Poultry feature monitoring-based precise feeding method and device for farm

Abstract

The invention discloses a livestock farm precise feeding method and device based on poultry feature monitoring, and relates to the technical field of data monitoring. The method comprises the following steps: acquiring image data and environment temperature of a target poultry house area in real time; extracting a state characteristic parameter of any poultry in the image data; importing the state characteristic parameters into a form evaluation model to obtain a form score; different feeding grades are divided according to the form scores, and different feeding modes are adopted for the different feeding grades; the feeding grade comprises a first feeding grade, a second feeding grade and a third feeding grade; periodically and dynamically adjusting and analyzing according to the environment temperature to obtain a feeding index; and readjusting the feeding grade according to the feeding index. Precise feeding of feed is achieved, the feed utilization rate is increased, and the breeding cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and specifically to a precise feeding method and device for a breeding farm based on poultry feature monitoring. Background Art

[0002] With the challenges faced by the traditional breeding mode, such as low efficiency, high cost and increased disease risk, poultry data monitoring has developed rapidly driven by the Internet of Things, sensors and artificial intelligence. Through standardized data collection and analysis, real-time assessment of the health status is achieved; combined with AI image recognition and Internet of Things technology, sick and weak chickens are accurately located and environmental management is optimized. The application of 5G technology further improves the data transmission efficiency, shortening the environmental control and disease warning response time of large-scale breeding to the minute level. This technological innovation not only reduces the dependence on manual labor and improves the survival rate, but also guarantees food safety through the whole-chain data traceability, promoting the industry to transform towards intelligence and sustainability.

[0003] The prior art (publication number CN119129880A) discloses a method and system for predicting the feed nutrient utilization rate of poultry, including: collecting characteristic monitoring data during the process of feeding poultry with feed; performing correlation analysis on the characteristic monitoring data to obtain a weight distribution result; performing data conversion processing on the corresponding characteristic monitoring data according to the weight distribution result; using the data after the conversion processing as input values and inputting them into a trained deep neural network integration model to obtain a model output result; performing data transformation on the model output to obtain the final feed nutrient utilization rate. In the present invention, the data set after data conversion processing is simultaneously input into a BP neural network model, a double-delay deep deterministic policy gradient neural network model and an LSTM neural network model that optimize the initial weights and thresholds through a genetic algorithm. The input data is analyzed by multiple neural network models, and the final analysis result is determined according to the result of majority voting.

[0004] However, in practical applications, it is impossible to accurately monitor the individual status of poultry in real time, and it is difficult to carry out differential feeding according to their growth needs, resulting in slow growth of some poultry; the influence of environmental temperature changes on poultry feeding is not considered either, and the feed delivery amount does not match the demand, causing waste, resulting in low feed utilization rate and high breeding cost. Summary of the Invention

[0005] The object of the present invention is to solve the problems that it is impossible to accurately monitor the individual status of poultry in real time, it is difficult to carry out differential feeding according to their growth needs, resulting in slow growth of some poultry; the influence of environmental temperature changes on poultry feeding is not considered either, and the feed delivery amount does not match the demand, causing waste, resulting in low feed utilization rate and high breeding cost, and to propose a method and device.

[0006] The object of the present invention can be achieved by the following technical solutions: First, a precise feeding method for a farm based on poultry feature monitoring is proposed. The method includes: Obtain the image data and environmental temperature of the target poultry house area in real time; extract the state characteristic parameters of any one poultry in the image data; the state characteristic parameters describe the state of the poultry at the current moment; import the state characteristic parameters into the morphological evaluation model to obtain a morphological score; the morphological score is used to describe the growth state of the poultry; Divide into different feeding levels according to the morphological score, and adopt different feeding methods for different feeding levels; the feeding levels include the first feeding level, the second feeding level, and the third feeding level; Perform periodic dynamic adjustment analysis based on the environmental temperature to obtain a feeding index; the feeding index is a measurement index for describing the feed fed to poultry; readjust the feeding level according to the feeding index.

[0007] Optionally, the pre-training process of the morphological evaluation model includes: Obtain multiple state characteristic parameters and morphological scores from the database; integrate the state characteristic parameters and their corresponding morphological scores into several training data and test data; Import several training data into an artificial intelligence model for training, and use the test data to test the trained artificial intelligence model; finally, obtain a morphological evaluation model with the state characteristic parameters and their corresponding morphological scores as the input and the morphological score as the output.

[0008] Optionally, the process of dividing into different feeding levels according to the morphological score and adopting different feeding methods for different feeding levels includes: Obtain a preset first morphological score threshold and a second morphological score threshold from the database; If the morphological score < the first morphological score threshold, it is divided into the first feeding level; If the first morphological score threshold ≤ the morphological score ≤ the second morphological score threshold, it is divided into the second feeding level; If the morphological score > the second morphological score threshold, it is divided into the third feeding level; The feeding method for the first feeding level adopts intensive feeding with high-nutrient-density feed; The feeding method for the second feeding level adopts standardized nutrition supply measures; The feeding method for the third feeding level adopts an energy restriction regulation mechanism.

[0009] Optionally, performing periodic dynamic adjustment analysis based on the environmental temperature to obtain a feeding index includes: Obtain the environmental temperature within a preset time period to obtain a temperature sequence T = {t1, t2, ……, t i ,……,t n}; T(i) represents the temperature corresponding to the i-th timestamp in the temperature sequence; i is the timestamp of the temperature sequence; n is the total number of timestamps in the temperature sequence; Through the formula where, DX(T(i)) represents the metabolic rate corresponding to the environmental temperature of t i ; D ref represents the metabolic rate corresponding to the optimal environmental temperature; T b represents the average temperature in the suitable temperature range; T min and T max respectively represent the minimum temperature and the maximum temperature in the suitable temperature range; k low and k high represent the low-temperature compensation factor and the high-temperature inhibition factor; k low and k high both have a value range of (0, 1); By calculating the metabolic rates corresponding to each environmental temperature in the temperature sequence, the average value of each metabolic rate is calculated and normalized to obtain the feeding index.

[0010] Optionally, readjusting the feeding level according to the feeding index includes: Comparing the feeding index with a preset lower threshold of the feeding index; If the feeding index < the preset lower threshold of the feeding index, the original feed level drops by one level; If the preset lower threshold of the feeding index ≤ the feeding index ≤ the preset upper threshold of the feeding index, the original feed level remains unchanged; If the feeding index > the preset upper threshold of the feeding index, the original feed level rises by one level.

[0011] Secondly, a precise feeding device for a farm based on poultry characteristic monitoring is proposed. The device includes: Data acquisition module: Real-time acquisition of image data and environmental temperature in the target poultry house area; extraction of the state characteristic parameters of any one poultry in the image data; the state characteristic parameters describe the state of the poultry at the current moment; State evaluation module: Importing the state characteristic parameters into the morphological evaluation model to obtain a morphological score; the morphological score is used to describe the growth state of the poultry; Feeding strategy module: Dividing into different feeding levels according to the morphological score, and adopting different feeding methods for different feeding levels; the feeding levels include the first feeding level, the second feeding level, and the third feeding level; Temperature control module: Performing periodic dynamic adjustment analysis according to the environmental temperature to obtain a feeding index; the feeding index is a measurement index for describing the feed fed to the poultry; readjusting the feeding level according to the feeding index.

[0012] Optionally, the state evaluation module includes a data training module and a data verification module: The data training module is used to obtain multiple state characteristic parameters and morphological scores from the database; integrate the state characteristic parameters and their corresponding morphological scores into several training data and verification data; The data verification module is used to import several training data into the artificial intelligence model for training, and verify the trained artificial intelligence model with the verification data; finally, a morphological evaluation model with the input being the state characteristic parameters and their corresponding morphological scores and the output being the morphological scores is obtained.

[0013] Optionally, the feeding strategy module includes a grading module and a feeding method module: The grading module is used to obtain a preset first threshold of morphological scores and a second threshold of morphological scores from the database; If the morphological score < the first threshold of morphological scores, it is classified into the first feeding grade; If the first threshold of morphological scores ≤ the morphological score ≤ the second threshold of morphological scores, it is classified into the second feeding grade; If the morphological score > the second threshold of morphological scores, it is classified into the third feeding grade; The feeding method module is used to adopt intensive feeding with high-nutrient-density feed for the first feeding grade; adopt standardized nutrition supply measures for the second feeding grade; and adopt an energy restriction regulation mechanism for the third feeding grade.

[0014] Optionally, the dynamic adjustment module includes a temperature control module and a data processing module: The temperature control module is used to obtain the ambient temperature within a preset time period to obtain a temperature sequence T = {t1, t2, ……, t i , ……, t n}; T(i) represents the temperature corresponding to the i-th timestamp in the temperature sequence; i is the timestamp of the temperature sequence; n is the total number of timestamps of the temperature sequence; Through the formula where DX(T(i)) represents the metabolic rate corresponding to the ambient temperature of t i ; D ref represents the metabolic rate corresponding to the optimum ambient temperature; T b represents the average temperature of the suitable temperature zone; T min and T max respectively represent the minimum temperature and the maximum temperature of the suitable temperature zone; k low and k high represent the low-temperature compensation factor and the high-temperature inhibition factor; k low and khigh The value ranges of all are (0, 1); The data processing module is configured to calculate the metabolic rate corresponding to each environmental temperature in the temperature sequence, calculate the average value of each metabolic rate and perform normalization processing to obtain a feeding index.

[0015] Optionally, the dynamic adjustment module includes a reset level module: The reset level module is configured to compare the feeding index with a preset lower threshold of the feeding index; If the feeding index < the preset lower threshold of the feeding index, the original feed level drops by one level; If the preset lower threshold of the feeding index < the feeding index < the preset upper threshold of the feeding index, the original feed level remains unchanged; If the feeding index > the preset upper threshold of the feeding index, the original feed level rises by one level.

[0016] Advantages of the present invention: The present invention obtains image data and environmental temperature of a target poultry house area in real time; extracts state characteristic parameters of any poultry in the image data; imports the state characteristic parameters into a morphological evaluation model to obtain a morphological score; divides into different feeding levels according to the morphological score, and adopts different feeding methods for different feeding levels; the feeding levels include a first feeding level, a second feeding level and a third feeding level; performs periodic dynamic adjustment analysis according to the environmental temperature to obtain a feeding index; readjusts the feeding level according to the feeding index. By monitoring the image data and environmental temperature of the target poultry house in real time, accurately extracting the state characteristics of poultry and scoring them, dividing feeding levels accordingly to implement differential feeding, and also being able to dynamically adjust the feeding strategy in combination with the environmental temperature, realizing precise feed delivery, improving the accuracy of evaluating the growth state of poultry, increasing the feed utilization rate, and reducing the breeding cost. Description of the Drawings

[0017] Figure 1 It is a flowchart of a precise feeding method for a farm based on poultry feature monitoring provided by an embodiment of the present invention; Figure 2 It is a structural schematic diagram of a precise feeding device for a farm based on poultry feature monitoring provided by an embodiment of the present invention. Detailed Embodiments

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the 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 the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] An embodiment of the present invention provides a precise feeding method for a farm based on poultry feature monitoring. Refer to Figure 1 , Figure 1 which is a flowchart of the precise feeding method for a farm based on poultry feature monitoring provided by an embodiment of the present invention. The method includes the following steps: Obtain image data and environmental temperature of a target poultry house area in real time; extract state characteristic parameters of any one poultry in the image data; the state characteristic parameters describe the state of the poultry at the current moment; import the state characteristic parameters into a morphology evaluation model to obtain a morphology score; the morphology score is used to describe the growth state of the poultry; Divide into different feeding levels according to the morphology score, and adopt different feeding methods for different feeding levels; the feeding levels include a first feeding level, a second feeding level, and a third feeding level; Perform periodic dynamic adjustment analysis according to the environmental temperature to obtain a feeding index; the feeding index is a measurement index for describing the feed fed to poultry; readjust the feeding level according to the feeding index.

[0020] Based on the precise feeding method for a farm based on poultry feature monitoring provided by an embodiment of the present invention, by monitoring the image data and environmental temperature of the target poultry house in real time, accurately extracting the state characteristics of poultry and scoring them, dividing feeding levels accordingly to implement differential feeding, and also dynamically adjusting the feeding strategy in combination with the environmental temperature, realizing precise feed delivery, improving the accuracy of evaluating the growth state of poultry, increasing the feed utilization rate, and reducing the breeding cost.

[0021] In one implementation, obtain image data and environmental temperature of a target poultry house area in real time; extract state characteristic parameters of any one poultry in the image data; the state characteristic parameters describe the state of the poultry at the current moment; import the state characteristic parameters into a morphology evaluation model to obtain a morphology score; the morphology score is used to describe the growth state of the poultry; Specifically, it should be noted that the image data of the target poultry house area is obtained through a camera device; the camera device includes a high-definition camera, etc.; the environmental temperature of the target poultry house area is obtained through a temperature sensor; the state change of any one poultry is obtained by collecting the image data; The morphology evaluation model is obtained through artificial intelligence model training; it includes obtaining multiple state characteristic parameters and morphology scores from a database; integrating the state characteristic parameters and their corresponding morphology scores into several training data and test data; specifically, the state characteristic parameters include the shape size, weight, glossiness, etc. of the poultry; by obtaining the state characteristic parameters of any one poultry in real time, and obtaining the shape change, weight change, and glossiness change of the poultry appearance, etc. through the state characteristic parameters, to comprehensively evaluate the morphological change situation, and its corresponding state score is used to describe the growth state of the poultry; integrating the state characteristic parameters and their corresponding morphology scores into several training data and test data; Import a number of training data into an artificial intelligence model for training, and use the test data to test the trained artificial intelligence model; specifically, input the state feature parameters and their corresponding morphological scores in the test data into the trained artificial intelligence model to output morphological scores. Check whether the absolute value of the difference between the morphological score and the morphological score recorded in the test data is within an acceptable range; if so, it means that this group of test data passes the test, and continue to test the next group of test data; if not, relevant parameters of the artificial intelligence model need to be adjusted, and continue to use this group of test data for testing; until a set proportion of test data passes the test; finally, obtain a morphological evaluation model with state feature parameters and their corresponding morphological scores as input and morphological scores as output; where the artificial intelligence model is a Transformer model, etc.

[0022] In one implementation, the process of dividing into different feeding levels according to morphological scores and adopting different feeding methods for different feeding levels includes: Obtain the preset first morphological score threshold and second morphological score threshold from the database; If the morphological score < the first morphological score threshold, it is divided into the first feeding level; If the first morphological score threshold ≤ morphological score ≤ the second morphological score threshold, it is divided into the second feeding level; If the morphological score > the second morphological score threshold, it is divided into the third feeding level; The feeding method for the first feeding level is to use high-nutrient-density feed for intensive feeding; The feeding method for the second feeding level is to adopt standardized nutrition supply measures; The feeding method for the third feeding level is to adopt an energy restriction regulation mechanism.

[0023] Specifically, it should be noted that the feeding level can describe that the higher the level, the more feed is fed; the preset first threshold and second threshold of morphological score are set by the staff based on historical experience; the feeding method of the first feeding level is to use high-nutrient-density feed for intensive feeding; for example, high-energy and high-protein feed (such as adding oil and fish meal) is used, and vitamins and minerals are supplemented simultaneously; the daily feeding frequency is increased to 3-4 times, and through free feeding or small and frequent meals, continuous nutrition supply is ensured to quickly improve the growth lag state of individuals with low scores; the feeding method of the second feeding level is to adopt standardized nutrition supply measures, for example, feed with a balanced ratio of energy, protein, and crude fiber is supplied according to the standard formula. Feed is fixed at 2-3 times a day, and the single amount is accurately set according to the animal's weight or growth stage. While avoiding overnutrition, the normal growth and development rhythm of the group is maintained; the feeding method of the third feeding level is to adopt an energy restriction regulation mechanism, for example, reduce the feeding amount by 10%-20%, and switch to low-energy and high-fiber feed. Reduce the proportion of fat and carbohydrates and increase dietary fiber. Feed strictly at a fixed time 2 times a day, and control weight gain by restricting the feeding time and amount to prevent health problems caused by overnutrition.

[0024] In one implementation, the feeding index obtained by performing periodic dynamic adjustment analysis according to the environmental temperature includes: Obtain the environmental temperature within a preset time period to obtain a temperature sequence T = {t1, t2, ……, t i , ……, t n}; T(i) represents the temperature corresponding to the i-th timestamp in the temperature sequence; i is the timestamp of the temperature sequence; n is the total number of timestamps of the temperature sequence; Through the formula where DX(T(i)) represents the metabolic rate corresponding to the environmental temperature of t i ; D ref represents the metabolic rate corresponding to the optimal environmental temperature; T b represents the average temperature of the suitable temperature zone; T min and T max respectively represent the minimum temperature and maximum temperature of the suitable temperature zone; k low and k high represent the low-temperature compensation factor and high-temperature inhibition factor; the value ranges of k low and k high are both (0, 1); By calculating the metabolic rates corresponding to each environmental temperature in the temperature sequence, the average value of each metabolic rate is calculated and normalized to obtain the feeding index; Specifically, the feeding index is an indicator for measuring the feed content based on the metabolic rate of poultry; the normalization operations include Z-Score, etc.; the preset time periods include one hour, one day, one week, etc.; k low and k high indicate that the low-temperature compensation factor and the high-temperature inhibition factor are obtained by the staff according to historical data; for example, poultry of the same batch (such as 28-day-old broilers) are divided into 3 to 5 groups, and the temperatures of the poultry houses are controlled to be 5°C, 3°C, 1°C below the minimum temperature of the suitable temperature zone and the optimum temperature of the suitable temperature zone (for example, if the suitable temperature zone is 18 - 23°C, then set T = 13°C, 15°C, 17°C, 20°C); the same amount of standard feed is fed to each group (based on 100% food intake at 20°C in the suitable temperature zone), and the 24-hour food intake is recorded; when the temperature drops by 1°C, the increased proportion of the food intake is recorded as k low ; for example, set the house temperature to 1°C, 3°C, 5°C above the upper limit of the suitable temperature zone and the upper limit of the suitable temperature zone (such as 23°C, 25°C, 27°C, 29°C); the feeding amount is the standard amount in the suitable temperature zone, and the 24-hour remaining feed amount is recorded; when the temperature rises by 1°C, the reduced proportion of the food intake is k high .

[0025] In this solution, the feeding index is calculated by averaging the metabolic rates corresponding to the environmental temperatures in the temperature sequence and then normalizing, which eliminates the individual data deviation caused by temperature fluctuations, makes the feeding index more representative and stable, effectively improves the feeding accuracy, and avoids the waste or insufficient feeding of feed.

[0026] In one implementation, readjusting the feeding level according to the feeding index includes: comparing the feeding index with the preset lower threshold of the feeding index; if the feeding index < the preset lower threshold of the feeding index, the original feed level drops by one level; if the preset lower threshold of the feeding index ≤ the feeding index ≤ the preset upper threshold of the feeding index, the original feed level remains unchanged; if the feeding index > the preset upper threshold of the feeding index, the original feed level rises by one level.

[0027] Specifically, the preset lower threshold and upper threshold of the feeding index are obtained by the staff based on the experience of historical data; This solution realizes the dynamic adjustment of the feed level by automatically comparing the preset threshold with the feeding index, reduces the cost of manual intervention; accurately matches the animal nutrition requirements, avoids feed waste or insufficient supply, and improves the utilization efficiency and breeding benefits; Based on the same inventive concept, the embodiment of the present invention also provides a precise feeding device for a farm based on poultry characteristic monitoring. See Figure 2 , Figure 2The structural schematic diagram of the precise feeding device for a farm based on poultry feature monitoring provided by the embodiment of the present invention includes: Data acquisition module: Real-time acquire the image data and environmental temperature of the target poultry house area; Extract the state characteristic parameters of any one poultry in the image data; The state characteristic parameters describe the state of the poultry at the current moment. State evaluation module: Import the state characteristic parameters into the morphology evaluation model to obtain a morphology score; The morphology score is used to describe the growth state of the poultry. Feeding strategy module: Divide into different feeding levels according to the morphology score, and adopt different feeding methods for different feeding levels; The feeding levels include the first feeding level, the second feeding level, and the third feeding level. Temperature regulation module: Perform periodic dynamic adjustment analysis according to the environmental temperature to obtain a feeding index; The feeding index is a measurement index for describing the feed fed to the poultry; Readjust the feeding levels according to the feeding index.

[0028] Based on the precise feeding device for a farm based on poultry feature monitoring provided by the embodiment of the present invention, by real-time monitoring the image data and environmental temperature of the target poultry house, accurately extracting the state characteristics of the poultry and scoring them, dividing the feeding levels accordingly to implement differential feeding, and also being able to dynamically adjust the feeding strategy in combination with the environmental temperature, realizing precise feed delivery, improving the accuracy of evaluating the growth state of the poultry, increasing the feed utilization rate, and reducing the breeding cost.

[0029] It should be noted that in this article, terms such as "including", "comprising" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device.

[0030] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. A precise feeding method for a farm based on poultry feature monitoring, characterized in that The method includes: Obtaining image data and environmental temperature of a target poultry house area in real time; extracting state characteristic parameters of any one poultry in the image data; the state characteristic parameters describe the state of the poultry at the current moment; importing the state characteristic parameters into a morphological evaluation model to obtain a morphological score; the morphological score is used to describe the growth state of the poultry; Dividing into different feeding levels according to the morphological score, and adopting different feeding methods for different feeding levels; the feeding levels include a first feeding level, a second feeding level, and a third feeding level; Performing periodic dynamic adjustment analysis according to the environmental temperature to obtain a feeding index; the feeding index is a measurement index for describing the feed fed to the poultry; readjusting the feeding level according to the feeding index.

2. The precise feeding method for a farm based on poultry feature monitoring according to claim 1, wherein The pre-training process of the morphological evaluation model includes: Obtaining a plurality of state characteristic parameters and morphological scores from a database; integrating the state characteristic parameters and their corresponding morphological scores into several training data and test data; Importing the several training data into an artificial intelligence model for training, and using the test data to test the trained artificial intelligence model; finally obtaining a morphological evaluation model with the state characteristic parameters and their corresponding morphological scores as inputs and the morphological score as the output.

3. The precise feeding method for a farm based on poultry feature monitoring according to claim 1, wherein The process of dividing into different feeding levels according to the morphological score and adopting different feeding methods for different feeding levels includes: Obtaining a preset first morphological score threshold and a second morphological score threshold from a database; If the morphological score < the first morphological score threshold, it is divided into the first feeding level; If the first morphological score threshold < the morphological score < the second morphological score threshold, it is divided into the second feeding level; If the morphological score > the second morphological score threshold, it is divided into the third feeding level; The feeding method for the first feeding level adopts intensive feeding with high-nutrient-density feed; The feeding method for the second feeding level adopts standardized nutrition supply measures; The feeding method for the third feeding level adopts an energy restriction regulation mechanism.

4. The precise feeding method for a farm based on poultry feature monitoring according to claim 1, wherein The process of performing periodic dynamic adjustment analysis according to the environmental temperature to obtain a feeding index includes: Obtain the ambient temperature within a preset time period to get a temperature sequence T = {t1, t2, ……, t i , ……, t n}; T(i) represents the temperature corresponding to the i-th timestamp in the temperature sequence; i is the timestamp of the temperature sequence; n is the total number of timestamps in the temperature sequence; Through the formula Among them, DX(T(i)) represents the metabolic rate when the environmental temperature is t i corresponding metabolic rate; D ref represents the metabolic rate corresponding to the optimum environmental temperature; T b represents the average temperature in the suitable temperature range; T min and T max respectively represent the minimum temperature and the maximum temperature in the suitable temperature range; k low and k high represent the low-temperature compensation factor and the high-temperature inhibition factor; k low and k high both have a value range of (0, 1); By calculating the metabolic rate corresponding to each environmental temperature in the temperature sequence, averaging each metabolic rate and performing normalization processing to obtain a feeding index.

5. The precise feeding method for a farm based on poultry feature monitoring according to claim 1, characterized in that, The process of readjusting the feeding level according to the feeding index includes: Comparing the feeding index with a preset lower threshold of the feeding index; If the feeding index < the preset lower threshold of the feeding index, the original feed level drops by one level; If the preset lower threshold of the feeding index ≤ the feeding index ≤ the preset upper threshold of the feeding index, the original feed level remains unchanged; If the feeding index > the preset upper threshold of the feeding index, the original feed level rises by one level.

6. The precise feeding device for a breeding farm based on poultry feature monitoring, characterized in that, The device includes: A data acquisition module: obtaining image data and environmental temperature of a target poultry house area in real time; extracting state characteristic parameters of any one poultry in the image data; the state characteristic parameters describe the state of the poultry at the current moment; A state evaluation module: importing the state characteristic parameters into a morphological evaluation model to obtain a morphological score; the morphological score is used to describe the growth state of the poultry; Feeding strategy module: Divided into different feeding levels according to morphological scores, and different feeding methods are adopted for different feeding levels; the feeding levels include the first feeding level, the second feeding level, and the third feeding level; Dynamic adjustment module: Periodically dynamically adjust and analyze according to the environmental temperature to obtain a feeding index; the feeding index is a measurement index describing the feeding of poultry feed; adjust the feeding level according to the feeding index.

7. The precision feeding device for a farm based on poultry feature monitoring according to claim 6, wherein The state evaluation module includes a data training module and a data verification module: The data training module is used to obtain multiple state characteristic parameters and morphological scores from the database; integrate the state characteristic parameters and their corresponding morphological scores into several training data and verification data; The data verification module is used to import several training data into an artificial intelligence model for training, and use the verification data to verify the trained artificial intelligence model; finally, obtain a morphological evaluation model with state characteristic parameters and their corresponding morphological scores as inputs and morphological scores as outputs.

8. The precision feeding device for a farm based on poultry feature monitoring according to claim 6, characterized in that, The feeding strategy module includes a level division module and a feeding method module: The level division module is used to obtain a preset first morphological score threshold and a second morphological score threshold from the database; If the morphological score < the first morphological score threshold, it is divided into the first feeding level; If the first morphological score threshold ≤ morphological score ≤ the second morphological score threshold, it is divided into the second feeding level; If the morphological score > the second morphological score threshold, it is divided into the third feeding level; The feeding method module is used to adopt intensive feeding with high-nutrient-density feed for the feeding method of the first feeding level; adopt standardized nutrition supply measures for the feeding method of the second feeding level; adopt an energy restriction regulation mechanism for the feeding method of the third feeding level.

9. The precision feeding device for a farm based on poultry feature monitoring according to claim 6, wherein The dynamic adjustment module includes a temperature control module and a data processing module: The temperature control module is used to obtain the ambient temperature within a preset time period to obtain a temperature sequence T = {t1, t2, ……, t i , ……, t n}; T(i) represents the temperature corresponding to the i-th timestamp in the temperature sequence; i is the timestamp of the temperature sequence; n is the total number of timestamps of the temperature sequence; Through the formula Among them, DX(T(i)) represents the metabolic rate when the environmental temperature is t i corresponding metabolic rate; D ref represents the metabolic rate corresponding to the optimum environmental temperature; T b represents the average temperature in the suitable temperature range; T min and T max respectively represent the minimum temperature and the maximum temperature in the suitable temperature range; k low and k high represent the low-temperature compensation factor and the high-temperature inhibition factor; k low and k high both have a value range of (0, 1); The data processing module is used to calculate the metabolic rate corresponding to each environmental temperature in the temperature sequence, calculate the average value of each metabolic rate and perform normalization processing to obtain the feeding index.

10. The precision feeding device for a farm based on poultry feature monitoring according to claim 6, characterized in that, The dynamic adjustment module includes a reset level module: The reset level module is used to compare the feeding index with a preset lower feeding index threshold; If the feeding index < the preset lower feeding index threshold, the original feed level drops by one level; If the preset lower feeding index threshold ≤ feeding index ≤ the preset upper feeding index threshold, the original feed level remains unchanged; If the feeding index > the preset upper feeding index threshold, the original feed level rises by one level.

Images