A method and system for detecting abnormal behavior of fish school

Abstract

The application discloses a fish school abnormal behavior detection method and system, and belongs to the field of aquaculture. The method comprises the following steps: acquiring fish school information based on an image acquisition module; acquiring fish school traces based on the fish school information; judging whether the fish school traces are abnormal based on the fish school traces; acquiring the following relationship of fish individuals in the fish school and obtaining the fish individual leading the fish school when it is determined that the fish school has abnormal behavior; acquiring the fish individual not in the fish school and obtaining the abnormal fish individual when it is determined that the fish school does not have abnormal behavior; acquiring the trace change parameters of the fish individual leading the fish school and / or the abnormal fish individual, acquiring the prediction cause of the abnormal behavior; acquiring the performance information of the fish individual leading the fish school and / or the abnormal fish individual, acquiring the actual cause of the fish school abnormal behavior based on the prediction cause of the abnormal behavior. The method solves the problem that it is difficult to realize real-time and accurate detection of fish school abnormal behavior in the prior art, and improves the detection efficiency and accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aquaculture, and particularly relates to a fish school abnormal behavior detection method and system. BACKGROUND

[0002] In aquaculture, due to the high density of fish, diseases can break out on a large scale in a very short time, causing important economic losses to the breeders, and before the fish get sick, they often have some characteristic performances, and for fish species with the habit of gathering, they will have certain performances in the gathering state, which provides the possibility for the early analysis of fish diseases. However, in the currently adopted technical solutions, the post-analysis method is usually used for disease analysis, and the fish individual is analyzed to analyze the disease performance. The problems of the current method are mainly as follows: firstly, it is difficult to establish a pre-analysis scheme, which leads to the easy spread of diseases in the fish school; secondly, the detection range of fish abnormal behavior is small, especially it is difficult to detect the abnormal behavior of the fish individual leading the fish school, and analyze the overall negative influence degree of the fish individual on the fish school.

[0003] Therefore, how to establish a pre-detection method for fish school abnormal behavior and realize comprehensive and accurate analysis of fish school abnormal behavior is a technical problem to be solved by the person skilled in the art. SUMMARY

[0004] In order to establish a pre-detection system in the detection of fish school abnormal behavior and realize accurate and comprehensive analysis of fish school abnormal behavior, the present application discloses a fish school abnormal behavior detection method and system, in particular:

[0005] First aspect:

[0006] A fish school abnormal behavior detection method, the detection method comprising:

[0007] acquiring fish school information based on an image acquisition module;

[0008] acquiring fish school traces based on the fish school information;

[0009] judging whether the fish school traces are abnormal based on the fish school traces;

[0010] when it is determined that the fish school has abnormal behavior, acquiring the following relationship of fish individuals in the fish school and obtaining the fish individual leading the fish school;

[0011] when it is determined that the fish school does not have abnormal behavior, acquiring the fish individual not in the fish school and obtaining the fish individual with abnormal behavior;

[0012] acquiring trace change parameters of the fish individual leading the fish school and / or the fish individual with abnormal behavior, and acquiring the predicted causes of abnormal behavior;

[0013] acquire the performance information of the leading fish and / or the fish with abnormal behavior, and acquire the actual cause of the abnormal behavior of the fish group based on the predicted cause of the abnormal behavior.

[0014] Optionally, the image acquisition module acquires the fish group information, including:

[0015] The image acquisition module includes a plurality of image acquisition devices arranged on the four side walls and the bottom surface of the space where the fish group is located.

[0016] The image acquisition module acquires the spatial coordinates of the fish.

[0017] The image acquisition module acquires the spatial coordinates of the fish.

[0018] The image acquisition module acquires the spatial coordinates of the fish.

[0019] Optionally, the image acquisition module acquires the fish group information, including:

[0020] The image acquisition module acquires the spatial coordinates of the fish.

[0021] The image acquisition module acquires the spatial coordinates of the fish.

[0022] Optionally, the image acquisition module acquires the fish group information, including:

[0023] The image acquisition module acquires the spatial coordinates of the fish.

[0024] The image acquisition module acquires the spatial coordinates of the fish.

[0025] The image acquisition module acquires the spatial coordinates of the fish.

[0026] The image acquisition module acquires the spatial coordinates of the fish.

[0027] The image acquisition module acquires the spatial coordinates of the fish.

[0028] The image acquisition module acquires the spatial coordinates of the fish.

[0029] when the spatial coverage of the fish school track and the spatial range data of the normal fish school track match, obtaining the characterization information of the fish school;

[0030] when the characterization information of the fish school and the normal characterization information obtained based on the fish school information do not match, determining that the fish school track is abnormal.

[0031] Optionally, when it is determined that the fish school has abnormal track behavior, obtaining the following relationship of fish individuals in the fish school and obtaining the fish individual leading the fish school, comprising:

[0032] obtaining the overall orientation of the fish school and obtaining the head-to-tail distance of adjacent fish individuals in the overall orientation;

[0033] obtaining the fish individual whose head-to-tail distance is not lower than the preset head-to-tail distance of adjacent fish individuals, and obtaining the two-tail fish individual with abnormal distance;

[0034] obtaining the fish individual in the rear position among the two-tail fish individual with abnormal distance in the overall orientation of the fish school;

[0035] obtaining the laterally adjacent fish individual of the fish individual in the rear position in the overall orientation and obtaining the distance of the laterally adjacent fish individual;

[0036] obtaining the fish individual whose distance of all the laterally adjacent fish individuals is not higher than the preset distance of adjacent fish individuals, and including the fish individual in the preset sub-school cluster;

[0037] based on the fish individual in the preset sub-school cluster, simultaneously obtaining the head-to-tail distance of the fish individual and the fish individual in the rear position in the overall orientation, the distance of the fish individual and the laterally adjacent fish individual, and obtaining all the fish individuals not lower than the preset head-to-tail distance of adjacent fish individuals and the preset distance of adjacent fish individuals, respectively, obtaining the edge fish individual of the sub-school cluster;

[0038] based on the range of all the edge fish individuals of the sub-school cluster, obtaining the sub-school cluster of the fish school;

[0039] obtaining the moving direction of the sub-school cluster of the fish school and obtaining the frontmost fish individual in the moving direction, obtaining the fish individual leading the fish school.

[0040] Optionally, when it is determined that the fish school does not have abnormal behavior, obtaining the fish individual not in the fish school and obtaining the behavior abnormal fish individual, comprising:

[0041] obtaining the fish individual outside the fish school, obtaining the free individual;

[0042] based on the fish school information, judging whether the variety information of the free individual is the same as the fish school information;

[0043] The free individual is an abnormal behavior fish individual when the breed information of the free individual is the same as the fish group information.

[0044] The free individual is an abnormal behavior fish individual when the breed information of the free individual is different from the fish group information.

[0045] Optionally, the method further comprises:

[0046] When there are multiple fish groups, obtaining all fish group information and fish group trajectories;

[0047] Obtaining fish groups with abnormal trajectories in all fish group trajectories, and obtaining fish individuals leading the abnormal fish groups;

[0048] Determining the fish individuals leading the abnormal fish groups as the abnormal behavior fish individuals.

[0049] Optionally, the method further comprises:

[0050] Obtaining trajectory change parameters of the fish individuals leading the fish groups and / or the abnormal behavior fish individuals, and obtaining a predicted cause of abnormal behavior.

[0051] Obtaining trajectory change parameters of the fish individuals leading the fish groups and / or the abnormal behavior fish individuals, and obtaining a predicted cause of abnormal behavior.

[0052] Obtaining a trajectory abnormality rate based on the trajectory change parameters.

[0053] Obtaining a predicted cause of abnormal behavior based on the trajectory abnormality rate.

[0054] Optionally, the method further comprises:

[0055] Obtaining possible behaviors of the fish individuals leading the fish groups and / or the abnormal behavior fish individuals based on the fish group information and the predicted cause of abnormal behavior.

[0056] Obtaining actual behaviors of the fish individuals leading the fish groups and / or the abnormal behavior fish individuals based on the image acquisition module.

[0057] Obtaining a behavior type that is the same as the possible behaviors and the actual behaviors, and obtaining a fish group behavior corresponding to the abnormal behavior.

[0058] Obtaining an actual cause of fish group abnormal behavior based on the fish group behavior corresponding to the abnormal behavior.

[0059] Second aspect:

[0060] The fish school abnormal behavior detection system is used for executing the fish school abnormal behavior detection method, and is characterized in that the detection system comprises an image acquisition module, a track analysis module, an abnormal behavior analysis module and an abnormal cause analysis module, wherein:

[0061] The image acquisition module and the track analysis module are connected and used for acquiring the spatial coordinates of the fish.

[0062] The track analysis module is further connected with the abnormal behavior analysis module, and is used for acquiring the track of the fish school after acquiring the spatial coordinates of the fish, and sending the track to the abnormal behavior analysis module.

[0063] The abnormal behavior analysis module is further connected with the abnormal cause analysis module, and is used for determining the predicted cause of the abnormal behavior based on the analysis data acquired by the abnormal behavior analysis module.

[0064] The abnormal cause analysis module is used for acquiring the actual cause of the fish school abnormal behavior.

[0065] The beneficial effects of the present application include:

[0066] 1. The pre-detection is realized. In the technical scheme of the present application, the track abnormal state of the fish school and the abnormal performance of the fish individual deviating from the fish school are analyzed, and then the predicted cause of the fish school abnormal behavior is determined based on the two types of abnormal performances, and the fish behavior with the abnormal behavior is further detected, so as to obtain the actual cause of the fish school abnormal behavior, thereby realizing the pre-detection of the fish school abnormal behavior.

[0067] 2. The detection precision is improved. In the technical scheme of the present application, the predicted cause of the fish school or fish individual abnormal behavior is analyzed based on the track abnormal parameters of the fish school and the fish individual, and then the actual cause of the fish school abnormal behavior is analyzed, thereby obtaining the specific cause of the current fish school abnormal behavior, and realizing the accurate analysis.

[0068] 3. The detection range is widened. In the technical scheme of the present application, for the abnormal behavior performance of the fish school, on the one hand, the analysis of the abnormal performance of the whole fish school is realized, and the fish individual causing the fish school abnormal behavior is determined, and on the other hand, the abnormal behavior performance of the fish individual deviating from the normal fish school is analyzed, thereby realizing the common analysis of the two types of fish abnormal behaviors, and widening the detection range. BRIEF DESCRIPTION OF DRAWINGS

[0069] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments of the present application or the prior art will be briefly introduced as follows. Obviously, the following description is only some embodiments of the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative effort based on these drawings are within the scope of the present application. The drawings are used to provide further understanding of the present disclosure and constitute a part of the specification, and are used to explain the present disclosure together with the following detailed embodiments, but do not constitute a limitation on the present disclosure. In the drawings:

[0070] Figure 1 A fish school abnormal behavior detection method flow chart provided by the embodiment of the present application;

[0071] Figure 2 A fish school abnormal behavior detection system schematic diagram provided by the embodiment of the present application. DETAILED DESCRIPTION

[0072] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort are within the scope of the present application. In addition, in the embodiments of the present application, "first", "second", etc. are used to distinguish similar objects, not necessarily to describe a specific order or sequence.

[0073] In aquaculture, the economic fish species for cultivation have the characteristics of high cultivation density and easy eutrophication of water body, which leads to the spread of some infectious diseases in a very short time in the fish school, resulting in large-area diseases in fish and causing significant economic losses to the breeders. In addition, for various fish species, the diseases and pests that occur usually include a latent period, a development period and an outbreak period. Many diseases will cause abnormal performance of the fish school in the latent period, and if the diseases are found in this period, key prevention can be made. Therefore, in the detection of abnormal behavior of the fish school, early detection needs to be made. However, in the current detection of abnormal behavior of the fish school, the abnormal behavior of the fish school can usually be recognized only in the development period or even the outbreak period, which is obviously not conducive to the prevention and treatment of diseases in aquaculture. Therefore, early detection of abnormal behavior of the fish school is needed.

[0074] In order to realize early detection of abnormal behavior of the fish school in aquaculture and improve the detection accuracy, the present application discloses a fish school abnormal behavior detection method, as shown in Figure 1 A fish school abnormal behavior detection method flow chart provided by the embodiment of the present application, specifically:

[0075] A fish school abnormal behavior detection method, characterized in that the detection method comprises:

[0076] S110, acquiring fish school information based on an image acquisition module.

[0077] S120, acquiring fish school traces based on the fish school information.

[0078] S130, judging whether the fish school traces are abnormal based on the fish school traces.

[0079] S140, acquiring fish individual following relationships in the fish school and obtaining a fish individual leading the fish school when it is determined that the fish school has abnormal trace behaviors.

[0080] S150, acquiring a fish individual not in the fish school and obtaining an abnormal behavior fish individual when it is determined that the fish school does not have abnormal behaviors.

[0081] S160, acquiring trace change parameters of the fish individual leading the fish school and / or the abnormal behavior fish individual and acquiring prediction causes of abnormal behaviors.

[0082] S170, acquiring performance information of the fish individual leading the fish school and / or the abnormal behavior fish individual, acquiring actual causes of fish school abnormal behaviors based on the prediction causes of abnormal behaviors.

[0083] The purpose of all the above steps is to analyze abnormal behaviors of the fish school itself and abnormal behaviors of fish individuals deviating from the normal fish school, then analyze abnormal behaviors of the fish individual leading the fish school, analyze fish abnormal behaviors, determine actual causes of fish school abnormal behaviors based on the abnormal behavior performance, and thus achieve all beneficial effects of the present application.

[0084] In the following, all the above steps will be specifically described as follows:

[0085] As described in step S110, the purpose of this step is to acquire overall information of the fish school based on the set image acquisition module, thereby obtaining fish school information, and then acquiring various key information based on the fish school information, which is used to analyze abnormal performance parameters of the fish school and analyze causes of abnormal behaviors. Specifically:

[0086] S111, the image acquisition module comprises a plurality of image acquisition devices, and the image acquisition devices are arranged on four side walls and a bottom surface of a space where the fish school is located.

[0087] The purpose of this step is that the fish population and the track of the fish individual are in three-dimensional space, so in the specific parameter analysis, the three-dimensional image information of the fish population needs to be collected, and then the three-dimensional track of the fish population is obtained according to the information. That is, in order to realize the analysis of the track of the fish population, at least one image collection device needs to be set to obtain the three-dimensional track.

[0088] Among them, for the image collection module, at least 5 image collection devices are arranged, and each image collection device is arranged on the four side walls and the bottom surface of the space where the fish population is located.

[0089] Among them, for the image collection device arranged, it can be arranged on the four side walls and the bottom surface, and the number of image collection devices on each surface can be more than one.

[0090] In some embodiments, a ranging device is also arranged in the image collection device to obtain the position of each fish.

[0091] S112, based on the image collection module, obtaining the spatial coordinates of the fish.

[0092] The purpose of this step is to obtain the spatial coordinates of the fish, and then determine the position of the fish population and the track parameters of the fish population based on the obtained spatial coordinates. Such parameters need to be obtained based on the spatial coordinates of the fish.

[0093] Among them, after the image collection module obtains the image of the fish, the spatial coordinates of the fish are obtained based on the image collection devices in five directions. The specific method is: based on the arranged image collection device, the position of the fish in the obtained image is obtained, and then the coordinates are obtained based on the positions in all directions.

[0094] Among them, the coordinate system can also be directly arranged on the four side walls and the bottom surface of the space where the fish is located, and then the spatial coordinates of the fish are obtained based on the coordinate coincidence parameters of the fish and the coordinate system.

[0095] In some embodiments, a plurality of cameras are arranged on each of the four side walls and the bottom surface of the space where the fish is located, and the fish is photographed when it reaches the front of the camera, so as to obtain the spatial coordinates of the fish based on the position of the camera.

[0096] In some embodiments, the spatial coordinates of the fish are obtained based on the ranging device.

[0097] S113, based on the spatial coordinates of the fish, obtaining the distance between adjacent fish.

[0098] The purpose of this step is to obtain the overall performance of the fish school, and it is obvious that the fish school needs to be obtained first, and then subsequent analysis can be based on the obtained fish school. Therefore, by obtaining the distance between adjacent fish, the fish school information can be obtained.

[0099] After obtaining the spatial coordinates of the fish, the distance between each fish and the surrounding fish needs to be obtained.

[0100] After obtaining the distance between the fish, the minimum distance is obtained, and the fish with the minimum distance is considered to be in the same fish school.

[0101] For the obtained fish school information, the minimum distance between each fish needs to be obtained, so as to obtain the aggregation state of the fish school.

[0102] For this step, it is not necessary to determine the relative position between the two adjacent fish, but only the distance between the two fish needs to be determined.

[0103] S114, obtain the fish individuals whose distance between adjacent fish is not higher than the preset distance between adjacent fish, the adjacent fish is in the same fish school, and obtain the fish school information.

[0104] The purpose of this step is that for the judgment process of the fish school, it is obvious that it is necessary to judge whether the detected fish is in the same fish school. In this case, it is necessary to compare the distance between adjacent fish with the preset distance to judge the fish school information.

[0105] For the preset distance between adjacent fish, it can be set based on technical experience.

[0106] In some embodiments, the mean distance between adjacent fish is obtained, and then the mean value is directly used as the preset distance between adjacent fish.

[0107] For the fish individuals whose distance is not higher than the preset distance between adjacent fish, it is considered to be in the same fish school.

[0108] After identifying the fish school, the fish school information needs to be obtained, which includes the fish species in the fish school, the number of fish, the common location of fish, etc.

[0109] The beneficial effect of step S110 is that by setting the image acquisition device on the four side walls and one bottom surface, the spatial position of the fish can be directly judged, and it is determined whether the detected fish is in the fish school, so as to realize accurate identification of the fish school and obtain the fish school information.

[0110] As step S120, the purpose of this step is that when the fish school appears abnormal behavior, each fish in the fish school may appear abnormal behavior, such as the fish school no longer maintaining, the fish school can only circle in one direction, etc. Such behavior belongs to the abnormal state of behavior, and the abnormal behavior is closely related to the abnormal behavior of the fish. Therefore, in the specific identification, the behavior of the fish school needs to be detected as a whole, so as to be used for the subsequent fish school abnormal behavior analysis process. Specifically:

[0111] S121, continuously acquiring the center point of the fish school based on the spatial coordinates of the center point of the fish school.

[0112] The purpose of this step is to consider the detection process of the fish school. If the whole fish school is in normal performance, it means that all fish individuals in the fish school are in normal state, and further abnormal behavior analysis is not needed. That is, by acquiring the center point coordinate of the fish school, the abnormal behavior judgment and analysis of the whole fish school can be realized, so as to confirm whether the fish school needs to be analyzed in the subsequent step.

[0113] Among them, for all fish individuals in the fish school, the spatial coordinates can be acquired based on the image acquisition module, so the center point coordinate of the fish school can be directly acquired based on the spatial coordinates.

[0114] Among them, the determination equation of the spatial coordinates of the center point of the fish school is:

[0115] ;

[0116] Among them, P （x，y，z） The spatial coordinates of the center point of the fish school are represented by X, Y and Z respectively. i The fish individual number index in the fish school is represented by i. j The total amount of the fish individual number index in the fish school is represented by N. x i , y i and z i respectively represent the x axis, y axis and z axis coordinate values of the fish individual in the fish school.

[0117] Among them, considering that the fish school is in frequent change in many cases, in the calculation of the center point coordinate, the center point coordinate needs to be calculated in real time.

[0118] In some embodiments, considering the change of the spatial coordinates of the center point of the fish school, it is possible that the change is not caused by the overall position change of the fish school, but simply by the accumulation of small behavioral changes of the fish individuals in the fish school, so further analysis is needed to reduce the interference term. Specifically:

[0119] wherein the spatial coordinates of all fish individuals in the fish school are obtained, and when the spatial coordinates of the center point of the fish school have a small change, the change amount of the spatial coordinates of all fish individuals is obtained;

[0120] wherein the vector of the change amount of the spatial coordinates of all fish individuals is obtained, and if the vector direction deviation degree is large, it is considered that the change of the spatial coordinates of the center point of the current fish school is caused by the discrete motion mode of the fish individuals, which is an interference term.

[0121] wherein, considering the schooling habit of the fish school, if the change of the spatial coordinates of the center point is caused by the overall motion of the fish school, and the vector deviation degree of the change amount of the spatial coordinates of the fish individuals is basically the same, whether it is moving in the same direction or circling around the center point, it is found that when the deviation degree is small, it is considered that the current change of the spatial coordinates of the center point is not an interference term and needs to be confirmed.

[0122] S122, obtaining the center point spatial coordinates of the center point of the fish school at different time nodes, and connecting the center point spatial coordinates at the different time nodes to obtain the track of the fish school.

[0123] The purpose of this step is to obtain the fish school track parameters when the center point spatial coordinates change, so that subsequent track judgment can be based on whether the fish school has abnormal behavior.

[0124] wherein, in the track judgment of the fish school, the parameters of the center point of the fish school at different time nodes need to be determined, and the corresponding relationship between the spatial coordinates and the time points is established.

[0125] wherein, after the change of the center point spatial coordinates is determined, the time node corresponding to the next center point spatial coordinates is determined, so as to obtain the corresponding relationship between the spatial coordinates of the center point of the fish school and the time nodes.

[0126] wherein, by connecting the obtained spatial coordinates of the center point, the track of the fish school can be obtained.

[0127] wherein, in the obtained track of the fish school, the corresponding relationship between the corresponding time nodes is also established, so as to obtain the corresponding relationship between the track and the time nodes.

[0128] The beneficial effect of step S120 is that, based on the connection of the spatial coordinates of the center points of the fish school, when the fish school track is obtained, the analysis of the overall movement state of the fish school can be realized, thereby realizing the analysis of the abnormal behavior state from the whole, reducing the operation resources, and realizing the analysis of the overall fish school state.

[0129] As described in step S130, the purpose of this step is to determine whether the detected fish school is in an abnormal behavior state based on its track information after obtaining the track information of the fish school. Specifically:

[0130] S131, obtain the fish school track occurrence time, and obtain the preset fish school track change rate interval based on the fish habit in the fish school.

[0131] The purpose of this step is that the overall activity mode of the fish school is different in different time periods, and accordingly, if the behavior of the fish school is different from the normal motion state in this time period, there is obviously an abnormal behavior of the fish school, so in the specific processing, the running motion state under the normal fish school state can be determined.

[0132] Among them, according to the habit of the fish school, the overall motion performance of the fish school in different time periods is obtained, for example, at night, fish usually swim less, so the track change rate is low, and during the day, the activity is more, so the track change rate is high.

[0133] Among them, according to the normal fish feeding process, the activity state of the fish school in different time periods can be obtained to obtain the track change rate.

[0134] In some embodiments, other scenarios such as feeding scenarios, water resource replacement scenarios, etc. are also considered, and the track change rate of the fish school in all such scenarios is obtained.

[0135] Among them, for the obtained fish school track change rate, the object of analysis is the fish school with normal behavior.

[0136] Among them, for the preset fish school track change rate interval to be obtained, the obtaining method is: for the fish school with normal performance, the lowest and highest fish school track change rates of the fish school in different time periods or scenarios under the normal motion state are obtained, and are taken as the starting point and the terminal point of the interval.

[0137] Among them, for the fish school track change rate, the calculation method is: the change parameters and directions of the spatial coordinates of the center points of the fish school are obtained, and the occurrence time length when the two spatial points change is obtained, and the ratio of the two is obtained. The fish school track change rate interval can be obtained. And for the preset fish school track change rate, it is obvious that the fish school track change rate with normal behavior is obtained.

[0138] S132, obtain the fish school track change rate, and compare the fish school track change rate with the preset fish school track change rate interval, if the fish school track change rate does not fall into the preset fish school track change rate interval, determine that the fish school is in a possible track abnormal state.

[0139] The purpose of this step is to determine whether the fish school is in an abnormal behavior state by considering the current scene and time of the fish school in the determination of the abnormal behavior of the fish school, and after determining, it is necessary to identify whether the detected fish school actually exists abnormal behavior.

[0140] Among them, the track change rate of the fish school needs to be obtained based on the measured results, so as to obtain the specific track change rate.

[0141] Among them, the obtained fish school track change rate and the obtained preset fish school track change rate interval are compared to analyze whether the former falls into the range of the latter.

[0142] Among them, when the former falls into the interval of the latter, it is considered that the current fish school may be in a track abnormal state.

[0143] Among them, the reason for setting it as a possible track abnormal state is that in the action of the fish school, it may be in a short-term track change and normal state mismatching condition in some cases, such as short-time stress, fish school collective floating caused by low air pressure, etc. These conditions actually belong to the normal activity state of the fish school, and if it is directly determined as fish school abnormality, it is obviously unreasonable, therefore, in the specific processing, the obtained result is considered to be in a possible abnormal state.

[0144] S133, when the fish school is in a possible track abnormal state, obtain the spatial coverage range data of the fish school track.

[0145] The purpose of this step is to further analyze whether the fish school is in a real abnormal behavior state when the track of the fish school indicates that it may be in an abnormal state, so other parameters need to be introduced for processing.

[0146] Among them, for the spatial coverage range, the coverage area of the central point spatial coordinate of the fish school in the whole fish space needs to be analyzed, so as to determine the spatial coverage range of the fish school track.

[0147] Among them, for the obtained spatial coverage range, the spatial coverage range data needs to be analyzed, so as to obtain the accurate spatial coverage range data.

[0148] Among them, in the process of obtaining the spatial coverage range data, the staying time of the fish school in the position also needs to be obtained.

[0149] In some embodiments, it can also be determined that the fish school is in activity, and the change rate of the track of the fish school in each spatial coverage area is analyzed.

[0150] In S134, spatial coverage range data of normal fish school track is obtained based on the fish school information.

[0151] The purpose of this step is to set the standard for whether the fish school is in an abnormal behavior state, and therefore the purpose of this step is to obtain the spatial coverage range data of the normal fish school.

[0152] In this step, after obtaining the fish school information, the fish species and the current growth cycle of the fish in the fish school information can be obtained, and then based on such information, the spatial coverage range of the normal fish school in the activity process is obtained.

[0153] In this step, the spatial coverage range of the normal fish school track can also be obtained based on historical data, so as to obtain the final result.

[0154] In this step, the spatial coverage range data of the normal fish school track is also established in correspondence with the coverage range and time, activity scene and other information, so as to obtain the spatial coverage range data of the normal fish school track.

[0155] In this step, the method of obtaining the spatial coverage range data of the normal fish school track is the same as that of step S133, and therefore will not be described here.

[0156] In S135, when the spatial coverage range data of the fish school track and the spatial coverage range data of the normal fish school track do not match, it is determined that the fish school track is abnormal.

[0157] The purpose of this step is to determine whether the detected fish school track is normal based on the track coverage range when it is determined that the fish school is in a possible abnormal track state.

[0158] In this step, the obtained spatial coverage range data of the fish school track and the spatial coverage range of the normal fish school track are compared, and when the ranges of the two do not obviously correspond, it is considered that the fish school track is abnormal.

[0159] In this step, the time range and scene data corresponding to the spatial coverage range data of the fish school track are also needed to find the normal fish school track hole spatial coverage range data corresponding to the two data, and when it is determined that the two do not match, it is considered that the fish school track is abnormal.

[0160] Based on the above steps, it can be basically determined whether the current fish school is in a track abnormal state, but there is a case where the fish school track spatial coverage data is in a normal state, but the fish school may still have abnormal behavior, therefore, further fish school track abnormal behavior judgment can be performed. Specifically:

[0161] S136, the judgment of whether the fish school is track abnormal, further comprises:

[0162] When the spatial coverage of the fish school track and the spatial range data of the normal fish school track match, the characteristic information of the fish school is obtained.

[0163] The purpose of this step is to analyze the state of the fish school. When the fish school track is considered normal based on the above steps, other parameters need to be introduced for analysis. Therefore, the purpose of this step is to determine other data that can be introduced.

[0164] The so-called characteristic information of the fish school refers to all performance information that can be used to describe the basic information of the fish school, such as fish school density, fish school movement direction, and further fish school grouping situation. All information that can be used to describe the state of the fish school can be applied as the characteristic information of the fish school.

[0165] The characteristic information of the fish school can be obtained based on the image acquisition module to obtain the characteristic information of the fish school.

[0166] S136, when the characteristic information of the fish school and the normal characteristic information obtained based on the fish school information do not match, it is determined that the fish school track is abnormal.

[0167] After obtaining the characteristic information of the fish school, the normal guarantee information under normal circumstances needs to be obtained, so as to compare the two types of information to determine whether it is in a fish school track abnormal state.

[0168] After obtaining the fish school information, the normal characteristic information of the fish school in the current fish school growth cycle can be obtained.

[0169] The normal characteristic information of the fish school needs to include all normal characteristic information, so that all normal characteristic information obtained is used as a normal guarantee information library.

[0170] The obtained characteristic information of the fish school and the normal characteristic information library are compared. When it is found that the characteristic information of the fish school and any characteristic information in the normal characteristic information library do not match, it is considered that the detected fish school is in a track abnormal state. For example, during feeding, the fish school will swim to the water surface and the swimming speed will increase, but if it is found that the fish school is still in the bottom area, it means that the fish school is in a track abnormal state.

[0171] The beneficial effect of step S130 is that, for the detected fish school, it is necessary to determine whether the fish school is likely to be abnormal in behavior according to the activity time and the scene of the fish, and then to determine whether it is in a truly abnormal behavior state based on the likely abnormal behavior, so as to realize accurate judgment of the abnormal state of the fish school based on multiple parameters. In addition, two types of methods are used to determine the abnormal behavior state. Specifically, when it is determined that the fish school is likely to be normal, it is further determined whether the fish school is in an abnormal behavior state based on the representation information of the fish school.

[0172] As described in step S140, the beneficial effect of this step is that, in the overall activity of the fish school, a following swimming relationship is usually automatically generated between fish individuals. If the fish individual leading the fish school has a disease, no matter which development cycle it is in, it will fundamentally affect the abnormal behavior state of the entire fish school. Therefore, in the specific processing, the fish individual leading the fish school is analyzed. Specifically:

[0173] S141, obtain the overall orientation of the fish school, and obtain the head-tail distance between adjacent fish individuals in the overall orientation.

[0174] The purpose of this step is that, in the determination of the fish individual leading the fish school, the fish school is obviously the analysis object. However, after the fish school shows abnormal behavior, the phenomenon of fish school division may occur. Before the division, some fish individuals may have a phenomenon of increased head-tail distance between fish individuals in the previous fish school, and after the division, the overall orientation of the fish school is known. Therefore, it is necessary to determine these two types of information to lay a foundation for subsequent analysis work.

[0175] In the analysis of the overall orientation of the fish school, the overall orientation of the fish school can be determined by obtaining the orientation of all fish individuals in the fish school.

[0176] For the fish school in the rotating state, the rotating direction of the fish school is the overall orientation.

[0177] After obtaining the overall orientation, the head-tail distance between adjacent fish individuals with the orientation relationship and in the front-back swimming distribution is obtained.

[0178] The head-tail distance refers to the distance between the tail of the previous fish and the head of the next fish.

[0179] S142, obtain the fish individuals whose head-tail distance is not less than the preset head-tail distance of adjacent fish individuals, and obtain the two-tail fish individuals with abnormal distance.

[0180] The purpose of this step is to analyze whether the school splitting problem exists after the abnormal school track is determined. When the school splitting occurs, it is obvious that there is a more serious abnormal performance.

[0181] After the interval is obtained, the interval is compared with the preset interval. When it is found that the former is not lower than the latter, it is considered that the interval obviously does not conform to the rule, and the school splitting problem may exist.

[0182] For the determination of the preset interval between adjacent fish individuals, the mean value obtained based on all intervals is obtained, and the mean value is the preset value.

[0183] S143, obtaining the fish individual in the rear position in the two fish individuals of the abnormal interval in the overall orientation of the fish school.

[0184] The purpose of this step is to determine the splitting range when the school splitting occurs. For the school splitting phenomenon, it is necessary to determine the leading fish individual of the school splitting and analyze the size of the school splitting. Therefore, in the specific processing process, the fish individual in the rear position is analyzed to form the basis of the school splitting.

[0185] The fish individual in the rear position means that, among the adjacent fish individuals in the same orientation, the fish individual in the rear position in the orientation direction is the fish individual in the rear position.

[0186] The fish individual in the rear position needs to be determined only when the interval between the two fish is determined to be abnormal.

[0187] S144, obtaining the laterally adjacent fish individual of the fish individual in the rear position in the overall orientation, and obtaining the interval of the laterally adjacent fish individual.

[0188] The purpose of this step is that although the interval between the two fish in the overall orientation of the fish school may be abnormal, this phenomenon has a high incidence and cannot be directly judged based on the result whether the school splitting tendency exists. Therefore, in the analysis of the school splitting tendency, the aggregation state of the laterally adjacent fish individual in the overall orientation is also needed to be determined.

[0189] In the determination of the laterally adjacent fish individual of the fish individual in the rear position in the overall orientation, the laterally adjacent fish individual includes the fish individual in the up, down, left and right directions of the fish individual in the rear position.

[0190] The fish individual in the lateral direction also does not need to be in a completely coincident state, but only needs to have a coincident area in the projection in each direction.

[0191] wherein, for the lateral adjacent fish individual distance, all fish individual distances in the lateral direction are directly obtained, to obtain the lateral adjacent fish individual distance.

[0192] S145, fish individuals with a distance not higher than the preset adjacent fish individual distance are obtained from all the lateral adjacent fish individual distances, and are included in the preset sub-group cluster.

[0193] The purpose of this step is that, after the obtained distance, if the fish school with abnormal trajectory is about to appear the sub-group phenomenon, the fish density in the sub-group fish school about to sub-group is larger, so in the processing, the sub-group cluster needs to be constructed based on the judgment of the distance.

[0194] wherein, after the lateral adjacent fish individual distance is determined, and compared with the preset distance, if the former is not higher than the latter, it is obvious that the adjacent fish individuals have a higher density, and there is a tendency to sub-group.

[0195] wherein, after the judgment, the fish individuals meeting the sub-group tendency are included in the same cluster, and the obtained cluster is the preset sub-group cluster.

[0196] wherein, for the preset adjacent fish individual, it can be set based on the experience of the technical personnel, or the mean value can be obtained based on all the obtained lateral adjacent fish individual distances.

[0197] S146, based on the fish individuals in the preset sub-group cluster, the fish individual tail and the fish individual head distance at the next position in the overall orientation, the fish individual and the lateral adjacent fish individual distance are obtained, and all fish individuals not lower than the preset adjacent fish individual head-tail distance and the preset adjacent fish individual distance are obtained, to obtain the sub-group cluster edge fish individual.

[0198] The purpose of this step is that, after the abnormal trajectory of the fish school is determined, it is necessary to analyze whether the fish school really contains a sub-group fish school, so that the fish individuals in the sub-group cluster are determined based on the analysis result, to obtain the sub-group result.

[0199] wherein, after the obtained preset sub-group cluster, the fish individuals therein are obtained.

[0200] wherein, for the fish individuals in the sub-group cluster, the fish individual head distance at the tail position and the next position is obtained, and compared with the preset distance in the preset overall orientation, if it is found that the former is greater than the latter, it means that the next fish individual may not be in the same sub-group cluster as the former fish individual.

[0201] Wherein, the fish individual in the sub-cluster is obtained, and the interval in the lateral direction is obtained, if the interval is not lower than the preset interval of the adjacent fish individual, it is considered that the fish individual and the adjacent fish individual in the lateral direction are not in the same sub-cluster.

[0202] Wherein, all the fish individuals in the preset sub-cluster are analyzed according to the above method, and all the fish individuals in the same sub-cluster in the lateral adjacent state are obtained.

[0203] Wherein, for the fish individual in the preset sub-cluster, the fish individual at the edge of the cluster is obtained, and the interval judgment is performed on the fish individual outside the cluster to determine whether the fish individual outside the preset sub-cluster needs to be included in the preset sub-cluster, until the fish individual at the edge of the cluster in the lateral direction of the overall fish school orientation is obtained.

[0204] Wherein, the fish individual in the overall fish school orientation is also analyzed based on the interval in the front-back direction, and the fish individual at the edge of the overall fish school orientation is obtained.

[0205] Wherein, after the fish individual at the edge of the overall fish school orientation and the lateral orientation is obtained, the fish individual is the fish individual at the edge of the sub-cluster, and the fish individual inside the fish individual is the fish individual of the sub-cluster.

[0206] S147, obtaining the sub-cluster of the fish school based on the range of all the fish individuals at the edge of the sub-cluster.

[0207] The purpose of this step is to establish the sub-cluster of the fish school after the fish individual at the edge is obtained.

[0208] Wherein, the fish individual at the edge of the sub-cluster is obtained, and the population coverage range is obtained.

[0209] Wherein, for the obtained population coverage range, all the fish individuals inside the population coverage range are included in the same cluster, and the sub-cluster of the fish school is obtained.

[0210] S148, obtaining the moving direction of the sub-cluster of the fish school, and obtaining the fish individual at the front of the moving direction to obtain the fish individual leading the fish school.

[0211] The purpose of this step is that for the fish school with abnormal trajectory, whether there is a sub-cluster or not, and for the sub-cluster itself, it essentially has a fish individual leading the fish school, and its influence on the fish school is the largest. Therefore, in order to realize the analysis of the cause of the fish school anomaly with less computing resources, it is necessary to obtain the fish individual leading the fish school.

[0212] Wherein, for the fish group, or for the whole fish group, the fish individual leading the fish group can be obtained based on the spatial coordinates of the fish individual.

[0213] Wherein, for the fish individual leading the fish group, if the fish group is in a sustained state of motion, the fish individual leading the fish group can also be directly identified based on the image acquisition module.

[0214] The beneficial effect of step S140 is to determine the cluster state of the fish group based on the analysis of the aggregation state of the fish individuals in the fish group in the front and lateral directions, and to determine the fish individual leading the fish group, so as to reduce the number of fish individuals that need to be analyzed in the fish group abnormal behavior detection, and to reduce the consumption of computing resources.

[0215] As described in step S150, the purpose of this step is that in some cases, some fish individuals will appear to be separated from the fish group, which usually means that there is abnormal behavior, especially when the fish group is in a normal state, so in the specific fish group abnormal behavior analysis, the fish individuals separated from the fish group also need to be detected. Specifically:

[0216] S151, obtaining the fish individual outside the fish group, obtaining the free individual.

[0217] The purpose of this step is that for the obtained fish group, there may be fish individuals separated from the fish group, which are likely to mean abnormal behavior, so it is necessary to identify such fish individuals.

[0218] Wherein, based on the fish individual outside the fish group, the spatial coordinates of the free individual are obtained in real time.

[0219] Wherein, the tendency of the free individual entering the fish group is analyzed, specifically by analyzing the trajectory vector of the free individual and comparing it with the overall direction of the fish group to determine the tendency. Wherein, the trajectory vector equation of the free individual is:

[0220] ;

[0221] Wherein, represents the trajectory vector of the free individual during the time node n to m ; represents the spatial coordinates of the free individual at time node m ; represents the spatial coordinates of the free individual at time node n .

[0222] Wherein, all trajectory vectors of the free individual during the activity are obtained, so as to obtain the sustained trajectory vector parameter.

[0223] Wherein, the included angle relationship between the trajectory vector of the free individual and the overall orientation of the fish school needs to be obtained, and the included angle equation is:

[0224] ;

[0225] Wherein, θ represents the trajectory vector of the free individual at time nodes m and n , and the included angle between the trajectory vector of the free individual and the overall orientation of the fish school at time nodes m and n , and represents the trajectory vector of the overall orientation of the fish school at time nodes m and n .

[0226] Wherein, based on the above included angle equation, the included angle between the trajectory vector of the free individual and the trajectory vector of the overall orientation of the fish school in different time periods is obtained, and if the included angle is in a continuous decreasing state, it is considered that the free individual has a high tendency to enter the fish school.

[0227] Wherein, other included angle judgment methods can also be used to analyze whether the free individual has a high tendency to integrate into the fish school.

[0228] Wherein, for the free individual with low tendency, it is considered as a free individual.

[0229] S152, based on the fish school information, judging whether the breed information of the free individual is the same as the fish school information.

[0230] The purpose of this step is that in some cases, fish of different breeds will be added during the breeding process to stimulate the fish school, and such fish usually will not mix with the fish school, therefore, based on the breed information, the interference term needs to be excluded.

[0231] Wherein, based on the fish school information, the fish breed information in the fish school is obtained.

[0232] Wherein, the fish school breed information and the breed information of the free individual are compared, and when it is found that the two are different, it is considered that the free individual is an interference term, which can be directly excluded.

[0233] S153, when the breed information of the free individual is the same as the fish school information, the free individual is an abnormal behavior fish individual.

[0234] The purpose of this step is to identify the abnormal individual of the same breed fish individual in the fish school, so as to determine whether the fish individual is in an abnormal state.

[0235] When the free individual is found to be of the same species as the fish species recorded in the fish population information, the free individual is considered to be an abnormal fish individual.

[0236] S154, when the species information of the free individual and the fish population information are different, the track of the free individual is obtained, and when the track of the free individual is abnormal, the free individual is an abnormal fish individual.

[0237] The purpose of this step is that in some cases, the abnormal behavior of the added fish of other species may also exist, and this abnormal behavior may mean that the pathogen existing in the set fish individual may spread to the fish population, therefore, the abnormal behavior of fish of different species also needs to be distinguished.

[0238] Wherein, when the species information of the free individual and the fish population information are different, the track of the free individual is obtained.

[0239] Wherein, the obtained track of the free individual and the normal track of the free individual are compared to determine whether the track of the free individual is abnormal.

[0240] The above is for the abnormal behavior of all obtained fish individuals, but in some cases, fish individuals with abnormal behavior may also exhibit clustering state, therefore, in specific processing, the possible clustering effect also needs to be determined. Specifically:

[0241] S155, when there are multiple fish populations, the information and track of all fish populations are obtained.

[0242] The purpose of this step is that in some cases, there may be multiple fish populations in the space where the whole fish population is located, and one of the fish populations may have abnormal behavior, therefore, in the determination, all existing fish populations need to be identified and the track is obtained.

[0243] Wherein, the method of obtaining all fish population information and fish population track is the same as that of steps S110 and S120, which will not be repeated here.

[0244] S156, obtain the fish population with abnormal track in all fish population tracks, and obtain the fish individual leading the abnormal fish population.

[0245] The purpose of this step is that for each fish population, if any one or more fish populations are found to have abnormal track, the fish individual leading the abnormal fish population needs to be obtained, so as to determine the abnormal fish individual.

[0246] Wherein, the method of judging the fish population with abnormal track and the fish individual leading the abnormal fish population is the same as that of steps S130 and S140, which will not be repeated here.

[0247] S157, determine the fish individual leading the abnormal fish group as the behavior abnormal fish individual.

[0248] The purpose of this step is to also regard the fish individual leading the abnormal fish group as an abnormal fish individual.

[0249] In which, the fish individual leading the abnormal fish group is directly treated as an abnormal fish individual.

[0250] The beneficial effect of step S150 is that the interference term in the abnormal behavior of the fish group is also analyzed, so as to realize the judgment of the abnormal behavior of all species of fish in the space where the fish is located, and at the same time, the abnormal fish individual is identified in the case of multiple fish groups.

[0251] As described in step S160, the purpose of this step is to analyze the cause of the abnormal behavior after obtaining the fish individual leading the fish group and / or the behavior abnormal fish individual based on the specific performance, so as to analyze the problems existing in the fish group. Specifically:

[0252] S161, obtain the track parameters of the fish individual leading the fish group and / or the behavior abnormal fish individual, and obtain the track change parameters.

[0253] The purpose of this step is that when the fish group has abnormal performance in the track, the cause of the abnormal behavior can be reflected in the specific state of the fish track, so in order to analyze the cause, the track change parameters need to be obtained.

[0254] In which, the track change parameters are judged according to the specific parameter types of the track, including: the change rate of the track, the turning radius of the track, the change frequency of the track, the common position of the fish group, the grouping probability of the fish group, etc.

[0255] In which, all the obtained parameters need to be obtained based on the time parameter, and can also be obtained according to the historical occurrence probability and other parameters.

[0256] S162, obtain the track change parameter weight based on the track change parameter.

[0257] The purpose of this step is that for the abnormal behavior of the fish group, usually multiple different track change parameters correspond to the severity of an abnormal behavior, so in the specific processing, the weight of the track change parameter needs to be obtained according to the corresponding relationship between the severity of the abnormal behavior and the track change parameter, which is used for subsequent cause analysis.

[0258] In which, the change rate of all the obtained track change parameters is obtained, and the proportion of the change rate is obtained, wherein the proportion of the change rate is:

[0259] ;

[0260] wherein, R s represents the proportion of the change rate of the first s trajectory change parameter; C s represents the change rate of the first s trajectory change parameter; C r represents the change rate of the first r trajectory change parameter; r represents the index of the trajectory change parameter; t represents the total amount of the index of the trajectory change parameter.

[0261] wherein, after obtaining the proportion of the change rate, the obtained proportion value can be directly applied as the weight.

[0262] wherein, for the calculated proportion parameter, the proportion parameter can be processed based on the way of setting the calculation time period, so as to realize the continuous acquisition of the proportion parameter.

[0263] S163, based on the trajectory change parameter weight, obtaining the trajectory abnormality rate.

[0264] The purpose of this step is to integrate all the obtained trajectory change parameters into the same parameter in the analysis of the trajectory abnormal behavior, so as to judge the final result based on the obtained result.

[0265] wherein, for the trajectory abnormality rate, a corresponding equation needs to be used for calculation, and the trajectory abnormality rate calculation equation is:

[0266] ;

[0267] wherein, σ represents the trajectory abnormality rate.

[0268] S164, based on the trajectory abnormality rate, obtaining the prediction cause of the abnormal behavior.

[0269] The purpose of this step is that, after obtaining the trajectory abnormality rate, the cause can be directly obtained based on the parameter.

[0270] wherein, in the prediction cause analysis of the abnormal behavior, the correlation between all the abnormal causes and the trajectory abnormality rate can be obtained based on the breeding experience, and based on this correlation, the corresponding abnormal rate can be calculated.

[0271] In the prediction cause of the abnormal behavior, the weight of the track abnormality parameter in each type of track abnormality rate needs to be determined according to the cause, for example, for the "scale explosion" phenomenon of the fish, the turning radius of the track gradually increases, and then the weight of the turning radius in the track abnormality rate needs to be increased in the established corresponding relationship.

[0272] In some embodiments, the prediction cause of the abnormal behavior can also be based on the combination of the track, the abnormal situation is constructed, and the track abnormality rate of each type of track change parameter that can cause the track abnormality is obtained, and the corresponding relationship between the track abnormality rate and the prediction cause of the corresponding abnormal behavior is established.

[0273] The beneficial effect of step S160 is that the cause of the abnormal behavior that may be caused can be predicted based on the track change rate, and only a small number of parameters need to be directly obtained in the process, which reduces the data amount while ensuring the accuracy of the cause analysis.

[0274] As described in step S170, the purpose of this step is to obtain multiple results based on the prediction cause of the abnormal behavior, and then determine the cause of the abnormal behavior of the fish school based on the obtained prediction cause. Specifically:

[0275] S171, based on the fish school information and the prediction cause of the abnormal behavior, obtaining the possible performance of the fish individual leading the fish school and / or the behavior abnormal fish individual.

[0276] The purpose of this step is to determine the basis for comparison in the analysis of the actual cause of the abnormal behavior of the fish school, so the purpose of this step is to obtain the numerical basis for comparison based on the obtained prediction cause.

[0277] In the case of obtaining the track abnormality rate, the prediction cause of each type of abnormal behavior corresponding to the track abnormality rate can be directly obtained, and the performance of the fish individual that may be caused by the obtained prediction cause can be obtained.

[0278] In the obtained possible performance, the corresponding relationship between the abnormal behavior and the growth cycle of the fish school needs to be established according to different growth cycles of the fish school,

[0279] S172, based on the image acquisition module, obtaining the performance of the fish individual leading the fish school and / or the behavior abnormal fish individual, and obtaining the actual performance.

[0280] The purpose of this step is to detect the actual performance of the fish school in the analysis of the cause of the abnormal behavior of the fish school.

[0281] Among them, for the actual performance, including the appearance performance of fish individuals, the change parameters of the track, etc., all types of performances can be regarded as actual performance, and the application is not limited.

[0282] S173, acquiring the performance type of the possible performance and the actual performance, obtaining the fish school performance corresponding to the abnormal behavior.

[0283] The purpose of this step is that, for the obtained possible performance and actual performance comparison result, after analysis, when it is found that the two match, it can be used to directly determine the cause of the fish school abnormal behavior.

[0284] Among them, the possible performance and the actual performance are compared item by item, and when the same is found, the fish school performance corresponding to the abnormal behavior is obtained.

[0285] Among them, the fish school performance corresponding to the abnormal behavior is essentially the possible performance mentioned above.

[0286] S174, based on the fish school performance corresponding to the abnormal behavior, obtaining the actual cause of the fish school abnormal behavior.

[0287] The purpose of this step is to determine the actual cause of the fish school abnormal behavior.

[0288] Among them, after the fish school performance corresponding to the abnormal behavior is determined, the actual cause of the specific fish school abnormal behavior can be determined based on the abnormal behavior cause corresponding to the performance.

[0289] The beneficial effect of step S170 is that, by separately obtaining the possible performance and the actual performance, the actual cause of the fish school abnormal behavior is directly determined, and by the way of item-by-item comparison, the item-by-item determination of the actual cause can be realized, and the analysis accuracy of the actual cause is improved.

[0290] In addition, the application also discloses a fish school abnormal behavior detection system for executing the fish school abnormal behavior detection method described in steps S110-S170. Figure 2 As shown in FIG. 1, it is a fish school abnormal behavior detection system schematic diagram provided by the embodiment of the application. The detection system comprises an image acquisition module, a track analysis module, an abnormal behavior analysis module and an abnormal cause analysis module, wherein:

[0291] The image acquisition module and the track analysis module are connected, and are used for acquiring the spatial coordinates of the fish.

[0292] The track analysis module is also connected with the abnormal behavior analysis module, and is used for acquiring the track of the fish school after acquiring the spatial coordinates of the fish, and sending the track to the abnormal behavior analysis module.

[0293] The abnormal behavior analysis module is further connected with the abnormal cause analysis module, and is configured to determine a predicted cause of the abnormal behavior based on the analysis data obtained by the abnormal behavior analysis module.

[0294] The abnormal cause analysis module is configured to obtain an actual cause of the abnormal behavior of the fish school.

[0295] The beneficial effects of the present application include:

[0296] 1. The present application realizes the pre-detection. In the technical solution of the present application, the abnormal state of the track of the fish school and the abnormal performance of the fish individual that is separated from the fish school are analyzed, and then the predicted cause of the abnormal behavior of the fish school is determined based on the two types of abnormal performances, and the fish behavior with the abnormal behavior is further detected, so as to obtain the actual cause of the abnormal behavior of the fish school, thereby realizing the pre-detection of the abnormal behavior of the fish school.

[0297] 2. The present application improves the detection accuracy. In the technical solution of the present application, the predicted cause of the abnormal behavior of the fish school or the fish individual is analyzed based on the track abnormal parameters of the fish school and the fish individual, and then the actual cause of the abnormal behavior of the fish school is analyzed by analyzing the appearance performance, so as to obtain the specific cause of the current abnormal behavior of the fish school, thereby realizing the accurate analysis.

[0298] 3. The present application expands the detection range. In the technical solution of the present application, for the abnormal behavior performance of the fish school, on the one hand, the analysis of the abnormal performance of the whole fish school is realized, and the fish individual that causes the abnormal behavior of the fish school is determined, and on the other hand, the abnormal behavior performance of the fish individual that is separated from the normal fish school is analyzed, thereby realizing the common analysis of the two types of fish abnormal behavior, and expanding the detection range.

[0299] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by computer program instruction related hardware, and the foregoing computer program can be stored in a non-volatile storage medium. When the computer program is executed, the steps of the method embodiments are executed. Alternatively, when the above-mentioned integrated units are realized in the form of software function modules and sold or used as independent products, they can also be stored in a non-volatile storage medium. Based on this understanding, the technical solutions of the embodiments of the present application can be embodied in the form of software products, which are stored in a non-volatile storage medium and include a plurality of instructions for causing an electronic device (which can be a personal computer, a server, a network device, etc.) to execute all or part of the methods described in the embodiments of the present application.

[0300] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A method of detecting abnormal behavior of a fish school, characterized by, The detection method comprises: Based on the image acquisition module, the fish school information is acquired; Based on the fish school information, the fish school track is acquired; Based on the fish school track, it is judged whether the fish school track is abnormal; When it is determined that the fish school has abnormal behavior, the fish individual following relationship in the fish school is acquired, and the fish individual leading the fish school is obtained; When it is determined that the fish school does not have abnormal behavior, the fish individual not in the fish school is acquired, and the behavior abnormal fish individual is obtained; The track change parameter of the fish individual leading the fish school and / or the behavior abnormal fish individual is acquired, the prediction cause of the abnormal behavior is acquired; The performance information of the fish individual leading the fish school and / or the behavior abnormal fish individual is acquired, the actual cause of the fish school abnormal behavior is acquired based on the prediction cause of the abnormal behavior, comprising: Based on the fish school information and the prediction cause of the abnormal behavior, the possible performance of the fish individual leading the fish school and / or the behavior abnormal fish individual is acquired; Based on the image acquisition module, the performance of the fish individual leading the fish school and / or the behavior abnormal fish individual is acquired, and the actual performance is acquired; The performance type same as the possible performance and the actual performance is acquired, and the fish school performance corresponding to the abnormal behavior is obtained; Based on the fish school performance corresponding to the abnormal behavior, the actual cause of the fish school abnormal behavior is acquired.

2. The method of claim 1, wherein, The fish school information is acquired based on the image acquisition module, comprising: The image acquisition module comprises a plurality of image acquisition devices, and the image acquisition devices are arranged on the four side walls and the bottom surface of the space where the fish school is located; The spatial coordinates of the fish are acquired based on the image acquisition module; The distance between adjacent fish is acquired based on the spatial coordinates of the fish; The fish individual whose distance with adjacent fish is not higher than the preset distance between adjacent fish is acquired, the adjacent fish is in the same fish school, and the fish school information is acquired.

3. The method of claim 1, wherein, The fish school track is acquired based on the fish school information, comprising: The center point of the fish school is continuously acquired, and the spatial coordinates of the center point of the fish school are acquired; The center point spatial coordinates of the center point of the fish school at different time nodes are acquired, and the center point spatial coordinates at different time nodes are connected to acquire the fish school track.

4. The method of claim 1, wherein, Based on the fish school track, it is judged whether the fish school track is abnormal, comprising: The fish school track occurrence time is acquired, and the preset fish school track change rate interval is acquired based on the fish habit in the fish school; The fish school track change rate is acquired, and compared with the preset fish school track change rate interval, if the fish school track change rate does not fall into the preset fish school track change rate interval, it is determined that the fish school is in the possible track abnormal state; When the fish school is in the possible track abnormal state, the spatial coverage range data of the fish school track is acquired; Based on the fish school information, the spatial coverage range data of the normal fish school track is acquired; When the spatial coverage range data of the fish school track and the spatial coverage range data of the normal fish school track do not match, it is determined that the fish school track is abnormal; The fish school track is also judged, comprising: When the spatial coverage range of the fish school track and the spatial range data of the normal fish school track match, the representation information of the fish school is acquired; When the representation information of the fish school and the normal representation information acquired based on the fish school information do not match, it is determined that the fish school track is abnormal.

5. The method of claim 1, wherein, The fish individual leading the fish school is obtained when it is determined that the fish school has abnormal behavior, including: Obtaining the overall orientation of the fish school and the head-to-tail distance between adjacent fish individuals in the overall orientation; Obtaining fish individuals with a head-to-tail distance not lower than a preset head-to-tail distance between adjacent fish individuals, and obtaining two fish individuals with abnormal distance; Obtaining the fish individual in the rear position among the two fish individuals with abnormal distance in the overall orientation of the fish school; Obtaining the laterally adjacent fish individual of the fish individual in the rear position in the overall orientation, and obtaining the distance between the laterally adjacent fish individual; Obtaining fish individuals with a distance between laterally adjacent fish individuals not higher than a preset distance between adjacent fish individuals, and including the fish individuals in a preset sub-school cluster; Based on the fish individuals in the preset sub-school cluster, simultaneously obtaining the distance between the tail of a fish individual and the head of a fish individual in the rear position in the overall orientation, the distance between the fish individual and the laterally adjacent fish individual, and obtaining all fish individuals not lower than the preset head-to-tail distance between adjacent fish individuals and the preset distance between adjacent fish individuals, respectively, to obtain fish individuals at the edge of the sub-school cluster; Based on the range of all fish individuals at the edge of the sub-school cluster, obtaining the sub-school cluster of the fish school; Obtaining the moving direction of the sub-school cluster of the fish school, and obtaining the fish individual at the forefront in the moving direction to obtain the fish individual leading the fish school.

6. The method of claim 1, wherein, When it is determined that the fish school has no abnormal behavior, the fish individual not in the fish school is obtained to obtain the behavior abnormal fish individual, including: Obtaining the fish individual outside the fish school to obtain the free individual; Based on the fish school information, determining whether the species information of the free individual is the same as the fish school information; When the species information of the free individual is the same as the fish school information, the free individual is the behavior abnormal fish individual; When the species information of the free individual is different from the fish school information, obtaining the track of the free individual, and when the track of the free individual is abnormal, the free individual is the behavior abnormal fish individual.

7. The method of claim 6, wherein the method further comprises: Further comprising: When there are multiple fish schools, obtaining all fish school information and fish school tracks; Obtaining fish schools with abnormal tracks among all fish school tracks, and obtaining fish individuals leading the abnormal fish schools; Determining the fish individuals leading the abnormal fish schools as the behavior abnormal fish individuals.

8. The method of claim 1, wherein, The track change parameter of the fish individual leading the fish school and / or the behavior abnormal fish individual is obtained to obtain the predicted cause of abnormal behavior, including: Obtaining the track parameter of the fish individual leading the fish school and / or the behavior abnormal fish individual to obtain the track change parameter; Based on the track change parameter, obtaining the track change parameter weight; Based on the track change parameter weight, obtaining the track abnormality rate; Based on the track abnormality rate, obtaining the predicted cause of abnormal behavior.

9. A fish school abnormal behavior detection system for performing the fish school abnormal behavior detection method according to any one of claims 1 to 8, characterized by The detection system comprises an image acquisition module, a track analysis module, an abnormal behavior analysis module, and an abnormal cause analysis module, wherein: The image acquisition module and the track analysis module are connected to obtain the spatial coordinates of the fish; The track analysis module is further connected with the abnormal behavior analysis module, and is configured to acquire the track of the fish school after acquiring the spatial coordinates of the fish, and send the track to the abnormal behavior analysis module; The abnormal behavior analysis module is further connected with the abnormal cause analysis module, and is configured to determine the predicted cause of the abnormal behavior based on the analysis data acquired by the abnormal behavior analysis module; The abnormal cause analysis module is configured to acquire the actual cause of the abnormal behavior of the fish school.

Images