Intelligent livestock and poultry breeding management system based on Internet of Things technology

Abstract

The invention relates to the technical field of breeding management, in particular to an intelligent livestock and poultry breeding management system based on the Internet of Things technology. The method comprises the following steps: acquiring a multi-dimensional feature monitoring data set of target detection livestock, analyzing the multi-dimensional feature monitoring data set, and determining direct feature information and indirect feature information; performing coupling analysis on the direct feature information and the indirect feature information, and determining a hoof disease pathological feature data set of the target detection livestock; and based on the direct feature information of each target detection livestock, tracing the hoof disease pathological expressions of all the target detection livestock according to the corresponding hoof disease pathological feature data set, and determining and outputting a hoof disease inspection report. According to the method, accurate hoof disease identification and tracing in the early stage of livestock suffering from hoof diseases are realized, sustainable development and diffusion of the livestock hoof diseases in breeding populations are effectively prevented, the automatic management efficiency of breeding management is improved, and the economic effect of user breeding is improved.

Description

Technical Field

[0001] The present application relates to the field of breeding management technology, and in particular to an intelligent livestock and poultry breeding management system based on Internet of Things technology. Background Art

[0002] As a product of the deep integration of modern agricultural technology and information technology, intelligent livestock and poultry breeding management technology is gradually changing the traditional breeding model. While improving the efficiency of livestock and poultry breeding management, intelligent livestock and poultry breeding management technology can effectively reduce dependence on manpower.

[0003] However, existing intelligent livestock and poultry breeding management technologies make it difficult to accurately identify and trace hoof diseases in livestock in the early stages of the disease, resulting in the continued development and spread of livestock hoof diseases in breeding populations, causing a decline in breeding production capacity and a negative impact on the economic effects of the user's breeding industry. Summary of the Invention

[0004] This application provides an intelligent livestock and poultry breeding management system based on Internet of Things technology to solve the above technical problems.

[0005] In a first aspect, the present application provides an intelligent livestock and poultry breeding management system based on Internet of Things technology, the system comprising: Acquiring a multidimensional feature monitoring data set of target detection livestock, analyzing the multidimensional feature monitoring data set to determine direct feature information and indirect feature information; performing coupled analysis on the direct feature information and the indirect feature information to determine a hoof disease pathology feature dataset of the target livestock; Based on the direct feature information of each target livestock, and according to the corresponding hoof disease pathology feature data set, the hoof disease pathology manifestations of all target livestock are traced, and a hoof disease inspection report is determined and output.

[0006] Through this solution, the multidimensional feature monitoring data set is analyzed to obtain direct feature information and indirect feature information for mapping livestock hoof diseases. On this basis, the direct feature information and indirect feature information are coupled and analyzed, and the hoof disease status of livestock is comprehensively judged to form a hoof disease pathology feature data set. Then, the hoof disease pathological manifestations of all the target livestock are traced, and a hoof disease inspection report containing the livestock disease status and investigation direction suggestions is generated. The hoof disease inspection report is provided to the corresponding breeding personnel, so that accurate hoof disease identification and tracing can be achieved in the early stage of livestock hoof disease, effectively preventing the continued development and spread of livestock hoof disease in the breeding population, improving the efficiency of automated breeding management, and improving the economic benefits of user breeding.

[0007] Optionally, the multi-dimensional feature monitoring data set includes livestock physiological habit data, livestock normal posture data, hoof state sensor data and livestock dynamic gait data; The livestock physiological habit data and the livestock normal posture data are obtained through a plurality of monitoring cameras arranged inside the breeding house; The hoof state sensing data and the livestock dynamic gait data are obtained through a number of Internet of Things sensors arranged in the livestock driving channel.

[0008] Through this solution, we characterize the multi-dimensional status of livestock from both static and dynamic dimensions, using livestock physiological habit data and normal livestock posture data collected from breeding houses, combined with hoof condition sensor data and livestock dynamic gait data collected from livestock driving lanes, to form a multi-dimensional feature monitoring data set, providing comprehensive scientific data for subsequent analysis of livestock hoof pathology characteristics.

[0009] Optionally, the livestock driving passage is a livestock passage for connecting the breeding house with the livestock activity area; The side walls of the livestock passage are provided with a movable gradually narrowing structure. The width of the passage decreases gradually from the entrance to the exit based on the average shoulder width of the livestock. Several IoT pressure-sensing fences are arranged on the left and right sides of the passage. The outlet end is provided with a slope transition section, the surface of which is coated with a high friction coefficient coating, and a number of Internet of Things three-dimensional motion capture sensors are arranged on the left and right sides of the slope transition section; The livestock dynamic gait data is constructed based on the data collected by the Internet of Things pressure sensing fence and the Internet of Things three-dimensional motion capture sensor.

[0010] Through this solution, the livestock driving channel is set as an overall progressively narrowing structure, and a slope transition section is set at the exit end of the livestock driving channel. While capturing the body avoidance tilting posture caused by hoof disease of livestock, the abnormal hoof lifting movement and abnormal hoof landing movement caused by hoof disease are more significantly exposed to the livestock, thereby improving the comprehensiveness and scientific nature of the livestock dynamic gait data, making the early hoof lesion characteristics more prominent, and thus improving the accuracy of subsequent judgment of the livestock hoof disease status.

[0011] Optionally, the ground of the livestock driving passage is configured as hard friction areas and soft buffer areas that are alternately distributed; The surface of the hard friction zone is provided with a hard raised texture array, and the surface of the soft buffer zone is provided with an elastic composite material; Several groups of IoT annular pressure-sensitive sensor arrays are embedded in the ground of the livestock driving passage, and the IoT annular pressure-sensitive sensor arrays are composed of several IoT pressure-sensitive sensors distributed in concentric circles; The hoof status sensing data is constructed based on the hoof pressure-sensitive data collected by several groups of the Internet of Things annular pressure-sensitive sensor arrays.

[0012] Through this scheme, hard friction areas with hard raised texture arrays and soft friction areas made of elastic composite materials are alternately set up on the ground of the livestock driving channel to simulate the stimulation of the hooves of livestock when walking on roads of different properties. In addition, an IoT annular pressure-sensitive sensor array composed of several IoT pressure-sensitive sensors distributed in concentric circles is used to capture the pressure deviation of the livestock hooves after being subjected to different stimuli, construct hoof status sensing data, accurately capture the early pathological characteristics of the livestock hooves, and improve the mapping sensitivity of hoof status sensing data to hoof lesions.

[0013] Optionally, analyzing the multidimensional feature monitoring data set to determine direct feature information includes: Decomposing and analyzing the dynamic gait characteristics of the target detected livestock in the livestock driving channel according to the livestock dynamic gait data, and determining a gait symmetry index and an avoidance imbalance degree; Analyzing the hoof state sensing data, and extracting hoof pressure distribution data and hoof pressure center offset data of the target livestock when the livestock passes through the hard friction zone and the soft buffer zone, respectively; The direct feature information is constructed according to the gait symmetry index, the avoidance imbalance degree, the hoof pressure distribution data, and the hoof pressure center offset data.

[0014] Through this solution, the dynamic gait data of livestock is analyzed, the gait symmetry index and avoidance imbalance degree are determined, and based on the hoof state sensor data, the hoof pressure distribution data and hoof pressure center offset data of the target detected livestock in different areas are extracted to jointly construct direct feature information. By utilizing the mechanical-motion coupling analysis mechanism, the direct feature information can accurately capture the characteristics exposed by early hoof lesions in livestock, realize the identification of mild gait compensation characteristics that are difficult to detect with existing technologies, and improve the accuracy of the analysis of livestock hoof lesions.

[0015] Optionally, analyzing the multidimensional feature monitoring data set to determine indirect feature information includes: Based on the livestock physiological habit data, periodic behavior modeling analysis is performed on the target livestock to determine the deviation of the frequency of eating and drinking and the length of standing and lying time; performing joint angle analysis on the target detected livestock based on the livestock normal posture data to determine trunk tilt tendency information; The indirect feature information is constructed based on the eating and drinking frequency, the standing and lying time deviation, and the trunk tilt tendency information.

[0016] Through this scheme, the dynamic gait data of livestock are analyzed to determine the gait symmetry index and avoidance imbalance of livestock when walking in the livestock driving channel. At the same time, combined with the deviation of the frequency of eating and drinking and the length of standing and lying time shown by livestock in daily life, indirect feature information is formed together, so that the indirect feature information can accurately characterize the indirect impact of early hoof disease on livestock, thereby improving the accuracy of early hoof disease analysis.

[0017] Optionally, the coupling analysis of the direct feature information and the indirect feature information to determine the hoof disease pathology feature dataset of the target livestock includes: Performing a risk-weighted assessment on each hoof of the target livestock according to the gait symmetry index, the avoidance imbalance degree, the hoof pressure distribution data, and the hoof pressure center offset data to generate a quantified hoof disease risk value for each hoof; Based on the trunk tilt tendency information, the hoof disease risk quantification value corresponding to each hoof is analyzed to determine whether the target livestock has hoof disease pathological characteristics, and if so, mark the target hoof with the disease; If the target livestock has the hoof disease pathological characteristics, the degree of progression of the hoof disease is determined based on the deviation of the eating and drinking frequency and the standing and lying time; The frequency of eating and drinking is negatively correlated with the degree of progression of the hoof disease, and the deviation of the standing and lying time is positively correlated with the degree of progression of the hoof disease; The hoof disease pathology feature dataset is constructed according to the target diseased hoof and the degree of progression of the hoof disease.

[0018] Through this scheme, based on the gait symmetry index, avoidance imbalance, hoof pressure distribution data and hoof pressure center offset data, a risk-weighted assessment is performed on each hoof of livestock, and a quantitative value of hoof disease risk is generated for each hoof. Combined with the trunk tilt tendency information, a comprehensive judgment is made on whether the livestock has hoof disease, and the target hoof with disease is marked. The deviation of eating and drinking frequency and standing and lying time is further used to indirectly map the degree of deterioration of the livestock's hoof disease. By integrating the target hoof with disease and the degree of deterioration of hoof disease, a hoof disease pathology feature dataset is constructed, which reduces the misjudgment of the early hoof disease status of livestock, and scientifically evaluates the development trend of the hoof, providing a scientific data basis for subsequent hoof disease tracing.

[0019] Optionally, the step of analyzing the hoof disease risk quantification value corresponding to each hoof based on the trunk tilt tendency information, determining whether the target livestock has hoof disease pathological characteristics, and if so, marking the target hoof with the disease, includes: According to the hoof disease risk quantification value of each hoof, target hooves whose hoof disease risk quantification value exceeds a preset risk threshold are screened, and the direction in which the target hoof is located is marked as a high-risk direction; The high-risk direction is compared with the trunk inclination direction reflected in the trunk inclination tendency information. If the trunk inclination direction is opposite to the high-risk direction, it is determined that the target detected livestock has hoof disease pathological characteristics, and the target hoof in the high-risk direction is marked as the diseased target hoof.

[0020] Through this scheme, the reverse verification mechanism of the trunk tilt direction is used to determine whether the target livestock has hoof disease pathological characteristics. If so, the target hoof in the high-risk direction is marked as a diseased target hoof, thereby improving the accuracy of livestock hoof disease assessment and further reducing the misjudgment rate.

[0021] Optionally, based on the direct feature information of each target livestock, and according to the corresponding hoof disease pathology feature dataset, the hoof disease pathology manifestations of all target livestock are traced, and a hoof disease examination report is determined and output, including: Performing spatial clustering feature analysis and temporal progression trend cluster analysis on the hoof disease pathology feature dataset corresponding to all the target detected livestock to determine the hoof disease manifestation feature type corresponding to each target detected livestock; The types of hoof disease manifestations include sporadic, localized, frequent, and large-scale outbreaks; According to the characteristic types of hoof disease, determine the path of tracing the cause of the disease; The path of inference of the cause of the disease includes individual behavioral abnormalities, local environmental inducements and group infection risks; The hoof disease inspection report is generated and outputted based on the unique identification of the target livestock, the characteristic type of the hoof disease manifestation and the cause tracing and inference path.

[0022] Through this scheme, spatial clustering feature analysis and temporal progression trend cluster analysis are performed on the hoof disease pathological characteristic datasets corresponding to different target livestock detection, and the hoof disease manifestation characteristic types corresponding to livestock are divided into sporadic, local frequent and large-scale attack types. A mapping relationship between the hoof disease manifestation characteristic types and the cause tracing inference path is established. The cause tracing inference path is used to guide the cause investigation direction of livestock breeders, improve the efficiency of hoof disease cause investigation of livestock breeders, effectively prevent the further development of hoof disease in livestock populations, and reduce the cost required by hoof disease impact.

[0023] Optionally, performing spatial clustering feature analysis and temporal progression trend cluster analysis on the hoof disease pathology feature dataset corresponding to all the target detected livestock to determine the hoof disease manifestation feature type corresponding to each target detected livestock includes: If the spatial correlation and temporal correlation of the data in the hoof disease pathological feature data set corresponding to different target livestock are both less than the preset correlation benchmark, it is determined that the hoof disease manifestation characteristic type of the current target livestock is the sporadic type; If the spatial correlation and the temporal correlation of the data in the hoof disease pathology feature data set corresponding to different target detection livestock are both greater than the preset correlation benchmark, it is determined that the hoof disease manifestation characteristic type of the current target detection livestock is the local frequent type; If the spatial correlation of the data in the hoof disease pathology feature data set corresponding to different target detection livestock is less than the corresponding preset correlation benchmark and the temporal correlation is greater than the corresponding preset correlation benchmark, it is judged that the hoof disease manifestation characteristic type of the current target detection livestock is the large-scale outbreak type.

[0024] Through this scheme, based on the spatial clustering characteristics and temporal progression trends of the hoof disease pathological characteristic data corresponding to different target livestock detection, and according to the numerical relationship between the spatial correlation and temporal correlation and the corresponding preset correlation benchmark, the hoof disease manifestation characteristic types corresponding to different target livestock detection are judged, so as to achieve scientific judgment and precise response to the hoof disease manifestation characteristic types, thereby improving the accuracy and effectiveness of the cause tracing inference path. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0026] Figure 1 A schematic diagram of a scenario of an intelligent livestock and poultry breeding management system based on Internet of Things technology provided in one embodiment of the present application; Figure 2 This is a flowchart of an intelligent livestock and poultry breeding management system based on Internet of Things technology provided in one embodiment of the present application. DETAILED DESCRIPTION

[0027] To make the purpose, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0028] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0029] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0030] Existing intelligent livestock and poultry breeding management technologies make it difficult to accurately identify and trace hoof diseases in livestock in the early stages of the disease, resulting in the continued development and spread of livestock hoof diseases in breeding populations, causing a decline in breeding production capacity and a negative impact on the economic effects of users' breeding industry.

[0031] Based on this, the present application provides an intelligent livestock and poultry breeding management system based on Internet of Things technology. The multi-dimensional feature monitoring data set is analyzed to obtain direct feature information and indirect feature information for mapping livestock hoof diseases. On this basis, the direct feature information and the indirect feature information are coupled and analyzed, and the hoof disease status of the livestock is comprehensively judged to form a hoof disease pathological feature data set. Then, the hoof disease pathological manifestations of all the target livestock are traced, and a hoof disease inspection report containing the livestock disease status and investigation direction suggestions is generated. The hoof disease inspection report is provided to the corresponding breeding personnel, so as to realize accurate hoof disease identification and tracing in the early stage of livestock hoof disease, effectively prevent the continuous development and spread of livestock hoof disease in the breeding population, improve the efficiency of automated breeding management, and improve the economic effect of user breeding.

[0032] Figure 1 This is a schematic diagram of an application scenario provided by this application. In the livestock breeding and management process, the system provided by this application can be used to accurately identify and trace hoof diseases in livestock in the early stages of the disease, effectively preventing the continued development and spread of hoof diseases in livestock breeding populations.

[0033] Specifically, the system provided in the present application is applied to any server, which communicates with a number of IoT sensors, obtains a multidimensional feature monitoring data set provided by the number of IoT sensors through the server, analyzes the multidimensional feature monitoring data set, and obtains direct feature information and indirect feature information for mapping livestock hoof diseases. On this basis, the direct feature information and the indirect feature information are coupled and analyzed, and the hoof disease status of the livestock is comprehensively judged to form a hoof disease pathological feature data set, and then the hoof disease pathological manifestations of all the target detected livestock are traced, and a hoof disease inspection report containing the livestock disease status and investigation direction suggestions is generated, and the hoof disease inspection report is provided to the corresponding breeding personnel, so as to realize accurate hoof disease identification and tracing in the early stage of livestock hoof disease, effectively prevent the continuous development and spread of livestock hoof disease in the breeding population, improve the efficiency of automated breeding management, and improve the economic effect of user breeding.

[0034] For specific implementation methods, please refer to the following embodiments.

[0035] Figure 2 This is a flowchart of an intelligent livestock and poultry breeding management system based on Internet of Things technology provided by an embodiment of this application. The system of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the system includes: S201. Obtain a multi-dimensional feature monitoring data set of target livestock, analyze the multi-dimensional feature monitoring data set, and determine direct feature information and indirect feature information.

[0036] The target livestock for detection may be livestock in a breeding farm that needs to be tested for hoof disease.

[0037] The multidimensional feature monitoring dataset can be a heterogeneous data set collected by IoT sensors that can reflect the status of livestock in many aspects.

[0038] Direct characteristic information may be characteristic data that is directly related to the health of livestock hooves.

[0039] Indirect characteristic information can be characteristic data reflecting abnormal overall behavior of livestock.

[0040] Specifically, the existing early diagnosis of livestock hoof diseases mostly relies on manual inspections, which are easily affected by subjective factors of inspectors, and it is difficult to capture the hidden symptoms caused by early hoof diseases, and it is difficult to accurately cover every individual livestock in the breeding population, resulting in difficulty in ensuring the accuracy of inspections. By deploying various types of Internet of Things sensors in breeding sites, data related to livestock hoof disease characteristics are collected from multiple dimensions to obtain a multidimensional feature monitoring data set. By screening the degree of correlation between various types of data in the multidimensional feature monitoring data set and hoof diseases, direct feature information and indirect feature information are obtained. Direct feature information can reflect the direct impact of hoof diseases on livestock behavior, and indirect feature information can reflect the indirect impact of the compensatory mechanism caused by hoof diseases on livestock habits, providing a scientific data basis for subsequent analysis and diagnosis of livestock hoof diseases.

[0041] S202: Perform coupling analysis on direct feature information and indirect feature information to determine a hoof disease pathology feature dataset of the target livestock.

[0042] The hoof disease pathology feature dataset may be a feature data set used to reflect the hoof disease status.

[0043] Specifically, after screening the direct and indirect characteristic information, the direct characteristic information is used as the basis for quantifying the risk of livestock hoof disease, and the indirect characteristic information is used as supporting evidence. The direct and indirect characteristic information are coupled and analyzed to comprehensively judge whether the current livestock suffers from hoof disease and the development trend of hoof disease, and to form a hoof disease pathology characteristic data set to accurately reflect the hoof disease status of individual livestock.

[0044] S203. Based on the hoof disease pathology characteristic dataset corresponding to each target livestock, trace the hoof disease pathology manifestations of all target livestock, determine and output a hoof disease inspection report.

[0045] The hoof disease inspection report may include the hoof disease status of each livestock in the current livestock population, the manifestation characteristics of hoof disease in the current livestock population, and corresponding disease tracing suggestions.

[0046] Specifically, the existing technology lacks the ability to trace the cause of hoof disease at the level of breeding groups, resulting in the treatment process focusing on individuals and ignoring the group, making it difficult to promptly eliminate the potential impact of hoof disease in the group, causing hoof disease to continue to occur frequently in the breeding group. By analyzing the onset pattern of the hoof disease pathological characteristic data set corresponding to each target livestock, the onset characteristics of hoof disease in the breeding group are obtained, and then it is determined whether the cause of the current livestock hoof disease manifestation is individual factors or group factors, the pathological manifestation of hoof disease is traced, and the corresponding investigation direction is obtained. By integrating the livestock disease status and the corresponding investigation direction, data visualization technology is used to generate the corresponding hoof disease inspection report, and the hoof disease inspection report is provided to the corresponding breeding personnel through human-computer interaction equipment, such as high-definition display screens, so that the breeding personnel can intuitively understand the livestock hoof disease status and the corresponding investigation direction.

[0047] Through this solution, the multidimensional feature monitoring data set is analyzed to obtain direct feature information and indirect feature information for mapping livestock hoof diseases. On this basis, the direct feature information and indirect feature information are coupled and analyzed, and the hoof disease status of livestock is comprehensively judged to form a hoof disease pathology feature data set. Then, the hoof disease pathological manifestations of all the target livestock are traced, and a hoof disease inspection report containing the livestock disease status and investigation direction suggestions is generated. The hoof disease inspection report is provided to the corresponding breeding personnel, so that accurate hoof disease identification and tracing can be achieved in the early stage of livestock hoof disease, effectively preventing the continued development and spread of livestock hoof disease in the breeding population, improving the efficiency of automated breeding management, and improving the economic benefits of user breeding.

[0048] In some embodiments, the multi-dimensional feature monitoring data set includes livestock physiological habit data, livestock normal posture data, hoof status sensor data and livestock dynamic gait data; livestock physiological habit data and livestock normal posture data are obtained through several monitoring cameras installed inside the breeding house; hoof status sensor data and livestock dynamic gait data are obtained through several Internet of Things sensors installed in the livestock driving channel.

[0049] Livestock physiological habit data can be basic behavioral data reflecting the daily periodic activity patterns of livestock.

[0050] The livestock normal posture data may be data used to characterize the posture characteristics of the livestock in a normal state.

[0051] The hoof state sensing data may be data reflecting changes in the state of the livestock's hoof during walking.

[0052] Livestock dynamic gait data may be data reflecting changes in the livestock's posture during walking.

[0053] The interior of the breeding house can be a fixed place for livestock to eat and rest in the farm.

[0054] A livestock passage can be a passage within a farm for livestock to move.

[0055] Specifically, physiological habit data is the core data reflecting the health status of livestock. For example, early-stage hoof disease can lead to painful behavioral inhibition, manifested as reduced eating frequency and abnormally prolonged lying time; livestock normal posture data can reflect the painful compensatory posture of livestock under the influence of hoof disease, such as center of gravity shift; hoof lesions (such as laminitis) will directly change the pressure distribution pattern of each hoof of livestock, and hoof status sensor data directly reflects the health status of livestock hooves; gait abnormality is the core external manifestation of hoof disease, and early hoof lesions will be unconsciously manifested through livestock dynamic gait data during the dynamic movement of livestock; since livestock habits and livestock normal postures need to be continuously monitored, livestock physiological habit data and livestock normal posture data are continuously obtained through several surveillance cameras installed inside the breeding house; hoof status sensor data and livestock dynamic gait data require livestock to be manifested during dynamic movement, so hoof status sensor data and livestock dynamic gait data are collected through several IoT sensors installed in the livestock driving channel.

[0056] Through this solution, we characterize the multi-dimensional status of livestock from both static and dynamic dimensions, using livestock physiological habit data and normal livestock posture data collected from breeding houses, combined with hoof condition sensor data and livestock dynamic gait data collected from livestock driving lanes, to form a multi-dimensional feature monitoring data set, providing comprehensive scientific data for subsequent analysis of livestock hoof pathology characteristics.

[0057] In some embodiments, the livestock driving channel is a livestock passage connecting the breeding house and the livestock activity area; the side wall of the livestock driving channel is provided with a movable gradually narrowing structure, and the channel width decreases gradually from the entrance to the exit based on the average shoulder width of the livestock, and a number of IoT pressure-sensing fences are arranged on the left and right sides of the channel; a slope transition section is provided at the exit end, and a high-friction coefficient coating is applied on the surface of the slope transition section, and a number of IoT three-dimensional motion capture sensors are arranged on the left and right sides of the slope transition section; based on the data collected by the IoT pressure-sensing fences and the IoT three-dimensional motion capture sensors, the dynamic gait data of the livestock is constructed.

[0058] The livestock activity area can be an area set up in the farm for livestock to exercise.

[0059] The movable gradually narrowing structure may be a movable channel structure whose lateral width gradually decreases from the inlet to the outlet.

[0060] The average shoulder width of livestock may be the average shoulder width of livestock corresponding to the current breeding population.

[0061] The gradient decrease may be a decrease in the channel width at a fixed ratio.

[0062] The IoT pressure sensing fence can be a movable fence that can be combined with an integrated pressure sensing unit.

[0063] The ramp transition section can be a channel segment with a fixed slope (such as 8 degrees).

[0064] A high coefficient of friction coating may be a coating that increases friction between the floor of the walkway and the hooves of livestock.

[0065] Specifically, the process of collecting multi-dimensional feature monitoring data sets is focused on the livestock passages between the breeding sheds and the livestock activity areas, so that the detection of livestock hoof diseases can be integrated into the daily management of the farm. When the livestock are moving between the breeding sheds and the livestock activity areas, the livestock are monitored for hoof diseases without any sense of contact. This can not only collect representative dynamic data of the livestock during their walking process, but also realize periodic automatic detection of the livestock, while avoiding the stress of the livestock under intrusive detection methods. By setting the livestock passage as a gradually narrowing structure, and The IoT pressure-sensing fences are set up on both sides to collect the lateral mechanical data of livestock during walking, so as to accurately capture the body avoidance tilt posture of livestock that may be caused by hoof disease; by setting a slope transition section at the exit end and applying a high-friction coefficient coating on the surface of the slope transition section, the livestock can be more significantly exposed to abnormal hoof lifting and abnormal hoof landing movements caused by hoof disease; the data collected by the IoT pressure-sensing fences and the IoT three-dimensional motion capture sensors are used to jointly construct the dynamic gait data of livestock, so as to construct the dynamic gait data of livestock, so as to make the dynamic gait data of livestock more visible. Through this solution, the livestock driving channel is set as an overall progressively narrowing structure, and a slope transition section is set at the exit end of the livestock driving channel. While capturing the body avoidance tilting posture caused by hoof disease of livestock, the abnormal hoof lifting movement and abnormal hoof landing movement caused by hoof disease are more significantly exposed to the livestock, thereby improving the comprehensiveness and scientific nature of the livestock dynamic gait data, making the early hoof lesion characteristics more prominent, and thus improving the accuracy of subsequent judgment of the livestock hoof disease status.

[0066] In some embodiments, the ground of the livestock driving passage is configured as alternating hard friction zones and soft buffer zones; the surface of the hard friction zone is provided with a hard raised texture array, and the surface of the soft buffer zone is provided with an elastic composite material; several groups of IoT annular pressure-sensitive sensor arrays are embedded inside the ground of the livestock driving passage, and the IoT annular pressure-sensitive sensor array is composed of several IoT pressure-sensitive sensors distributed in concentric circles; based on the hoof pressure-sensitive data collected by several groups of IoT annular pressure-sensitive sensor arrays, hoof status sensing data is constructed.

[0067] The hard friction zone may be an area in the ground area of ​​a livestock driving path used to simulate a high-resistance hard road surface, so as to induce a pressure response of the livestock's hooves.

[0068] The soft buffer zone can be an area in the ground area of ​​the livestock runway used to simulate soft ground and reflect the cushioning adaptability of the hoof.

[0069] The IoT annular pressure-sensitive sensor array may be an annular pressure-sensitive detection module composed of pressure-sensitive sensor units arranged in concentric circles.

[0070] The hoof pressure-sensitive data may be data used to characterize the pressure characteristics applied by each hoof to the ground during the walking of livestock.

[0071] Specifically, the early signs of hoof disease are not obvious, and the pain stimulation caused to livestock is not strong. It is only reflected through the livestock's movement reflexes when the affected area is specifically stimulated. If the floor of the livestock driving path uses a flat structure with a single material, it is difficult to differentiate and stimulate the biomechanical response of the livestock's hoof. (For example, if a local wound on the sole of the livestock's hoof is sunken due to physical damage, the flat structure cannot directly stimulate the wound, and the abnormal behavior of the livestock is difficult to capture.) As a result, the characteristics of early hoof lesions are not obvious and difficult to be captured by the corresponding sensors. By alternating hard friction zones with hard raised texture arrays and soft friction zones made of elastic composite materials on the floor of the livestock driving path, the stimulation of livestock's hooves while walking on different road properties is simulated. The hard friction zones expose hoof lesions through local high-voltage stimulation, while the soft friction zones reflect the uniformity of hoof pressure distribution through damping structures. By deploying an IoT ring pressure-sensitive sensor array composed of several concentrically distributed IoT pressure-sensitive sensors, the pressure deviation of the livestock's hoof after being stimulated by different stimuli is captured, and hoof state sensing data is constructed to accurately capture the characteristics of early hoof lesions in social livestock.

[0072] Through this scheme, hard friction areas with hard raised texture arrays and soft friction areas made of elastic composite materials are alternately set up on the ground of the livestock driving channel to simulate the stimulation of the hooves of livestock when walking on roads of different properties. In addition, an IoT annular pressure-sensitive sensor array composed of several IoT pressure-sensitive sensors distributed in concentric circles is used to capture the pressure deviation of the livestock hooves after being subjected to different stimuli, construct hoof status sensing data, accurately capture the early pathological characteristics of the livestock hooves, and improve the mapping sensitivity of hoof status sensing data to hoof lesions.

[0073] In some embodiments, based on the dynamic gait data of livestock, the dynamic gait characteristics of the target livestock in the livestock driving channel are decomposed and analyzed to determine the gait symmetry index and avoidance imbalance degree; the hoof state sensor data is analyzed to extract the hoof pressure distribution data and hoof pressure center offset data of the target livestock when passing through the hard friction area and the soft buffer area respectively; direct feature information is constructed based on the gait symmetry index, avoidance imbalance degree, hoof pressure distribution data and hoof pressure center offset data.

[0074] The gait symmetry index can be used to characterize the consistency of left and right limb movements during livestock walking.

[0075] The avoidance imbalance degree can reflect the degree of instability in the movement of livestock when they avoid being stimulated by the pathological part of the hoof.

[0076] The hoof pressure distribution data can be a pressure topology map of the hoof sole contact surface recorded by an IoT annular pressure-sensitive sensor array.

[0077] The hoof pressure center offset data may be an offset distance of the detected hoof sole pressure center relative to the geometric center.

[0078] Specifically, based on the livestock walking motion data collected by the IoT three-dimensional motion capture sensor, the difference in the landing time of the left and right hooves and the difference in the joint movement amplitude are compared, and a weighted gait symmetry index is generated ((left and right hooves landing time difference / gait cycle) × (Δ joint movement amplitude difference / maximum joint angle)). Based on the bilateral pressure change amplitude recorded by the pressure-sensing fence and the positive correlation coefficient between the bilateral pressure change amplitude and the avoidance imbalance degree, the avoidance imbalance degree is quantified. The hoof pressure of the target detected livestock when passing through the hard friction area and the soft buffer zone is obtained through the IoT annular pressure-sensitive sensor array. Force distribution data and hoof pressure center offset data (after the livestock hoof diseased area is stimulated by the corresponding ground, the livestock will instinctively avoid the area to be stimulated by pressure again. At this time, the pressure distribution of each hoof on the corresponding ground will change significantly. At the same time, the livestock's avoidance action will cause the pressure center of the corresponding diseased hoof to shift, and it will no longer be the corresponding geometric structure center). By integrating gait symmetry index, avoidance imbalance, hoof pressure distribution data and hoof pressure center offset data, direct feature information is constructed, so that the direct feature information can accurately capture the characteristics exposed by early hoof lesions in livestock.

[0079] Through this solution, the dynamic gait data of livestock is analyzed, the gait symmetry index and avoidance imbalance degree are determined, and based on the hoof state sensor data, the hoof pressure distribution data and hoof pressure center offset data of the target detected livestock in different areas are extracted to jointly construct direct feature information. By utilizing the mechanical-motion coupling analysis mechanism, the direct feature information can accurately capture the characteristics exposed by early hoof lesions in livestock, realize the identification of mild gait compensation characteristics that are difficult to detect with existing technologies, and improve the accuracy of the analysis of livestock hoof lesions.

[0080] In some embodiments, based on the livestock's physiological habit data, periodic behavior modeling analysis is performed on the target detected livestock to determine the frequency of eating and drinking and the deviation of the standing and lying time; based on the livestock's normal posture data, the target detected livestock is analyzed for joint angles to determine the trunk tilt tendency information; based on the frequency of eating and drinking, the deviation of the standing and lying time and the trunk tilt tendency information, indirect feature information is constructed.

[0081] Periodic behavior modeling analysis can be the process of analyzing the daily periodic behavior patterns of livestock.

[0082] The frequency of eating and drinking can be the frequency with which livestock eat feed and drink water.

[0083] The standing and lying time deviation can be the degree of deviation between the actual standing and lying time ratio distribution of livestock and the normal standing and lying time ratio distribution within a fixed time period (such as 24 hours).

[0084] Joint angle parsing can be the process of analyzing the key leg angle changes of target detection livestock in daily standing postures.

[0085] The trunk tilt tendency information may be information used to characterize the trunk tilt direction of the target detected livestock in a standing posture.

[0086] Specifically, when livestock suffer from hoof disease, in addition to directly affecting the livestock's walking posture, it will also have a potential impact on the livestock's living habits, such as loss of appetite, dislike of standing, etc.; based on the livestock's physiological habit data, a semi-supervised learning strategy is adopted to manually label a certain proportion (such as 10%) of the monitoring data fragments (eating, drinking, standing, lying), and train the livestock periodic behavior recognition model. The trained model realizes automatic classification on the full amount of data, and converts the identified eating, drinking, standing, and lying events into pulse signals with a 24-hour cycle (the time of event occurrence is 1, and the rest are 0), and the statistical analysis is carried out. The frequency of eating and drinking and the proportion of standing and lying time were distributed, and the difference between the distribution of standing and lying time and the healthy standing and lying time distribution benchmark of the corresponding livestock population was quantified (the incremental difference in the proportion of lying time was used as the abnormal assessment benchmark, that is, when the proportion of lying time exceeded the corresponding benchmark value, the higher the proportion of lying time, the greater the corresponding deviation). The deviation of standing and lying time was obtained, and the three-dimensional coordinates of the hip-knee-ankle joint were established based on the normal posture data of livestock. The projection of the livestock spine midline was used as the benchmark to identify the trunk tilt tendency information. By integrating the information of eating and drinking frequency, standing and lying time deviation and trunk tilt tendency, indirect feature information was constructed.

[0087] Through this scheme, the dynamic gait data of livestock are analyzed to determine the gait symmetry index and avoidance imbalance of livestock when walking in the livestock driving channel. At the same time, combined with the deviation of the frequency of eating and drinking and the length of standing and lying time shown by livestock in daily life, indirect feature information is formed together, so that the indirect feature information can accurately characterize the indirect impact of early hoof disease on livestock, thereby improving the accuracy of early hoof disease analysis.

[0088] In some embodiments, a risk-weighted assessment is performed on each hoof of the target detected livestock based on the gait symmetry index, avoidance imbalance, hoof pressure distribution data, and hoof pressure center offset data to generate a hoof disease risk quantification value for each hoof; based on the trunk tilt tendency information, the hoof disease risk quantification value corresponding to each hoof is analyzed to determine whether the target detected livestock has hoof disease pathological characteristics, and if so, the diseased target hoof is marked; if the target detected livestock has hoof disease pathological characteristics, the degree of hoof disease deterioration is determined based on the frequency of eating and drinking and the deviation of standing and lying time; the frequency of eating and drinking is negatively correlated with the degree of hoof disease deterioration, and the deviation of standing and lying time is positively correlated with the degree of hoof disease deterioration; a hoof disease pathology feature dataset is constructed based on the diseased target hoof and the degree of hoof disease deterioration.

[0089] Risk-weighted assessment can be a process of weighted assessment of hoof disease risk based on multidimensional characteristics.

[0090] A hoof disease risk quantification value can be a numerical indicator used to characterize the likelihood of a single hoof lesion.

[0091] A hoof pathology signature may be a characteristic that indicates that the livestock has a hoof disease.

[0092] The target hoof may be a specific diseased hoof of an animal.

[0093] The degree of progression of lameness can be an indicator of the stage of lameness development in livestock.

[0094] Specifically, the gait symmetry index, avoidance imbalance, hoof pressure distribution data and hoof pressure center offset data are converted into Z-score standardized values, loaded into the pre-trained random forest model, and the standardized gait symmetry index, avoidance imbalance, hoof pressure distribution data and hoof pressure center offset data are input into the pre-trained random forest model. A risk-weighted assessment is performed on each hoof of the target livestock, and the quantitative value of the hoof disease risk of each hoof is output. On this basis, combined with the trunk tilt tendency information, it is judged whether there is a correlation between the current high-risk hoof and the corresponding livestock trunk tilt direction. If there is a correlation, it can be determined that the livestock has hoof disease pathological characteristics, and further The first step is to quantify the degree of hoof disease deterioration by combining the frequency of eating and drinking and the deviation of standing and lying time. Among them, the frequency of eating and drinking is negatively correlated with the degree of hoof disease deterioration. The lower the eating frequency, the higher the degree of hoof disease deterioration. The deviation of standing and lying time is positively correlated with the degree of hoof disease deterioration. The higher the deviation of standing and lying time, the higher the proportion of lying time of livestock, and the higher the degree of hoof disease deterioration. The negative correlation coefficient between the frequency of eating and drinking and the degree of hoof disease deterioration is combined with the positive correlation coefficient between the deviation of standing and lying time and the degree of hoof disease deterioration to comprehensively evaluate the degree of hoof disease deterioration. By integrating the diseased target hoof and the degree of hoof disease deterioration, a hoof disease pathology characteristic dataset is constructed.

[0095] Through this scheme, based on the gait symmetry index, avoidance imbalance, hoof pressure distribution data and hoof pressure center offset data, a risk-weighted assessment is performed on each hoof of livestock, and a quantitative value of hoof disease risk is generated for each hoof. Combined with the trunk tilt tendency information, a comprehensive judgment is made on whether the livestock has hoof disease, and the target hoof with disease is marked. The deviation of eating and drinking frequency and standing and lying time is further used to indirectly map the degree of deterioration of the livestock's hoof disease. By integrating the target hoof with disease and the degree of deterioration of hoof disease, a hoof disease pathology feature dataset is constructed, which reduces the misjudgment of the early hoof disease status of livestock, and scientifically evaluates the development trend of the hoof, providing a scientific data basis for subsequent hoof disease tracing.

[0096] In some embodiments, based on the hoof disease risk quantification value of each hoof, target hooves whose hoof disease risk quantification value exceeds a preset risk threshold are screened, and their directions are marked as high-risk directions; the high-risk direction is compared with the trunk tilt direction reflected in the trunk tilt tendency information. If the trunk tilt direction is opposite to the high-risk direction, it is determined that the target detected livestock has hoof disease pathological characteristics, and the target hoof in the high-risk direction is marked as a diseased target hoof.

[0097] The preset risk threshold may be a critical value for determining that the hoof has entered a high-risk pathological state.

[0098] The high-risk direction may be the relative position of the hoof where the risk quantification value exceeds the limit, such as the left front, right rear, etc.

[0099] The trunk tilt direction may be a dominant offset orientation in the trunk tilt tendency information.

[0100] Specifically, after livestock suffers from hoof disease, affected by pain and discomfort, the livestock will reflexively reduce the force on the diseased hoof during daily standing, causing the livestock's trunk to deviate in the opposite direction of the position of the diseased hoof. For example, if the left front hoof of the livestock is diseased, its trunk will tend to tilt to the right and rear. Based on the numerical comparison results between the hoof disease risk quantification value of each hoof and the corresponding preset risk threshold, the direction of the target hoof whose hoof disease risk quantification value exceeds the preset risk threshold is defined as a high-risk direction, and the high-risk direction is compared with the trunk tilt direction reflected in the trunk tilt tendency information to evaluate whether there is a reverse mapping relationship between the high-risk direction and the trunk tilt direction. If so, it can be determined that the target detected livestock has hoof disease pathological characteristics, and the target hoof in the high-risk direction is marked as a diseased target hoof.

[0101] Through this scheme, the reverse verification mechanism of the trunk tilt direction is used to determine whether the target livestock has hoof disease pathological characteristics. If so, the target hoof in the high-risk direction is marked as a diseased target hoof, thereby improving the accuracy of livestock hoof disease assessment and further reducing the misjudgment rate.

[0102] In some embodiments, spatial clustering feature analysis and temporal progression trend clustering analysis are performed on the hoof disease pathological feature datasets corresponding to all target detected livestock to determine the hoof disease manifestation characteristic type corresponding to each target detected livestock; the hoof disease manifestation characteristic types include sporadic, local frequent and large-scale attack types; based on the hoof disease manifestation characteristic types, the etiology tracing and inference path is determined; the etiology tracing and inference path includes individual behavioral abnormalities, local environmental inducements and group infection risks; based on the unique identification of the target detected livestock, the hoof disease manifestation characteristic type and the etiology tracing and inference path, a hoof disease inspection report is generated and output.

[0103] Spatial aggregation characteristic analysis can be the process of analyzing the spatial aggregation patterns of livestock with different hoof diseases corresponding to the locations of breeding houses.

[0104] Cluster analysis of temporal progression trends can be a process of analyzing the temporal distribution patterns of hoof disease progression in livestock with different hoof diseases.

[0105] The hoof disease manifestation characteristic type may be a type label used to characterize the disease condition of the livestock hoof and then corresponding to the distribution characteristics in the breeding population.

[0106] The unique identifier may be identification information used to identify and distinguish individual livestock, and the unique identifier may be the livestock unique code corresponding to the livestock electronic ear tag.

[0107] Sporadic disease can be characterized as an independent random case of hoof disease in a population.

[0108] The local frequency type can be a sign that the current livestock hoof disease situation has a pattern of being concentrated in a local area within the breeding house.

[0109] The mass outbreak pattern may represent the widespread distribution of livestock hoof disease in the breeding population.

[0110] The etiology tracing inference path may be the tracing path guidance information used to guide the direction of etiology investigation.

[0111] Individual behavioral abnormalities can be a typical attribution path for sporadic cases.

[0112] Local environmental triggers can be the core inference direction for local recurrent conditions.

[0113] The risk of mass infection can be the main tracing target for large-scale outbreaks.

[0114] Specifically, after judging the hoof disease conditions of different livestock individuals, if only further diagnosis and treatment are carried out for individual cases, it is impossible to identify the possible group transmission pattern of hoof disease, which makes it difficult to reduce the occurrence of hoof disease from the root; through the spatial clustering algorithm, the spatial distribution of cases and the coverage area map of the breeding house are spatially superimposed and analyzed to determine whether the livestock corresponding to the cases have spatial clustering characteristics in the breeding house. At the same time, through the time series clustering algorithm, it is determined whether the development trend of hoof disease between different cases has temporal continuity characteristics. According to the spatial clustering characteristics and temporal continuity characteristics shown by different cases, the characteristic type of hoof disease manifestation is determined, and a hoof disease table is established. The mapping information between the current feature type and the cause tracing inference path is as follows: the traceability investigation direction for the sporadic type is individual behavioral abnormalities (for individual sporadic cases, it is sufficient to focus on individual conditions); the traceability investigation direction for the local frequent type is local environmental inducements (local intensive attacks are usually related to the influence of local environmental factors); and the traceability investigation direction for the large-scale attack type is group infection risk (widespread attacks are usually related to infectious factors). On this basis, the unique identifier of the target livestock is used as the data index, and the corresponding hoof disease manifestation feature type and cause tracing inference path are used as the data content to construct the hoof disease report information, and then the corresponding hoof disease inspection report is generated and output through data visualization technology.

[0115] Through this scheme, spatial clustering feature analysis and temporal progression trend cluster analysis are performed on the hoof disease pathological characteristic datasets corresponding to different target livestock detection, and the hoof disease manifestation characteristic types corresponding to livestock are divided into sporadic, local frequent and large-scale attack types. A mapping relationship between the hoof disease manifestation characteristic types and the cause tracing inference path is established. The cause tracing inference path is used to guide the cause investigation direction of livestock breeders, improve the efficiency of hoof disease cause investigation of livestock breeders, effectively prevent the further development of hoof disease in livestock populations, and reduce the cost required by hoof disease impact.

[0116] In some embodiments, if the spatial correlation and temporal correlation of data in the hoof disease pathology feature data set corresponding to different target detected livestock are both less than a preset correlation benchmark, then the hoof disease manifestation characteristic type of the current target detected livestock is judged to be sporadic; if the spatial correlation and temporal correlation of data in the hoof disease pathology feature data set corresponding to different target detected livestock are both greater than the preset correlation benchmark, then the hoof disease manifestation characteristic type of the current target detected livestock is judged to be local frequent; if the spatial correlation of data in the hoof disease pathology feature data set corresponding to different target detected livestock is less than the corresponding preset correlation benchmark and the temporal correlation is greater than the corresponding preset correlation benchmark, then the hoof disease manifestation characteristic type of the current target detected livestock is judged to be large-scale outbreak.

[0117] Spatial correlation can be an indicator used to characterize the degree to which different livestock cases are spatially clustered.

[0118] Temporal correlation can be an indicator that represents the degree of temporal correlation between the severity of lameness disease corresponding to different livestock cases.

[0119] The preset association benchmark can be a baseline association index used to assess whether different livestock cases are strongly associated in space or time.

[0120] Specifically, the coverage area of ​​the breeding house is divided into several local areas, and the case distribution heat map is drawn by the kernel density estimation algorithm to identify the case distribution density in each local area. On this basis, the spatial autocorrelation is quantified by the Moran's I index to obtain the spatial correlation; the time window in the monitoring period chassis is divided according to the fixed unit analysis time, and the changes in the degree of deterioration of hoof lesions corresponding to different cases in different time windows are extracted. The temporal correlation is obtained by quantifying the direct Pearson correlation coefficient of the degree of deterioration of hoof lesions of different cases between adjacent time windows; the corresponding spatial correlation and temporal correlation are compared with the corresponding preset correlation benchmarks: when the spatial correlation and temporal correlation of the data in the hoof disease pathological feature data set corresponding to different target livestock are both less than the preset correlation benchmark, it means that there is no significant correlation between different cases, and the hoof disease manifestation characteristic type of the current target livestock is judged to be occasional Hair style; when the spatial correlation and temporal correlation of the data in the hoof disease pathology characteristic dataset corresponding to different target livestock detection are greater than the preset correlation benchmark, it means that the cases are concentrated in the local area and the disease development trend has a temporal correlation, and the hoof disease manifestation characteristic type of the current target livestock detection is judged to be a local frequent type; when the spatial correlation of the data in the hoof disease pathology characteristic dataset corresponding to different target livestock detection is less than the corresponding preset correlation benchmark and the temporal correlation is greater than the corresponding preset correlation benchmark, it means that the cases are not concentrated in the local area, but are distributed in all areas, and there is a temporal correlation between the cases in different areas and the disease development, and the hoof disease manifestation characteristic type of the current target livestock detection is judged to be a large-scale outbreak type.

[0121] Through this scheme, based on the spatial clustering characteristics and temporal progression trends of the hoof disease pathological characteristic data corresponding to different target livestock detection, and according to the numerical relationship between the spatial correlation and temporal correlation and the corresponding preset correlation benchmark, the hoof disease manifestation characteristic types corresponding to different target livestock detection are judged, so as to achieve scientific judgment and precise response to the hoof disease manifestation characteristic types, thereby improving the accuracy and effectiveness of the cause tracing inference path.

Claims

1. An intelligent livestock and poultry breeding management system based on Internet of Things technology, characterized in that: include: Acquiring a multidimensional feature monitoring data set of target detection livestock, analyzing the multidimensional feature monitoring data set to determine direct feature information and indirect feature information; performing coupled analysis on the direct feature information and the indirect feature information to determine a hoof disease pathology feature dataset of the target livestock; Based on the direct feature information of each target livestock, and according to the corresponding hoof disease pathology feature data set, the hoof disease pathology manifestations of all target livestock are traced, and a hoof disease inspection report is determined and output.

2. The system according to claim 1, wherein: The multi-dimensional feature monitoring data set includes livestock physiological habit data, livestock normal posture data, hoof state sensor data and livestock dynamic gait data; The livestock physiological habit data and the livestock normal posture data are obtained through a plurality of monitoring cameras arranged inside the breeding house; The hoof state sensing data and the livestock dynamic gait data are obtained through a number of Internet of Things sensors arranged in the livestock driving channel.

3. The system according to claim 2, characterized in that The livestock driving passage is a livestock passage for connecting the breeding house and the livestock activity area; The side walls of the livestock passage are provided with a movable gradually narrowing structure. The width of the passage decreases gradually from the entrance to the exit based on the average shoulder width of the livestock. Several IoT pressure-sensing fences are arranged on the left and right sides of the passage. The outlet end is provided with a slope transition section, the surface of which is coated with a high friction coefficient coating, and a number of Internet of Things three-dimensional motion capture sensors are arranged on the left and right sides of the slope transition section; The livestock dynamic gait data is constructed based on the data collected by the Internet of Things pressure sensing fence and the Internet of Things three-dimensional motion capture sensor.

4. The system according to claim 3, characterized in that The ground of the livestock driving passage is configured as hard friction areas and soft buffer areas that are alternately distributed; The surface of the hard friction zone is provided with a hard raised texture array, and the surface of the soft buffer zone is provided with an elastic composite material; Several groups of IoT annular pressure-sensitive sensor arrays are embedded in the ground of the livestock driving passage, and the IoT annular pressure-sensitive sensor arrays are composed of several IoT pressure-sensitive sensors distributed in concentric circles; The hoof status sensing data is constructed based on the hoof pressure-sensitive data collected by several groups of the Internet of Things annular pressure-sensitive sensor arrays.

5. The system according to claim 4, characterized in that The analyzing the multi-dimensional feature monitoring data set to determine direct feature information includes: Decomposing and analyzing the dynamic gait characteristics of the target detected livestock in the livestock driving channel according to the livestock dynamic gait data, and determining a gait symmetry index and an avoidance imbalance degree; Analyzing the hoof state sensing data, and extracting hoof pressure distribution data and hoof pressure center offset data of the target livestock when the livestock passes through the hard friction zone and the soft buffer zone, respectively; The direct feature information is constructed according to the gait symmetry index, the avoidance imbalance degree, the hoof pressure distribution data, and the hoof pressure center offset data.

6. The system according to claim 5, characterized in that The analyzing the multi-dimensional feature monitoring data set to determine indirect feature information includes: Based on the livestock physiological habit data, periodic behavior modeling analysis is performed on the target livestock to determine the deviation of the frequency of eating and drinking and the length of standing and lying time; performing joint angle analysis on the target detected livestock based on the livestock normal posture data to determine trunk tilt tendency information; The indirect feature information is constructed based on the eating and drinking frequency, the standing and lying time deviation, and the trunk tilt tendency information.

7. The system according to claim 6, characterized in that The coupling analysis of the direct feature information and the indirect feature information to determine the hoof disease pathology feature dataset of the target livestock includes: Performing a risk-weighted assessment on each hoof of the target livestock according to the gait symmetry index, the avoidance imbalance degree, the hoof pressure distribution data, and the hoof pressure center offset data to generate a quantified hoof disease risk value for each hoof; Based on the trunk tilt tendency information, the hoof disease risk quantification value corresponding to each hoof is analyzed to determine whether the target livestock has hoof disease pathological characteristics, and if so, mark the target hoof with the disease; If the target livestock has the hoof disease pathological characteristics, the degree of progression of the hoof disease is determined based on the deviation of the eating and drinking frequency and the standing and lying time; The frequency of eating and drinking is negatively correlated with the degree of progression of the hoof disease, and the deviation of the standing and lying time is positively correlated with the degree of progression of the hoof disease; The hoof disease pathology feature dataset is constructed according to the target diseased hoof and the degree of progression of the hoof disease.

8. The system according to claim 7, characterized in that The method of analyzing the hoof disease risk quantification value corresponding to each hoof based on the trunk tilt tendency information, determining whether the target livestock has hoof disease pathological characteristics, and if so, marking the target hoof with the disease, includes: According to the hoof disease risk quantification value of each hoof, target hooves whose hoof disease risk quantification value exceeds a preset risk threshold are screened, and the direction in which the target hoof is located is marked as a high-risk direction; The high-risk direction is compared with the trunk inclination direction reflected in the trunk inclination tendency information. If the trunk inclination direction is opposite to the high-risk direction, it is determined that the target detected livestock has hoof disease pathological characteristics, and the target hoof in the high-risk direction is marked as the diseased target hoof.

9. The system according to claim 8, characterized in that Based on the direct feature information of each target livestock, and according to the corresponding hoof disease pathology feature dataset, the hoof disease pathology manifestations of all target livestock are traced, and a hoof disease inspection report is determined and output, including: Performing spatial clustering feature analysis and temporal progression trend cluster analysis on the hoof disease pathology feature dataset corresponding to all the target detected livestock to determine the hoof disease manifestation feature type corresponding to each target detected livestock; The types of hoof disease manifestations include sporadic, localized, frequent, and large-scale outbreaks; According to the characteristic types of hoof disease, determine the path of tracing the cause of the disease; The path of inference of the cause of the disease includes individual behavioral abnormalities, local environmental inducements and group infection risks; The hoof disease inspection report is generated and outputted based on the unique identification of the target livestock, the characteristic type of the hoof disease manifestation and the cause tracing and inference path.

10. The system according to claim 9, characterized in that The spatial clustering feature analysis and temporal progression trend clustering analysis of the hoof disease pathology feature dataset corresponding to all the target detected livestock are performed to determine the hoof disease manifestation feature type corresponding to each target detected livestock, including: If the spatial correlation and temporal correlation of the data in the hoof disease pathological feature data set corresponding to different target livestock are both less than the preset correlation benchmark, it is determined that the hoof disease manifestation characteristic type of the current target livestock is the sporadic type; If the spatial correlation and the temporal correlation of the data in the hoof disease pathology feature data set corresponding to different target detection livestock are both greater than the preset correlation benchmark, it is determined that the hoof disease manifestation characteristic type of the current target detection livestock is the local frequent type; If the spatial correlation of the data in the hoof disease pathology feature data set corresponding to different target detection livestock is less than the corresponding preset correlation benchmark and the temporal correlation is greater than the corresponding preset correlation benchmark, it is judged that the hoof disease manifestation characteristic type of the current target detection livestock is the large-scale outbreak type.

Images