Animal welfare automatic monitoring method and monitoring system based on finite state machine model

Abstract

The application provides an animal welfare automatic monitoring method and monitoring system based on a finite state machine model, and the monitoring method comprises the following steps: constructing a finite state machine model; acquiring influence factor data of animal welfare evaluation; adopting the finite state machine model to perform state conversion on the influence factors of the welfare evaluation in a hierarchical manner; acquiring evaluation data of the animal welfare corresponding to the hierarchical state conversion; judging the grade of the evaluation data under the influence factor condition corresponding to the hierarchical state conversion; and replacing the influence factor data corresponding to the hierarchical state conversion to a grade that satisfies a predetermined grade standard in the case that the grade does not satisfy the predetermined grade standard. The influence factors of the evaluation animal welfare are converted in a hierarchical manner through the finite state machine model, the influence factors affecting the animal welfare are automatically adjusted, so that the evaluation data used for evaluating the animal welfare satisfies the predetermined grade standard, and the automatic adjustment level of the animal welfare is improved, so that the animal welfare is in a good state.

Description

Technical Field

[0001] This invention relates to the field of modern intelligent animal husbandry, and in particular to an automatic animal welfare monitoring method and system based on a finite state machine model. Background Technology

[0002] With the development of computer and network technology and the improvement of informatization, the issue of animal welfare in animal husbandry (referring to how animals adapt to their environment and meet their basic natural needs) has attracted more and more attention. Some scholars have studied the issues related to animal welfare in pastures to a certain extent, such as: [1] The relationship between environmental quality and animal welfare: Logistic regression was used to model the compliance of three animal-based checkpoints, including the random farm effect to explain repeated checks. The independent variables were divided into three categories: (1) main predictors; (2) prior confounding factors; (3) potential confounding factors; [2] He Jincheng et al. studied the influence of environmental temperature and humidity on the infrared thermal imaging temperature of dairy cows through field measurement, and the method of using local surface infrared thermal imaging temperature to replace the internal (rectal) temperature measurement. Using regression analysis algorithm, it was found that the infrared thermal imaging temperature of the eyes, nose, and epidermis of dairy cows was significantly correlated with the rectal temperature, indicating that the temperature and humidity of the external environment have an important impact on the body temperature of cattle; [3] Liao Zhisheng, in view of the current status and problems of the animal welfare legislative system, proposed legislative directions from five aspects: formulating a basic animal welfare law, improving the legal responsibility accountability mechanism, expanding the scope of animal welfare protection, enhancing the operability of legal provisions, and unifying the powers of animal protection law enforcement agencies, so as to promote the improvement and development of the corresponding animal welfare legislative system. However, in terms of the intelligent, automated, and digital regulation of animal welfare in ranches, there are still problems such as manual regulation as the main method, long time consumption, and low efficiency. Moreover, long working hours will also consume the energy of staff, resulting in the inability to adjust external influencing factors in a timely manner, leading to poor long-term animal welfare status, which in turn affects animal immunity, breeding level, parity, etc. Summary of the Invention

[0003] The purpose of this invention is to provide an automatic animal welfare monitoring method and system based on a finite state machine model, in order to solve the problem that animal welfare cannot be effectively automated in related technologies.

[0004] The embodiments of the present invention adopt the following technical solution: an automatic animal welfare monitoring method based on a finite state machine model, comprising:

[0005] Construct a finite state machine model;

[0006] Obtain data on factors influencing animal welfare assessment;

[0007] The finite state machine model is used to stratify the influencing factors of welfare evaluation and perform state transitions accordingly.

[0008] Obtain animal welfare evaluation data for the corresponding stratum, and determine the level of the evaluation data under the influencing factors of the corresponding stratum state transition; if the level does not meet the predetermined level standard, replace the influencing factor data for the corresponding stratum until the level meets the predetermined level standard.

[0009] In some embodiments, constructing the finite state machine model includes constructing a Millitype finite state machine model:

[0010] M = (S, I, O, F, G, S0)

[0011] Where S is a finite set of influencing factor data states; I = {l1, l2, ..., l...} n} represents a finite set of inputs; O represents the output data; F represents the state transition function; G represents the output function; and S0 represents the initial state of the influencing factor data.

[0012] In some embodiments, the finite state machine model is used to hierarchically transform the influencing factors of welfare evaluation into state transitions, including: using the finite state machine model to divide the influencing factors of welfare evaluation into at least feeding state transition control StateA and environmental state transition control StateB.

[0013] In some embodiments, the feeding state transition control StateA has the following finite state machine model:

[0014] M A =(S A I A O A F A G A S0)

[0015] Among them, S A S is a finite set of influencing factor data states based on State A. A ={S0, S1, ..., S i}, S0, S1, ..., S i At least different influencing factor data states should be constructed; I A ={a}, where a represents the first input evaluation data; O A For output data based on State A; F A For the state transition function based on State A, corresponding to S A The number of internal influencing factor data states is used to evaluate the influencing factor data states using F. A Iteration to determine the iterated StateB; G A This is the output function of the finite state machine model based on State A when it reaches the final state.

[0016] In some embodiments, the feeding state transition control StateA outputs at least a feeding control strategy based on the feeding state transition control StateA at the final state of its finite state machine model.

[0017] In some embodiments, the environmental state transition control StateB is at least configured as a feeding information level state, whose finite state machine model is as follows:

[0018] M B =(S B I B O B F B G B S0)

[0019] Among them, S B It is a finite set of influencing factor data states based on State B, i.e.

[0020] S B ={S (i+1)a S (i+1)b S (i+1)j}

[0021] Among them, S (i+1)a S (i+1)b S (i+1)j At least different influencing factor data states should be constructed; I B ={c}, where c represents the second evaluation data input, O B For output data based on State B; F B For the state transition function based on State B, corresponding to S B The number of internal influencing factor data states is used to evaluate the influencing factor data states using F. B Iteration to obtain the iterated StateA;G B This is the output function for the final state of the finite state machine model based on State B.

[0022] In some embodiments, the level corresponding to the second evaluation data is obtained based on the Yolov5 model.

[0023] In some embodiments, the environmental state transition control StateB outputs at least an environmental control strategy based on the environmental state transition control StateB at the final state of its finite state machine model.

[0024] In some embodiments, the M B =(S B I B O B F BG B In S0), S0 is F based on StateA. A The S determined in the last iteration during the iteration process i .

[0025] The present invention also provides a monitoring system that uses the above-mentioned automatic animal welfare monitoring method based on a finite state machine model to monitor animal welfare.

[0026] The beneficial effects of the embodiments of the present invention are as follows:

[0027] By using a finite state machine model to stratify and transform the influencing factors of animal welfare evaluation, the factors affecting animal welfare are automatically adjusted so that the evaluation data used to evaluate animal welfare meet the predetermined level standards, thereby improving the level of automated adjustment of animal welfare and ensuring that animal welfare is in a good state. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart of the automatic animal welfare monitoring method based on a finite state machine model according to the present invention.

[0030] Figure 2 This is a finite state machine model diagram of the present invention.

[0031] Figure 3 This is a schematic diagram of the control transition process of the automatic animal welfare monitoring method based on the finite state machine model of the present invention. Detailed Implementation

[0032] Various aspects and features of the present invention are described herein with reference to the accompanying drawings.

[0033] It should be understood that various modifications can be made to the embodiments described herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of the invention will be apparent to those skilled in the art.

[0034] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the invention and, together with the general description of the invention given above and the detailed description of the embodiments given below, serve to explain the principles of the invention.

[0035] These and other features of the invention will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0036] It should also be understood that although the invention has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of the invention, which have the features of the claims and are therefore all within the scope of protection defined herein.

[0037] The above and other aspects, features and advantages of the invention will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0038] Specific embodiments of the invention are described below with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of the invention, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the invention. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely to serve as the basis and representative basis for the claims to teach those skilled in the art to use the invention in various ways with substantially any suitable detailed structure.

[0039] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments of the present invention.

[0040] To address the problems in the background art, this invention provides an automatic animal welfare monitoring method based on a finite state machine model, such as... Figure 1 The flowchart of the present invention is shown.

[0041] The automatic animal welfare monitoring method based on a finite state machine model, according to embodiments of the present invention, mainly includes:

[0042] Construct a finite state machine model;

[0043] Obtain data on factors influencing animal welfare assessment;

[0044] The finite state machine model is used to stratify the influencing factors of welfare evaluation and perform state transitions accordingly.

[0045] Obtain animal welfare evaluation data for the corresponding stratum, and determine the level of the evaluation data under the influencing factors of the corresponding stratum state transition; if the level does not meet the predetermined level standard, replace the influencing factor data for the corresponding stratum until the level meets the predetermined level standard.

[0046] The main objective of this invention includes, but is not limited to, providing an automated analysis method for animal welfare regulation based on finite state machines. This method employs a hierarchical finite state machine model to stratify and transform the influencing factors of animal welfare evaluation in smart ranches, thereby achieving the effect of automated adjustment of animal welfare in ranches.

[0047] like Figure 2 and combined Figure 3 As shown, the automatic monitoring method includes: constructing a finite state machine model based on the principles of finite state machines. Especially when there are many states in the system, traditional finite state machines may have many state transitions, significantly increasing complexity and leading to problems such as poor maintainability, poor scalability, and poor reusability. This invention mainly applies the Millitype finite state machine model for animal welfare evaluation. A mathematical model of a finite state machine can be represented by a six-tuple, i.e.

[0048] M = (S, I, O, F, G, S0)

[0049] Where S is a finite set of influencing factor data states; I = {l1, l2, ..., l...} n} represents a finite set of inputs; O represents the output data; F represents the state transition function; G represents the output function; and S0 represents the initial state of the influencing factor data.

[0050] Obtain data on factors influencing animal welfare assessment. This data generally refers to external factors that affect animal welfare.

[0051] The finite state machine model is used to stratify the influencing factors of welfare evaluation and perform state transitions. State transitions include: using the finite state machine model, the influencing factors of welfare evaluation are divided into at least two state transition controls: feeding (State A) and environment (State B). Specifically, the corresponding state transition controls can be set according to the number of states in the system.

[0052] The following example uses a dairy cow as an illustration. Figure 2 As shown:

[0053] The feeding state transition control StateA, whose finite state machine model can be described as follows:

[0054] M A =(S A I A O A F A G A S0)

[0055] (1) Wherein, S A S is a finite set of influencing factor data states based on State A. A={S0, S1, ..., S i}, S0, S1, ..., S i At least different influencing factor data states should be constructed, S i This represents any intermediate state. Based on the pasture environment level, let's take i = 4 as an example. The specific states can be described as follows: S0 is feeding a fully mixed total ration three times a day, S1 is feeding small amounts frequently at fixed times and in fixed quantities, S2 is gradually changing the feed, S3 is the feed is clean, fresh, and odorless, and S4 is adding an appropriate amount of salt to the feed. S0 is the initial state.

[0056] (2)I A Let I be the input set of the finite state machine model of State A. A ={a}, where a represents the first input evaluation data. Let a represent the input video of a cow's ear. By tracking the dynamic keypoint model, the ear tracking trajectory of the cow is obtained, and thus the fluctuation of the cow's ear is obtained.

[0057] (3)O A This refers to the output data based on State A, specifically the output data at the end of each transition in State A.

[0058] (4)F A For the state transition function based on State A, corresponding to S A The number of internal influencing factor data states is used to evaluate the influencing factor data states using F. A Iteration is used to determine the iterated StateB;

[0059] State transition function F A Its iterative process is as follows:

[0060] F A (S0, a0) = S1, F A (S0, a1) = S2, F A (S0, s2) = State B

[0061] F A (S1, a0) = S2, F A (S1, a1) = S3, F A (S1, a2) = State B

[0062] F A (S2, a0) = S3, F A (S2, a1) = S4, F A (S2, a2) = State B

[0063] F A (S3, a0) = S4, F A(S3, a1) = S0, F A (S3, a2) = State B

[0064] F A (S4, a0) = S0, F A (S4, a1) = S1, F A (S4, a2) = State B

[0065] (5)G A This is the output function of the finite state machine model based on State A when it reaches its final state. This model only has an output in its final state, which represents the feeding control strategy of the pasture based on the feeding state transition control of State A.

[0066] The environmental state transition control StateB is at least constructed as an environmental information level state, and its finite state machine model can be described as follows:

[0067] M B =(S B I B O B F B G B S0)

[0068] Among them, S B It is a finite set of influencing factor data states based on State B, i.e.

[0069] S B ={S (i+1)d S (i+1)b S (i+1)j}

[0070] (1) Wherein, S (i+1)d S (i+1)b S (i+1)j At least different influencing factor data states should be constructed. Corresponding to i being 4 in the above feeding state transition control StateA, according to the pasture environment state transition control, the specific state can be described as follows: S 5a To enable the fan to turn on at set times, S 5b To enable the timed shower function, S 5c To enable the timed activation of the fill light function, S 5d To enable the ventilation function to be turned on at set times, S 5e To enable the heater to be turned on at a set time, S 5f This is to disable the function. The initial state S0 of StateB is the final state S4 in StateA.

[0071] (2)I B Let I be the input set of the inner finite state machine model of State B. B={c}, where c represents the second evaluation data input. Taking c as an example representing the input real-time fur color image feature signal, the fur color level is obtained after pre-training the improved Yolov5 model.

[0072] (3)O B This refers to the output data based on StateB, specifically the output data at the end of each transition in StateB.

[0073] (4)F B For the state transition function based on StateB, corresponding to S B The number of internal influencing factor data states is used to evaluate the influencing factor data states using F. B Iteration is used to obtain the iterated StateA. The state transition function F... B Its iterative process is as follows:

[0074] F B (S 5a b0)=S 5b F B (S 5a b1) = S 5c F B (S 5a b2) = S 5f

[0075] F B (S 5b b0)=S 5c F B (S 5b b1) = S 5d F B (S 5b b2) = S 5f

[0076] F B (S 5c b0)=S 5d F B (S 5c b0)=S 5e F B (S 5c b2) = S 5f

[0077] F B (S 5d b0)=S 5e F B (S 5d b1) = S 5a F B (S 5d b2) = S 5f

[0078] F B (S 5e b0) = StateA, F B (S 5e b1) = S 5b F B (S 5e b2) = S 5f

[0079] (5)G B This is the output function of the StateB-based finite state machine model when it reaches the final state. This model only has an output in the final state, and the output is based on the environment control strategy of StateB, which controls the environment state transition.

[0080] Table 1. Meaning of each state in StateA and StateB

[0081]

[0082] In the table, under feeding status transition control, S0 indicates feeding a fully mixed total diet three times a day; S1 indicates feeding small amounts frequently at fixed times and in fixed quantities; S2 indicates gradually changing the feed; S3 indicates the feed is clean, fresh, and odorless; S4 indicates adding an appropriate amount of salt to the feed. Under environmental status transition control, S... 5a Indicates a timed fan activation function; S 5b S indicates that the shower function will be turned on at a set time. 5c This indicates that the fill light function will be turned on at regular intervals; S 5d This indicates that the ventilation function will be turned on at set times; S 5e S indicates the timed heater activation function. 5f This represents the optimal environmental strategy.

[0083] Obtain animal welfare evaluation data for the corresponding strata and determine the level of the evaluation data under the influencing factors of the corresponding strata state transition. Based on the above description, the evaluation data for the feeding state transition control State A is the ear movement of cattle, which can be classified as slow ear movement, moderate ear movement, and flexible ear movement. When the ear movement is slow or moderate, the corresponding feeding control strategy is changed. The evaluation data for the environmental state transition control State B is the coat color of cattle, which can be classified as poor coat color, medium coat color, and excellent coat color. When the coat color is poor, another feeding strategy is adopted to obtain the optimal environmental strategy under the optimal environmental state of the corresponding environmental state transition control State B.

[0084] Table 2. Meaning of each input to StateA and StateB

[0085]

[0086] In the table, the input a0 for StateA layer represents slow ear undulation; a1 represents slow ear undulation; and a2 represents slow ear undulation. The input b0 for StateB layer represents poor coat color grade; b1 represents medium coat color grade; and b2 represents excellent coat color grade.

[0087] If the level does not meet the predetermined level standard, replace the data of the influencing factors in the corresponding layer until the level meets the predetermined level standard.

[0088] The present invention also provides a monitoring system that uses the above-mentioned automatic animal welfare monitoring method based on a finite state machine model to monitor animal welfare.

[0089] The foregoing has described in detail several embodiments of the present invention, but the present invention is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications based on the concept of the present invention, and all such variations and modifications should fall within the scope of protection claimed by the present invention.

Claims

1. An automatic animal welfare monitoring method based on a finite state machine model, characterized in that, include: Construct a finite state machine model; Obtain data on factors influencing animal welfare assessment; The finite state machine model is used to stratify the influencing factors of welfare evaluation and perform state transitions accordingly. Obtain animal welfare evaluation data for the corresponding stratum, and determine the level of the evaluation data under the influencing factors of the corresponding stratum state transition; if the level does not meet the predetermined level standard, replace the influencing factor data for the corresponding stratum until the level meets the predetermined level standard. The construction of the finite state machine model includes constructing a Millitype finite state machine model: in, S It is a finite set of influencing factor data states; For a finite set of inputs; For output data; F This is the state transition function. G This is the output function; S 0 represents the initial state of the influencing factor data; The finite state machine model is used to hierarchically transform the influencing factors of welfare evaluation into state transitions, including: using the finite state machine model to divide the influencing factors of welfare evaluation into at least feeding state transition control StateA and environmental state transition control StateB. The feeding state transition control StateA has the following finite state machine model: in, For a finite set of influencing factor data states based on State A, i.e. , At least different influencing factor data states should be constructed; , This represents the first evaluation data input; A This is the output data based on State A; For the state transition function based on State A, the corresponding The number of internal influencing factor data states is used to determine the influencing factor data states. Iteration is used to determine the iterated StateB; This is the output function of the finite state machine model based on StateA when it reaches the final state; The environmental state transition control StateB is at least constructed as a feeding information level state, and its finite state machine model is as follows: in, It is a finite set of influencing factor data states based on State B, i.e. in, At least different influencing factor data states should be constructed; , c This represents the second evaluation data input. B This is the output data based on State B; For the state transition function based on State B, corresponding to The number of internal influencing factor data states is used to determine the influencing factor data states. Iterate to obtain the iterated StateA; This is the output function for the final state of the finite state machine model based on State B.

2. The automatic animal welfare monitoring method based on a finite state machine model according to claim 1, characterized in that, The feeding state transition control StateA outputs at least a feeding control strategy based on the feeding state transition control StateA at the final state of its finite state machine model.

3. The automatic animal welfare monitoring method based on a finite state machine model according to claim 1, characterized in that, The level corresponding to the second evaluation data is obtained based on the Yolov5 model.

4. The automatic animal welfare monitoring method based on a finite state machine model according to claim 1, characterized in that, The environmental state transition control StateB outputs at least an environmental control strategy based on the environmental state transition control StateB at the final state of its finite state machine model.

5. The automatic animal welfare monitoring method based on a finite state machine model according to claim 1, characterized in that, The In For StateA-based The last iteration during the iteration process determines .

6. A monitoring system, characterized in that, Animal welfare monitoring is performed using the automatic animal welfare monitoring method based on the finite state machine model as described in any one of claims 1 to 5.

Images