A bass and treble sound equipment control system

Abstract

The application relates to the technical field of sound intelligent control, in particular to a high and low sound equipment control system; target biological recognition of a monitoring area is carried out to obtain target parameter data and member motion data, a target aggregation group and target aggregation parameters are obtained through clustering algorithm recognition; a target aggregation danger coefficient is calculated according to the target aggregation parameters, and a critical aggregation group is determined according to the target aggregation danger coefficient; the environment condition between the critical aggregation group and a monitoring point is recognized to obtain monitoring environment data; a sound pattern recognition model is constructed, the target aggregation parameters and the monitoring environment data of the critical aggregation group are recognized, sound pattern data and sound parameter data are obtained, and expulsion is carried out; an expulsion effect coefficient is calculated according to the change of the motion data of the critical aggregation group before and after, and the sound pattern recognition model is fed back and re-recognized. The target biological recognition, expulsion and feedback in the monitoring area effectively improve the biological expulsion effect.

Description

Technical Field

[0001] This invention relates to the field of intelligent audio control technology, specifically a control system for high and low frequency audio equipment. Background Technology

[0002] With technological advancements, acoustic repulsion technology, as an environmentally friendly, safe, and efficient biological repulsion technique, has gradually gained attention. This technology utilizes sound stimuli of specific frequencies, volumes, and waveforms to disrupt the normal behavior of organisms, causing them to move away from a designated area.

[0003] Compared to traditional repulsion methods, acoustic repulsion technology offers significant advantages. For example, it is non-invasive: acoustic repulsion does not involve direct harm to organisms, avoiding damage to the ecosystem. It is highly efficient: by precisely adjusting the frequency and intensity of sound, a strong repulsion effect can be produced targeting specific species, avoiding unnecessary impact on other harmless organisms. It is sustainable: unlike chemical repulsion methods, acoustic repulsion does not lead to environmental pollution or problems with antibiotic resistance.

[0004] However, existing sound repulsion technologies are limited in their modes and cannot accurately set repulsion parameters based on biological aggregation, movement, and environmental conditions; they also fail to identify and adjust the repulsion effect, resulting in an imprecise biological repulsion effect of sound.

[0005] Therefore, a control system for high and low frequency audio equipment is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a control system for high and low frequency audio equipment. This system obtains target parameter data and member movement data through target biometric identification in a monitored area. A clustering algorithm is used to identify target clusters and target cluster parameters. A target cluster danger coefficient is calculated based on the target cluster parameters, and a critical cluster is identified based on the target cluster danger coefficient. The environmental conditions between the critical cluster and the monitoring point are identified, obtaining monitoring environment data. An audio pattern recognition model is constructed to identify the target cluster parameters and monitoring environment data of the critical cluster, obtaining audio pattern data and audio parameter data for dispersal. Based on the changes in the movement data of the critical cluster, a dispersal effectiveness coefficient is calculated, and the audio pattern recognition model is fed back and re-identified, effectively improving the biometric dispersal effect.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A high-frequency and low-frequency audio equipment control system includes:

[0009] The target distribution recognition module uses the location of the audio equipment as the monitoring point, and then obtains the monitoring area according to the set monitoring radius; it identifies target organisms entering the monitoring area and collects target parameter data and member movement data.

[0010] The clustering identification module identifies target parameter data and member movement data through a clustering algorithm, and divides the target organisms into target clusters based on the identification results; based on the target parameter data and member movement data of the target organisms in the target clusters, the target clustering parameters are obtained.

[0011] The risk assessment module calculates the target cluster risk coefficient based on the target cluster parameters; and identifies the most critical clusters based on the target cluster risk coefficient.

[0012] The audio parameter acquisition module identifies the environmental conditions between the critical gathering group and the monitoring point, and obtains monitoring environment data; it constructs an audio pattern recognition model to identify the target gathering parameters of the critical gathering group and the monitoring environment data, and obtains audio pattern data and audio parameter data for biological expulsion; the audio pattern data includes low-frequency mode and high-frequency mode; the audio parameter data includes sound frequency, sound pressure level, sound waveform, sound duration and sound direction.

[0013] The sound effect feedback module obtains the sound wave time point based on the target aggregation parameters and monitoring environment data; calculates the expulsion effect coefficient based on the changes in the movement data of the critical aggregation group before and after the sound wave time point; and performs feedback and re-recognition on the sound pattern recognition model based on the expulsion effect coefficient.

[0014] The target parameter data includes: organism type, organism quantity, organism size coefficient, and organism hazard coefficient; wherein, the organism hazard coefficient is obtained based on historical data on the losses caused by organism types to the protected targets;

[0015] The member movement data includes: relative position of the organism, movement speed of the organism, direction of movement of the organism, and relative distance of the organism; the relative position of the organism is the position and orientation data of the organism relative to the monitoring point; the relative distance of the organism is the distance between the organism and the monitoring point.

[0016] The target aggregation parameters include the species of aggregation organisms, biohazard factor, number of aggregation organisms, size factor of aggregation organisms, relative position of aggregation organisms, movement speed of aggregation organisms, movement direction of aggregation organisms, relative distance of aggregation organisms, and density of aggregation organisms;

[0017] The process of obtaining the relative position of the aggregated organisms is as follows: obtain the position data of the target organisms in the target aggregate, construct the minimum circumcircle based on the position data, and use the relative position data of the center of the minimum circumcircle and the monitoring point as the relative position of the aggregated organisms.

[0018] The movement speed of the aggregated organisms is the average movement speed of the target organisms in the target aggregate;

[0019] The direction of movement of the aggregated organisms is the average of the direction of movement of the target organisms in the target aggregate;

[0020] The relative distance between aggregated organisms is the average relative distance between the target organisms in the target aggregate and the monitoring point;

[0021] The process of obtaining the aggregation density is as follows: the minimum circumcircle is constructed based on the location data of the target organisms in the target aggregation, and the aggregation density is obtained based on the volume of the minimum circumcircle and the number of aggregation organisms.

[0022] The formula for calculating the target aggregation risk factor is as follows:

[0023]

[0024] Where Dans represents the target aggregation risk factor; cla i T represents the biohazard coefficient corresponding to the species of organisms that aggregate; i denoted by α, where n represents the size coefficient of the aggregated organism; θ() represents the angle function in space; dir represents the direction of movement of the aggregated organism; D1 represents the direction of the line connecting the target aggregated organism and the monitoring point; DT represents the angle threshold; vel represents the movement speed of the aggregated organism; VT represents the movement speed threshold; tan represents the relative distance of the aggregated organism; TT represents the relative distance threshold; den represents the density of the aggregated organism; ET represents the density threshold; α1, α2, α3 and α4 represent the hazard coefficients; exp represents the exponential function with the natural constant as the base.

[0025] The monitoring environment data includes environmental media composition data, media movement data, environmental temperature data, and environmental noise data; the environmental media composition data is the media data between the animal and the monitoring point.

[0026] The sound pattern recognition model includes a parameter data input layer, a parameter data recognition layer, a sound pattern recognition layer, and a sound parameter output layer. The parameter data input layer inputs target aggregation parameters and monitoring environment data into the model. The parameter data recognition layer extracts and recognizes features from the target aggregation parameters and monitoring environment data. The sound pattern recognition layer identifies sound pattern data, including bass and treble modes. The sound parameter output layer identifies sound parameter data, including sound frequency, sound pressure level, sound waveform, sound duration, and sound direction.

[0027] The training process for the audio pattern recognition model is as follows:

[0028] Biological expulsion tests were conducted based on audio, resulting in an expulsion test dataset and expulsion test labels.

[0029] The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound mode data, and test sound parameter data; the expulsion test label includes the test expulsion effectiveness coefficient.

[0030] The sound pattern recognition model was trained using the eviction test dataset and eviction test labels.

[0031] The sound wave time point is the moment when the sound emitted by the sound source has an effect on the organism.

[0032] The method for calculating the sound wave time point is as follows: based on the environmental medium composition data, medium movement data, environmental temperature data, and environmental noise data in the monitoring environment data, the time when the sound emitted by the speaker intersects with the movement trajectory of the critical gathering group is determined as the first time period; the movement trajectory of the critical gathering group is obtained through target aggregation parameters.

[0033] The reaction time of the aggregated organisms to the sound repulsion was obtained from experimental tests and used as the second time period.

[0034] The sound wave time-effect point is calculated based on the time point when the sound is emitted, the first time period, and the second time period.

[0035] The process of acquiring the movement data of the critical cluster before and after the aforementioned acoustic time point is as follows:

[0036] Set a time threshold; acquire motion data of the critical cluster within the time threshold range before the sound wave time point to obtain first motion data, which includes the movement speed, movement direction, relative distance, and density of the clustered organisms;

[0037] The motion data of the critical cluster within the time threshold range after the sound wave time point is obtained to obtain the second motion data, which includes the movement speed, movement direction, relative distance and density of the expelling organisms.

[0038] The expulsion effect coefficient was calculated based on the first and second motion data.

[0039]

[0040] Where Eff represents the expulsion effect coefficient; tce represents the relative distance of the expelled organisms; tan represents the relative distance of the aggregated organisms; θ() represents the angle function of directions in space; dir represents the direction of movement of the aggregated organisms; D1 represents the direction of the line connecting the target aggregate and the monitoring point; qir represents the direction of movement of the expelled organisms; D2 represents the direction of the line connecting the target aggregate and the monitoring point after the sound wave time point; f() represents the disorder function of the direction data; qsd represents the speed of movement of the expelled organisms; vel represents the speed of movement of the aggregated organisms; smd represents the density of the expelled organisms; den represents the density of the aggregated organisms; β1, β2, β3 and β4 represent the expulsion coefficients.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0042] 1. This invention obtains target aggregation parameters of a target cluster and calculates the target aggregation hazard coefficient of the target cluster from four dimensions: the biological hazard coefficient and size weight of the target organisms in the target cluster, the movement direction and speed of the target organisms in the target cluster, the relative distance between the target cluster and the monitoring point, and the density of the target cluster. The hazard level of the target cluster is accurately and comprehensively measured through these four dimensions. Then, the highest hazard level critical cluster is accurately determined through the target aggregation hazard coefficient.

[0043] 2. This invention uses sound to conduct biological expulsion tests, obtaining an expulsion test dataset and expulsion test labels. The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound pattern data, and test sound parameter data. The expulsion test labels include test expulsion effect coefficients. A sound pattern recognition model is obtained by training the expulsion test dataset and expulsion test labels. The sound pattern recognition model can accurately identify the sound pattern data and sound parameter data with the optimal expulsion effect based on the target aggregation parameters and monitoring environment data.

[0044] 3. This invention obtains the sound wave time point based on the time when the sound emitted by the speaker intersects with the movement trajectory of the dangerous cluster, and the reaction time of the clustered organisms to the sound; it acquires the movement data of the dangerous cluster within the time threshold range before and after the sound wave time point as the first movement data and the second movement data; it calculates the repulsion effect coefficient based on the relative distance change, movement angle, movement speed change, and organism density change between the first movement data and the second movement data; and it accurately measures the organism repulsion effect through the repulsion effect coefficient. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the structure of a high and low frequency audio device control system according to the present invention;

[0046] Figure 2This is a schematic diagram of the audio pattern recognition model of the present invention;

[0047] Figure 3 This is a schematic diagram of the audio device structure of the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Example 1

[0050] A control system for high and low frequency audio equipment, the structure of which is as follows: Figure 1 As shown, it includes:

[0051] The target distribution recognition module uses the location of the audio equipment as the monitoring point and obtains the monitoring area based on the set monitoring radius; it identifies target organisms entering the monitoring area and collects target parameter data and member movement data.

[0052] The target parameter data includes: organism type, organism quantity, organism size coefficient, and organism hazard coefficient; wherein, the organism hazard coefficient is obtained from historical data on the losses caused by organism type to the protected target, including economic losses, quantity losses, etc.; the organism size coefficient is obtained by comparing the actual size data of the target organism with standard size data; the standard size data is obtained by averaging samples collected from the target organism.

[0053] The member movement data includes: relative position of the organism, movement speed of the organism, direction of movement of the organism, and relative distance of the organism; the relative position of the organism is the position and orientation data of the organism relative to the monitoring point; the relative distance of the organism is the distance between the organism and the monitoring point.

[0054] The clustering identification module identifies target parameter data and member movement data through a clustering algorithm, and divides the target organisms into target clusters based on the identification results; and obtains the target clustering parameters based on the target parameter data and member movement data of the target organisms in the target clusters.

[0055] The target aggregation parameters include the species of aggregation organisms, biohazard factor, number of aggregation organisms, size factor of aggregation organisms, relative position of aggregation organisms, movement speed of aggregation organisms, movement direction of aggregation organisms, relative distance of aggregation organisms, and density of aggregation organisms;

[0056] The process of obtaining the relative position of the aggregated organisms is as follows: obtain the position data of the target organisms in the target aggregate, construct the minimum circumcircle based on the position data, and use the relative position data of the center of the minimum circumcircle and the monitoring point as the relative position of the aggregated organisms.

[0057] The movement speed of the aggregated organisms is the average movement speed of the target organisms in the target aggregate;

[0058] The direction of movement of the aggregated organisms is the average value of the movement directions of the target organisms in the target aggregate, which is calculated by using a direction vector;

[0059] The relative distance between aggregated organisms is the average relative distance between the target organisms in the target aggregate and the monitoring point;

[0060] The process of obtaining the aggregation density is as follows: the minimum circumcircle is constructed based on the location data of the target organisms in the target aggregation, and the aggregation density is obtained based on the volume of the minimum circumcircle and the number of aggregation organisms.

[0061] This invention identifies target parameter data and member movement data through a clustering algorithm, and divides the target organisms into target clusters based on the identification results; based on the target parameter data and member movement data of the target organisms in the target clusters, the target cluster parameters are obtained; the target organisms are accurately divided, providing a basis for subsequent hazard identification.

[0062] The danger level measurement module calculates the danger coefficient of the target cluster based on the target cluster parameters; and determines the most dangerous critical cluster based on the target cluster danger coefficient.

[0063] The formula for calculating the target aggregation risk factor is as follows:

[0064]

[0065] Where Dans represents the target aggregation risk factor; cla i T represents the biohazard coefficient corresponding to the species of organisms that aggregate; i denoted by α, where α represents the size coefficient of the aggregated organism; n represents the number of organisms in the target aggregate; θ() represents the angle function in space; dir represents the direction of movement of the aggregated organism; D1 represents the direction of the line connecting the target aggregate and the monitoring point; DT represents the angle threshold; vel represents the movement speed of the aggregated organism; VT represents the movement speed threshold; tan represents the relative distance of the aggregated organism; TT represents the relative distance threshold; den represents the density of the aggregated organism; ET represents the density threshold; α1, α2, α3 and α4 represent the hazard coefficients; exp represents the exponential function with the natural constant as the base.

[0066] This invention obtains target aggregation parameters of a target cluster and calculates the target aggregation hazard coefficient of the target cluster from four dimensions: the biohazard coefficient and size weight of the target organisms in the target cluster, the movement direction and speed of the target organisms in the target cluster, the relative distance between the target cluster and the monitoring point, and the density of the target cluster. Through these four dimensions, the hazard level of the target cluster is accurately and comprehensively measured. Then, through the target aggregation hazard coefficient, the most critical cluster with the highest hazard level is accurately identified.

[0067] The audio parameter acquisition module identifies the environmental conditions between the critical gathering group and the monitoring point, and obtains monitoring environment data; it constructs an audio pattern recognition model to identify the target gathering parameters of the critical gathering group and the monitoring environment data, and obtains audio pattern data and audio parameter data for biological expulsion; the audio pattern data includes low-frequency mode and high-frequency mode; the audio parameter data includes sound frequency, sound pressure level, sound waveform, sound duration and sound direction.

[0068] The monitoring environment data includes environmental medium composition data, medium movement data, environmental temperature data, and environmental noise data; the environmental medium composition data is the medium data between the animal and the monitoring point, including gas, liquid, solid, etc.

[0069] The sound pattern recognition model is built on a deep neural network model, and its structure is as follows: Figure 2 As shown; it includes a parameter data input layer, a parameter data recognition layer, a sound pattern recognition layer, and a sound parameter output layer;

[0070] The parameter data input layer inputs the target aggregation parameters and monitoring environment data into the model;

[0071] The parameter data identification layer performs feature extraction and identification on target aggregation parameters and monitoring environment data;

[0072] The audio pattern recognition layer identifies audio pattern data, including bass mode and treble mode;

[0073] The audio parameter output layer is used to identify audio parameter data, including sound frequency, sound pressure level, sound waveform, sound duration, and sound direction.

[0074] The training process of the sound pattern recognition model is as follows:

[0075] Biological expulsion tests were conducted based on audio, resulting in an expulsion test dataset and expulsion test labels.

[0076] The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound mode data, and test sound parameter data; the expulsion test label includes the test expulsion effectiveness coefficient.

[0077] The sound pattern recognition model was trained using the eviction test dataset and eviction test labels.

[0078] This invention conducts biological expulsion tests based on sound, obtaining an expulsion test dataset and expulsion test labels. The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound pattern data, and test sound parameter data. The expulsion test labels include test expulsion effect coefficients. A sound pattern recognition model is obtained by training the expulsion test dataset and expulsion test labels. The sound pattern recognition model can accurately identify the sound pattern data and sound parameter data with the optimal expulsion effect based on the target aggregation parameters and monitoring environment data.

[0079] The sound effect feedback module obtains the sound wave time point based on the target aggregation parameters and monitoring environment data; calculates the expulsion effect coefficient based on the changes in the movement data of the critical aggregation group before and after the sound wave time point; and performs feedback and re-recognition on the sound pattern recognition model based on the expulsion effect coefficient.

[0080] The sound wave time point is the moment when the sound emitted by the sound source has an effect on the organism.

[0081] The method for calculating the sound wave time point is as follows: based on the environmental medium composition data, medium movement data, environmental temperature data, and environmental noise data in the monitoring environment data, the time when the sound emitted by the speaker intersects with the movement trajectory of the critical gathering group is determined as the first time period; the movement trajectory of the critical gathering group is obtained through target aggregation parameters.

[0082] The reaction time of the aggregated biological species to sound was obtained based on experimental tests and used as the second time period;

[0083] The sound wave time-effect point is calculated based on the time point when the sound is emitted, the first time period, and the second time period.

[0084] The process of acquiring the movement data of the critical cluster before and after the sound wave time point is as follows: setting a time threshold;

[0085] The motion data of the target cluster within a time threshold range before the sound wave time point is obtained to obtain the first motion data, which includes the movement speed of the clustered organisms, the movement direction of the clustered organisms, the relative distance between the clustered organisms, and the density of the clustered organisms.

[0086] The motion data of the target cluster within a time threshold range after the sound wave time point is obtained to obtain the second motion data, which includes the speed of the expelling organisms, the direction of the expelling organisms, the relative distance of the expelling organisms, and the density of the expelling organisms.

[0087] The expulsion effect coefficient was calculated based on the first and second motion data.

[0088]

[0089] Where Eff represents the expulsion effect coefficient; tce represents the relative distance of the expelled organisms; tan represents the relative distance of the aggregated organisms; θ() represents the angle function of directions in space; dir represents the direction of movement of the aggregated organisms; D1 represents the direction of the line connecting the target aggregate and the monitoring point; qir represents the direction of movement of the expelled organisms; D2 represents the direction of the line connecting the target aggregate and the monitoring point after the sound wave time point; f() represents the disorder function of the direction data, which is implemented through the covariance matrix; qsd represents the movement speed of the expelled organisms; vel represents the movement speed of the aggregated organisms; smd represents the density of the expelled organisms; den represents the density of the aggregated organisms; β1, β2, β3 and β4 represent the expulsion coefficients.

[0090] This invention obtains the sound wave time point by considering the time when the sound emitted by the speaker intersects with the movement trajectory of the critically endangered population, and the reaction time of the species in the population to the sound. It then acquires the movement data of the critically endangered population within a time threshold range before and after the sound wave time point, using it as the first movement data and the second movement data. Based on the changes in relative distance, movement angle, movement speed, and biological density between the first and second movement data, it calculates the expulsion effect coefficient. The expulsion effect coefficient accurately measures the biological expulsion effect.

[0091] This invention performs target biometric identification in a monitored area, obtaining target parameter data and member movement data. A clustering algorithm is used to identify target clusters and their parameters. A target cluster danger coefficient is calculated based on the cluster parameters, and a critical cluster is identified based on this coefficient. The environmental conditions between the critical cluster and the monitoring point are identified, yielding monitoring environment data. An acoustic pattern recognition model is constructed to identify the target cluster parameters and monitoring environment data of the critical cluster, obtaining acoustic pattern data and acoustic parameter data for dispersal. Based on the changes in the critical cluster's movement data, a dispersal effectiveness coefficient is calculated, and the acoustic pattern recognition model is fed back and re-identified, effectively improving the biometric dispersal effect.

[0092] Example 2

[0093] With the deepening exploration of the ocean, the development and maintenance of shipping routes, the construction of offshore resource extraction platforms, and the development of tourism models, human society is increasingly intersecting with the ocean.

[0094] When humans operate at sea, they often encounter disturbances caused by the aggregation of marine life. For example, birds gathering on or circling ships may collide with them, obstruct the navigator's vision, and cause damage. Furthermore, bird droppings containing acidic substances can accumulate and corrode the hull over time. Large schools of fish can also impair navigational clarity, affecting ship propulsion and the efficiency of underwater equipment. Even large fish appearing in areas with high population density can pose safety risks.

[0095] Furthermore, excessive aggregation of organisms can negatively impact certain marine ecosystems, leading to localized ecological imbalances. For example, some birds can overfeed, affecting the populations of certain fish or marine organisms and thus disrupting the marine food chain, necessitating biological expulsion.

[0096] Using sound for biological repulsion is a relatively convenient and environmentally friendly method. The high and low frequency audio equipment control system described in this invention can effectively achieve the effect of repelling marine organisms.

[0097] The structure of the high and low frequency audio equipment control system is as follows: Figure 1 As shown, it includes:

[0098] The target distribution recognition module uses the location of the audio equipment as the monitoring point and obtains the monitoring area based on the set monitoring radius; it identifies target organisms entering the monitoring area and collects target parameter data and member movement data.

[0099] The target parameter data includes: organism type, organism quantity, organism size coefficient, and organism hazard coefficient; wherein, the organism hazard coefficient is obtained based on historical data on the losses caused by organism types to the protected targets;

[0100] The member movement data includes: relative position of the organism, movement speed of the organism, direction of movement of the organism, and relative distance of the organism; the relative position of the organism is the position and orientation data of the organism relative to the monitoring point; the relative distance of the organism is the distance between the organism and the monitoring point.

[0101] The clustering identification module identifies target parameter data and member movement data through a clustering algorithm, and divides the target organisms into target clusters based on the identification results; and obtains the target clustering parameters based on the target parameter data and member movement data of the target organisms in the target clusters.

[0102] The target aggregation parameters include the species of aggregation organisms, biohazard factor, number of aggregation organisms, size factor of aggregation organisms, relative position of aggregation organisms, movement speed of aggregation organisms, movement direction of aggregation organisms, relative distance of aggregation organisms, and density of aggregation organisms;

[0103] The process of obtaining the relative position of the aggregated organisms is as follows: obtain the position data of the target organisms in the target aggregate, construct the minimum circumcircle based on the position data, and use the relative position data of the center of the minimum circumcircle and the monitoring point as the relative position of the aggregated organisms.

[0104] The movement speed of the aggregated organisms is the average movement speed of the target organisms in the target aggregate;

[0105] The direction of movement of the aggregated organisms is the average of the direction of movement of the target organisms in the target aggregate;

[0106] The relative distance between aggregated organisms is the average relative distance between the target organisms in the target aggregate and the monitoring point;

[0107] The process of obtaining the aggregation density is as follows: the minimum circumcircle is constructed based on the location data of the target organisms in the target aggregation, and the aggregation density is obtained based on the volume of the minimum circumcircle and the number of aggregation organisms.

[0108] The marine biological aggregation patterns were identified, and the data is shown in Table 1.

[0109] Table 1. Marine bioaccumulation parameter data

[0110]

[0111] The target clusters in Table 1 are sorted in descending order by the relative distance between the aggregated organisms; and the density of the aggregated organisms is kept to an integer value during the calculation process.

[0112] This invention identifies target parameter data and member movement data through a clustering algorithm, and divides the target organisms into target clusters based on the identification results; based on the target parameter data and member movement data of the target organisms in the target clusters, the target cluster parameters are obtained; the target organisms are accurately divided, providing a basis for subsequent hazard identification.

[0113] The danger level measurement module calculates the danger coefficient of the target cluster based on the target cluster parameters; and determines the most dangerous critical cluster based on the target cluster danger coefficient.

[0114] The formula for calculating the target aggregation risk factor is as follows:

[0115]

[0116] Where Dans represents the target aggregation risk factor; cla i T represents the biohazard coefficient corresponding to the species of organisms that aggregate; i The following values ​​represent the size coefficient of the aggregated organisms; n represents the number of organisms in the target aggregate; θ() represents the angle function in space; dir represents the direction of movement of the aggregated organisms; D1 represents the direction of the line connecting the target aggregate and the monitoring point; DT represents the angle threshold; vel represents the movement speed of the aggregated organisms; VT represents the movement speed threshold; tan represents the relative distance of the aggregated organisms; TT represents the relative distance threshold; den represents the density of the aggregated organisms; ET represents the density threshold; α1, α2, α3 and α4 represent the hazard coefficients; exp represents the exponential function with the natural constant as the base; wherein, the hazard coefficients are obtained through data verification and optimization.

[0117] This invention obtains target aggregation parameters of a target cluster and calculates the target aggregation hazard coefficient of the target cluster from four dimensions: 1. the biohazard coefficient and size weight of the target organisms in the target cluster; 2. the movement direction and speed of the target organisms in the target cluster; 3. the relative distance between the target cluster and the monitoring point; and 4. the density of the target cluster. By accurately and comprehensively measuring the hazard level of the target cluster through these four dimensions, and then accurately identifying the most dangerous critical clusters through the target aggregation hazard coefficient.

[0118] The audio parameter acquisition module identifies the environmental conditions between the critical gathering group and the monitoring point, and obtains monitoring environment data; it constructs an audio pattern recognition model to identify the target gathering parameters of the critical gathering group and the monitoring environment data, and obtains audio pattern data and audio parameter data for biological expulsion; the audio pattern data includes low-frequency mode and high-frequency mode; the audio parameter data includes sound frequency, sound pressure level, sound waveform, sound duration and sound direction.

[0119] The monitoring environment data includes environmental media composition data, media movement data, environmental temperature data, and environmental noise data; the environmental media composition data is the media data between the animal and the monitoring point.

[0120] The environmental media composition data of the target cluster were identified, resulting in Table 2.

[0121] Table 2. Environmental Media Composition Data of the Target Cluster

[0122]

[0123] The sound pattern recognition model includes a parameter data input layer, a parameter data recognition layer, a sound pattern recognition layer, and a sound parameter output layer. The parameter data input layer inputs target aggregation parameters and monitoring environment data into the model. The parameter data recognition layer extracts and recognizes features from the target aggregation parameters and monitoring environment data. The sound pattern recognition layer identifies sound pattern data, including bass and treble modes. The sound parameter output layer identifies sound parameter data, including sound frequency, sound pressure level, sound waveform, sound duration, and sound direction.

[0124] The training process of the sound pattern recognition model is as follows:

[0125] Biological expulsion tests were conducted based on audio, resulting in an expulsion test dataset and expulsion test labels.

[0126] The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound mode data, and test sound parameter data; the expulsion test label includes the test expulsion effectiveness coefficient.

[0127] The sound pattern recognition model was trained using the eviction test dataset and eviction test labels.

[0128] This invention conducts biological expulsion tests based on sound, obtaining an expulsion test dataset and expulsion test labels. The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound pattern data, and test sound parameter data. The expulsion test labels include test expulsion effect coefficients. A sound pattern recognition model is obtained by training the expulsion test dataset and expulsion test labels. The sound pattern recognition model can accurately identify the sound pattern data and sound parameter data with the optimal expulsion effect based on the target aggregation parameters and monitoring environment data.

[0129] The sound effect feedback module obtains the sound wave time point based on the target aggregation parameters and monitoring environment data; calculates the expulsion effect coefficient based on the changes in the movement data of the critical aggregation group before and after the sound wave time point; and performs feedback and re-recognition on the sound pattern recognition model based on the expulsion effect coefficient.

[0130] The feedback and re-identification include adjusting the sound pattern recognition model based on the expulsion effect coefficient; and using the adjusted sound pattern recognition model to re-identify the critical cluster.

[0131] The sound wave time point is the moment when the sound emitted by the sound source has an effect on the organism.

[0132] The method for calculating the sound wave time point is to determine the time when the sound emitted by the speaker intersects with the movement trajectory of the critical gathering group based on the environmental medium composition data, medium movement data, environmental temperature data, and environmental noise data in the monitoring environment data, and use this time as the first time period.

[0133] The reaction time of the aggregated biological species to sound was obtained based on experimental tests and used as the second time period;

[0134] The sound wave time-effect point is calculated based on the time point when the sound is emitted, the first time period, and the second time period.

[0135] The process of acquiring the movement data of the critical cluster before and after the sound wave time point is as follows: setting a time threshold;

[0136] The motion data of the target cluster within a time threshold range before the sound wave time point is obtained to obtain the first motion data, which includes the movement speed of the clustered organisms, the movement direction of the clustered organisms, the relative distance between the clustered organisms, and the density of the clustered organisms.

[0137] The motion data of the target cluster within a time threshold range after the sound wave time point is obtained to obtain the second motion data, which includes the speed of the expelling organisms, the direction of the expelling organisms, the relative distance of the expelling organisms, and the density of the expelling organisms.

[0138] The data acquisition method in the second motion dataset is the same as that in the first motion dataset.

[0139] The expulsion effect coefficient was calculated based on the first and second motion data.

[0140]

[0141] Where Eff represents the expulsion effect coefficient; tce represents the relative distance of the expelled organisms; tan represents the relative distance of the aggregated organisms; θ() represents the angle function of directions in space; dir represents the direction of movement of the aggregated organisms; D1 represents the direction of the line connecting the target aggregate and the monitoring point; qir represents the direction of movement of the expelled organisms; D2 represents the direction of the line connecting the target aggregate and the monitoring point after the sound wave time point; f() represents the disorder function of the direction data; qsd represents the speed of movement of the expelled organisms; vel represents the speed of movement of the aggregated organisms; smd represents the density of the expelled organisms; den represents the density of the aggregated organisms; β1, β2, β3 and β4 represent the expulsion coefficients.

[0142] The expulsion coefficient is obtained through data verification and optimization.

[0143] This invention obtains the sound wave time point by considering the time when the sound emitted by the speaker intersects with the movement trajectory of the critically endangered population, and the reaction time of the species in the population to the sound. It then acquires the movement data of the critically endangered population within a time threshold range before and after the sound wave time point, using it as the first movement data and the second movement data. Based on the changes in relative distance, movement angle, movement speed, and biological density between the first and second movement data, it calculates the expulsion effect coefficient. The expulsion effect coefficient accurately measures the biological expulsion effect.

[0144] The aforementioned high and low frequency audio equipment control system is applied to audio equipment; the structure of the audio equipment is as follows: Figure 3 As shown, it includes a data processing unit, a high-frequency control unit, and a low-frequency control unit. In the data processing unit, target organism identification and early warning are performed to obtain sound mode data and sound parameter data. The high-frequency control unit and the low-frequency control unit are controlled by the sound mode data and sound parameter data. At the same time, the data processing unit can also identify and provide feedback on the expulsion effect.

[0145] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A control system for high and low frequency audio equipment, characterized in that, include: The target distribution recognition module uses the location of the audio equipment as the monitoring point, and then obtains the monitoring area based on the set monitoring radius; Target organisms entering the monitored area are identified, and target parameter data and member movement data are collected; The clustering identification module identifies target parameter data and member movement data through a clustering algorithm, and divides the target organisms into target clusters based on the identification results; based on the target parameter data and member movement data of the target organisms in the target clusters, the target clustering parameters are obtained. The risk level assessment module calculates the target cluster risk coefficient based on the target clustering parameters. The most critical clusters with the highest risk level are identified based on the target cluster risk coefficient; The audio parameter acquisition module identifies the environmental conditions between the critical gathering group and the monitoring point, and obtains monitoring environment data; it constructs an audio pattern recognition model to identify the target gathering parameters of the critical gathering group and the monitoring environment data, and obtains audio pattern data and audio parameter data for biological expulsion; the audio pattern data includes low-frequency mode and high-frequency mode; the audio parameter data includes sound frequency, sound pressure level, sound waveform, sound duration and sound direction. The sound effect feedback module obtains the sound wave time point based on the target aggregation parameters and monitoring environment data; calculates the repulsion effect coefficient based on the changes in the movement data of the critical aggregation group before and after the sound wave time point; and performs feedback and re-recognition on the sound pattern recognition model based on the repulsion effect coefficient. The sound wave time point is the moment when the sound emitted by the sound source has an effect on the organism. The method for calculating the sound wave time point is as follows: based on the environmental medium composition data, medium movement data, environmental temperature data, and environmental noise data in the monitoring environment data, the time when the sound emitted by the speaker intersects with the movement trajectory of the critical gathering group is determined as the first time period; the movement trajectory of the critical gathering group is obtained through target aggregation parameters. The reaction time of the aggregated organisms to the sound repulsion was obtained from experimental tests and used as the second time period. The sound wave time-effect point is calculated based on the time point when the sound is emitted, the first time period, and the second time period. The process of acquiring the movement data of the critical cluster before and after the aforementioned acoustic time point is as follows: Set a time threshold; acquire motion data of the critical cluster within the time threshold range before the sound wave time point to obtain first motion data, which includes the movement speed, movement direction, relative distance, and density of the clustered organisms; The motion data of the critical cluster within the time threshold range after the sound wave time point is obtained to obtain the second motion data, which includes the movement speed, movement direction, relative distance and density of the expelling organisms.

2. The high and low frequency audio equipment control system according to claim 1, characterized in that: The target parameter data includes: organism type, organism quantity, organism size coefficient, and organism hazard coefficient; wherein, the organism hazard coefficient is obtained based on historical data on the losses caused by organism types to the protected targets; The member movement data includes: relative position of the organism, movement speed of the organism, direction of movement of the organism, and relative distance of the organism; the relative position of the organism is the position and orientation data of the organism relative to the monitoring point; the relative distance of the organism is the distance between the organism and the monitoring point.

3. The high and low frequency audio equipment control system according to claim 1, characterized in that: The target aggregation parameters include the species of aggregation organisms, biohazard factor, number of aggregation organisms, size factor of aggregation organisms, relative position of aggregation organisms, movement speed of aggregation organisms, movement direction of aggregation organisms, relative distance of aggregation organisms, and density of aggregation organisms; The process of obtaining the relative position of the aggregated organisms is as follows: obtain the position data of the target organisms in the target aggregate, construct the minimum circumcircle based on the position data, and use the relative position data of the center of the minimum circumcircle and the monitoring point as the relative position of the aggregated organisms. The movement speed of the aggregated organisms is the average movement speed of the target organisms in the target aggregate; The direction of movement of the aggregated organisms is the average of the direction of movement of the target organisms in the target aggregate; The relative distance between aggregated organisms is the average relative distance between the target organisms in the target aggregate and the monitoring point; The process of obtaining the aggregation density is as follows: the minimum circumcircle is constructed based on the location data of the target organisms in the target aggregation, and the aggregation density is obtained based on the volume of the minimum circumcircle and the number of aggregation organisms.

4. The high and low frequency audio equipment control system according to claim 3, characterized in that: The formula for calculating the target aggregation risk factor is as follows: ； in, Indicates the target aggregation risk factor; This indicates the biohazard coefficient corresponding to the species of organisms that aggregate; Indicates the size coefficient of aggregated organisms; Indicates the number of aggregated organisms in the target cluster; An angle function representing directions in space; Indicates the direction of movement of aggregated organisms; Indicates the direction of the line connecting the target cluster to the monitoring point; Indicates the included angle threshold; Indicates the speed of movement of aggregated organisms; Indicates the threshold of motion speed; Indicates the relative distance between aggregated organisms; Indicates the relative distance threshold; Indicates the density of aggregated organisms; Indicates the density threshold; , , and Indicates the risk factor; This represents an exponential function with the natural constant as its base.

5. A high and low frequency audio equipment control system according to claim 4, characterized in that: The monitoring environment data includes environmental media composition data, media movement data, environmental temperature data, and environmental noise data; the environmental media composition data is the media data between the animal and the monitoring point.

6. The high and low frequency audio equipment control system according to claim 1, characterized in that: The sound pattern recognition model includes a parameter data input layer, a parameter data recognition layer, a sound pattern recognition layer, and a sound parameter output layer. The parameter data input layer inputs target aggregation parameters and monitoring environment data into the model. The parameter data recognition layer extracts and recognizes features from the target aggregation parameters and monitoring environment data. The sound pattern recognition layer identifies sound pattern data, including bass and treble modes. The sound parameter output layer identifies sound parameter data, including sound frequency, sound pressure level, sound waveform, sound duration, and sound direction. The training process for the audio pattern recognition model is as follows: Biological expulsion tests were conducted based on audio, resulting in an expulsion test dataset and expulsion test labels. The expulsion test dataset includes test target aggregation parameters, test monitoring environment data, test sound mode data, and test sound parameter data; the expulsion test label includes the test expulsion effectiveness coefficient. The sound pattern recognition model was trained using the eviction test dataset and eviction test labels.

7. The high and low frequency audio equipment control system according to claim 1, characterized in that: The expulsion effect coefficient was calculated based on the first and second motion data. ； in, Indicates the expulsion effect coefficient; Indicates the relative distance between the expelled organisms; Indicates the relative distance between aggregated organisms; An angle function representing directions in space; Indicates the direction of movement of aggregated organisms; Indicates the direction of the line connecting the target cluster to the monitoring point; Indicates the direction of movement of the expelling organisms; Indicates the direction of the line connecting the target cluster and the monitoring point after the sound wave time point; A function representing the degree of disorder in directional data; Indicates the speed of movement of the expelling organisms; Indicates the speed of movement of aggregated organisms; Indicates the density of expelled organisms; Indicates the density of aggregated organisms; , , and This represents the expulsion coefficient.

Images