Free-range chicken health monitoring method and system based on excrement detection

Abstract

The invention relates to the field of free-range chicken health monitoring, in particular to a free-range chicken health monitoring method and system based on excrement detection. The method comprises the following steps: acquiring an excrement image of free-range chickens through monitoring equipment, analyzing the excrement image to obtain a health index of a free-range chicken flock, constructing a matrix through chicken excrement obtained in an activity route of the free-range chickens on feeding management of the chicken flock and the health index, and constructing a monitoring sequence through the health index and temperature change of the chicken flock. And constructing a matrix through the data in the monitoring sequence, and obtaining a health monitoring result through matrix analysis. The health of the chicken flocks is further judged according to the health state of the chicken flocks determined according to body temperature changes and excrement, feeding management is adjusted according to the dynamic change of the health of the chicken flocks, the health state of the free-range chicken flocks is fed back more visually by monitoring the health of the chicken flocks, meanwhile, the feeding mode can be adjusted in time when the health state of the chicken flocks fluctuates, and the feeding efficiency is improved. A large range of diseases of fed chicken flocks are avoided.

Description

Technical Field

[0001] This invention relates to the field of health monitoring of free-range chickens, and specifically to a method and system for health monitoring of free-range chickens based on fecal detection. Background Technology

[0002] Chicken droppings are closely linked to the health of chickens; different colors and textures of droppings are associated with different chicken diseases, making chickens a primary concern for poultry farmers. Currently, monitoring chicken droppings mainly relies on manual inspections by experienced staff or veterinarians, which is highly subjective, inaccurate, and prone to failing to detect early-stage diseases. Therefore, it is essential to adopt advanced detection technologies to analyze and assess chicken droppings, improving the efficiency and accuracy of droppings analysis and providing technical support for intelligent and unmanned farming.

[0003] Meanwhile, when raising free-range chickens, it is necessary not only to monitor the health status of the chickens, but also to pay attention to the changes in the health status of the chickens caused by the feeding conditions. In the current technology, only the health of the chickens is detected, which ignores the conditions of the chickens' feeding environment that affect the health status of the flock. Therefore, there is an urgent need for a free-range chicken health monitoring method and system based on fecal detection to solve the above-mentioned problems. Summary of the Invention

[0004] In view of the limitations of the existing methods described above, the purpose of this invention is to propose a method and system for monitoring the health of free-range chickens based on fecal detection, so as to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions.

[0005] To achieve the above objectives, according to one aspect of the present invention, a method and system for monitoring the health of free-range chickens based on fecal detection are provided, the method comprising the following steps: A method for monitoring the health of free-range chickens based on fecal analysis, the method comprising the following steps: S100: Acquire images of free-range chicken droppings through monitoring equipment; S200: Analyze the fecal images to obtain the health index of the free-range chicken flock; S300: Constructing a matrix of chicken flock feeding management and health index through chicken manure; S400: A monitoring sequence is constructed by linking health index with changes in flock temperature; S500: A matrix is ​​constructed by monitoring data in the sequence, and the health monitoring results are obtained through matrix analysis.

[0006] Further, in step S100, the location of feces during the free-range chickens' activities is obtained through monitoring, and a sample of feces images of the free-range chickens are periodically acquired at close range using a drone. Monitoring data is obtained through the feces images, and the type of feces is determined by comparing the color and consistency of the feces with known feces-disease relationships. The health of the free-range chicken flock is monitored, and monitoring data is obtained through health monitoring. The monitoring data includes the flock's behavioral characteristics, flock temperature status, and the health status of the flock's feces.

[0007] Further, in step S200, edge noise reduction processing is performed on some sample images of free-range chickens acquired by the drone. Based on the shape, color, and surface features in the acquired sample images, the chicken feces are analyzed. The preliminary analysis distinguishes the chicken feces samples into normal chicken feces samples and abnormal chicken feces samples based on the surface features of the chicken feces. The sample ratio of normal chicken feces samples and abnormal chicken feces samples is recorded. By performing data preprocessing on the spectral data of the two types of feces samples using four methods, namely no processing, multivariate scattering correction, SG convolution smoothing, and standard deviation standardization, samples with small differences between the groups are obtained. The sample processing reduces the impact of abnormal feces samples on the health assessment of the chicken flock.

[0008] Furthermore, background clutter in the collected fecal images is treated as noise, and the background of the fecal images is removed by noise reduction. Since the background in the fecal images differs greatly in color and texture from the feces, this study uses the maximum inter-class variance method to achieve adaptive removal of the image background. Let the image after removing the background be defined as image P. For image P(x,y), let T be the threshold used to separate feces and background; and let be the proportion of foreground pixels to the total number of pixels in the image. The average gray level of the fecal image is denoted as . The proportion of background image pixels to the total number of pixels in the image is denoted as . The average gray level of the background image is denoted as . Assume the overall average gray level of the image pixels is ƞ, and its inter-class variance is denoted as g; if the size of the image P(x,y) is X×Y, and the threshold T of the number of pixels for all gray levels in the image is denoted as . The number of pixels whose grayscale values ​​are greater than a set threshold T is recorded as . g= The adaptive segmentation was completed by calculating the inter-class variance g and the maximum threshold T using an iterative method. Image segmentation was performed using T as the segmentation threshold to obtain a binary image with the background removed. This binary image, processed by the OTSU algorithm, was then overlaid with the original image to obtain the image with the background removed. Texture features of feces in the image were obtained using the least squares method and the sparrow search algorithm. These texture features and color were used to perform a preliminary health assessment of the chickens, and a health assessment score was output.

[0009] Further, in step S300, the obtained flock behavior characteristics, flock temperature status, and flock fecal health status are labeled as SP_Status, BT_Status, and CM_Status. The data are instantaneous values ​​obtained at the same time interval within a certain period. The flock behavior characteristics include the feeding, drinking, posture, agility, and gait of the entire free-range flock. The flock temperature status is the temperature change of the entire free-range flock in this period compared to the previous period. The flock fecal health status is the health assessment score obtained by judging the flock feces. The difference is calculated by adjacent data to obtain the change value of the flock health status at different times. Based on the magnitude of the change value observed at different times, the change values ​​are defined as change_SP, change_BT, and change_CM. By using the same acquisition cycle as the feces images of free-range chickens, the temperature of the flock when returning to the pen is collected by a temperature sensor to obtain the average temperature of the flock. The average temperature of the flock is recorded by the average temperature of the flock collected in the previous cycle, and chickens with abnormal temperatures are marked. An empty set temp with a time series is set. The collected temperature values ​​are added to the set temp in chronological order to obtain a set temp with a time series of temperature values, temp=[t1, t2, ..., tn], where n is the number of acquisition nodes in the cycle, and avg(temp) represents the average temperature. The temperature value ti is standardized with the flock temperature status BT_Status, and data at the same time are mapped. The ti is the i-th element of the set temp. The health status CM_Status of the chicken flock's feces is sorted by the time series of the collected data to form a health monitoring sequence MON. The differences between all adjacent CM_Status values ​​in the health monitoring sequence MON are calculated in chronological order, and the resulting differences are used to construct a difference sequence Dist. The difference sequence Dist is then divided according to the size of its elements: when Dist[i] ≥ 0, the element is classified as an increasing sequence GDist; when Dist[i] < 0, the element is classified as a decreasing sequence LDist. Dist[i] is the i-th element of the difference sequence. The elements in the increasing sequence GDist are then filtered, and the growth potential coefficient ATE is calculated. The calculation method for the growth potential coefficient ATE is as follows: ATE=log( )- ; Max(GDist) represents the maximum value of the elements in the increasing sequence GDist, Min(GDist) represents the minimum value of the elements in the increasing sequence GDist, and len(GDist) represents the length of the increasing sequence GDist. That is, to obtain the maximum value of the reduced sequence LDist, the That is, to obtain the minimum value of the reduced sequence LDist, the To obtain the arithmetic mean of the reduced sequence LDist, log() is the logarithmic function with base 2, and ln() is the natural logarithm. The potential coefficient ATE is used to indicate the stable state of the free-range chicken flock when the potential coefficient ATE approaches 0 within a period of free-range raising. If the potential coefficient ATE fluctuates, it indicates that there is a problem with the free-range chicken flock's raising status.

[0010] Further, in step S400, record For the earliest adjacent health status CM_Status difference in the difference sequence Dist obtained by the i-th collection node, By comparing with the minimum value in the reduced sequence LDist The ratio is obtained by comparison. / ,remember For the latest adjacent health status CM_Status difference in the difference sequence Dist obtained by the i-th acquisition node, the minimum value in the reduction sequence LDist is... Through with The ratio is obtained by comparison. The ratio / and The stable value STA was obtained through calculation. STA= , The To reduce the sum of the minimum, average, and maximum values ​​of the LDist sequence; By calculating the decreasing sequence, adjustments are made for fluctuations, and the stable value STA is then incorporated into the ATE calculation: Output the updated growth coefficient .

[0011] A health monitoring system for free-range chickens based on fecal detection, the system comprising: an acquisition module for acquiring flock temperature, flock fecal samples, and flock behavioral characteristics; The identification module identifies and analyzes fecal images to obtain the health index of free-range chickens. The processing module is used to process the monitoring data acquired by the acquisition module and output the monitoring results.

[0012] A non-volatile storage medium storing a computer program, wherein the device containing the non-volatile storage medium executes a method for monitoring the health of free-range chickens based on fecal detection as described in any one of the above methods by running the computer program.

[0013] A computer program product includes: a computer program, wherein when the computer program is executed by a processor, it implements a method for monitoring the health of free-range chickens based on fecal detection as described in any one of the above methods.

[0014] Preferably, the three-dimensional data containing the change values ​​change_SP, change_BT, and change_CM are mapped to a three-dimensional Cartesian coordinate system to obtain a three-dimensional discrete data point set; Time series, flow velocity series and temperature series are generated based on a three-dimensional discrete data point set. After standardizing the time series, flow velocity series and temperature series respectively, the DBSCAN clustering algorithm is used to cluster them and retain the valid data points of each series. The intersection of the valid data points of all series is re-aligned to obtain a three-dimensional ordered data point set. Multiple constraint equations are formed based on the condition that adjacent segmented data point sets satisfy multi-order continuity at corresponding connection points. A set of coefficient equations is formed based on multiple constraint equations, and the set of coefficient equations is solved to obtain the interpolation coefficients of each segmented data point set. A standardized three-dimensional curve function is obtained based on the interpolation coefficients of all segmented data points. The health status of the chicken flock is optimized by obtaining the three-dimensional curve function through three changes.

[0015] The beneficial effects of this invention are as follows: by identifying and analyzing feces, the health status of the chicken flock is determined; after the chickens are free-range, their body temperature changes are recorded; and the health status of the chicken flock is further judged based on the changes in body temperature and the feces. At the same time, the feeding management is adjusted according to the dynamic changes in the health of the chicken flock. By monitoring the health of the chicken flock, the health status of the free-range chicken flock is more intuitively reflected. At the same time, when the health status of the chicken flock fluctuates, animal health care can be strengthened in a timely manner to avoid large-scale diseases in the chicken flock. Attached Figure Description

[0016] The above and other features of the present invention will become more apparent from the detailed description of the embodiments shown in conjunction with the accompanying drawings. In the accompanying drawings, the same reference numerals denote the same or similar elements. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort. In the drawings: Figure 1 The diagram shows a flowchart of a method for monitoring the health of free-range chickens based on fecal detection. Detailed Implementation

[0017] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with the embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0018] In the description of this invention, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0019] like Figure 1 The diagram shown is a flowchart of a method for monitoring the health of free-range chickens based on fecal detection according to the present invention. The following is a summary of the method. Figure 1 This paper describes a method for monitoring the health of free-range chickens based on fecal detection according to an embodiment of the present invention.

[0020] S100: Acquire images of free-range chicken droppings through monitoring equipment; S200: Analyze the fecal images to obtain the health index of the free-range chicken flock; S300: Constructing a matrix of chicken feeding management and health index based on chicken manure obtained from the activity routes of free-range chickens; S400: A monitoring sequence is constructed by linking health index with changes in flock temperature; S500: A matrix is ​​constructed by monitoring data in the sequence, and the health monitoring results are obtained through matrix analysis.

[0021] Further, in step S100, the location of feces during the free-range chickens' activity is obtained through monitoring. When the free-range chickens leave the activity area, the activity area includes the inside of the chicken coop and the free-range chickens' activity area. Typically, monitoring cameras are set up at the locations where the free-range chickens have been active. After the monitoring cameras capture the chickens leaving, images of the free-range chicken feces are automatically collected via a preset drone path. At the same time, a camera is set in the drone to collect images of chicken feces. A sample of free-range chicken feces images is obtained at close range by the drone. Monitoring data is obtained from the feces images. The monitoring data cycle is generally set to once in the morning and once in the afternoon each day. A sample of chicken feces is obtained from the same location for analysis. The type of feces is determined by comparing the color and consistency of the feces with known feces-disease relationships. The health of the free-range chickens is monitored by observing the texture features, moisture content, consistency, color, and proportion of each feature in the feces images. Monitoring data is obtained through health monitoring. The monitoring data includes chicken flock behavior characteristics, chicken flock temperature status, and chicken flock feces health status.

[0022] Further, in step S200, edge noise reduction processing is performed on some sample images of free-range chickens acquired by the drone. Based on the shape, color, and surface features in the acquired sample images, the chicken feces are analyzed. The preliminary analysis distinguishes the chicken feces samples into normal chicken feces samples and abnormal chicken feces samples based on the surface features of the chicken feces. The sample ratio of normal chicken feces samples and abnormal chicken feces samples is recorded. By performing data preprocessing on the spectral data of the two types of feces samples using four methods, namely no processing, multivariate scattering correction, SG convolution smoothing, and standard deviation standardization, samples with small differences between the groups are obtained. The sample processing reduces the impact of abnormal feces samples on the health assessment of the chicken flock.

[0023] Preferably, it predicts the category of unknown samples by establishing a response function between the original sample data and its category, making it suitable for sample sets with a small number of samples and a large number of variables. Therefore, the PLS-DA method is used to establish a classification and discrimination model, and the most suitable preprocessing method for each type of feces is selected according to the evaluation criteria. The preprocessed data is then subjected to data dimensionality reduction using three variable optimization methods: principal component analysis, competitive adaptive reweighted sampling, and an improved hybrid frog leaping algorithm, to improve the lightweight nature of the model. The optimal data dimensionality reduction method for each type of feces is determined based on the model evaluation index, and finally, a lightweight classification and discrimination model for chicken feces based on PLS-DA is established.

[0024] Model optimization aims to reduce redundant wavelengths and improve computational performance. While maintaining accuracy, it extracts feature wavelengths to distinguish fecal types, reducing the number of feature wavelengths and achieving model lightweighting. Preprocessed chicken fecal sample data is subjected to dimensionality reduction using PCA, CARS, and an improved ISFLA method, significantly reducing the original feature wavelength variables. The dimensionality-reduced data is then used to build a PLS-DA classification model, achieving model lightweighting. Compared to the original ISFLA algorithm, the improved ISFLA algorithm uses five-fold cross-validation combined with PLS model regression to determine a more accurate number of principal factors, *l*. The PLS-DA discriminant model is then built, and the discrimination results are output. The discrimination error is calculated, and the group with the smallest discrimination error is selected as the final number of principal factors. The number of iterations, previously determined empirically, is now determined based on minimizing the root mean square error of the model's cross-validation.

[0025] Furthermore, background clutter in the collected fecal images is treated as noise, and the background of the fecal images is removed by noise reduction. Since the background in the fecal images differs greatly in color and texture from the feces, this study uses the maximum inter-class variance method to achieve adaptive removal of the image background. Let the image after removing the background be defined as image P. For image P(x,y), let T be the threshold used to separate feces and background; and let be the proportion of foreground pixels to the total number of pixels in the image. The average gray level of the fecal image is denoted as . The proportion of background image pixels to the total number of pixels in the image is denoted as . The average gray level of the background image is denoted as . Assume the overall average gray level of the image pixels is ƞ, and its inter-class variance is denoted as g; if the size of the image P(x,y) is X×Y, and the threshold T of the number of pixels for all gray levels in the image is denoted as . The number of pixels whose grayscale values ​​are greater than a set threshold T is recorded as . g= The adaptive segmentation was completed by calculating the inter-class variance g and the maximum threshold T using an iterative method. Image segmentation was performed using T as the segmentation threshold to obtain a binary image with the background removed. This binary image, processed by the OTSU algorithm, was then overlaid with the original image to obtain the image with the background removed. Texture features of feces in the image were obtained using the least squares method and the sparrow search algorithm. These texture features and color were used to perform a preliminary health assessment of the chickens, and a health assessment score was output.

[0026] Further, in step S300, the obtained flock behavior characteristics, flock temperature status, and flock fecal health status are labeled as SP_Status, BT_Status, and CM_Status. The data are instantaneous values ​​obtained at the same time interval within a certain period. The flock behavior characteristics include the feeding, drinking, posture, agility, and gait of the entire free-range flock. The flock temperature status is the temperature change of the entire free-range flock in this period compared to the previous period. The flock fecal health status is the health assessment score obtained by judging the flock feces. The difference is calculated by adjacent data to obtain the change value of the flock health status at different times. Based on the magnitude of the change value observed at different times, the change values ​​are defined as change_SP, change_BT, and change_CM. By using the same acquisition cycle as the feces images of free-range chickens, the temperature of the flock when returning to the pen is collected by a temperature sensor to obtain the average temperature of the flock. The average temperature of the flock is recorded by the average temperature of the flock collected in the previous cycle, and chickens with abnormal temperatures are marked. An empty set temp with a time series is set. The collected temperature values ​​are added to the set temp in chronological order to obtain a set temp with a time series of temperature values, temp=[t1, t2, ..., tn], where n is the number of acquisition nodes in the cycle, and avg(temp) represents the average temperature. The temperature value ti is standardized with the flock temperature status BT_Status, and data at the same time are mapped. The ti is the i-th element of the set temp. The health status CM_Status of the chicken flock's feces is sorted by the time series of the collected data to form a health monitoring sequence MON. The differences between all adjacent CM_Status values ​​in the health monitoring sequence MON are calculated in chronological order, and the resulting differences are used to construct a difference sequence Dist. The difference sequence Dist is then divided according to the size of its elements: when Dist[i] ≥ 0, the element is classified as an increasing sequence GDist; when Dist[i] < 0, the element is classified as a decreasing sequence LDist. Dist[i] is the i-th element of the difference sequence. The elements in the increasing sequence GDist are then filtered, and the growth potential coefficient ATE is calculated. The calculation method for the growth potential coefficient ATE is as follows: ATE=log( )- ; Max(GDist) represents the maximum value of the elements in the increasing sequence GDist, Min(GDist) represents the minimum value of the elements in the increasing sequence GDist, and len(GDist) represents the length of the increasing sequence GDist. That is, to obtain the maximum value of the reduced sequence LDist, the That is, to obtain the minimum value of the reduced sequence LDist, the To obtain the arithmetic mean of the reduced sequence LDist, log() is the logarithmic function with base 2, and ln() is the natural logarithm. The potential coefficient ATE is used to indicate the stable state of the free-range chicken flock when the potential coefficient ATE approaches 0 within a period of free-range raising. If the potential coefficient ATE fluctuates, it indicates that there is a problem with the free-range chicken flock's raising status.

[0027] Further, in step S400, record For the earliest adjacent health status CM_Status difference in the difference sequence Dist obtained by the i-th collection node, By comparing with the minimum value in the reduced sequence LDist The ratio is obtained by comparison. / ,remember For the latest adjacent health status CM_Status difference in the difference sequence Dist obtained by the i-th acquisition node, the minimum value in the reduction sequence LDist is... Through with The ratio is obtained by comparison. The ratio / and The stable value STA was obtained through calculation. STA= , The To reduce the sum of the minimum, average, and maximum values ​​of the LDist sequence; By calculating the decreasing sequence, adjustments are made for fluctuations, and the stable value STA is then incorporated into the ATE calculation: Output the updated growth coefficient .

[0028] A health monitoring system for free-range chickens based on fecal detection, the system comprising: an acquisition module for acquiring flock temperature, flock fecal samples, and flock behavioral characteristics; The identification module identifies and analyzes fecal images to obtain the health index of free-range chickens. The processing module is used to process the monitoring data acquired by the acquisition module and output the monitoring results.

[0029] A non-volatile storage medium storing a computer program, wherein the device containing the non-volatile storage medium executes a method for monitoring the health of free-range chickens based on fecal detection as described in any one of the above methods by running the computer program.

[0030] A computer program product includes: a computer program, wherein when the computer program is executed by a processor, it implements a method for monitoring the health of free-range chickens based on fecal detection as described in any one of the above methods.

[0031] Preferably, the three-dimensional data containing the change values ​​change_SP, change_BT, and change_CM are mapped to a three-dimensional Cartesian coordinate system to obtain a three-dimensional discrete data point set; Time series, flow velocity series and temperature series are generated based on a three-dimensional discrete data point set. After standardizing the time series, flow velocity series and temperature series respectively, the DBSCAN clustering algorithm is used to cluster them and retain the valid data points of each series. The intersection of the valid data points of all series is re-aligned to obtain a three-dimensional ordered data point set. Multiple constraint equations are formed based on the condition that adjacent segmented data point sets satisfy multi-order continuity at corresponding connection points. A set of coefficient equations is formed based on multiple constraint equations, and the set of coefficient equations is solved to obtain the interpolation coefficients of each segmented data point set. A standardized three-dimensional curve function is obtained based on the interpolation coefficients of all segmented data points. The health status of the chicken flock is optimized by obtaining the three-dimensional curve function through three changes.

[0032] The computer program product may be a processor, which can be a Central Processing Unit (CPU), or 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, transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of the system, connecting various sub-regions of the system via various interfaces and lines.

[0033] The non-volatile storage medium can be used to store the computer program and / or modules. The processor implements various system functions by running or executing the computer program and / or modules stored in the memory, and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0034] Although the invention has been described in considerable detail and particularly with regard to several of the described embodiments, it is not intended to limit itself to any of these details or embodiments or any particular embodiment, thereby effectively covering the intended scope of the invention. Furthermore, the invention has been described above with respect to embodiments foreseeable by the inventors in order to provide a useful description, and non-substantial modifications to the invention that have not yet been foreseen may still represent equivalent modifications.

Claims

1. A method for monitoring the health of free-range chickens based on fecal detection, characterized in that, The method includes the following steps: S100: Acquire images of free-range chicken droppings through monitoring equipment; S200: Analyze the fecal images to obtain the health index of the free-range chicken flock; S300: Constructing a matrix of chicken feeding management and health index based on chicken manure obtained from the activity routes of free-range chickens; S400: A monitoring sequence is constructed by linking health index with changes in flock temperature; S500: A matrix is ​​constructed by monitoring data in the sequence, and the health monitoring results are obtained through matrix analysis.

2. The method for monitoring the health of free-range chickens based on fecal detection according to claim 1, characterized in that, In step S100, the location of feces during the free-range chickens' activities is obtained through monitoring, and feces image samples outside the activity range of the free-range chickens are periodically acquired by drone at close range. Monitoring data is obtained through feces images, and the type of feces is determined by comparing the color and consistency of the feces with known feces-disease relationships. The health of the chicken flock is monitored, and monitoring data is obtained through health monitoring. The monitoring data includes the flock's behavioral characteristics, flock temperature status, and the health status of the flock's feces.

3. The method for monitoring the health of free-range chickens based on fecal detection according to claim 1, characterized in that, In step S200, edge noise reduction processing is performed on some sample images of free-range chickens acquired by the UAV. Based on the shape, color, and surface features in the acquired sample images, the chicken feces are analyzed. The preliminary analysis distinguishes the chicken feces samples into normal chicken feces samples and abnormal chicken feces samples based on the surface features of the chicken feces. The sample ratio of normal chicken feces samples and abnormal chicken feces samples is recorded. By performing data preprocessing on the spectral data of the two types of feces samples using four methods, namely no processing, multivariate scattering correction, SG convolution smoothing, and standard deviation standardization, samples with small differences between the groups are obtained. The sample processing reduces the impact of abnormal feces samples on the assessment of chicken health.

4. The method for monitoring the health of free-range chickens based on fecal detection according to claim 3, characterized in that, Background clutter in the collected fecal images is treated as noise. The background of the fecal images is removed by noise reduction. The background in the fecal images has a large difference in color and texture from the feces. Therefore, this study uses the maximum inter-class variance method to achieve adaptive removal of the image background. Let the image after removing the background be defined as image P. For image P(x,y), let T be the threshold used to separate feces and background; and let be the proportion of foreground pixels to the total number of pixels in the image. The average gray level of the fecal image is denoted as . The proportion of background image pixels to the total number of pixels in the image is denoted as . The average gray level of the background image is denoted as . Assume the overall average gray level of the image pixels is ƞ, and its inter-class variance is denoted as g; if the size of the image P(x,y) is X×Y, and the threshold T of the number of pixels for all gray levels in the image is denoted as . The number of pixels whose grayscale values ​​are greater than a set threshold T is recorded as g= The adaptive segmentation was completed by calculating the inter-class variance g and the maximum threshold T using an iterative method. Image segmentation was performed using T as the segmentation threshold to obtain a binary image with the background removed. This binary image, processed by the OTSU algorithm, was then overlaid with the original image to obtain the image with the background removed. Texture features of feces in the image were obtained using the least squares method and the sparrow search algorithm. These texture features and color were used to perform a preliminary health assessment of the chickens, and a health assessment score was output.

5. The method for monitoring the health of free-range chickens based on fecal detection according to claim 1, characterized in that, In step S300, the obtained flock behavior characteristics, flock temperature status, and flock fecal health status are labeled as SP_Status, BT_Status, and CM_Status. The data are instantaneous values ​​obtained at the same time interval within a certain period. The flock behavior characteristics include the feeding, drinking, posture, agility, and gait of the entire free-range flock. The flock temperature status is the temperature change of the entire free-range flock in this period compared to the previous period. The flock fecal health status is the health assessment score obtained by judging the flock feces. The difference between adjacent data is calculated to obtain the change value of the flock health status at different times. Based on the magnitude of the change value observed at different times, the change values ​​are defined as change_SP, change_BT, and change_CM. By using the same acquisition cycle as the feces images of free-range chickens, the temperature of the flock when returning to the pen is collected by a temperature sensor to obtain the average temperature of the flock. The average temperature of the flock is recorded by the average temperature of the flock collected in the previous cycle, and chickens with abnormal temperatures are marked. An empty set temp with a time series is set. The collected temperature values ​​are added to the set temp in chronological order to obtain a set temp with a time series of temperature values, temp=[t1, t2, ..., tn], where n is the number of acquisition nodes in the cycle, and avg(temp) represents the average temperature. The temperature value ti is standardized with the flock temperature status BT_Status, and data at the same time are mapped. The ti is the i-th element of the set temp. The health status CM_Status of the chicken flock's feces is sorted by the time series of the collected data to form a health monitoring sequence MON. The differences between all adjacent CM_Status values ​​in the health monitoring sequence MON are calculated in chronological order, and the resulting differences are used to construct a difference sequence Dist. The difference sequence Dist is then divided according to the size of its elements: when Dist[i] ≥ 0, the element is classified as an increasing sequence GDist; when Dist[i] < 0, the element is classified as a decreasing sequence LDist. Dist[i] is the i-th element of the difference sequence. The elements in the increasing sequence GDist are then filtered, and the growth potential coefficient ATE is calculated. The calculation method for the growth potential coefficient ATE is as follows: ATE=log( )- ; Max(GDist) represents the maximum value of the elements in the increasing sequence GDist, Min(GDist) represents the minimum value of the elements in the increasing sequence GDist, and len(GDist) represents the length of the increasing sequence GDist. That is, to obtain the maximum value of the reduced sequence LDist, the That is, to obtain the minimum value of the reduced sequence LDist, the To obtain the arithmetic mean of the reduced sequence LDist, log() is the logarithmic function with base 2, and ln() is the natural logarithm. The potential coefficient ATE is used to indicate the stable state of the free-range chicken flock when the potential coefficient ATE approaches 0 within a period of free-range raising. If the potential coefficient ATE fluctuates, it indicates that there is a problem with the free-range chicken flock's raising status.

6. The method for monitoring the health of free-range chickens based on fecal detection according to claim 5, characterized in that, In step S400, record For the earliest adjacent health status CM_Status difference in the difference sequence Dist obtained by the i-th collection node, By comparing with the minimum value in the reduced sequence LDist The ratio is obtained by comparison. / ,remember For the latest adjacent health status CM_Status difference in the difference sequence Dist obtained by the i-th acquisition node, the minimum value in the reduction sequence LDist is... Through with The ratio is obtained by comparison. The ratio / and The stable value STA was obtained through calculation. STA= , The To reduce the sum of the minimum, average, and maximum values ​​of the LDist sequence; By calculating the decreasing sequence, adjustments are made for fluctuations, and the stable value STA is then incorporated into the ATE calculation: Output the updated growth coefficient .

7. A health monitoring system for free-range chickens based on fecal detection, characterized in that, The system includes: an acquisition module for acquiring chicken flock temperature, chicken flock feces samples, and chicken flock behavioral characteristics; The identification module identifies and analyzes fecal images to obtain the health index of free-range chickens. The processing module is used to process the monitoring data acquired by the acquisition module and output the monitoring results.

8. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores a computer program, wherein the device containing the non-volatile storage medium executes a method for monitoring the health of free-range chickens based on fecal detection as described in any one of claims 1 to 6 by running the computer program.

9. A computer program product, characterized in that, include: A computer program, wherein when executed by a processor, the computer program implements a method for monitoring the health of free-range chickens based on fecal detection as described in any one of claims 1 to 6.

Images