Tilapia mossambica fry intelligent screening separate culture identification method, system and equipment

Abstract

The invention belongs to the field of automatic control optimization, and provides a tilapia mossambica fry intelligent screening and separate breeding identification method, system and equipment, behavior image data of fry are collected, and judgment is performed according to the time shifting direction difference corresponding to the feeding frequency and the time shifting direction difference corresponding to the fin motion amplitude; judging according to the time-shifting direction difference corresponding to the movement track deviation and the time-shifting direction difference corresponding to the local density so as to judge whether to generate behavior dissimilatory regulatory factors by using the time-shifting direction differences of the data of the ingestion frequency, the fin movement amplitude, the movement track deviation and the local density in different time windows or not; and regulating the water flow by using the behavioral dissimilatory regulatory factor. More accurate and real-time sorting control is realized, and equipment misoperation and fry stress reaction are effectively reduced. Therefore, the separate cultivation precision is improved, and the adaptability and stability of the system are enhanced.

Description

Technical Field

[0001] The present invention belongs to the field of optimization of automatic control, and particularly relates to a method, system and device for intelligent screening, sorting and identification of tilapia fry. Background Art

[0002] With the continuous development of the aquaculture industry, intelligent and automated aquaculture technologies have become an important trend in the industry's development. Especially in the tilapia fry sorting process, the application of intelligent sorting technologies is gradually replacing traditional manual screening and physical sorting methods, improving the aquaculture efficiency and accuracy. Currently, many aquaculture systems use intelligent sorting devices based on computer vision and sensors to collect the behavioral data of fry in real time through devices such as high-frame-rate cameras and infrared sensors, and use algorithms such as deep learning to analyze the characteristics of fry such as body length, swimming speed, and feeding activity. These technologies have been able to identify the size differences of fry to a certain extent for classification and sorting. For example, deep learning technology can be used to analyze behavioral characteristics such as the feeding frequency and movement trajectory of fry to determine whether tilapia fry are suitable for being assigned to a specific aquaculture environment. At the same time, many sorting devices have started to apply automatic control systems. For example, servo motors are used to control the mesh aperture of the sieve and the water flow velocity to meet the passing requirements of fry of different sizes. Especially intelligent control systems based on mechanical devices, combined with automatic adjustment functions, can adjust device parameters according to the real-time data of fry to improve the sorting efficiency and accuracy.

[0003] Although the prior art has made certain progress in the intelligent screening and sorting of fry, there is still an over-reliance on manual or preset parameter control. Many existing sorting devices rely on manual intervention or fixed mesh aperture and water flow velocity parameters for adjustment. Although some devices can dynamically adjust the mesh aperture and water flow velocity through sensors, for example, a fry screening and counting device described in the patent document with publication number CN109042453B, its adjustment is usually based on preset fixed parameters or simple rules, lacking intelligent analysis based on the behavioral changes of fry. Therefore, it is impossible to flexibly respond to the behavioral differences between different tilapia fry during the aquaculture process, resulting in the inability to further improve the sorting accuracy and aquaculture efficiency.

[0004] In the prior art, although there are devices for sorting through behavioral data, they often cannot accurately capture the subtle behavioral changes of small-sized fry caused by the extrusion of large-sized fry. For example, if an intelligent fry sorting system and its usage method described in the patent document with publication number CN117502356A are used, among the tilapia group, small-sized fry may exhibit abnormal behaviors such as reduced feeding frequency and limited movement under the extrusion of large-sized fry. These subtle changes are often difficult to identify through simple threshold judgments or traditional single-behavior analysis methods, thus affecting the sorting accuracy.

[0005] Although many current sorting devices have automated functions, most are based on static behavior patterns and simple threshold judgments, lacking the ability to make real-time and dynamic adjustments according to the behavioral differences of fry. For example, as described in the patent document with the publication number CN117859689A for a fry grading device for aquaculture, when the device faces changes in fry behavior, it often cannot flexibly adjust the sieve aperture, sieve inclination, or water flow velocity according to the degree of change, and can only operate relying on fixed rules or preset parameters. Therefore, it cannot adapt to the complex environmental changes and the diversity of fry behavior in aquaculture.

[0006] In traditional fry sorting devices, fry are often subjected to mechanical physical effects or water flow impacts during the screening process, resulting in an increase in stress responses. The damage to the health of fry not only affects the growth rate but may also lead to a decrease in sorting accuracy. In addition, in the prior art, the means to alleviate the stress response of fry are relatively limited, often improving by adjusting a single device parameter, rather than making precise adjustments by fully utilizing dynamic monitoring data.

[0007] Current intelligent sorting systems often lack adaptability when dealing with complex aquaculture environments such as water quality, light changes, and differences in fry populations. The prior art usually relies on fixed parameters for automatic adjustment, which limits the flexibility and accuracy of the system when dealing with different environmental and water conditions. Summary of the Invention

[0008] The purpose of the present invention is to provide an intelligent screening, separating, and identifying method, system, and device for tilapia fry to solve one or more technical problems existing in the prior art, and at least provide a beneficial choice or create conditions.

[0009] To achieve the above purpose, according to one aspect of the present invention, there is provided an intelligent screening, separating, and identifying method for tilapia fry, the method comprising the following steps: Collect behavioral image data of fry, mark the area occupied by each fry in the image, and sort the fry according to the area size of the fry; Within different time windows, respectively obtain data of the fry including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density; Calculate the time-shift direction differences between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density; Judgment is made based on the difference in time shift direction corresponding to the feeding frequency and the difference in time shift direction corresponding to the fin movement amplitude, and judgment is made based on the difference in time shift direction corresponding to the deviation of the movement trajectory and the difference in time shift direction corresponding to the local density. Based on this judgment, it is identified whether it is necessary to use the difference in time shift direction between different time windows of the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density to generate a behavior alienation regulation factor. If so, the behavior alienation regulation factor is used to regulate the water flow.

[0010] Further, an underwater camera device is used with a multi-angle camera layout to collect the behavioral image data of fry in real time. A computer vision model is used to identify and mark each fry, and the area occupied by each fry in the image frame of the behavioral image data is marked.

[0011] Further, the method of regulating the water flow using the behavior alienation regulation factor is: regulating the water flow by controlling the adjustment of the screen aperture with the behavior alienation regulation factor.

[0012] Further, the method of regulating the water flow using the behavior alienation regulation factor may also be: Regulating the water flow by controlling the adjustment of the screen inclination angle with the behavior alienation regulation factor.

[0013] Further, the method of regulating the water flow using the behavior alienation regulation factor may also be: Regulating the water flow by controlling the adjustment of the water flow velocity with the behavior alienation regulation factor.

[0014] Further, the method of calculating the difference in time shift direction between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density respectively is: Sort the sample fry according to the area size they occupy in the image. In each time window, according to the sorting of the sample fry, form a feeding frequency vector with the numerical values of the feeding frequency of each sample fry, form a fin movement amplitude vector with the numerical values of the fin movement amplitude of each sample fry, form a movement trajectory deviation vector with the numerical values of the movement trajectory deviation of each sample fry, and form a local density vector with the numerical values of the local density of each sample fry; The cosine similarity of the feeding frequency vector in the previous time window with respect to the feeding frequency vector in the subsequent time window is regarded as the time-shift direction difference of the feeding frequency. The cosine similarity of the fin movement amplitude vector in the previous time window with respect to the fin movement amplitude vector in the subsequent time window is regarded as the time-shift direction difference of the fin movement amplitude. The cosine similarity of the movement trajectory deviation vector in the previous time window with respect to the movement trajectory deviation vector in the subsequent time window is regarded as the time-shift direction difference of the movement trajectory deviation. The cosine similarity of the local density vector in the previous time window with respect to the local density vector in the subsequent time window is regarded as the time-shift direction difference of the local density.

[0015] Further, the method for generating the behavior alienation adjustment factor using the time-shift direction differences of the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density between different time windows is as follows: The exponential representation value after combining the time-shift direction difference of the movement trajectory deviation with the time-shift direction difference of the local density, compared to the exponential representation value after combining the time-shift direction difference of the feeding frequency with the time-shift direction difference of the fin movement amplitude, the obtained ratio is the behavior alienation adjustment factor.

[0016] Further, in the step of condition judgment, it specifically includes: Judge whether the absolute value G1 of the difference obtained by subtracting the value of the time-shift direction of the fin movement amplitude from the value of the time-shift direction of the feeding frequency is greater than 1 / 2. If not, there is no need to regulate the water flow; If so, continue to judge whether the absolute value G2 of the difference obtained by subtracting the value of the time-shift direction of the local density from the value of the time-shift direction of the movement trajectory deviation is greater than 1 / 2. If not, there is no need to regulate the water flow, but if so, it is necessary to calculate the behavior alienation adjustment factor.

[0017] The present invention also provides an intelligent screening, separating, and identifying system for tilapia fry. The intelligent screening, separating, and identifying system for tilapia fry includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the intelligent screening, separating, and identifying method for tilapia fry. The intelligent screening, separating, and identifying system for tilapia fry can run on computing devices such as desktop computers, laptop computers, palmtop computers, and cloud data centers. The operable system may include, but is not limited to, a processor, a memory, and a server cluster. The processor executes the computer program and runs in the following units of the system: A data acquisition unit, configured to collect the behavior image data of the fry, mark the area occupied by each fry in the image, and sort the fry according to the area size of the fry; A feature calculation unit for respectively obtaining data including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density of fry in different time windows; A time shift direction unit for respectively calculating the time shift direction differences between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density; A judgment and adjustment unit for judging based on the time shift direction difference corresponding to the feeding frequency and the time shift direction difference corresponding to the fin movement amplitude, and judging based on the time shift direction difference corresponding to the movement trajectory deviation and the time shift direction difference corresponding to the local density, so as to judge whether to use the time shift direction differences between different time windows of the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density to generate a behavior alienation adjustment factor, and if so, using the behavior alienation adjustment factor to control the water flow.

[0018] Correspondingly, the present invention also provides an electronic device, a readable storage medium, and a computer program product: An electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the intelligent screening, separation, and identification method for tilapia fry and the methods of each step therein.

[0019] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to make the computer execute the intelligent screening, separation, and identification method for tilapia fry and the methods of each step therein.

[0020] A computer program product, comprising a computer program, wherein the computer program realizes the intelligent screening, separation, and identification method for tilapia fry and the methods of each step therein when executed by a processor.

[0021] The beneficial effects of the present invention are as follows: The present invention provides an intelligent screening, separation, and identification method, system, and device for tilapia fry, collects the behavioral image data of fry, judges based on the time shift direction difference corresponding to the feeding frequency and the time shift direction difference corresponding to the fin movement amplitude, and judges based on the time shift direction difference corresponding to the movement trajectory deviation and the time shift direction difference corresponding to the local density, so as to judge whether to use the time shift direction differences between different time windows of the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density to generate a behavior alienation adjustment factor, and uses the behavior alienation adjustment factor to control the water flow. It realizes more accurate and real-time sorting control, and effectively reduces equipment misoperation and fry stress response. It not only improves the separation accuracy, but also enhances the adaptability and stability of the system. Brief Description of the Drawings

[0022] By describing in detail the embodiments shown in the accompanying drawings, the above and other features of the present invention will become more obvious. The same reference numerals in the drawings of the present invention represent the same or similar elements. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. In the drawings: Figure 1 Shown is a flowchart of an intelligent screening, sorting, and identification method for tilapia fry; Figure 2 Shown is a system structure diagram of an intelligent screening, sorting, and identification system for tilapia fry. Detailed Embodiments

[0023] The following will clearly and completely describe the concept, specific structure, and technical effects of the present invention in combination with the embodiments and the drawings to fully understand the purpose, solution, and effects of the present invention. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0024] In the description of the present invention, the meaning of several is one or more, the meaning of multiple is more than two, greater than, less than, exceeding, etc. are understood as not including the present number, above, below, within, etc. are understood as including the present number. If the first and second are described only for the purpose of distinguishing technical features, they cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0025] As Figure 1 Shown is a flowchart of an intelligent screening, sorting, and identification method for tilapia fry according to the present invention. The following will describe an intelligent screening, sorting, and identification method, system, and device for tilapia fry according to the embodiments of the present invention in combination with Figure 1 to elaborate.

[0026] The present invention provides an intelligent screening, sorting, and identification method for tilapia fry. The method specifically includes the following steps: Collect the behavioral image data of the fry, mark the area occupied by each fry in the image, and sort the fry according to the area size of the fry; Within different time windows, respectively obtain the data of the fry including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density; calculate the time-shift direction difference between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density; Perform a conditional judgment based on whether the absolute difference between the time shift direction difference corresponding to the feeding frequency and the time shift direction difference corresponding to the fin movement amplitude is greater than 1 / 2. If not, do not initiate the regulation of the water flow; If so, continue to perform a conditional judgment based on whether the absolute difference between the time shift direction difference corresponding to the deviation of the movement trajectory and the time shift direction difference corresponding to the local density is still greater than 1 / 2. If not, do not initiate the regulation of the water flow. If it is still greater than 1 / 2, generate a behavior alienation adjustment factor using the time shift direction differences of the feeding frequency, fin movement amplitude, movement trajectory deviation, and local density data between different time windows, and use the behavior alienation adjustment factor to regulate the water flow.

[0027] In some embodiments, specifically: In the fish pond of tilapia fry to be detected, set up the adjustment device and its control and activation device; Separate shoot, identify and mark each sample fry in the fish group, and identify the area size occupied by each sample fry in the image; For the fish group at different time windows, with the earlier time window as the previous time window and the later time window as the subsequent time window, obtain the data of the feeding frequency, fin movement amplitude, movement trajectory deviation, and local density of each sample fry in each time window.

[0028] Sort each sample fry according to the area size occupied in the image. In each time window, according to the sorting of each sample fry, form a feeding frequency vector with the numerical values of the feeding frequency of each sample fry, form a fin movement amplitude vector with the numerical values of the fin movement amplitude of each sample fry, form a movement trajectory deviation vector with the numerical values of the movement trajectory deviation of each sample fry, and form a local density vector with the numerical values of the local density of each sample fry.

[0029] The cosine similarity of the feeding frequency vector in the previous time window relative to the feeding frequency vector in the subsequent time window is used as the feeding frequency time shift direction F, the cosine similarity of the fin movement amplitude vector in the previous time window relative to the fin movement amplitude vector in the subsequent time window is used as the fin movement amplitude time shift direction A, the cosine similarity of the movement trajectory deviation vector in the previous time window relative to the movement trajectory deviation vector in the subsequent time window is used as the movement trajectory deviation time shift direction T, and the cosine similarity of the local density vector in the previous time window relative to the local density vector in the subsequent time window is used as the local density time shift direction D.

[0030] Then, perform a conditional judgment: Judge whether the absolute value G1 of the difference obtained by subtracting the numerical value of the fin movement amplitude time shift direction from the numerical value of the feeding frequency time shift direction is greater than 1 / 2. If not, there is no need to regulate the water flow; If so, continue to determine whether the absolute value G2 of the difference obtained by subtracting the value of the local density time shift direction from the value of the movement trajectory deviation time shift direction is greater than 1 / 2. If not, there is no need to regulate the water flow; However, if it is still greater, then it is necessary to calculate BDaF: BDaF = [Exp(movement trajectory deviation time shift direction T) * exp(local density time shift direction D)] / [exp(feeding frequency time shift direction F) * exp(fin movement amplitude time shift direction A)]; The formula is simplified to BDaF = [Exp(movement trajectory deviation time shift direction T + local density time shift direction D)] / [exp(feeding frequency time shift direction F + fin movement amplitude time shift direction A)]; In specific cases where the effect is uncertain, it can be further simplified to BDaF = (movement trajectory deviation time shift direction T + local density time shift direction D) / exp(feeding frequency time shift direction F + fin movement amplitude time shift direction A); Use 1 - BDaF as the weight to regulate the water flow.

[0031] Furthermore, an underwater camera device is used with a multi - angle camera layout to collect real - time behavioral image data of fry. A computer vision model is used to identify and mark each fry, and mark the area occupied by each fry in the image frames of the behavioral image data.

[0032] In some embodiments, a high - frame - rate and high - resolution underwater camera device can be used, preferably supplemented with a multi - angle and multi - perspective camera layout to obtain data from different perspectives and capture image data of the movement details of the behavior of each fry. A pre - trained large - scale vision model including but not limited to the YOLO model is used to perform object detection of each fry in the behavioral image data of the fry, and identify and mark the identity of each fry.

[0033] In the video sequence that has been identified and marked, a 3D CNN model is used to directly capture fine - grained motion features in the spatial and temporal dimensions on the video frame sequence, and extract the motion features of the local fish body including the fins and tail and its overall motion features for each fry. Including but not limited to, detecting the amplitude of the fin swing of each fry between the image frames of the video sequence as the fin movement amplitude, calculating the feeding frequency by detecting the proportion of the number of times each fry is detected to have a feeding behavior in the total number of frames between the image frames of the video sequence, calculating the movement trajectory deviation by detecting the distance each fry moves from the initial frame to the end frame between the image frames of the video sequence, and detecting the density of fry within a range of, for example, a radius of 10 cm - 15 cm centered on each fry as the local density between the image frames of the video sequence.

[0034] Preferably, in the case of insufficient light or turbidity in the underwater environment, infrared thermal imaging can be used to supplement visible light data.

[0035] Preferably, for the four items of data including the feeding frequency, fin movement amplitude, movement trajectory deviation, and local density of each fry, each item is first normalized separately and then used for subsequent calculations.

[0036] Furthermore, the method for regulating the water flow using the behavior alienation regulation factor is: by controlling the adjustment of the screen aperture with the behavior alienation regulation factor to regulate the water flow.

[0037] In some embodiments, the adjustment of the screen aperture can be achieved through an automated control of a screening and grading device described in a patent document such as CN116158395A. This device consists of multiple screening valves and screens to form partitions with different apertures, thereby realizing the screening of fry of different specifications. Among them, the screen aperture decreases sequentially to meet the passing requirements of fry of different specifications. Preferably, when the value of the behavior alienation regulation factor denoted as BDaF is higher than 1, it indicates that the behavior of small-sized fry is abnormal due to the extrusion by large-sized fry. At this time, it is necessary to reduce the screen aperture to ensure that small-sized fry are not mis-screened away. Generally, the screen aperture can be dynamically adjusted based on BDaF. For example, when BDaF is 1.23, the screen aperture is reduced by the value of 1 - BDaF so that small fry can pass smoothly. -0.23 can be rounded to the nearest fifth, and the screen aperture is reduced approximately in this proportion to prevent it from being filtered out by a large screen aperture. When the value of low BDaF is lower than 1, it indicates that the behavior difference of the fry is small. At this time, the screen aperture can be appropriately increased to improve the sorting efficiency and reduce the interference during fry screening. Among them, the adjustment of the screen aperture is completed by a servo motor or an automatic adjustment device. By controlling the motor and the transmission system, the aperture of the screen is automatically adjusted according to the BDaF value. For high BDaF, the motor control system reduces the aperture to ensure that the fry are not squeezed or mis-screened; for low BDaF, the aperture is increased to improve the screening efficiency.

[0038] Furthermore, the method for regulating the water flow using the behavior alienation regulation factor may also be: By controlling the adjustment of the screen inclination angle with the behavior alienation regulation factor to regulate the water flow.

[0039] In some embodiments, by changing the inclination angle of the sieve, the speed and path of fry passing through the sieve can be adjusted, thereby improving the sorting accuracy. The change in the inclination angle of the sieve affects the speed of the water flow and the flow trajectory of the fry, and further affects the stratification and sorting of the fry. When the BDaF value is greater than 1, it indicates a strong squeezing effect of the fry. Increasing the inclination angle can help small-sized fry pass through the sieve smoothly. For example, when the BDaF is 1.09, the original inclination angle is adjusted to the nearest integer of about 109%, reducing the residence time of the fry on the sieve and avoiding the influence of long-term squeezing on the fry. When the BDaF value is not higher than 1, which is used to indicate that the behavioral differences of the fry are small and the squeezing effect is weak, the inclination angle of the sieve can be appropriately reduced to improve the natural fluidity of the fry and reduce the burden on the sieve system. The inclination angle of the sieve is controlled by a servo motor or a hydraulic system and is dynamically adjusted according to the specific value of the BDaF rounded to the nearest integer. The system described in the present invention adjusts the inclination angle of the sieve according to the real-time monitored BDaF value, thereby optimizing the sorting path and speed of the fry.

[0040] Furthermore, the method of regulating the water flow using the behavior alienation adjustment factor may also be: Regulate the water flow by controlling the adjustment of the water flow velocity with the behavior alienation adjustment factor.

[0041] In some embodiments, by adjusting the speed of the water flow, different-sized fry can be effectively guided to stratify according to their behavioral characteristics, thereby improving the sorting efficiency. The water flow velocity has a direct impact on the stratification, flow velocity, and position of the fry. When the BDaF is greater than 1, it shows a large squeezing effect, and the behavior of tilapia fry changes significantly. The water flow velocity can be preferably increased to help large-sized fry move quickly, while slowing down the flow velocity of small-sized fry to make it easier to be correctly stratified. When the BDaF is less than 1, it shows that the behavior change of the warning fry is small, and too large a flow velocity may affect the natural flow of the fry. At this time, the water flow velocity can be controlled to decrease to reduce the interference to the fry. Among them, in one implementation, the water flow velocity is adjusted by a water pump control system, and the power or flow rate of the water pump can be automatically adjusted according to the BDaF. At a higher BDaF value, the power of the water pump increases proportionally to increase the flow velocity; at a lower BDaF value, the water flow velocity decreases linearly to ensure the normal flow of the fry.

[0042] Furthermore, the method of calculating the time-shift direction difference between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density is: Sort the fry samples according to the area they occupy in the image. In each time window, according to the sorting of the fry samples, form a feeding frequency vector with the numerical values of the feeding frequencies of the fry samples, form a fin movement amplitude vector with the numerical values of the fin movement amplitudes of the fry samples, form a movement trajectory deviation vector with the numerical values of the deviations of the movement trajectories of the fry samples, and form a local density vector with the numerical values of the local densities of the fry samples; Use the cosine similarity of the feeding frequency vector of the previous time window relative to the feeding frequency vector of the next time window as the feeding frequency time shift direction difference, use the cosine similarity of the fin movement amplitude vector of the previous time window relative to the fin movement amplitude vector of the next time window as the fin movement amplitude time shift direction difference, use the cosine similarity of the movement trajectory deviation vector of the previous time window relative to the movement trajectory deviation vector of the next time window as the movement trajectory deviation time shift direction difference, and use the cosine similarity of the local density vector of the previous time window relative to the local density vector of the next time window as the local density time shift direction difference.

[0043] The area size of tilapia fry in the image reflects their actual occupied position and relative size in the field of view. The size of tilapia fry is usually closely related to their body shape, health status, and activity level. In the behavioral analysis of tilapia fry, there may be significant differences in the behavioral characteristics of large-sized fry and small-sized fry. For example, large-sized tilapia fry usually swim faster and have a higher feeding frequency, while small-sized fry may be relatively sluggish or have weaker feeding behavior. Sorting the fry according to their area helps to track the behavioral patterns of the fry based on their body shape changes and avoid data noise caused by the distribution of fry in different proportions. The sorted feature vectors can place the fry in order of size, which can effectively reduce the interference caused by fry in different proportions, enabling the analysis to focus only on the behavioral changes of groups of fry with similar body shapes rather than simply mixing the data of all fry. This data organization method makes the subsequent behavioral analysis more accurate and efficient. After sorting, when calculating the time shift direction difference of behaviors, time series analysis can be performed on groups of fry with the same body shape, avoiding the interference of differences caused by the mixing of fry with different body shapes. During the behavioral change process of fry, small-sized fry are usually the most affected group by extrusion. After sorting and performing behavioral feature vector analysis, it is possible to more accurately capture the behavioral abnormalities of small-sized fry caused by the extrusion of large-sized fry. Since the behavioral differences of small-sized fry are relatively small, after sorting, it is possible to more precisely capture the behavioral abnormalities caused by the extrusion of large-sized fry, such as a decrease in the feeding frequency. This makes the calculation of the behavior alienation regulation factor more accurate and automatically adjusts equipment such as water flow and sieves to ensure precise separation of the fry and avoid mis-sieving and stress responses.

[0044] This method captures the behavioral changes of fry by calculating the time-shift direction differences within different time windows based on feeding frequency, fin movement amplitude, movement trajectory deviation, and local density data, so as to obtain the time-shift direction of the fry's behavior. Existing technologies usually rely only on a single behavioral feature such as feeding frequency or movement trajectory deviation, while the present invention conducts a comprehensive behavioral analysis by introducing multi-dimensional features including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density. By comparing the behavioral changes within different time windows, the behavioral differences of fry can be accurately captured, especially the behavioral changes caused by large-sized fry squeezing small-sized fry. This method can dynamically calculate and compare the behavioral differences of fry during the real-time process. Compared with the traditional screening method based on static parameters, it has higher real-time performance and flexibility. By monitoring the dynamic behavioral changes of fry within different time windows, it can better adapt to the complex breeding environment and the diversity of fry behavior. Generally, existing methods often rely on static thresholds or simple behavioral criteria and cannot capture subtle behavioral changes. However, by calculating the time-shift direction differences, the present invention can accurately evaluate the subtle behavioral changes of fry within a short period, which is crucial for timely identifying the behavioral abnormalities of small-sized fry caused by the squeezing of large-sized fry. The method described in the present invention combines the direction differences of multiple key features in each dimension of time change, including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density, etc. Compared with a single index, it can more comprehensively reflect the behavioral state of fry. By calculating the time-shift direction differences of behavioral features at different times, the present invention can dynamically monitor the changes in fry behavior and avoid misjudgment caused by the limitation of static thresholds.

[0045] Further, the method for generating a behavior alienation adjustment factor using the time-shift direction differences of the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density between different time windows is as follows: Taking the exponentialized representation value after combining the time-shift direction difference of movement trajectory deviation and the time-shift direction difference of local density, compared with the exponentialized representation value after combining the time-shift direction difference of feeding frequency and the time-shift direction difference of fin movement amplitude, the obtained ratio is the behavior alienation adjustment factor.

[0046] Among them, the behavior alienation adjustment factor, denoted as BDaF in the embodiments of the present invention, is an adjustment coefficient based on the behavioral feature changes of fry within different time windows, and is used to accurately adjust the parameters of the subculture device, including the sieve aperture, sieve inclination angle, and water flow velocity, so as to optimize the sorting effect of fry. The calculation of BDaF is based on the time-shift direction differences of the feeding frequency, fin movement amplitude, movement trajectory deviation, and local density of fry, and represents the degree of change in the behavior of fry under the influence of environmental factors such as squeezing. The magnitude of the BDaF value directly affects the adjustment intensity of the subculture device, and is used to automatically adjust the working states of the water flow, sieve, and screening device to ensure the accurate grading and healthy growth of fry.

[0047] The local density increases and the trajectory deviation increases; at the same time, the feeding frequency decreases and the amplitude of fin movement decreases, indicating the appearance of a compression signal. Trajectory perturbation and density increase are in the numerator; while the decrease in feeding and fin activity is in the denominator. It represents the overall change trend of the behavioral state from the previous window to the next window, not judging static values, but judging the structural relationship between behavioral change trends; without preset weights and independent of model training, but constructing a structural index that can measure the alienation effect.

[0048] The behavioral alienation regulator is used to dynamically adjust the working state of the fry screening device including the screen aperture, screen inclination angle, and water flow velocity. Its value is based on the time shift direction of the feeding frequency, fin movement amplitude, movement trajectory deviation, and local density of the fry in the front and back time windows, and calculates the change difference of these behavioral characteristics according to the cosine similarity algorithm, and generates a BDaF value based on this difference.

[0049] The magnitude of the BDaF value determines the adjustment amplitude of the subculture device. The larger the value, the stronger the degree of extrusion of the fry, and stronger adjustment is required to ensure the sorting accuracy and the health of the fry, and the value of the behavioral alienation regulator is calculated. According to the real-time change of BDaF, the system can automatically adjust the parameters of the subculture device, such as reducing the screen aperture, increasing the screen inclination angle, or adjusting the water flow velocity, so as to achieve precise classification and stratification of the fry and effectively reduce the stress response of the fry caused by extrusion.

[0050] The present invention generates a behavior dissimilation adjustment factor BDaF by calculating the differences in the time-shift directions of various behavioral characteristics such as feeding frequency, fin movement amplitude, movement trajectory deviation, and local density at different times. This factor is used to automatically adjust the screen aperture, screen inclination, and water flow velocity, thereby adjusting the working state of the grading device. The automatic adjustment of BDaF based on behavioral differences is dynamically calculated according to the differences in the time-shift directions of fry behaviors, representing the degree of change in fry behaviors. By using BDaF to adjust the equipment parameters, the automatic adjustment and precise management of fry sorting are achieved. The prior art usually relies on fixed screen apertures or fixed sorting strategies, while the present invention can perform adaptive adjustments based on real-time data, greatly improving the sorting accuracy. Traditional methods adjust the equipment through manual control or fixed parameters, which are easily affected by environmental factors and difficult to cope with the changing aquaculture environment. The present invention reflects the dynamic changes in fry behaviors through the BDaF, can precisely adjust the screen aperture, screen inclination, and water flow velocity, and reduce the stress response caused by fry extrusion or over-sorting. By calculating BDaF and adjusting the working state of the grading device in real time, the sorting accuracy can be improved and the stress response of fry can be reduced, enabling the present invention to minimize the stress response of fry to the greatest extent and ensuring that small-sized fry will not be damaged or mis-sorted due to the extrusion of large-sized fry. Compared with the existing methods, the present invention can provide a higher-precision fry classification and stratification effect. Among them, the adjustment mechanism of BDaF based on dynamic behavioral differences reflects the dynamic trend of fry behavior changes, enabling the adjustment of equipment parameters to respond to the behavior changes of fry in real time, thereby ensuring accurate sorting. The adaptive mechanism of the present invention can automatically optimize the equipment parameters according to the behavior data collected in real time, improve the sorting accuracy, and reduce the stress response.

[0051] Further, in the step of performing the condition judgment, it specifically includes: Determine whether the absolute value G1 of the difference obtained by subtracting the value of the time-shift direction of the fin movement amplitude from the value of the time-shift direction of the feeding frequency is greater than 1 / 2. If not, there is no need to regulate the water flow; If so, continue to determine whether the absolute value G2 of the difference obtained by subtracting the value of the time-shift direction of the local density from the value of the time-shift direction of the movement trajectory deviation is greater than 1 / 2. If not, there is no need to regulate the water flow, but if so, it is necessary to calculate the behavior dissimilation adjustment factor.

[0052] In this embodiment, preferably, 1 / 2 is selected to define the critical value of behavioral changes. Especially in the analysis of the behavioral patterns of fry, such as feeding frequency and the amplitude of fin movement, the behavioral differences between fry usually show a certain regularity. Taking 1 / 2 as the demarcation point can ensure that the system will only make adjustments when the behavioral difference exceeds a certain critical value, and when it is less than that, it is considered that the difference is not sufficient to affect the separation of cultivation, thus avoiding the system from overreacting. If the threshold is set too low, such as 1 / 4 or smaller, it may lead to overly sensitive adjustments, that is, even minor behavioral fluctuations will cause the system to make adjustments, resulting in unnecessary equipment adjustments and energy waste. Using 1 / 2 as the threshold can reduce ineffective adjustments and ensure that the system only makes necessary responses when the behavioral difference is large.

[0053] Among them, the design of making two judgments respectively for G1 and G2 is a multi-level filtering mechanism. The first judgment is whether G1 is greater than 1 / 2 to ensure whether the difference between the feeding frequency and the amplitude of fin movement is significant, and then by judging whether G2 is greater than 1 / 2, the situations that do not require adjustment are further filtered out. This hierarchical judgment can improve the robustness of the system and ensure that delicate adjustments can be made in a complex environment. The first layer of judgment is to judge whether the difference between the feeding frequency and the amplitude of fin movement is obvious. Usually, these two characteristics directly reflect the basic behavioral states of the fry, such as whether there is sufficient food and whether they are in an active state. The second layer of judgment is that if the first layer of judgment considers the behavioral difference to be obvious, continue to compare the deviation of the movement trajectory and the local density. The second layer of judgment strengthens the detailed analysis of the behavior and avoids over-adjustment. Through the two judgments, the response of the system becomes more precise. For example, if only G1 > 1 / 2 and G2 <= 1 / 2, it indicates that the behavioral difference is small and does not require a full adjustment. The two judgments can effectively filter out some unnecessary adjustments, making the system more accurate and real-time, and avoiding over-response to unimportant changes. If there is no such secondary judgment, the system may start to adjust the device due to minor fluctuations such as the difference between the feeding frequency and the amplitude of fin movement, resulting in frequent changes of the equipment, energy waste and over-adjustment. Therefore, by setting the judgment threshold of 1 / 2, this problem is avoided, and the working efficiency and stability of the equipment are improved.

[0054] In this embodiment, this step makes conditional judgments on the time-shift direction differences of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density, ensuring that the regulation mechanism is only activated when the sub-culturing equipment needs to be adjusted, thus avoiding unnecessary operations. Through the conditional judgments of G1 and G2, unnecessary equipment adjustments are effectively avoided. The present invention can ensure the fry sorting effect while avoiding frequent or unnecessary equipment adjustments. Traditional automated sub-culturing devices may adjust the equipment immediately after each data collection, while the present invention first determines the significance of behavior changes and only activates equipment regulation when necessary, reducing energy consumption and equipment wear. By comprehensively judging the time-shift direction differences of multiple behavioral characteristics, the present invention can ensure that adjustments are only made when the fry behavior changes significantly, thus avoiding incorrect equipment adjustments caused by minor fluctuations. This design effectively reduces ineffective adjustment operations, improves the operation efficiency and stability of the equipment. At the same time, the operation cycle and service life of the equipment are also extended, saving the breeding cost. Through multi-level conditional judgments, the present invention ensures that the equipment is only adjusted when needed, avoiding over-regulation or unnecessary operations caused by errors. This method can accurately judge the degree of change in fry behavior, ensure the adjustment accuracy of the equipment, and avoid over-reaction.

[0055] The intelligent screening, sub-culturing and identification system for tilapia fry runs on any computing device such as a desktop computer, a laptop computer, a palm computer or a cloud data center. The computing device includes: a processor, a memory, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps in the intelligent screening, sub-culturing and identification method for tilapia fry. The operable system may include, but is not limited to, a processor, a memory, and a server cluster.

[0056] An intelligent screening, sub-culturing and identification system for tilapia fry provided by an embodiment of the present invention, as Figure 2 shown, the intelligent screening, sub-culturing and identification system for tilapia fry in this embodiment includes: a processor, a memory, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, it implements the steps in the above-mentioned embodiment of the intelligent screening, sub-culturing and identification method for tilapia fry. When the processor executes the computer program, it runs in the following units of the system: A data acquisition unit, configured to collect behavioral image data of fry, mark the area occupied by each fry in the image, and sort the fry according to the area size of the fry; A feature calculation unit, configured to respectively obtain data of the fry including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density within different time windows; A time-shift direction unit for calculating the time-shift direction differences between different time windows based on data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density respectively; A judgment and adjustment unit for judging based on the time-shift direction differences corresponding to the feeding frequency and the time-shift direction differences corresponding to the fin movement amplitude, and judging based on the time-shift direction differences corresponding to the movement trajectory deviation and the time-shift direction differences corresponding to the local density, so as to judge whether to use the time-shift direction differences between different time windows of the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density to generate a behavior alienation adjustment factor, and if so, use the behavior alienation adjustment factor to control the water flow.

[0057] Among them, in order to better unify the linear relationship and probability connection of the numerical values between physical quantities of different units, dimensionless processing can be performed on different physical quantities.

[0058] Among them, preferably, for all undefined variables in the present invention, if there is no clear definition, they can all be artificially set thresholds.

[0059] The intelligent screening, sub-culturing and identification system for tilapia fry can run on computing devices such as desktop computers, laptop computers, palmtop computers, and cloud data centers. The intelligent screening, sub-culturing and identification system for tilapia fry includes, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above examples are only examples of an intelligent screening, sub-culturing and identification method, system and device for tilapia fry, and do not constitute a limitation on an intelligent screening, sub-culturing and identification method, system and device for tilapia fry. It may include more or fewer components than the examples, or combine some components, or different components. For example, the intelligent screening, sub-culturing and identification system for tilapia fry may also include input / output devices, network access devices, buses, etc.

[0060] The present invention also provides an electronic device, a readable storage medium, and a computer program product: An electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the intelligent screening, sub-culturing and identification method for tilapia fry and the methods of each step therein.

[0061] A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the intelligent screening, sub-culturing and identification method for tilapia fry and the methods of each step therein.

[0062] A computer program product includes a computer program which, when executed by a processor, implements the intelligent screening, separation and identification method for tilapia fry and the methods of the various steps therein.

[0063] Among them, the electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0064] Various embodiments of the systems and techniques described above in this article can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0065] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to the processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, so that when the program codes are executed by the processor or controller, the functions / operations specified in the flowchart and / or block diagram are implemented. The program codes can be executed entirely on the machine, partially on the machine, executed partially on the machine and partially on a remote machine as an independent software package, or executed entirely on a remote machine or server.

[0066] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0067] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0068] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0069] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship to each other.

[0070] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete component gate circuits, or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the intelligent tilapia fry screening, separating, and identifying system, and connects various sub-regions of the entire intelligent tilapia fry screening, separating, and identifying system through various interfaces and lines.

[0071] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the intelligent tilapia fry screening, separating, and identifying method, system, and device by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0072] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present invention can be achieved, and no limitations are imposed herein.

[0073] The present invention provides an intelligent screening, sorting and identification method, system and device for tilapia fry. The method collects behavioral image data of the fry, and makes judgments based on the difference in time shift direction corresponding to the feeding frequency and the difference in time shift direction corresponding to the movement amplitude of the fin part, and makes judgments based on the difference in time shift direction corresponding to the deviation of the movement trajectory and the difference in time shift direction corresponding to the local density, so as to determine whether to generate a behavior alienation adjustment factor using the data of the feeding frequency, the movement amplitude of the fin part, the deviation of the movement trajectory and the local density in different time windows, and use the behavior alienation adjustment factor to regulate the water flow. More accurate and real-time sorting control is achieved, and equipment misoperation and fry stress response are effectively reduced. Not only the sorting accuracy is improved, but also the adaptability and stability of the system are enhanced.

[0074] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent screening, sorting and identification method for tilapia fry, characterized in that, The method includes: Collecting behavioral image data of fry, marking the area occupied by each fry in the image, and sorting the fry according to the area size of the fry; within different time windows, respectively obtaining data including feeding frequency, fin movement amplitude, movement trajectory deviation, and local density of the fry; calculating the time-shift direction differences between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density respectively; making judgments based on the time-shift direction differences corresponding to the feeding frequency and the time-shift direction differences corresponding to the fin movement amplitude, and making judgments based on the time-shift direction differences corresponding to the movement trajectory deviation and the time-shift direction differences corresponding to the local density, so as to identify whether to generate a behavior alienation adjustment factor using the time-shift direction differences between different time windows for the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density respectively. If so, use the behavior alienation adjustment factor to regulate the water flow.

2. The intelligent screening, sorting and identification method for tilapia fry according to claim 1, wherein Wherein, Using an underwater camera device supplemented with a multi-angle camera layout to collect behavioral image data of fry in real time, using a computer vision model to identify and mark each fry, and marking the area occupied by each fry in the image frame of the behavioral image data.

3. The intelligent screening, separating and identifying method for tilapia fry according to claim 1, characterized in that, The method of using the behavior alienation adjustment factor to regulate the water flow is: Regulating the water flow by controlling the adjustment of the screen aperture with the behavior alienation adjustment factor.

4. The intelligent screening, sorting and identification method for tilapia fry according to claim 1, characterized in that The method of using the behavior alienation adjustment factor to regulate the water flow may also be: Regulating the water flow by controlling the adjustment of the screen inclination angle with the behavior alienation adjustment factor.

5. The intelligent screening, sorting and identification method for tilapia fry according to claim 1, wherein The method of using the behavior alienation adjustment factor to regulate the water flow may also be: Regulating the water flow by controlling the adjustment of the water flow velocity with the behavior alienation adjustment factor.

6. The intelligent screening, sorting and identification method for tilapia fry according to claim 1, characterized in that, The method of calculating the time-shift direction differences between different time windows based on the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density respectively is: Sorting each sample fry according to the area size occupied in the image. In each time window, according to the sorting of each sample fry, forming a feeding frequency vector with the numerical values of the feeding frequency of each sample fry, forming a fin movement amplitude vector with the numerical values of the fin movement amplitude of each sample fry, forming a movement trajectory deviation vector with the numerical values of the movement trajectory deviation of each sample fry, and forming a local density vector with the numerical values of the local density of each sample fry; Taking the cosine similarity of the feeding frequency vector in the previous time window relative to the feeding frequency vector in the subsequent time window as the time-shift direction difference of the feeding frequency, taking the cosine similarity of the fin movement amplitude vector in the previous time window relative to the fin movement amplitude vector in the subsequent time window as the time-shift direction difference of the fin movement amplitude, taking the cosine similarity of the movement trajectory deviation vector in the previous time window relative to the movement trajectory deviation vector in the subsequent time window as the time-shift direction difference of the movement trajectory deviation, and taking the cosine similarity of the local density vector in the previous time window relative to the local density vector in the subsequent time window as the time-shift direction difference of the local density.

7. The intelligent screening, sorting and identification method for tilapia fry according to claim 6, characterized in that, The method for generating a behavior alienation regulation factor using the differences in the time shift directions between different time windows for the data of feeding frequency, fin movement amplitude, movement trajectory deviation, and local density respectively is as follows: Taking the exponential representation value after combining the time shift direction difference of the movement trajectory deviation with the time shift direction difference of the local density, compared with the exponential representation value after combining the time shift direction difference of the feeding frequency with the time shift direction difference of the fin movement amplitude, the obtained ratio is the behavior alienation regulation factor.

8. The intelligent screening, sorting and identification method for tilapia fry according to claim 7, characterized in that, Among them, In the step of conditional judgment, it specifically includes: Judging whether the absolute value of the difference obtained by subtracting the value of the time shift direction of the fin movement amplitude from the value of the time shift direction of the feeding frequency is greater than 1 / 2. If not, there is no need to regulate the water flow; If so, continue to judge whether the absolute value of the difference obtained by subtracting the value of the time shift direction of the local density from the value of the time shift direction of the movement trajectory deviation is greater than 1 / 2. If not, there is no need to regulate the water flow, but if so, it is necessary to calculate the behavior alienation regulation factor.

9. An intelligent screening, sorting and identification system for tilapia fry, characterized in that, The intelligent screening, sorting, and identification system for tilapia fry runs on any computing device such as a desktop computer, a laptop computer, or a cloud data center. The computing device includes: a processor, a memory, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps in the intelligent screening, sorting, and identification method for tilapia fry according to any one of claims 1 to 8.

10. An electronic device, comprising: At least one processor; And a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor. The instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1 to 8.

Images