Dynamic injection posture adjusting method based on fry size information

Abstract

The invention discloses a dynamic injection posture adjustment method based on fry size information, and relates to the technical field of aquaculture, and the method comprises the steps: carrying out the image collection of a fry injection region, and carrying out the preprocessing of the collected image; identifying fry size information according to the preprocessed fry image; establishing a mathematical model between the injection parameter and the fry size according to the fry size information, and calculating to obtain a predicted fry injection parameter according to the mathematical model; calculating to obtain an injection parameter error sum-of-squares function according to the predicted fry injection parameters in combination with the injection angle under the optimal injection effect; planning a movement track of an injection needle by utilizing a path planning algorithm, and predicting a fry injection position by combining the movement speed of the fry; according to the predicted fry injection position and the actual fry position, fry injection accuracy is calculated, and the injection posture is dynamically adjusted in combination with the injection angle error. According to the invention, the injection accuracy and efficiency are improved, the damage to fish fries is reduced, and the automation level of the production process is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aquaculture, and more specifically, to a method for dynamically adjusting injection postures based on fry size information. Background Art

[0002] In the field of aquaculture, in order to prevent and treat fry diseases and improve the survival rate and growth quality of fry, it is often necessary to inject drugs into the fry. The traditional fry injection method mainly relies on manual operation. This method not only has low efficiency and high labor intensity, but also due to the differences in the technical levels of operators, it is difficult to ensure the accuracy of the injection dose and the consistency of the injection posture, which is likely to cause harm to the fry and affect the healthy growth of the fry.

[0003] Deficiencies of the prior art:

[0004] With the continuous development of computer technology, image processing technology and automatic control technology, automatic fry injection equipment has gradually been applied. However, most of the existing automatic injection equipment does not consider the differences in the individual sizes of fry when injecting drugs, and adopts fixed injection parameters and postures, and cannot be adjusted according to the actual situation of the fry, resulting in poor injection effects and low drug utilization rates. Therefore, how to dynamically adjust the injection posture according to the size information of the fry to improve the accuracy and efficiency of injection is an urgent problem to be solved in the current field of aquaculture.

[0005] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for dynamically adjusting injection postures based on fry size information, which improves the accuracy and efficiency of injection by dynamically adjusting the injection posture according to the size information of the fry, so as to solve the problems raised in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for dynamically adjusting injection postures based on fry size information includes the following steps:

[0009] Collect images of the fry injection area and preprocess the collected images;

[0010] Identify the fry size information according to the preprocessed fry images;

[0011] Establish a mathematical model between the injection parameters and the fry size according to the fry size information, and calculate the predicted fry injection parameters according to the mathematical model;

[0012] Calculate the sum of squared errors function of injection parameters based on the predicted fry injection parameters and the injection angle under the best injection effect;

[0013] Use the path planning algorithm to plan the movement trajectory of the injection needle, and predict the fry injection position in combination with the movement speed of the fry;

[0014] Calculate the fry injection accuracy according to the predicted fry injection position and the actual fry position, and dynamically adjust the injection posture in combination with the injection angle error.

[0015] In a preferred embodiment, the preprocessing process of the collected image is as follows:

[0016] Convert the collected color image into a grayscale image to reduce the data volume. The grayscale conversion expression is:

[0017] H = 0.299×R + 0.587×G + 0.114×B;

[0018] In the formula, H is the converted grayscale value, and R, G, and B are the red, green, and blue channel values of the image pixels respectively;

[0019] Calculate each pixel in the image through this formula to obtain the grayscale image;

[0020] Use the Gaussian filtering algorithm to denoise the image. The process of removing noise interference in the image is as follows:

[0021] For each pixel in the image, calculate the weighted average of the pixel values in the pixel neighborhood to obtain the filtered pixel value;

[0022] Construct a Gaussian kernel and perform a convolution operation on each pixel in the image and its neighboring pixels;

[0023] Multiply the Gaussian kernel by the pixel values in the corresponding neighborhood in the image and sum them to obtain the filtered pixel value, thereby realizing the removal of image noise;

[0024] Use the histogram equalization algorithm to enhance the image, improve the contrast of the image, and provide high-quality image data for subsequent size recognition.

[0025] In a preferred embodiment, the process of identifying the fry size information according to the preprocessed fry image is as follows:

[0026] Construct a two-dimensional grid map, obtain the coordinates of the two points on both sides of the head, tail, and widest part of the body of the fry, and calculate the length and width of the fry through the distance formula between two points. The calculation formula is as follows:

[0027]

[0028] In the formula, Li is the fry length, K i is the fry width, (x1, y1) are the coordinates of the fry head,

[0029] (x2, y2)

[0030] are the coordinates of the fry tail, (x3, y3), (x4, y4) are the two points on both sides of the widest part of the fry respectively.

[0031] In a preferred embodiment, the process of calculating the predicted fry injection parameters according to the mathematical model is as follows:

[0032] The injection parameters include the injection angle;

[0033] Obtain the injection angles of fry of different sizes under the best injection effect;

[0034] Use the least squares method to fit the fry size information and establish a mathematical model between the injection parameters and the fry size to predict the injection parameters of fry of different sizes. The expression of the mathematical model is:

[0035]

[0036] In the formula, is the predicted fry injection angle, L i is the fry length, K i is the fry width, a is the fry length coefficient, b is the fry width coefficient, and c is the constant coefficient.

[0037] In a preferred embodiment, the process of calculating the sum of squared errors function of the injection parameters according to the predicted fry injection parameters combined with the injection angle under the best injection effect is as follows:

[0038] Use the least squares method to find a set of coefficients a, b, c to minimize the error between the predicted fry injection angle and the injection angle under the best injection effect. The formula is:

[0039]

[0040] In the formula, S(a, b, c) is the sum of squared errors function;

[0041] To find the minimum value of S(a, b, c), take the partial derivatives of a, b, c respectively and set the partial derivatives to zero. The specific calculation formulas of the partial derivatives are as follows:

[0042]

[0043] Solve the system of equations to obtain the coefficients a, b, c, thereby determining the mathematical model between the injection angle and the fry size and obtaining the sum of squared errors function of the injection parameters.

[0044] In a preferred embodiment, the process of planning the movement trajectory of the injection needle using a path planning algorithm is as follows:

[0045] Set the current position of the injection device as the starting point and the predicted injection position of the fry as the ending point;

[0046] The cost of each grid includes the movement cost g(n) from the starting point to the current grid and the estimated cost h(n) from the current grid to the ending point. The total cost f(n) = g(n) + h(n);

[0047] By comparing the f(n) values of adjacent grids, select the grid with the minimum cost, and gradually search for the optimal path from the starting point to the ending point to obtain the movement trajectory of the injection needle.

[0048] In a preferred embodiment, the process of predicting the injection position of the fry is as follows:

[0049] Obtain the initial position coordinates (x0, y0) of the fry, and use the center point coordinates of the fry as the initial position coordinates;

[0050] Use the Kalman filter algorithm to predict the movement speed of the fry and decompose the movement speed to obtain the speeds of the fry in the x and y directions respectively;

[0051] Obtain the injection speed and injection dose of the fry, and calculate the injection time by multiplying the injection speed by the injection dose;

[0052] Predict the injection position of the fry by comprehensively calculating the initial position, movement speed, and injection time of the fry. The calculation formula is as follows:

[0053]

[0054] In the formula, (x f , y f ) is the predicted injection position of the fry, v fx , v fy are the speeds of the fry in the x and y directions respectively, J is the injection dose, v zs is the injection speed of the fry, is the injection time.

[0055] In a preferred embodiment, according to the predicted injection position of the fry combined with the actual position of the fry, calculate the injection accuracy of the fry. The specific calculation formula is:

[0056]

[0057] In the formula, TP is the injection accuracy of the fry, (x f , y f ) is the predicted injection position of the fry, (x s , y s) is the actual fry position.

[0058] In a preferred embodiment, the process of dynamically adjusting the injection posture according to the fry injection accuracy and the injection angle error is as follows:

[0059] Calculate the adjustment coefficient based on the sum of squared errors function and the weighted fry injection accuracy;

[0060] Compare and analyze the adjustment coefficient with a preset adjustment threshold. If the adjustment coefficient is greater than the preset adjustment threshold, it is necessary to dynamically adjust the injection posture by adjusting the injection parameters;

[0061] If the adjustment coefficient is less than the preset adjustment threshold, there is no need to dynamically adjust the injection posture.

[0062] In a preferred embodiment, the adjustment coefficient is calculated based on the sum of squared errors function and the weighted fry injection accuracy. The calculation formula is as follows:

[0063] P = u1 * S(a, b, c) - u2 * TP

[0064] Where P is the adjustment coefficient, S(a, b, c) is the sum of squared errors, u1 is the sum of squared errors weight factor, TP is the fry injection accuracy, and u2 is the fry injection accuracy weight factor.

[0065] The technical effects and advantages of the dynamic injection posture adjustment method based on fry size information of the present invention:

[0066] 1. By dynamically adjusting the injection posture according to the size information of fry, the present invention can accurately control the operation parameters during injection, avoiding uneven or inaccurate injection caused by inconsistent fry sizes. This can significantly improve the accuracy and effect of injection, ensuring that each fry is evenly injected. The method of dynamically adjusting the injection posture can adapt to fry of different sizes in real time, effectively reducing the adjustment time caused by inconsistent fry sizes and avoiding the repetition and waste of manual intervention. This can significantly improve production efficiency and reduce the workload of operators in large-scale production. By adjusting the injection posture for fry of different sizes, it can ensure that the fry do not receive excessive or insufficient pressure or damage during injection, thereby improving the safety of the injection process and enhancing the survival rate and survival ratio of the fry. This is of crucial significance for fry breeding and reproduction. This method can dynamically adjust the posture and parameters of the equipment according to the size information of the fry, with a high level of automation, reducing the possibility of human operation errors and improving the consistency of operation. For different species or sizes of fry, the system can be adjusted according to real-time data, with good adaptability. The dynamic injection posture adjustment method can dynamically adjust the injection parameters according to the size of the fry, thus avoiding waste of resources such as excessive medicaments, time, and manpower. This optimization can effectively reduce costs and improve breeding efficiency. Since the system can monitor and adjust the injection posture in real time, it reduces the risks and losses caused by improper operation. For example, inappropriate injection angles or forces may cause damage to the fry. Through precise control, risks can be reduced and the healthy development of the fry can be ensured.

[0067] 2. By collecting and analyzing fry images, the present invention accurately identifies the fry size, and then dynamically adjusts the injection angle and speed according to the size information. This personalized injection parameter setting ensures that the drug can be accurately injected into the target position in the fry body, enabling the drug to play a more effective role and improving the effect of treating and preventing diseases. Compared with the traditional injection method with fixed parameters, the accuracy of injection is greatly improved, and the treatment failure caused by improper injection position and dosage is reduced. The mathematical model established in combination with the principles of fluid mechanics makes the injection angle and speed match the fry size and physiological structure, which is conducive to the uniform diffusion of the drug in the fish body, avoiding local accumulation or uneven diffusion of the drug caused by improper injection, and improving the utilization rate of the drug. For fry of different sizes, parameters such as injection force and depth are reasonably controlled to avoid physical damage to the fry due to unreasonable injection parameters. For example, for smaller fry, the injection speed and force are appropriately reduced to minimize the impact on the fragile tissues of the fry and reduce the mortality rate of the fry during the injection process, thereby improving the survival rate and health status of the fry. Using the path planning algorithm, with the height of the injection device position as a reference, combined with the fry size and movement state, the optimal movement trajectory of the injection needle is planned. This makes the injection process more efficient, reduces the ineffective travel of the injection needle during movement, greatly shortens the injection time for a single fry, thereby improving the overall injection efficiency and reducing the labor cost and labor intensity. By real-time monitoring the movement state of the fry, when the change in the fry movement speed exceeds the preset threshold, the injection parameters and the movement trajectory are recalculated in a timely manner to ensure that the injection process is not interfered by the fry movement and to ensure the continuity and efficiency of the injection process. In the image acquisition link, through reasonable camera layout, light source configuration, and image preprocessing technology, it can adapt to different lighting and background conditions to ensure accurate acquisition of fry images in various aquaculture environments. In terms of model construction and parameter calculation, various influencing factors are considered, making the method highly versatile and applicable to the injection of fry of different varieties and different growth stages. Through a large amount of data collection and model training, as well as the analysis and fitting of experimental data, the injection parameters and algorithms are continuously optimized, enabling the system to continuously improve the injection effect with the accumulation of data and the increase of experience and adapt to the changing aquaculture needs. The present invention integrates a variety of advanced technologies such as computer vision, deep learning, and automation control, providing an innovative solution for the intelligent transformation of the aquaculture industry and promoting the development of aquaculture from traditional manual operation to automation and intelligence. During the entire injection process, the system collects a large amount of data on fry size, movement state, injection parameters, etc. These data can provide a basis for subsequent aquaculture decisions, such as optimizing the aquaculture environment and adjusting the feed feeding strategy, to promote the refined management of the aquaculture industry. Description of the Drawings

[0068] Figure 1Schematic structural diagram of the dynamic injection attitude adjustment method based on fry size information of the present invention. Detailed implementation manners

[0069] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0070] Embodiment 1 Figure 1 The dynamic injection attitude adjustment method based on fry size information of the present invention is given.

[0071] Collect images of fry in the injection area and preprocess the collected images;

[0072] Use a high-resolution camera to be arranged above the fry injection area to collect multi-angle images of the fry to be injected. During the collection process, ensure uniform light and avoid the influence of shadows on the image quality;

[0073] Convert the collected color image into a grayscale image to reduce the amount of data. The grayscale conversion expression is:

[0074] H = 0.299×R + 0.587×G + 0.114×B;

[0075] In the formula, H is the converted grayscale value, and R, G, and B are the red, green, and blue channel values of the image pixels respectively;

[0076] Calculate each pixel in the image through this formula to obtain a grayscale image;

[0077] Adopt the Gaussian filtering algorithm to denoise the image and remove the noise interference in the image. The process is as follows:

[0078] For each pixel in the image, calculate the weighted average of the pixel values in the pixel neighborhood to obtain the filtered pixel value;

[0079] Construct a Gaussian kernel and perform a convolution operation on each pixel in the image and its neighborhood pixels;

[0080] Multiply the Gaussian kernel by the pixel values in the corresponding neighborhood in the image and sum to obtain the filtered pixel value, thereby realizing the removal of image noise;

[0081] Then, use the histogram equalization algorithm to enhance the image, improve the contrast of the image, and provide high-quality image data for subsequent size recognition.

[0082] Identify the fry size information based on the preprocessed fry images;

[0083] Construct a two-dimensional grid map, and use the image annotation tool Label Img to manually annotate the head, tail, and two side points of the widest part of the fry, obtain the coordinates of the head, tail, and two sides of the widest part of the fry's body, and calculate the length and width of the fry through the distance formula between two points. The calculation formula is as follows:

[0084]

[0085]

[0086] In the formula, L i is the length of the fry, K i is the width of the fry, (x1, y1) is the coordinate of the fry's head,

[0087] (x2, y2)

[0088] is the coordinate of the fry's tail, (x3, y3), (x4, y4) are the two side points of the widest part of the fry respectively;

[0089] Establish a mathematical model between the injection parameters and the fry size based on the fry size information, and calculate the predicted fry injection parameters according to the mathematical model;

[0090] The injection parameters include the injection angle;

[0091] Obtain the injection angles of fry with different sizes under the best injection effect;

[0092] Use the least squares method to fit the fry size information, establish a mathematical model between the injection parameters and the fry size, and predict the injection parameters of fry with different sizes. The expression of the mathematical model is:

[0093]

[0094] In the formula, is the predicted fry injection angle, L i is the length of the fry, K i is the width of the fry, a is the fry length coefficient, b is the fry width coefficient, and c is the constant coefficient;

[0095] Calculate the sum of squared errors function of the injection parameters according to the predicted fry injection parameters combined with the injection angle under the best injection effect;

[0096] Use the least squares method to find a set of coefficients a, b, c to minimize the error between the predicted fry injection angle and the injection angle under the best injection effect. The formula is:

[0097]

[0098] Where S(a, b, c) is the sum of squared error function;

[0099] To find the minimum value of S(a, b, c), partial derivatives are taken with respect to a, b, and c respectively, and the partial derivatives are set to zero. The specific calculation formulas for the partial derivatives are as follows:

[0100]

[0101] Solve the above system of equations to obtain the coefficients a, b, and c, thereby determining the mathematical model between the injection angle and the fry size, and calculating the predicted fry injection parameters based on the fry size information.

[0102] Use the path planning algorithm to plan the movement trajectory of the injection needle, and predict the fry injection position in combination with the movement speed of the fry;

[0103] Set the current position of the injection device as the starting point and the predicted fry injection position as the ending point. The cost of each grid includes the movement cost g(n) from the starting point to the current grid and the estimated cost h(n) from the current grid to the ending point. The total cost f(n) = g(n) + h(n); by comparing the f(n) values of adjacent grids, select the grid with the minimum cost, and gradually search for the optimal path from the starting point to the ending point;

[0104] Obtain the initial position coordinates (x0, y0) of the fry, and use the center point coordinates of the fry as the initial position coordinates;

[0105] Use the Kalman filter algorithm to predict the movement speed of the fry, and decompose the movement speed to obtain the speeds of the fry in the x - direction and y - direction respectively;

[0106] Obtain the fry injection speed and injection dose, and calculate the injection time from the injection speed and injection dose;

[0107] Predict the fry injection position by comprehensively calculating the initial position of the fry, movement speed, and injection time. The calculation formula is as follows:

[0108]

[0109] Where (x f , y f ) is the predicted fry injection position, v fx , v fy are the speeds of the fry in the x - direction and y - direction respectively, J is the injection dose, v zs is the fry injection speed, is the injection time.

[0110] Calculate the fry injection accuracy based on the predicted fry injection position and the actual fry position, and dynamically adjust the injection posture in combination with the injection angle error.

[0111] Based on the predicted fry injection position and the actual fry position, the fry injection accuracy is calculated. The specific calculation formula is as follows:

[0112]

[0113] In the formula, TP is the fry injection accuracy, (x f , y f ) is the predicted fry injection position, and (x s , y s ) is the actual fry position.

[0114] The process of dynamically adjusting the injection posture according to the fry injection accuracy and the injection angle error is as follows:

[0115] The adjustment coefficient is calculated by weighted calculation based on the sum of squared errors function and the fry injection accuracy. The calculation formula is as follows:

[0116] P = u1 * S(a, b, c) - u2 * TP

[0117] In the formula, P is the adjustment coefficient, S(a, b, c) is the sum of squared errors, u1 is the weight factor of the sum of squared errors, TP is the fry injection accuracy, and u2 is the weight factor of the fry injection accuracy.

[0118] The adjustment coefficient is compared and analyzed with the preset adjustment threshold. If the adjustment coefficient is greater than the preset adjustment threshold, it is necessary to dynamically adjust the injection posture by adjusting the injection parameters;

[0119] If the adjustment coefficient is less than the preset adjustment threshold, there is no need to dynamically adjust the injection posture.

[0120] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0121] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0122] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0123] In addition, in each embodiment of the present application, each functional module may be integrated into one processing module, or each module may exist physically alone, or two or more modules may be integrated into one module.

[0124] As described above, the above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0125] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall all be included in the protection scope of the present invention.

Claims

1. A dynamic injection attitude adjustment method based on fry size information, characterized in that, It includes the following steps: Collect images of the fry injection area and preprocess the collected images; Identify the fry size information based on the preprocessed fry images; Establish a mathematical model between the injection parameters and the fry size according to the fry size information, and calculate the predicted fry injection parameters based on the mathematical model; Calculate the sum of squared errors function of the injection parameters according to the predicted fry injection parameters combined with the injection angle under the best injection effect; Use the path planning algorithm to plan the movement trajectory of the injection needle, and predict the fry injection position in combination with the movement speed of the fry; Calculate the fry injection accuracy according to the predicted fry injection position combined with the actual fry position, and dynamically adjust the injection posture in combination with the injection angle error.

2. The dynamic injection attitude adjustment method based on fry size information according to claim 1, wherein The process of preprocessing the collected images is as follows: Convert the collected color image into a grayscale image to reduce the data volume. The grayscale conversion expression is: H = 0.299×R + 0.587×G + 0.114×B; Where, H is the converted grayscale value, and R, G, and B are the red, green, and blue channel values of the image pixels respectively; Calculate each pixel in the image through this formula to obtain the grayscale image; Use the Gaussian filtering algorithm to denoise the image. The process of removing the noise interference in the image is as follows: For each pixel in the image, calculate the weighted average of the pixel values in the pixel neighborhood to obtain the filtered pixel value; Construct a Gaussian kernel and perform a convolution operation on each pixel in the image and its neighborhood pixels; Multiply the Gaussian kernel by the pixel values in the corresponding neighborhood in the image and sum them to obtain the filtered pixel value, thereby realizing the removal of image noise; Use the histogram equalization algorithm to enhance the image, improve the contrast of the image, and provide high-quality image data for subsequent size recognition.

3. The dynamic injection attitude adjustment method based on fry size information according to claim 2, characterized in that The process of identifying the fry size information based on the preprocessed fry images is as follows: Construct a two-dimensional grid map, obtain the coordinates of the head, tail, and two points on both sides of the widest part of the fry body, and calculate the length and width of the fry through the distance formula between two points. The calculation formula is as follows: where L i is the length of the fry, K i is the width of the fry, and (x1, y1) are the coordinates of the fry's head (x2, y2) is the fry tail coordinate, and (x3, y3), (x4, y4) are the two points on both sides of the widest part of the fry respectively.

4. The dynamic injection attitude adjustment method based on fry size information according to claim 3, characterized in that The process of calculating the predicted fry injection parameters according to the mathematical model is as follows: The injection parameters include the injection angle; Obtain the injection angles of fry of different sizes under the best injection effect; Use the least squares method to fit the fry size information and establish a mathematical model between the injection parameters and the fry size to predict the injection parameters of fry of different sizes. The expression of the mathematical model is: In the formula, is the predicted injection angle of fry, L i is the length of fry, K i is the width of fry, a is the length coefficient of fry, b is the width coefficient of fry, and c is a constant coefficient.

5. The dynamic injection attitude adjustment method based on fry size information according to claim 4, wherein The process of calculating the sum of squared errors function of the injection parameters according to the predicted fry injection parameters combined with the injection angle under the best injection effect is as follows: Use the least squares method to find a set of coefficients a, b, c to minimize the error between the predicted fry injection angle and the injection angle under the best injection effect. The formula is: Where, S(a, b, c) is the sum of squared errors function; To find the minimum value of S(a, b, c), take the partial derivatives of a, b, c respectively and set the partial derivatives to zero. The specific calculation formulas of the partial derivatives are as follows: Solve the system of equations to obtain the coefficients a, b, and c, thereby determining the mathematical model between the injection angle and the fry size, and obtaining the sum-of-squares function of the injection parameter error.

6. The dynamic injection attitude adjustment method based on fry size information according to claim 5, characterized in that The process of using the path planning algorithm to plan the motion trajectory of the injection needle is as follows: Set the current position of the injection device as the starting point and the predicted fry injection position as the ending point; The cost of each grid includes the movement cost g(n) from the starting point to the current grid and the estimated cost h(n) from the current grid to the ending point. The total cost f(n) = g(n) + h(n); By comparing the f(n) values of adjacent grids, select the grid with the minimum cost, and gradually search for the optimal path from the starting point to the ending point to obtain the motion trajectory of the injection needle.

7. The dynamic injection attitude adjustment method based on fry size information according to claim 6, wherein The process of predicting the fry injection position is as follows: Obtain the initial position coordinates (x0, y0) of the fry, and use the center point coordinates of the fry as the initial position coordinates; Use the Kalman filter algorithm to predict the movement speed of the fry, and decompose the movement speed to obtain the speeds of the fry in the x and y directions respectively; Obtain the injection speed and injection dose of the fry, and calculate the injection time by dividing the injection speed by the injection dose; Predict the fry injection position by comprehensively calculating the initial position, movement speed, and injection time of the fry. The calculation formula is as follows: Wherein, (x f , y f ) is the predicted injection position of the fry, v fx , v fy are the speeds of the fry in the x - direction and y - direction respectively, J is the injection dose, v zs is the injection speed of the fry, is the injection time.

8. The dynamic injection attitude adjustment method based on fry size information according to claim 7, wherein Based on the predicted fry injection position and the actual fry position, calculate the fry injection accuracy. The specific calculation formula is: Wherein, TP is the injection accuracy of fry, (x f , y f ) is the predicted injection position of fry, and (x s , y s ) is the actual position of fry.

9. The dynamic injection attitude adjustment method based on fry size information according to claim 8, characterized in that, The process of dynamically adjusting the injection posture according to the fry injection accuracy and the injection angle error is as follows: Calculate the adjustment coefficient by weighted calculation based on the sum-of-squares function of the error and the fry injection accuracy; Compare and analyze the adjustment coefficient with the preset adjustment threshold. If the adjustment coefficient is greater than the preset adjustment threshold, it is necessary to dynamically adjust the injection posture by adjusting the injection parameters; If the adjustment coefficient is less than the preset adjustment threshold, there is no need to dynamically adjust the injection posture.

10. The dynamic injection attitude adjustment method based on fry size information according to claim 9, characterized in that Calculate the adjustment coefficient by weighted calculation based on the sum-of-squares function of the error and the fry injection accuracy. The calculation formula is as follows: P = u1 * S(a, b, c) - u2 * TP In the formula, P is the adjustment coefficient, S(a, b, c) is the sum of squares of errors, u1 is the weight factor of the sum of squares of errors, TP is the fry injection accuracy, and u2 is the weight factor of the fry injection accuracy.

Images