, in an aquaculture farm, acoustic detection equipment is configured to emit acoustic signals at a frequency of 80 kHz. These sound waves propagate through the water and are received by directional array sensors arranged at different locations. The sensors then record the timestamp information for each acoustic reflection signal. The acoustic reflection signals are processed using an independent component analysis algorithm to separate signals from different sound sources. After processing, acoustic reflection signals from two primary sources are obtained: one from a school of fish and the other from underwater facilities or environmental structures. A short-time Fourier transform algorithm is used to convert the time-domain acoustic reflection signals into frequency-domain data. After processing, the following spectral characteristics are obtained: the frequency of the acoustic reflection signal from the school of fish is 75 kHz, the frequency peak is at 80 kHz, and the amplitude is 0.2 volts. The acoustic reflection monitoring database records the historical acoustic emission frequency and amplitude data, as well as the spectral characteristics of other acoustic reflection signals. A support vector machine algorithm is used to train a model based on the spectral characteristic data to construct an acoustic reflection source type recognition model to identify whether the acoustic reflection signal originates from a school of fish or other objects in the water. By recording the echo time of the sound wave reflection signal, the position of the fish school relative to the sensor array is estimated based on the time difference. If the echo time of the sound wave reflection is 0.6 seconds and the speed of sound wave propagation in water is 1500 meters per second, then the position of the fish school can be calculated to be about 900 meters away from the sensor.

[0036] Step S102 : calculating the similarity of the frequency spectrum characteristics of the fish school acoustic wave reflection signals based on the fish school acoustic wave reflection signals, identifying and eliminating multipath reflection signals from the same acoustic wave reflection source.

[0037] The acoustic reflection signals of fish schools are obtained from the acoustic reflection monitoring database. The similarity of the spectral characteristics of the fish school acoustic reflection signals is calculated using the Pearson correlation coefficient method. If the similarity is greater than a preset similarity threshold and the echo time difference is less than a preset difference threshold, the acoustic reflection signals of the fish school are determined to be multipath reflection signals from the same acoustic reflection source. Multipath reflection signals from the same acoustic reflection source are eliminated, and the unique acoustic reflection signal from the acoustic reflection source is retained.

[0038] For example, acoustic reflection signal data from two schools of fish, Signal A and Signal B, are obtained from an acoustic reflection monitoring database. Signals A and B are converted from the time domain to the frequency domain using a short-time Fourier transform (SFT). The spectral characteristics of Signals A and B, including frequency, frequency peak, and amplitude, are extracted. If Signal A has a frequency of 75 kHz, a frequency peak at 80 kHz, and an amplitude of 0.3 F, and Signal B has a frequency of 76 kHz, a frequency peak at 81 kHz, and an amplitude of 0.29 F, then the similarity between the spectral characteristics of Signals A and B is calculated using the Pearson correlation coefficient. If the calculated Pearson correlation coefficient is 0.95, which is greater than the preset minimum similarity threshold of 0.9, then the echo time difference between Signals A and B is checked. If the echo time of Signal A is 0.6 seconds and the echo time of Signal B is 0.59 seconds, then the echo time difference between Signals A and B is determined to be within the allowable range if it is less than 0.1 seconds, based on the preset difference threshold of 0.05. The spectral feature similarity between Signal A and Signal B is higher than the preset similarity threshold of 0.9, and the echo time difference between Signal A and Signal B is lower than the preset difference threshold of 0.05 seconds. Therefore, Signal A and Signal B are determined to be multipath reflection signals from the same acoustic wave reflection source.

[0039] Step S103, based on the echo width, spectral characteristics, phase difference and phase offset of the fish school acoustic reflection signal, a fish school density prediction model is constructed to predict the fish school density, and the fish school activity index is determined based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish school in the water area.

[0040] Based on the unique acoustic reflection signal from the acoustic reflection source, the phase difference and phase offset of the acoustic emission signal are determined. A recurrent neural network algorithm is used for model training based on the echo width, spectral characteristics, phase difference, and phase offset of the acoustic reflection signal from the fish school to construct a fish density prediction model and predict fish density. Fish density is continuously monitored. If the rate of change in fish density exceeds a preset density threshold, an image of the fish body is captured using a drone-mounted camera and saved to a fish appearance image database. If the rate of change in fish density is below the preset density threshold, the echo signals at different time points are compared to determine the fish's position and swimming speed. The fish activity index is determined based on the swimming speed, density, swimming speed volatility, and speed change rate of the fish in the water area.

[0041] For example, in an aquaculture farm, acoustic detection equipment transmits a 200kHz acoustic signal. The received acoustic reflection signal from a school of fish has an echo width of 2 milliseconds. The spectral characteristics show a peak frequency of 195kHz, an amplitude of 0.8 units, a phase difference of 15 degrees, and a phase offset of 5 degrees. Based on this data, a recurrent neural network algorithm is used for model training to construct a fish density prediction model, predicting the fish density in the area to be 50 fish per cubic meter. Fish density is continuously monitored. If the rate of change in fish density exceeds the preset density threshold of 10 fish per cubic meter, a drone-mounted camera is used to capture images of the fish's appearance and save them to the fish appearance image database. If the fish density change rate falls below the preset density threshold of 10 per cubic meter, by comparing the position changes of the fish reflection signal at different time points, such as every 5 seconds, the fish swimming speed is determined to be 0.2 meters per second. The fluctuation of the fish swimming speed in this area is calculated to be 0.03 meters per second, and the speed change rate is 0.01 meters per second squared. Combining these parameters, the fish activity index is calculated to be 0.75.

[0042] Among them, the fish activity index is determined based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish in the water area.

[0043] According to the swimming speed of the fish school in the water area, the fish school swimming speed fluctuation formula is used Calculate the fluctuation of fish swimming speed σ V (t), where V(t) is the swimming speed of the fish school at time t, It is the average swimming speed in the time period T. According to the swimming speed, density, swimming speed fluctuation and speed change rate of the fish school in the water area, the fish school activity index formula is used Calculate the fish activity index FAI(t), where V(t) is the average swimming speed of the fish school, D(t) is the fish school density, and σ V (t) is the fluctuation of the swimming speed of the fish school, is the rate of change of the swimming speed of the fish school.

[0044] For example, in an aquaculture farm, an acoustic detection system was used to record the swimming data of a school of fish. At a certain time t, the instantaneous swimming speed V(t) of the school of fish was 1.5 m / s, and the fish density D(t) was 8 fish per cubic meter. By analyzing the speed data for the last 10 seconds, the average swimming speed of the school of fish was calculated. is 1.3 m / s. According to the formula of the fluctuation of the swimming speed of fish schools Calculate the fluctuation of fish swimming speed σ V (t) is 0.25 m / s, reflecting the variation of the fish school's swimming speed during this period. The greater the fluctuation of the fish school's swimming speed, the stronger the activity of the fish school. At the same time, the speed change rate of the fish school is recorded. The fish activity index formula is 0.1 m / s2 and is calculated based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish in the water area. The calculated fish activity index (FAI(t)) is 6.8. A higher value of the FAI generally means more frequent fish activity, reflecting the health and activity of fish in the waters.

[0045] In step S104, the health status of the fish school is determined based on the reference fish school activity index threshold and the fish school activity index, and the appearance images of the fish schools with abnormal health status are obtained using the camera mounted on the drone.

[0046] Based on the calculated fish activity index, water environment sensors are used to obtain real-time water environment parameters at the time of the fish activity index detection. These parameters include flow rate, dissolved oxygen content, water temperature, and turbidity, and are stored in the fish monitoring database. The baseline fish activity index threshold is adjusted based on the real-time water environment parameters. The fish health status is determined based on the baseline fish activity index threshold and the fish activity index, with the fish health status being classified as normal or abnormal. If the difference between the fish activity index and the baseline fish activity index threshold is greater than a preset difference threshold, the fish health status is determined to be abnormal. Appearance images of fish with abnormal health are captured using a drone-mounted camera and stored in the fish appearance image database.

[0047] For example, based on the calculated fish activity index, the environmental parameters of the water body corresponding to the detection time of the fish activity index are obtained in real time through the water environment sensor, and these data are saved in the fish monitoring database. If at the time of monitoring, the flow rate is 0.3 m / s, the dissolved oxygen content is 8 mg / L, the water temperature is 22 degrees Celsius, and the turbidity is 5 NTU. If under the current environmental conditions, the baseline fish activity index threshold is set to 7.0. The current fish activity index FAI(t) = 6.8 is compared with the baseline threshold of 7.0, and the difference is 0.2, which is lower than the preset difference threshold of 1.0. Therefore, the health status of the fish school can be judged to be normal. If the difference between the monitored fish activity index and the baseline threshold exceeds the difference threshold of 1.0, such as the fish activity index reaches 8.2, the difference is 1.2, which is greater than 1.0, then the health status of the fish school is automatically judged to be abnormal. If the health status of the fish school is determined to be abnormal, the fish body appearance image of the abnormal fish school is captured by a camera mounted on a drone, and the fish body appearance image of the fish school is uploaded and saved in a fish school appearance image database.

[0048] Among them, the baseline fish activity index threshold is adjusted according to the water environment parameters obtained in real time.

[0049] Using a fish monitoring database, we obtain historical water environment parameters and annotate the baseline fish activity index thresholds corresponding to these parameters. We then use a random forest regression algorithm for model training to construct a fish activity index threshold adjustment model. Using this model, we adjust the baseline fish activity index threshold based on real-time water environment parameters. We then use the adjusted baseline fish activity index threshold and the real-time fish activity index to determine fish health.

[0050] For example, at an aquaculture farm, a large amount of historical water environment data and corresponding fish activity data is collected through a monitoring database. This water environment data includes environmental parameters such as flow rate, dissolved oxygen content, water temperature, and turbidity. For example, during a certain historical period, the water temperature was 20 degrees Celsius, the flow rate was 0.4 m / s, the dissolved oxygen content was 9 mg / L, and the turbidity was 6 NTU, corresponding to a baseline fish activity index threshold of 6.5. The historical water environment data is annotated with a baseline fish activity index threshold, and a random forest regression algorithm is used for model training to construct a fish activity index threshold adjustment model. In actual use, when the system obtains real-time water environment parameters, such as the current water temperature of 23 degrees Celsius, the flow rate of 0.3 m / s, the dissolved oxygen content of 7 mg / L, and the turbidity of 4 NTU, these real-time environmental parameters are input into the trained fish activity index threshold adjustment model, resulting in a baseline fish activity index threshold of 7.2 suitable for the current water conditions. The real-time fish activity index is compared with the adjusted baseline fish activity index threshold. If the difference is less than the difference threshold of 1.0, the fish health is considered normal. For example, if the real-time fish activity index is 6.8, the difference between the real-time fish activity index and the baseline fish activity index threshold of 7.2 is 0.4, which is less than the preset difference threshold of 1.0, so the fish health is considered normal. If the fish activity index is 5.5 at a certain moment, the difference between the real-time fish activity index and the baseline fish activity index threshold is 1.7, which is greater than the preset difference threshold of 1.0, the fish health is considered abnormal. The system will immediately issue an alarm and activate the drone's onboard camera to capture images of the abnormal fish. The images are uploaded and saved to the fish appearance image database.

[0051] Step S105 , identifying the behavioral characteristics and external symptoms of the fish school based on the appearance images of the fish whose health status is abnormal, and determining the disease category of the fish school.

[0052] Using a database of fish appearance images, we collect historical images of fish with abnormal health status, annotate their behavioral characteristics and external symptoms, and use a convolutional neural network for model training to construct a fish identification model. Based on real-time images of fish with abnormal health status and images of fish with a density change rate exceeding a preset density threshold, we identify fish behavioral characteristics and external symptoms. These include, but are not limited to, abnormal swimming and floating heads. External symptoms include, but are not limited to, surface lesions, discoloration, scale abnormalities, ulcers, and parasites. Based on these behavioral characteristics and external symptoms, we determine the disease category of the fish. Fish health monitoring equipment transmits the fish health status and disease category to the fish health management center in real time.

[0053] For example, a database of fish appearance images was used to obtain 5,000 images of fish schools, annotated with behavioral characteristics and external symptoms. Of these, 2,000 images showed varying degrees of abnormal swimming patterns, such as erratic movements and lateral swimming, 1,500 showed fish floating. The remaining 1,500 images revealed external symptoms of the fish, including surface lesions, discoloration, scale abnormalities, ulcers, and parasite attachment. A convolutional neural network was used to train a model to construct a fish feature recognition model. This model was trained and validated on images annotated with behavioral characteristics and external symptoms, and was able to identify abnormal swimming patterns. The accuracy of the model was 93% for abnormal swimming patterns and 90% for floating behavior. Regarding external symptoms, the model achieved 95% accuracy for surface lesions, 92% for scale abnormalities, and 91% for other symptoms, such as discoloration. During actual monitoring, fish health monitoring equipment collects real-time images of fish with abnormal health and fish density changes exceeding a preset threshold. These images are then analyzed using a trained convolutional neural network model. If, during a particular monitoring session, abnormal swimming behavior is identified in certain fish, image analysis also reveals that 10 of these fish exhibit obvious ulcers. Based on these behavioral characteristics and external symptoms, these fish are suspected to be suffering from typical bacterial canker, a condition typically caused by deteriorating water quality or infection with pathogens.

[0054] Step S106, formulate fish health intervention measures based on the fish disease category, and send them to the automated water quality control equipment and automatic drug delivery equipment for fish health management, real-time monitoring of fish activity index, and judgment of the effectiveness of fish health intervention measures.

[0055] Through the Fish Health Management Center's dispatching system, fish health intervention measures are developed based on the disease type. These intervention measures include water quality adjustment requirements, medication type, and dosage. These intervention measures are recorded and sent to automated water quality control and automatic medication delivery equipment. Oxygen aeration and pH adjustment equipment are used to adjust water quality parameters. Automatic medication delivery equipment injects medication into the water at a preset delivery rate and cycle, based on the disease prevention or treatment requirements specified in the fish health adjustment work order. This delivery information is then transmitted to the Fish Health Management Center. Fish activity indexes are monitored in real time, compared to baseline activity index thresholds, to determine the effectiveness of the fish health intervention measures. Feedback is then sent to the Fish Health Management Center. If the intervention measures are effective and the fish health returns to normal, the Fish Health Management Center's dispatching system shuts down the water quality control and automatic medication delivery equipment, halting water quality adjustments and medication delivery. If the fish health intervention measures are not effective, new fish health intervention measures will be generated and the water quality control equipment and automatic drug delivery equipment will be readjusted until the fish health status returns to normal.

[0056] For example, if abnormal fish health is identified in a particular area, and it is inferred that these fish may be suffering from bacterial canker, combined with water quality data, the current dissolved oxygen content in the water is 5 mg / L and the pH is 6.5. The optimal water quality conditions for this type of fish require a dissolved oxygen content of at least 8 mg / L and a pH between 7.0 and 7.5. Based on the external symptoms of the fish, treatment with an antibiotic such as florfenicol is recommended, with a recommended dose of 0.5 grams per cubic meter of water. Based on this, the Fish Health Management Center's dispatch system generates a set of intervention measures, including water quality adjustment and drug treatment. These measures include raising the dissolved oxygen level to at least 8 mg / L, aerating the water using oxygen aerators, adjusting the pH to 7.0, neutralizing the water with sodium bicarbonate using a pH adjustment device, and administering florfenicol at a dose of 0.5 grams per cubic meter using an automated dosing device, with dosing set every 12 hours for three days. The intervention measures are recorded in the fish health management work order and automatically sent to the corresponding equipment control port. The oxygen enrichment equipment and pH adjustment equipment start to operate according to the set water quality targets, and the automatic dosing equipment starts the drug injection process according to the preset dosage and time period. After each dosing is completed, the equipment will transmit the dosing time, dosage and real-time parameters of the water body to the fish health management center. During the intervention process, the activity index of the fish is monitored in real time. If the baseline fish activity index threshold is 7.0, the initial monitored fish activity index is 5.5, which reflects the low activity state of the fish. 24 hours after the intervention measures were implemented, the fish activity index increased to 6.8, and the fish behavior also tended to be normal, indicating that the intervention measures are initially effective. 48 hours after the intervention, if the activity index rises to 7.2, exceeding the baseline activity index threshold of 7.0, and the external symptoms of the fish are significantly alleviated, the fish's health status returns to normal. The fish health management center's dispatch system shuts down the oxygen aeration equipment, pH adjustment equipment, and automatic drug delivery equipment, halts all water quality adjustments and drug delivery, and records the process and results to the fish health management center, forming a complete health intervention report. If the fish's health status fails to improve effectively during the intervention, such as if the activity index remains at 5.5 or decreases after 24 hours, new intervention measures are generated based on current environmental and health data, such as increasing the antibiotic dosage, changing the drug type, or adjusting water quality targets, until the fish's health status returns to normal.

[0057] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the concept of this application. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A fish density monitoring method based on big data and acoustic signals, characterized in that: The method comprises: According to the sound wave reflection signal of the water area to be detected, a sound wave reflection source type identification model is constructed to determine the type of the sound wave reflection source of the sound wave reflection signal and the location of the sound wave reflection source; Based on the acoustic wave reflection signals of the school of fish, the similarity of the spectral characteristics of the acoustic wave reflection signals of the school of fish is calculated, and the multipath reflection signals from the same acoustic wave reflection source are identified and eliminated; Based on the echo width, spectral characteristics, phase difference and phase offset of the fish school acoustic reflection signal, a fish school density prediction model is constructed to predict the fish school density. The fish school activity index is determined based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish school in the water area. The health status of the fish school is determined based on the baseline fish school activity index threshold and the fish school activity index. The drone-mounted camera is used to obtain images of the fish body appearance of fish with abnormal health status. Based on images of fish with abnormal health status, identify fish behavioral characteristics and external symptoms, and determine the type of fish disease; Develop fish health intervention measures based on fish disease categories and send them to automated water quality control equipment and automatic drug dosing equipment for fish health management. Monitor fish activity indexes in real time to determine the effectiveness of fish health intervention measures. The method comprises: constructing a fish density prediction model based on the echo width, spectral characteristics, phase difference and phase offset of the fish acoustic reflection signal to predict the fish density, and determining the fish activity index based on the swimming speed, density, swimming speed volatility and speed change rate of the fish in the water area, including: Based on the unique sound wave reflection signal from the sound wave reflection source, the phase difference and phase offset of the sound wave transmission signal are determined; based on the echo width, spectral characteristics, phase difference and phase offset of the sound wave reflection signal of the fish school, a recurrent neural network algorithm is used for model training to construct a fish school density prediction model to predict the fish school density; the fish school density is continuously monitored, and if the fish school density change rate is higher than the preset density threshold, the drone-mounted camera is used to obtain the appearance image of the fish body and save it to the fish school appearance image database; if the fish school density change rate is lower than the preset density threshold, the echo signals at different time points are compared to judge the position change of the fish school and determine the swimming speed of the fish school; based on the swimming speed, density, swimming speed volatility and speed change rate of the fish school in the water area, the fish school activity index is determined; The method of determining the fish activity index based on the swimming speed, density, swimming speed fluctuation and speed change rate of the fish in the water area includes: According to the swimming speed of the fish school in the water area, the fish school swimming speed fluctuation formula is used , calculate the fluctuation of fish swimming speed ,in, is the swimming speed of the fish school at time t, is the average swimming speed in the time period T; based on the swimming speed, density, swimming speed volatility and speed change rate of the fish school in the water area, the fish school activity index formula is used , calculate the fish activity index ,in, is the average swimming speed of the fish school, is the fish density, is the fluctuation of the swimming speed of fish schools, is the rate of change of the swimming speed of the fish school.

2. The method according to claim 1, wherein The method of constructing a sound wave reflection source type identification model based on the sound wave reflection signal of the water area to be detected, determining the type of the sound wave reflection source of the sound wave reflection signal, and determining the position of the sound wave reflection source includes: The acoustic detection equipment emits acoustic waves to the water area to be detected according to the set acoustic wave emission parameters, and the directional array sensor obtains the acoustic wave reflection signal and the corresponding timestamp information; the acoustic wave reflection signal is separated according to the acoustic wave reflection signal of the water area to be detected by the independent component analysis algorithm to obtain the acoustic wave reflection signals of different acoustic wave reflection sources, and save them to the acoustic wave reflection monitoring database; the short-time Fourier transform algorithm is used to convert the time domain acoustic wave reflection signal into frequency domain data, and its spectral characteristics are extracted, which include frequency, frequency peak and amplitude; the historical acoustic wave emission frequency, amplitude, and spectral characteristics of the acoustic wave reflection signal are obtained through the acoustic wave reflection monitoring database, and the type of acoustic wave reflection source is marked. The support vector machine algorithm is used for model training to construct an acoustic wave reflection source type recognition model to determine the acoustic wave reflection source type of the acoustic wave reflection signal, including fish school acoustic wave reflection sources and non-fish school acoustic wave reflection sources; the echo time of the acoustic wave reflection signal of the fish school is obtained, and the position of the fish school relative to the sensor array is calculated through time difference estimation to determine the position of the fish school.

3. The method according to claim 1, wherein The method comprises calculating the similarity of the frequency spectrum characteristics of the fish school acoustic wave reflection signals based on the fish school acoustic wave reflection signals, and identifying and eliminating multipath reflection signals in the same acoustic wave reflection source, including: The acoustic wave reflection signals of fish schools are obtained through the acoustic wave reflection monitoring database. The similarity of the spectral characteristics of the acoustic wave reflection signals of fish schools is calculated using the method of calculating the Pearson correlation coefficient. If the similarity is greater than a preset similarity threshold and the echo time difference is less than a preset difference threshold, the acoustic wave reflection signals of the fish school are determined to be multipath reflection signals of the same acoustic wave reflection source; the multipath reflection signals in the same acoustic wave reflection source are eliminated, and the unique acoustic wave reflection signal of the acoustic wave reflection source is retained.

4. The method according to claim 1, wherein The method of determining the health status of a fish school based on the baseline fish school activity index threshold and the fish school activity index, and obtaining an image of the fish body of a fish school with an abnormal health status using a camera mounted on a drone, includes: Based on the calculated fish activity index, the water environment parameters corresponding to the detection time of the fish activity index are obtained in real time through the water environment sensor and saved in the fish monitoring database. The water environment parameters include flow rate, dissolved oxygen content, water temperature, and turbidity. The baseline fish activity index threshold is adjusted based on the real-time water environment parameters. The fish health status is determined based on the baseline fish activity index threshold and the fish activity index, and the fish health status is classified as normal or abnormal. If the difference between the fish school activity index and the baseline fish school activity index threshold is greater than the preset difference threshold, the health status of the fish school is judged to be abnormal. The camera mounted on the drone is used to obtain the appearance image of the fish body of the fish school with abnormal health status and save it to the fish school appearance image database.

5. The method according to claim 4, wherein The method of adjusting the threshold of the baseline fish activity index according to the water environment parameters obtained in real time includes: Through the fish monitoring database, historical water environment parameters are obtained, and the baseline fish activity index thresholds corresponding to the historical water environment parameters are marked. The random forest regression algorithm is used for model training to construct a fish activity index threshold adjustment model. According to the real-time water environment parameters, the fish activity index threshold adjustment model is used to adjust the baseline fish activity index threshold. The health status of the fish population is judged based on the adjusted baseline fish activity index threshold and the real-time fish activity index.

6. The method according to claim 1, wherein The method of identifying the behavioral characteristics and external symptoms of the fish school based on the appearance images of the fish with abnormal health status, and determining the disease category of the fish school, includes: Through the fish appearance image database, historical fish appearance images with abnormal health status are obtained, and behavioral characteristics and external symptoms are marked. A convolutional neural network is used for model training to build a fish feature recognition model; based on the real-time acquired fish appearance images with abnormal health status and fish appearance images of fish with a density change rate higher than a preset density threshold, the fish behavioral characteristics and fish external symptoms are identified. The fish behavioral characteristics include but are not limited to abnormal swimming and floating heads, and the fish external symptoms include but are not limited to surface lesions, discoloration, scale abnormalities, ulcers, and parasites; the fish disease category is determined based on the fish behavioral characteristics and fish external symptoms; and the fish health status and fish disease category are sent to the fish health management center in real time through fish health monitoring equipment.

7. The method according to claim 1, wherein According to the fish disease category, fish health intervention measures are formulated and sent to the automated water quality control equipment and automatic drug delivery equipment for fish health management, real-time monitoring of fish activity index, and judgment of the effectiveness of fish health intervention measures, including: Through the Fish Health Management Center's dispatching system, fish health intervention measures are formulated based on the disease type. These intervention measures include water quality adjustment requirements, the type and dosage of medication required, and are recorded and sent to automated water quality control and automatic drug delivery equipment. Water quality parameters are adjusted using oxygen enrichment and pH adjustment equipment. Automatic drug delivery equipment injects drugs into the water at a preset delivery rate and cycle based on the disease prevention or treatment requirements specified in the fish health adjustment work order, and this delivery information is transmitted to the Fish Health Management Center. The fish activity index is monitored in real time, and the current fish activity index is compared with the baseline fish activity index threshold to determine the effectiveness of the fish health intervention measures. Feedback is then sent to the Fish Health Management Center. If the fish health intervention measures are effective and the fish health status returns to normal, the Fish Health Management Center's dispatching system shuts down the water quality control and automatic drug delivery equipment, halting water quality adjustment and drug delivery. If the fish health intervention measures are ineffective, new fish health intervention measures are generated, and the water quality control and automatic drug delivery equipment are readjusted until the fish health status returns to normal.

Images