Poultry state monitoring method and monitoring system based on motion trail

Abstract

The invention discloses a poultry state monitoring method and system based on an activity track, and the method comprises the steps: firstly obtaining reference receiving time, first receiving time and second receiving time, and calculating a movement track point set of a chicken according to the reference receiving time, the first receiving time and the second receiving time, intercepting a plurality of track point subsets from the moving track point set based on the time windows, performing feature extraction on each track point subset to obtain a feature parameter set of each time window, and generating a time-sharing health state monitoring result of each time window according to each feature parameter set. And finally, generating a health monitoring result according to each time-sharing monitoring result. Precise positioning of the chicken is achieved through the propagation distance difference of the electromagnetic waves, then monitoring of the health state is achieved through analysis of the track of the chicken, monitoring automation is achieved, meanwhile, basic data errors caused by factors such as chicken shielding can be avoided, and monitoring accuracy is effectively improved.

Description

Technical Field

[0001] This application relates to the field of data analysis technology, specifically to a method and system for monitoring the status of poultry based on activity trajectories. Background Technology

[0002] In large-scale poultry farms, it is often necessary to monitor the health status of poultry (e.g., chickens) in order to ensure their health. Current technology often relies on manual observation for monitoring. This method is highly dependent on the experience of the staff, is subjective, has limited accuracy, and cannot achieve 24-hour uninterrupted monitoring.

[0003] Some large-scale poultry farms use cameras to collect visual images of chickens, analyzing these images to assess their mental state and health. However, this method is not only technically demanding, but also susceptible to problems in the farm environment. Obstruction between chickens and the placement of cameras can easily affect the quality of the captured images, thus impacting the accuracy of the underlying data and resulting in low monitoring precision. Summary of the Invention

[0004] The main purpose of this application is to provide a method and system for monitoring the status of poultry based on activity trajectories, aiming to solve the shortcomings of low monitoring accuracy in the existing technology.

[0005] This application achieves the above objectives through the following technical solutions: A method for monitoring the status of poultry based on activity trajectories includes the following steps: Obtain the reference reception time t 0-i First reception time t 1-i Second receiving time t 2-i Where i represents the timestamp number; According to the reference reception time t 0-i First reception time t 1-i Second receiving time t 2-i The algorithm generates a set of movement trajectory points. Based on a time window, a subset of trajectory points is randomly selected from the set of moving trajectory points; Feature extraction is performed on each of the aforementioned trajectory point subsets to obtain a feature parameter set for each time window; the feature parameter set includes average motion speed, activity range parameter, and activity frequency parameter. Based on the respective feature parameter sets, time-division monitoring results are generated for each time window; Health monitoring results are generated based on the monitoring results at each time point.

[0006] Optionally, based on the reference reception time t 0-i First reception time t1-i Second receiving time t 2-i The process of generating a set of moving trajectory points includes the following steps: Based on the timestamp number, the reference reception time t is determined. 0-i First reception time t 1-i Second receiving time t 2-i The calculations are combined into several groups, where the expression for each group is (t). 0-i t 1-i t 2-i ), i represents the timestamp number; Obtain any calculation group and calculate the first time difference and the second time difference respectively; wherein the expression for calculating the first time difference is: The expression for calculating the second time difference is: ; The first distance difference is calculated based on the first time difference; wherein the expression for calculating the first distance difference is: c represents the propagation speed of electromagnetic waves; The second distance difference is calculated based on the second time difference; wherein, the expression for calculating the second distance difference is: c represents the propagation speed of electromagnetic waves; A hyperbolic positioning model is constructed based on the first distance difference and the second distance difference, and the real-time coordinates are obtained by solving the hyperbolic positioning model; wherein the expression of the hyperbolic positioning model is: (x, y) represent the coordinates of the chicken, and (x0, y0), (x1, y1) and (x2, y2) represent the position coordinates of the reference signal receiver, the first signal receiver and the second signal receiver, respectively. Repeat the steps of obtaining any calculation group and calculating the first time difference and the second time difference respectively to obtain the set of moving trajectory points.

[0007] Optionally, a subset of trajectory points may be randomly selected from the set of movement trajectory points based on a time window, including the following steps: Set the total duration of the time window T, the step size ΔT, and the number of time windows Q; Select a trajectory point as the starting point from the set of moving trajectory points, and then generate the first subset of trajectory points based on the total duration T of the time window and the time interval between adjacent trajectory points; Several subsets of trajectory points are generated sequentially based on the step size ΔT and the number of time windows Q.

[0008] Optionally, feature extraction is performed on each of the aforementioned trajectory point subsets to obtain a feature parameter set for each time window; the feature parameter set includes average motion speed, activity range parameters, and activity frequency parameters, and includes the following steps: Obtain a subset of trajectory points for each time window; Calculate the average velocity of each time window based on each of the aforementioned subsets of trajectory points; Based on each subset of trajectory points, a minimum bounding circle is generated for each time window, and the radius of each minimum bounding circle is taken as the activity range parameter. Where j represents the time window number; Set a displacement threshold, and calculate the activity frequency parameters for each time window based on the displacement threshold and each subset of trajectory points; Output feature parameter sets for each time window , ,..., , where j represents the time window number.

[0009] Optionally, the expression for calculating the average velocity is as follows: Where j represents the time window number, B j s represents the number of trajectory points within time window j. j and e j These represent the starting and ending trajectory points of time window j, respectively. The sampling time interval between two adjacent trajectory points is represented by k, which represents the trajectory point number. k y k ) represents the coordinates of the trajectory point numbered k.

[0010] Optionally, a displacement threshold is set, and the activity frequency parameters for each time window are calculated based on the displacement threshold and each subset of trajectory points, including the following steps: Set the displacement threshold D0; Based on each of the aforementioned subsets of trajectory points, calculate the actual displacement d between any two adjacent points. k-k+1 Generate the actual displacement set {d 1-2 d 2-3 ... d k-k+1} j Where k and k+1 represent the numbers of two adjacent trajectory points, and j represents the number of the time window; Based on the dwell interval screening formula, the number of dwell intervals A in the actual displacement set is screened and counted. j The expression for the retention interval screening formula is d. k-k+1 ≤D0; The activity frequency parameter is calculated according to the formula, wherein the expression for the activity frequency parameter is as follows: ,in This represents the sampling time interval between two adjacent trajectory points.

[0011] Optionally, time-division monitoring results are generated for each time window based on each of the aforementioned feature parameter sets, including the following steps: Obtain any set of feature parameters, perform normalization calculation on it, and obtain the feature normalization parameter set; The feature normalization parameter set is weighted to obtain a state-weighted score; wherein the calculation expression for the state-weighted score is as follows: and ; A state determination model is defined, and time-sharing health monitoring results are output based on the state determination model and the state weighted score; wherein the expression of the state determination model is: S0 represents the state determination threshold, which is a constant.

[0012] Repeat the steps of obtaining any set of feature parameters, performing normalization calculations on it, and obtaining the feature normalization parameter set to obtain the time-division monitoring results for each time window.

[0013] Optionally, the expression for calculating the feature normalization parameter set is as follows: ,in and Let r represent the maximum average speed and the minimum average speed, respectively. max and r min These represent the maximum and minimum activity radii, respectively. and These represent the maximum and minimum values ​​of the activity frequency parameter, respectively.

[0014] Optionally, health monitoring results are generated based on the monitoring results at each time slot, including the following steps: The monitoring results for each time period are sorted according to the timestamp, and a status determination sequence {P1, P2, ..., P} is generated. j}; Set the threshold N0 for continuous abnormal cycles and the threshold M0 for cumulative abnormal cycles; Count the maximum number of consecutive abnormal cycles N in the state determination sequence; Count the actual cumulative number of abnormal cycles M in the state determination sequence; If N≥N0 or / and M≥M0, the state is determined to be abnormal; otherwise, the state is determined to be normal.

[0015] Accordingly, this application also discloses a monitoring system based on the above-mentioned monitoring method, including... The data acquisition module is used to obtain the reference reception time t. 0-i First reception time t 1-i Second receiving time t 2-i Where i represents the timestamp number; The trajectory point calculation module is used to calculate the trajectory point based on the reference receiving time t. 0-i First reception time t 1-i Second receiving time t 2-i The algorithm generates a set of movement trajectory points. The decomposition module is used to randomly extract several subsets of trajectory points from the set of moving trajectory points based on a time window; The feature extraction module is used to extract features from each of the said trajectory point subsets to obtain the feature parameter set for each time window; the feature parameter set includes average motion speed, activity range parameter, and activity frequency parameter. The first calculation module is used to generate time-sharing monitoring results for each time window based on the respective feature parameter sets; The health assessment module is used to generate health monitoring results based on the monitoring results at each time period.

[0016] Compared with the prior art, this application has the following beneficial effects: This application first obtains the baseline reception time, the first reception time, and the second reception time, and calculates the set of moving trajectory points based on the baseline reception time, the first reception time, and the second reception time. Then, based on the time window, it extracts several subsets of trajectory points from the set of moving trajectory points. Subsequently, it extracts features from each subset of trajectory points to obtain the feature parameter set for each time window. Then, it generates the time-division monitoring results for each time window based on each feature parameter set. Finally, it generates the health monitoring results based on each time-division monitoring result.

[0017] Compared with the prior art, this application has the following beneficial effects: This application first obtains the baseline reception time, first reception time and second reception time of the chicken, and calculates the set of movement trajectory points of the chicken based on the baseline reception time, first reception time and second reception time. Then, based on the time window, it extracts several trajectory point subsets from the set of movement trajectory points, and then performs feature extraction on each trajectory point subset to obtain the feature parameter set of each time window. Then, it generates the time-division health status monitoring results of each time window based on each feature parameter set. Finally, it outputs the health monitoring results of the chicken based on each time-division health status monitoring result. Compared with existing technologies, this application achieves fully automatic calculation of trajectory points through fully automatic data acquisition and calculation, which not only improves monitoring efficiency, but also eliminates the subjectivity of human experience judgment by calculating all parameters through an objective model, thereby improving the objectivity of monitoring results and improving the accuracy of monitoring results. Secondly, this application uses a network of radio transmitters and multiple receivers. As the trajectory points change, the distance between the radio transmitter and each receiver will change. The distance difference can be calculated by the time difference of the received signals, and then multiple hyperbolic functions can be constructed. The intersection of each hyperbolic function is the real-time coordinate. The above technical solution not only has simple hardware equipment, which helps to reduce hardware costs, but also its positioning accuracy is not affected by the strength of the positioning signal, which can achieve high-precision positioning, providing a basis for subsequent accurate calculation of parameters and improving the accuracy of monitoring. Finally, the trajectory point acquisition method of this application can minimize the impact of occlusion and installation location, thereby improving monitoring accuracy; at the same time, the trajectory point acquisition described in this application is not affected by light, so it can achieve 24-hour monitoring, thus effectively ensuring the accuracy and continuity of the underlying data, and thus ensuring the accuracy and reliability of the final monitoring results. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a poultry status monitoring method based on activity trajectory provided in Embodiment 1 of this application; Figure 2 The diagram illustrates the principle of coordinate calculation. Figure 3 This is a schematic diagram of the structure of a monitoring system provided in Embodiment 2 of this application; The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection or an electrical connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Furthermore, if the embodiments of this invention involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0022] Implementation Method 1

[0023] Reference Figures 1 to 2 This embodiment, as an optional implementation of this application, discloses a method for monitoring the status of poultry based on activity trajectories, including the following steps: S1. Obtain the reference reception time t 0-i First reception time t 1-i Second receiving time t 2-i Where i represents the timestamp number; First, the signal transmission period is set, and then the reception time of each signal receiver is obtained, i.e., the reference reception time t. 0-i First reception time t 1-i Second receiving time t 2-i Where i represents the timestamp number; It should be noted that the technical solution described in this application assumes that all received data originates from the same monitored object. In actual operation, it is also necessary to classify the data according to the code of the monitored object; the code of the monitored object is transmitted through a signal transmitting device. S2, Based on the reference reception time t 0-i First reception time t 1-i Second receiving time t 2-i The algorithm generates a set of movement trajectory points. S21. Based on the timestamp number, the reference receiving time t of each of the above-mentioned reference receiving timestamps is... 0-i First reception time t 1-i Second receiving time t 2-i The calculations are combined into several groups, where the expression for each group is (t). 0-i t 1-i t 2-i ), i represents the timestamp number; To facilitate the later statistics and organization of data, the signal emitted by the wireless signal transmitting device includes not only the code of the monitored object, but also the signal transmission time information. For signals with the same signal transmission time information, their reception time is assigned the same timestamp number. Subsequently, signals with the same timestamp number are grouped into the same calculation group; as the reception time increases, a series of calculation groups will be generated, and the expression for each calculation group is (t... 0-i t 1-i t 2-i ), i represents the timestamp number; S22. Obtain any calculation group and calculate the first time difference and the second time difference respectively; wherein the calculation expression for the first time difference is: The expression for calculating the second time difference is: ; Obtain any calculation group, and calculate the first time difference and the second time difference according to the parameters of the calculation group. The expression for calculating the first time difference is as follows: The expression for calculating the second time difference is: ; It should be noted that if four signal receivers are set up, a third time difference also needs to be calculated, and the calculation formula is exactly the same as the formula above. S23. Calculate the first distance difference based on the first time difference; wherein the calculation expression for the first distance difference is: c represents the propagation speed of electromagnetic waves; The first distance difference is calculated based on the first time difference obtained in step S22. The expression for the calculation of the first distance difference is as follows: c represents the propagation speed of electromagnetic waves; It should be noted that the propagation speed of electromagnetic waves is taken as their propagation speed in air, which is a constant. S24. Calculate the second distance difference based on the second time difference; wherein, the calculation expression for the second distance difference is: c represents the propagation speed of electromagnetic waves; S25. Construct a hyperbolic positioning model based on the first distance difference and the second distance difference, and solve the hyperbolic positioning model to obtain real-time coordinates; wherein the expression of the hyperbolic positioning model is: (x, y) represent the coordinates of the chicken, and (x0, y0), (x1, y1) and (x2, y2) represent the position coordinates of the reference signal receiver, the first signal receiver and the second signal receiver, respectively. First, a two-dimensional coordinate system is established as the standard coordinate system, and the coordinates of each signal receiver are calibrated in the standard coordinate system. At a fixed moment, the position of the chicken being monitored is fixed, so the distance between it and each signal receiver is fixed. For example, if the distance between the chicken, i.e. the monitored object, and the reference signal receiver and the first signal receiver is fixed, then the distance difference is a constant value, which conforms to the definition of a hyperbola. Therefore, two hyperbolic functions can be constructed separately through the distance difference, and the corresponding coordinate points can be obtained by solving the hyperbolic functions together. S26. Repeat the steps of obtaining any calculation group and calculating the first time difference and the second time difference respectively to obtain the set of moving trajectory points; Repeating steps S22 to S25 will calculate all trajectory points. Marking each parameter according to the timestamp will yield the set of moving trajectory points, specifically expressed as {(x, y)1, (x, y)2, (x, y)3, ..., (x, y)}. i ...}; S3. Randomly select a subset of trajectory points from the set of moving trajectory points based on a time window; S31. Set the total duration of the time window T, the movement step size ΔT, and the number of time windows Q; First, set the total duration of the time window T, the movement step size ΔT, and the number of time windows Q. It should be noted that the movement step size ΔT represents the time interval between two adjacent time windows. The above time interval can be the time interval between the starting trajectory point, the time interval between the ending trajectory point, or the time interval between the starting trajectory point and the ending trajectory point. S32. Select a trajectory point as the starting point from the set of moving trajectory points, and then generate the first subset of trajectory points according to the total duration T of the time window and the time interval between adjacent trajectory points. Then, a trajectory point is randomly selected from the set of moving trajectory points as the starting trajectory point of the first trajectory point subset. Then, each trajectory point within the corresponding time period is extracted according to the total duration T of the time window, thereby generating the first trajectory point subset. It should be noted that the time interval between adjacent trajectory points is actually the signal transmission cycle, that is, the time interval between two adjacent signal transmissions, which is a fixed value set by the device. S33. Generate several subsets of trajectory points sequentially based on the step size ΔT and the number of time windows Q.

[0024] Once the first subset of trajectory points is selected, the starting trajectory point of the second subset of trajectory points can be found by moving along the time axis according to the step size ΔT. Then, the second subset of trajectory points can be generated by combining the total duration of the time window. Repeat the above steps to obtain a specified number of trajectory point subsets; Since the monitored object is in continuous motion, the set of movement trajectory points of the monitored object will continue to expand and become numerous, which is not conducive to later calculation and evaluation. By randomly selecting a subset of trajectory points, not only can the above-mentioned technical problem of high computational difficulty be solved, but also the physiological state of the monitored object at different times can be analyzed in a targeted manner by dividing the time into time periods, thereby understanding the physiological state of the monitored object at different time periods. Secondly, by analyzing physiological states at different times, it is possible to effectively determine the changing trends of the physiological states of the monitored subjects, thereby allowing for early intervention and ensuring the effectiveness of disease prevention for the monitored subjects.

[0025] S4. Extract features from each of the said trajectory point subsets to obtain the feature parameter set for each time window; the feature parameter set includes average motion speed, activity range parameter and activity frequency parameter. S41. Obtain a subset of trajectory points for each time window; S42. Calculate the average velocity of each time window based on each of the aforementioned subsets of trajectory points; To obtain a subset of trajectory points, the specific expression can be listed as {s j ... e j Then, the distance between any two adjacent trajectory points is calculated. It should be noted that in large-scale farms, the ground is relatively flat, and the coordinates of the monitored object in the height direction can be ignored when it moves on the ground. Therefore, only its planar coordinates need to be calculated. Then, the distance between two coordinate points can be quickly calculated using Euclidean geometry, thus obtaining a series of distance values. Secondly, since the time interval between two trajectory points is fixed, the speed between any two adjacent trajectory points can be calculated by combining the above distance values. Finally, the average speed of each subset of trajectory points can be calculated. The formula for calculating the average velocity is as follows: Where j represents the time window number, B j s represents the number of trajectory points within time window j. j and e j These represent the starting and ending trajectory points of time window j, respectively. The sampling time interval between two adjacent trajectory points is represented by k, which represents the trajectory point number. k y k () represents the coordinates of the trajectory point numbered k; S43. Based on each subset of trajectory points, generate a minimum bounding circle for each time window, and take the radius of each minimum bounding circle as the activity range parameter. Where j represents the time window number; Welzl algorithm is used to process each subset of trajectory points to generate minimum bounding circles for each time window and obtain the radius of each minimum bounding circle. Then, the radius of each smallest enclosing circle is taken as the activity range parameter. Where j represents the time window number; S44. Set a displacement threshold, and calculate the activity frequency parameters for each time window based on the displacement threshold and each subset of trajectory points. S441. Set the displacement threshold D0; The displacement threshold refers to the minimum distance a health-monitored subject can move within a fixed time period (the time interval between two adjacent trajectory points); the above parameters can be flexibly set according to the different species and ages of the monitored subjects. S442. Based on each of the aforementioned subsets of trajectory points, calculate the actual displacement d between any two adjacent points. k-k+1 Generate the actual displacement set {d 1-2 d 2-3 ... d k-k+1} j Where k and k+1 represent the numbers of two adjacent trajectory points, and j represents the number of the time window; Given any subset of trajectory points, calculate the actual displacement between each pair of adjacent trajectory points using the Euclidean distance formula. Then, aggregate all the actual displacements to obtain the actual displacement set {d}. 1-2 d 2-3 ... d k-k+1} j Where k and k+1 represent the numbers of two adjacent trajectory points, and j represents the number of the time window; S443. Based on the dwell interval screening formula, filter and count the number of dwell intervals A within the actual displacement set. j The expression for the retention interval screening formula is d. k-k+1 ≤D0; When the actual displacement is lower than the set displacement threshold D0, the activity level of the chickens is considered to be lower than that of healthy chickens. Therefore, the monitored object is considered to be in a stagnant state within a certain area. The expression for the stagnation interval screening formula is d. k-k+1 ≤D0; By checking each actual displacement using the above-mentioned dwell zone screening formula, it can be determined whether it is a dwell zone. Finally, the total number of dwell zones can be counted to obtain the number of dwell zones A. j ; S444. Calculate the activity frequency parameter according to the calculation formula, wherein the calculation expression for the activity frequency parameter is as follows: ,in This represents the sampling time interval between two adjacent trajectory points.

[0026] By statistically analyzing the cumulative duration of each dwelling area and the ratio of that cumulative duration to the total duration of the time window, the proportion of the monitored object in a dwelling state within a time window can be obtained. This proportion is the activity frequency parameter. Therefore, the calculation expression for the activity frequency parameter is as follows: ,in This represents the sampling time interval between two adjacent trajectory points; The monitored object in a normal state is always in a state of irregular movement. Its stay time in a certain location is limited. As reflected in the activity frequency parameter, the lower the value, the higher the activity frequency parameter is, and the closer it is to 1, the more serious the disease condition of the monitored object is. S45, Output feature parameter sets for each time window. , ,..., , where j represents the time window number.

[0027] The parameters of the same time window are collected into a unified set, thereby outputting a set of feature parameters for each time window; S5. Generate time-division monitoring results for each time window based on the set of characteristic parameters described above; S51. Obtain any set of feature parameters, perform normalization calculation on it, and obtain the set of feature normalization parameters. The calculation expression for the feature normalization parameter set is as follows: ,in and r represents the maximum and minimum average velocities, respectively. max and r min These represent the maximum and minimum activity radii, respectively. and These represent the maximum and minimum values ​​of the activity frequency parameter, respectively.

[0028] The normalization calculation described above can unify parameters with different dimensions, clearing obstacles for subsequent calculations. On the other hand, it can map all parameters to the range of 0-1, thus more vividly depicting the meaning of each parameter. S52. Perform a weighted calculation on the feature normalization parameter set to obtain a state-weighted score; wherein the calculation expression for the state-weighted score is: and ω1 is the weighting parameter for average motion speed, ω2 is the weighting parameter for range of motion, and ω1 is the weighting parameter for activity frequency. It should be noted that the above weight parameters can be flexibly set according to preferences.

[0029] S53. Set a state determination model, and output time-sharing health monitoring results based on the state determination model and the state weighted score; wherein the expression of the state determination model is: S0 represents the state determination threshold, which is a constant.

[0030] The state-weighted score calculated in step S52 is input into the state determination model to obtain the time-division monitoring result corresponding to the time window. S54. Repeat the steps of obtaining any set of feature parameters, performing normalization calculations on it, and obtaining the feature normalization parameter set, and obtain the time-division monitoring results for each time window respectively.

[0031] S6. Generate health monitoring results based on the monitoring results at each time period.

[0032] The monitoring results for each time period are sorted according to the timestamp, and a status determination sequence {P1, P2, ..., P} is generated. j}; The health status monitoring results for each time period are sorted according to the timestamp of each time window, thus obtaining the status determination sequence {P1, P2, ..., P...}. j}; Specifically, the state determination sequence can be expressed as {healthy, sick, healthy, healthy, sick, sick, healthy, healthy, healthy, sick, healthy}; S62. Set the continuous abnormal cycle judgment threshold N0 and the cumulative abnormal cycle judgment threshold M0. For example, set the threshold for continuous abnormal cycles to N0=4, and the threshold for cumulative abnormal cycles to M0=6. S63. Count the maximum number of consecutive abnormal cycles N in the state determination sequence; The actual number of consecutive abnormal maximum cycles N in the state determination sequence is counted. For example, according to the example in step S61, the actual number of consecutive abnormal maximum cycles N=2. S64. Count the actual cumulative number of abnormal cycles M in the state determination sequence; According to the example in step S61, the actual cumulative number of abnormal cycles M = 4. S65. If N≥N0 or / and M≥M0, the state is determined to be abnormal; otherwise, the state is determined to be normal.

[0033] The above parameters are compared. If N≥N0 or / and M≥M0, the state is determined to be abnormal; otherwise, the state is determined to be normal. (This is in conjunction with the previous...) If none of the distance parameters satisfy the above formula, the monitored object is determined to be in a healthy state.

[0034] By comparing the results of health monitoring at different times, the influence of random factors on the monitoring results can be effectively eliminated, thereby improving the accuracy of the monitoring. Secondly, the state determination sequence can also effectively monitor the health change trend of the monitored object, so as to detect it in the early stage when the health of the monitored object begins to deteriorate and take corresponding measures, which is conducive to reducing the incidence of the monitored object.

[0035] Accordingly, this application also discloses a monitoring device based on the above monitoring method, including a radio signal transmitting device, several radio signal receiving devices and a control computer; The radio signal transmitting device described herein adopts a mature radio signal transmitting module in the prior art and is integrated into a module, such as being integrated into the ankle bracelet of the monitored object; preferably, the radio signal transmitting module is integrated into a module that can be fitted onto the neck of a chicken. It should be noted that the signals emitted by the radio signal transmitting device contain unique coded information of the monitored object; The radio signal receiving device shall be provided with at least three sets, namely a reference signal receiver, a first signal receiver, and a second signal receiver. The above receivers have the same structure and their names can be interchanged. They can be specified in advance when in use. It should be noted that the spacing between each signal receiver should be as large as possible to improve the accuracy of data calculation. Preferably, they are placed at the four corners of the farm. The number of signal receivers can also be set to more than four to improve positioning accuracy through calculation of multiple equations. To improve positioning accuracy, the radio signal used in this application is preferably an electromagnetic wave in the 2.4 GHz frequency band; It should be noted that the above trajectory monitoring method can be used for chicken farming, that is, for large-scale free-range chicken farms: First, devices such as leg bands are put on the chickens. A signal transmitter is installed inside the leg band, and a reference signal receiver and other signal receivers are placed at the four corners of the farm. As the chickens move, the signal transmitter continuously emits signals, which are then received by various signal receivers. Simultaneously, based on the order of reception, a reference reception time t is generated sequentially. 0-i First reception time t 1-i Second receiving time t 2-i Where i represents the timestamp number; Combining the aforementioned methods, using the reference reception time t 0-i First reception time t 1-i Second receiving time t 2-i The movement trajectory of the chickens can be calculated, and the health status of the chickens can be judged by the above monitoring methods, while ensuring the accuracy of the judgment; It should be noted that in large-scale farms, the ground is relatively flat, and the coordinates of chickens in the height direction can be ignored when they are moving on the ground. Compared with existing technologies, this application realizes fully automated monitoring of chicken health status through fully automated data collection and calculation. Not only is the monitoring efficiency higher, but also all parameters are calculated through an objective model, thereby eliminating the subjectivity of human experience judgment, improving the objectivity of monitoring results, and helping to improve the accuracy of monitoring results. Secondly, this application involves attaching a radio signal transmitter to the chicken's body and setting up at least three receivers within the chicken farm. As the chicken moves, the distance between the radio signal transmitter and each receiver changes. The distance difference can be calculated using the time difference of the received signals, thereby constructing multiple hyperbolic functions. The intersection of these hyperbolic functions represents the chicken's real-time coordinates. Compared to traditional three-dimensional coordinate positioning modules, this technical solution not only has simpler hardware, reducing hardware costs, but also offers better positioning accuracy within the limited space of a chicken farm. It is unaffected by the strength of the positioning signal, enabling high-precision positioning of the chicken and providing a foundation for accurate subsequent calculations of parameters, thus improving monitoring accuracy. Finally, this application uses the trajectory points of chickens to quantitatively analyze the behavior of chickens. Compared with normal chickens, sick chickens have reduced movement speed due to the decline of physiological functions. At the same time, due to the reduced desire to move, the distance between each trajectory point will be reduced, and they will be highly concentrated in a certain area. The time spent in a certain area will also be significantly prolonged. Therefore, through the above comprehensive analysis, sick chickens can be quickly and accurately screened from the flock, thereby achieving accurate monitoring of the health status of chickens.

[0036] Meanwhile, the trajectory point acquisition described in this application is not affected by light, so it can achieve 24-hour monitoring, and it is not affected by chickens blocking the view. Therefore, it can effectively ensure the accuracy of the underlying data, thereby ensuring the accuracy of the final monitoring results.

[0037] Implementation Method 2

[0038] This embodiment, as an optional embodiment of this application, discloses a monitoring system, including a data acquisition module and a trajectory point calculation module, wherein the output end of the data acquisition module is communicatively connected to the input end of the trajectory point calculation module. The output of the trajectory point calculation module is sequentially connected to the decomposition module and the feature extraction module; the output of the feature extraction module is sequentially connected to the first calculation module and the health determination module.

[0039] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for monitoring the status of poultry based on activity trajectories, characterized in that, Includes the following steps: acquiring a reference reception time t 0-i , a first reception time t 1-i , and a second reception time t 2-i ; wherein i denotes a timestamp number; According to the reference receiving time t 0-i , the first receiving time t 1-i , and the second receiving time t 2-i , the mobile trajectory point set is calculated and generated. Based on a time window, a subset of trajectory points is randomly selected from the set of moving trajectory points; Feature extraction is performed on each of the aforementioned trajectory point subsets to obtain the feature parameter set for each time window; The set of feature parameters includes average movement speed, range of motion parameters, and activity frequency parameters; Based on the respective feature parameter sets, time-division monitoring results are generated for each time window; Health monitoring results are generated based on the monitoring results at each time point.

2. The poultry status monitoring method based on activity trajectory according to claim 1, characterized in that, The reference receiving time t 0-i , the first receiving time t 1-i , and the second receiving time t 2-i The solving and generating mobile trajectory point set includes the following steps: Based on the timestamp number, the reference reception time t is determined. 0-i First reception time t 1-i Second receiving time t 2-i The calculations are combined into several groups, where the expression for each group is (t). 0-i t 1-i t 2-i ), i represents the timestamp number; Obtain any calculation group and calculate the first time difference and the second time difference respectively; wherein the expression for calculating the first time difference is: The expression for calculating the second time difference is: ; The first distance difference is calculated based on the first time difference; wherein the expression for calculating the first distance difference is: c represents the propagation speed of electromagnetic waves; The second distance difference is calculated based on the second time difference; wherein, the expression for calculating the second distance difference is: c represents the propagation speed of electromagnetic waves; A hyperbolic positioning model is constructed based on the first distance difference and the second distance difference, and the real-time coordinates are obtained by solving the hyperbolic positioning model; wherein the expression of the hyperbolic positioning model is: (x, y) represent the coordinates of the chicken, and (x0, y0), (x1, y1) and (x2, y2) represent the position coordinates of the reference signal receiver, the first signal receiver and the second signal receiver, respectively. Repeat the steps of obtaining any calculation group and calculating the first time difference and the second time difference respectively to obtain the set of movement trajectory points of the chickens.

3. The poultry status monitoring method based on activity trajectory according to claim 1, characterized in that, The step of randomly selecting a subset of trajectory points from the set of movement trajectory points based on a time window includes the following steps: Set the total duration of the time window T, the step size ΔT, and the number of time windows Q; Select a trajectory point as the starting point from the set of moving trajectory points, and then generate the first subset of trajectory points based on the total duration T of the time window and the time interval between adjacent trajectory points; Several subsets of trajectory points are generated sequentially based on the step size ΔT and the number of time windows Q.

4. The poultry status monitoring method based on activity trajectory according to claim 1, characterized in that, The feature extraction is performed on each of the trajectory point subsets to obtain the feature parameter set for each time window; The feature parameter set includes average movement speed, range of motion parameters, and activity frequency parameters, and includes the following steps: Obtain a subset of trajectory points for each time window; Calculate the average velocity of each time window based on each of the aforementioned subsets of trajectory points; Based on each subset of trajectory points, a minimum bounding circle is generated for each time window, and the radius of each minimum bounding circle is taken as the activity range parameter. Where j represents the time window number; Set a displacement threshold, and calculate the activity frequency parameters for each time window based on the displacement threshold and each subset of trajectory points; Output feature parameter sets for each time window , ,..., , where j represents the time window number.

5. The poultry status monitoring method based on activity trajectory according to claim 4, characterized in that, The formula for calculating the average velocity is as follows: Where j represents the time window number, B j s represents the number of trajectory points within time window j. j and e j These represent the starting and ending trajectory points of time window j, respectively. The sampling time interval between two adjacent trajectory points is represented by k, which represents the trajectory point number. k y k ) represents the coordinates of the trajectory point numbered k.

6. The poultry status monitoring method based on activity trajectory according to claim 4, characterized in that, The process of setting a displacement threshold and calculating the activity frequency parameters for each time window based on the displacement threshold and each subset of trajectory points includes the following steps: Set the displacement threshold D0; Based on each of the aforementioned subsets of trajectory points, calculate the actual displacement d between any two adjacent points. k-k+1 Generate the actual displacement set {d 1-2 d 2-3 ... d k-k+1 } j Where k and k+1 represent the numbers of two adjacent trajectory points, and j represents the number of the time window; Based on the dwell interval screening formula, the number of dwell intervals A in the actual displacement set is screened and counted. j The expression for the retention interval screening formula is d. k-k+1 ≤D0; The activity frequency parameter is calculated according to the formula, wherein the expression for the activity frequency parameter is as follows: ,in This represents the sampling time interval between two adjacent trajectory points.

7. The poultry status monitoring method based on activity trajectory according to claim 1, characterized in that, The step of generating time-division health status monitoring results for each time window based on each of the aforementioned feature parameter sets includes the following steps: Obtain any set of feature parameters, perform normalization calculation on it, and obtain the feature normalization parameter set; The feature normalization parameter set is weighted to obtain a state-weighted score; wherein the calculation expression for the state-weighted score is as follows: and ; A state determination model is defined, and time-sharing health monitoring results are output based on the state determination model and the state weighted score; wherein the expression of the state determination model is: S0 represents the state determination threshold, which is a constant; Repeat the steps of obtaining any set of feature parameters, performing normalization calculations on it, and obtaining the feature normalization parameter set to obtain the time-division monitoring results for each time window.

8. The poultry status monitoring method based on activity trajectory according to claim 7, characterized in that, The calculation expression for the feature normalization parameter set is as follows: ,in and Let r represent the maximum average speed and the minimum average speed, respectively. max and r min These represent the maximum and minimum activity radii, respectively. and These represent the maximum and minimum values ​​of the activity frequency parameter, respectively.

9. The poultry status monitoring method based on activity trajectory according to claim 1, characterized in that, The process of generating health monitoring results based on the monitoring results at different times includes the following steps: The monitoring results for each time period are sorted according to the timestamp, and a status determination sequence {P1, P2, ..., P} is generated. j }; Set the threshold N0 for continuous abnormal cycles and the threshold M0 for cumulative abnormal cycles; The maximum number of consecutive abnormal cycles, N, in the statistical state determination sequence; Count the actual cumulative number of abnormal cycles M in the state determination sequence; If N≥N0 or / and M≥M0, the state is determined to be abnormal; otherwise, the state is determined to be normal.

10. A monitoring system for the poultry status monitoring method based on activity trajectory as described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to obtain the reference reception time t. 0-i First reception time t 1-i Second receiving time t 2-i ; Where i represents the timestamp number; The trajectory point calculation module is used to calculate the trajectory point based on the reference receiving time t. 0-i First reception time t 1-i Second receiving time t 2-i The algorithm generates a set of movement trajectory points. The decomposition module is used to randomly extract several subsets of trajectory points from the set of moving trajectory points based on a time window; The feature extraction module is used to extract features from each of the said trajectory point subsets to obtain the feature parameter set for each time window; The set of feature parameters includes average movement speed, range of motion parameters, and activity frequency parameters; The first calculation module is used to generate time-sharing monitoring results for each time window based on the respective feature parameter sets; The health assessment module is used to generate health monitoring results based on the monitoring results at each time period.

Images