Risk identification method and device using dairy cow as collateral
Through a cow health status recognition model based on blockchain and machine learning, combining the shooting images of cows and forage/treatment data, the accuracy of cow collateral risk identification is solved, and the risk assessment and data of cow collateral are realized. Improvement of sex.
Patent Information
- Application Number
- CN202411272453.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-06
AI Technical Summary
In the prior art, it is difficult to accurately assess the risk identification of dairy cows as collateral, and there are problems with data authenticity and reliability, which leads to uncertainty and risks when providing mortgage loans.
By identifying the captured images of the target cows based on the preset cow health status identification model, combining the forage and treatment data in the blockchain, the actual mortgage amount is corrected, and the accumulated milk production and health status are identified through the machine learning model. Generate risk warning messages to ensure the safety and reliability of the data.
Accurate identification of the risks of dairy cow collateral is achieved, reducing the risks of banks in mortgage loans, improving the authenticity and reliability of data, and helping to discover and avoid risks in breeding management.
Smart Images

Figure CN119090626A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of blockchain technology, and in particular to a method and device for identifying risks using dairy cows as collateral. Background Art
[0002] The ranch owner who raises dairy cows hopes to obtain mortgage funds by using the cows as collateral. However, although the fund provider, such as a bank, can assess and determine the mortgage amount of the collateral through a valuation model, there are many uncertainties in the process of raising dairy cows, which brings certain risks to the bank's provision of mortgage loans. The bank hopes to facilitate business development under the premise of controllable risks.
[0003] In addition, risk identification of collateral requires a large amount of relevant data, and the authenticity and reliability of the data are also particularly important. In the existing technology, ranch owners can use farm management systems to standardize breeding management and avoid risks in the breeding process. However, relying solely on farm management systems is not enough to avoid a large number of breeding process risks. Summary of the invention
[0004] In view of the problems in the prior art, the embodiments of the present invention provide a method and device for identifying risks using dairy cows as collateral, which can at least partially solve the problems in the prior art.
[0005] On the one hand, the present invention provides a risk identification method using dairy cows as collateral, comprising:
[0006] Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times used as collateral are recognized to obtain the health status recognition result of the target cow;
[0007] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0008] According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount;
[0009] If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0010] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0011] The preset cow health status recognition model is obtained by training a machine learning model according to the cow health status recognition sample data, including:
[0012] Acquire a dairy cow data set; the dairy cow data set includes a plurality of photographed image samples corresponding to the cow faces of a plurality of dairy cows corresponding to each breed, and physiological parameter detection data of a plurality of dairy cows corresponding to each breed;
[0013] Perform health status analysis on physiological parameter detection data of multiple dairy cows corresponding to each breed, use multiple image samples with normal health status as positive samples, and use multiple image samples with abnormal health status as negative samples;
[0014] The preset cow health status recognition model is obtained by training a machine learning model according to the positive samples and the negative samples.
[0015] The risk identification method using dairy cows as collateral also includes:
[0016] Acquire target physiological parameter detection data of the target cow and second treatment data actually provided to the target cow from a target farm management system, and recognize photographed images of forage fed to the target cow when feeding the target cow, to obtain second forage data actually fed to the target cow;
[0017] According to the target physiological parameter detection data, the second forage data and the second treatment data, the predetermined actual mortgage amount of the target dairy cow is revised to obtain a second revised mortgage amount;
[0018] If it is determined that the difference between the second revised mortgage amount and the first revised mortgage amount is greater than a preset difference threshold, a second risk warning message is generated to indicate the risk caused by dairy cow breeding management.
[0019] Wherein, after the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral further includes:
[0020] Determine a target cow health identification status result corresponding to the target physiological parameter detection data according to a pre-established correspondence relationship between preset physiological parameter detection data and preset cow health identification status result;
[0021] The target cow health identification result is compared with the health identification result, and if it is determined that the difference between the comparison results is outside a preset difference range, a first reminder message is generated for manual troubleshooting of related equipment.
[0022] Wherein, after the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral further includes:
[0023] Acquire a first forage data item list corresponding to the first forage data, a photographed image of feeding forage when feeding the target cow corresponding to the second forage data, and forage feeding operation record information of a forage feeder;
[0024] A second reminder message is generated for manual verification of the forage feeding data.
[0025] Wherein, after the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral further includes:
[0026] Acquire a first treatment data item list corresponding to the first treatment data, a photographed image of the drug feeding and / or operation process when treating the target dairy cow corresponding to the second treatment data, and drug feeding operation record information of a drug feeding person;
[0027] A third reminder message is generated for manual verification of drug feeding data.
[0028] On the one hand, the present invention provides a risk identification device using dairy cows as collateral, comprising:
[0029] An acquisition unit is used to identify the photographed images corresponding to the target cow face at different times of the target cow used as collateral based on a preset cow health status recognition model to obtain a health status recognition result of the target cow;
[0030] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0031] a correction unit, configured to correct a predetermined actual mortgage amount of the target cow according to the health identification status result, first forage data provided by a supplier to a target breeding farm where the target cow is located and first treatment data provided by a medical staff to the target cow obtained from the blockchain, so as to obtain a first corrected mortgage amount;
[0032] an identification unit, configured to generate a first risk warning message indicating that the amount of the collateral is insufficient if it is determined that the sum of the first revised collateral amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual collateral amount;
[0033] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0034] In another aspect, an embodiment of the present invention provides an electronic device, including: a processor, a memory, and a bus, wherein:
[0035] The processor and the memory communicate with each other via the bus;
[0036] The memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the following method:
[0037] Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times used as collateral are recognized to obtain the health status recognition result of the target cow;
[0038] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0039] According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount;
[0040] If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0041] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0042] An embodiment of the present invention provides a non-transitory computer-readable storage medium, including:
[0043] The non-transitory computer-readable storage medium stores computer instructions, which cause the computer to execute the following method:
[0044] Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times used as collateral are recognized to obtain the health status recognition result of the target cow;
[0045] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0046] According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount;
[0047] If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0048] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0049] The risk identification method and device for using a cow as collateral provided by an embodiment of the present invention identify the photographed images corresponding to the target cow face at different times of the target cow used as collateral based on a preset cow health status identification model to obtain the health identification status result of the target cow; wherein the preset cow health status identification model is obtained by training a machine learning model based on cow health status identification sample data; according to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located obtained from the blockchain, and the first treatment data provided by the medical staff to the target cow, the actual collateral amount of the target cow determined in advance is corrected to obtain a first corrected collateral amount; if it is determined that the sum of the first corrected collateral amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual collateral amount, a first risk warning message indicating that the collateral amount is insufficient is generated; wherein the cumulative milk production is the cumulative milk production of the target cow from the date on which the target cow is used as collateral, which can accurately identify the risk of using the cow as collateral, ensure the safety and reliability of relevant data, and help to discover and avoid risks in breeding management. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:
[0051] Figure 1 It is a flow chart of a risk identification method using dairy cows as collateral provided by an embodiment of the present invention.
[0052] Figure 2 It is a schematic diagram of the structure of a risk identification device using dairy cows as collateral provided by an embodiment of the present invention.
[0053] Figure 3 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0054] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention. It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other arbitrarily.
[0055] Figure 1 FIG. 1 is a flow chart of a method for identifying risks using dairy cows as collateral provided by an embodiment of the present invention. Figure 1 As shown, the risk identification method using dairy cows as collateral provided by the embodiment of the present invention includes:
[0056] Step S1: based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times used as collateral are recognized to obtain the health status recognition result of the target cow;
[0057] The preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data.
[0058] Step S2: Based on the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the actual mortgage amount of the target cow determined in advance is revised to obtain a first revised mortgage amount.
[0059] Step S3: if it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0060] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0061] In the above step S1, the device identifies the photographed images corresponding to the target cow face at different times used as collateral based on the preset cow health status recognition model to obtain the health status recognition result of the target cow;
[0062] The preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data. The device may be a computer device that executes the method, for example, it may include a server, and specifically may include a server node in a blockchain. It should be noted that the acquisition and analysis of data involved in the embodiments of the present invention are authorized by the user.
[0063] The nodes in the blockchain may include bank nodes, dairy farming company nodes, and third-party service nodes, wherein the third-party service nodes may include forage supply company nodes and veterinary medicine company nodes, etc. Some steps in the method of the present invention may be executed in the bank node server, and other steps may be executed in the dairy farming company node. The corresponding server node may be authorized to access the access rights to obtain relevant data and provide the authority to provide relevant data. The relevant data exists in the blockchain, which can ensure that the data will not be tampered with. When data tracing is required, real and accurate data can be obtained safely and reliably.
[0064] The target cow can be one cow or multiple cows, and the time can be determined according to the actual situation. For example, the face of each target cow can be photographed by a camera at every hour of the 24 hours a day, and then the photographed image can be recognized.
[0065] Taking a target cow as an example, multiple cow face pictures taken at 0 o'clock of the cow can be input into a preset cow health status recognition model, and the health recognition status result output by the preset cow health status recognition model can be used as the health recognition status result of the cow at that moment.
[0066] Alternatively, multiple cow face pictures taken at 0 o'clock may be saved, multiple cow face pictures taken at 1 o'clock may be saved, and so on, multiple cow face pictures taken at 24 o'clock may be obtained and saved, and these 24 sets of pictures may be simultaneously input into a preset cow health status recognition model, and the health recognition status result output by the preset cow health status recognition model may be used as the health recognition status result of the cow on that day. The health recognition status result may specifically include excellent, good, average, slightly poor, and very poor, without specific limitation.
[0067] There is no limitation on the specific model of the machine learning model, and the training method can be selected as a conventional training method in the field.
[0068] The preset cow health status recognition model is obtained by training a machine learning model according to the cow health status recognition sample data, including:
[0069] Acquire a cow data set; the cow data set includes a plurality of captured image samples corresponding to the cow faces of a plurality of cows corresponding to each breed, and physiological parameter detection data of a plurality of cows corresponding to each breed; acquire a plurality of captured image samples corresponding to the cow faces of a plurality of cows corresponding to each breed, including:
[0070] Perform initial image recognition on the original captured images, remove noise images, and use the retained images as multiple captured image samples corresponding to the cow faces of multiple dairy cows corresponding to each breed. Initial image recognition may include identifying whether the initial outline of the cow face is clear and visible. If it is not clear, it means that it is a noise image and is removed. Initial image recognition can be achieved using existing technologies.
[0071] Obtain physiological parameter test data for multiple cows corresponding to each breed, including:
[0072] By reading the data in the smart neck ring pre-installed on the neck of each cow, the physiological parameter detection data of multiple cows corresponding to each breed are obtained. The data in the smart neck ring may include the cow's movement information, heart rate, blood pressure and other data. The physiological parameter detection data may include the data in the above-mentioned smart neck ring, and may also include data calculated by using the data in the smart neck ring as intermediate data.
[0073] The health status analysis is performed on the physiological parameter detection data of multiple dairy cows corresponding to each breed, and multiple image samples with normal health status are used as positive samples, and multiple image samples with abnormal health status are used as negative samples. The ratio of the number of positive samples and negative samples can be set independently according to actual conditions to prevent overfitting in data training.
[0074] The preset cow health status recognition model is obtained by training the machine learning model according to the positive samples and the negative samples. After the preset cow health status recognition model is trained, the cow health status can be directly identified by photographing the cow's face, without the need to use a smart neck ring, which each cow needs to wear and is expensive, while the camera is cheap, and multiple cows can share one camera, which reduces the overall feeding and management costs.
[0075] In addition, not using smart neck collars can also make cows more comfortable and help increase milk production.
[0076] In the above step S2, the device amends the predetermined actual mortgage amount of the target cow according to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located obtained from the blockchain, and the first treatment data provided by the medical staff to the target cow, to obtain a first amended mortgage amount. Determining the actual mortgage amount of the target cow may include:
[0077] The basic static information and basic dynamic information of the target cow are obtained, wherein the basic static information may include the family pedigree of the cow that bred the target cow and the age of the target cow. It can be understood that the family pedigree of the target cow is fixed data, and the age of the target cow is also relatively static data. During the mortgage period, the basic static information can be considered to be fixed data.
[0078] The basic dynamic information may include the target physiological parameter detection data of the target cow, and the category evaluation index for measuring the evaluation price of the target cow, such as the GTPI index and the PTAM index, etc. It can be understood that the target physiological parameter detection data is changing dynamically every day or even every moment, and the GTPI index and the PTAM index, etc., are not fixed data during the mortgage period.
[0079] The basic static information and basic dynamic information of the target cow can be input into the cow valuation calculation model, and the actual mortgage amount of the target cow can be output through the cow valuation calculation model.
[0080] In order to fully consider the risks that the collateral brings to the bank, it can usually be assumed that the actual collateral amount will not increase over time, but will decrease over time. However, the speed of decrease over time is affected by many factors, including the health status of the target dairy cow, the quality of the forage food it eats, whether it is sick, whether the treatment plan is appropriate after the disease occurs, and if drugs are used in the treatment, the side effects of the drugs, etc.
[0081] Therefore, a properly constructed first-modification mortgage amount will usually be less than the actual mortgage amount.
[0082] The first forage data may include forage types, such as ryegrass and Sudan grass, and may also include forage supply time. It can be understood that the closer the forage supply time is to the current time, the fresher the forage is and the higher the forage quality is.
[0083] The first treatment data may include the type of disease confirmed for the target dairy cow, the severity, the treatment drugs provided, and the dosage, etc.
[0084] The health identification status results may specifically include excellent, good, average, slightly poor and very poor, without specific limitation.
[0085] For the health identification status result, the first forage data, and the first treatment data, corresponding weights can be set respectively to evaluate the first deduction point corresponding to the health identification status, the second deduction point corresponding to the forage, and the third deduction point corresponding to the treatment. For each data item in the health identification status result, the first forage data, and the first treatment data, corresponding sub-weights can also be set to accurately evaluate the first deduction point, the second deduction point, and the third deduction point. The first modified mortgage amount is calculated according to the following expression:
[0086] The first revised mortgage amount = actual mortgage amount - (first deduction + second deduction + third deduction).
[0087] In the above step S3, if the device determines that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0088] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0089] You can generate a digital cow on the display interface and add a digital cow ID and the prompt "Insufficient collateral amount".
[0090] A digital cow refers to a virtual cow image corresponding to a real dairy cow, and the digital cow ID may include a digital cow number for distinguishing different digital cows.
[0091] It is understandable that the ranch owner raising dairy cows can negotiate with the bank to use the sales of milk produced by the target dairy cows used for mortgage to repay the actual mortgage amount in advance. The target dairy cows used for mortgage can still be raised by the ranch owner who originally raised the dairy cows at the original breeding place, but the target dairy cows are now used for mortgage. The ranch owner who raises dairy cows can sell the milk produced by the target dairy cows and use the selling price to offset the actual mortgage amount of the target dairy cows.
[0092] It can be understood that as long as the target cow can produce milk normally, the cumulative milk production is a variable that increases over time.
[0093] Because uncertainty increases over time, the first amendment mortgage amount is a variable that decreases over time.
[0094] When the target cow is just determined as collateral, the first revised mortgage amount is closer to the actual mortgage amount, but the cumulative milk production is also a small value. Therefore, the sum of the first revised mortgage amount and the corresponding sales price of the target cow's cumulative milk production will usually be greater than or equal to the actual mortgage amount.
[0095] As time goes by, as long as the target cow can produce milk normally, the corresponding selling price of the cumulative milk production of the target cow will continue to increase. As long as the target cow is in good health, there is no problem with the quality of the supplied forage, and the target cow is not sick, the first revised mortgage amount will not be significantly reduced. Therefore, the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow can be maintained greater than the actual mortgage amount.
[0096] However, if the above-mentioned target cows cannot produce milk normally, the target cows are in poor health, there are problems with the quality of the supplied forage, or the target cows are sick or have one or more illnesses, the sum of the first revised mortgage amount and the cumulative sales price corresponding to the target cows' milk production will be less than the actual mortgage amount, and there will be a risk of insufficient mortgage amount.
[0097] It is understandable that one or more of the above items may fluctuate to a certain extent. For example, the target cow may not produce milk for one or two days, but may return to normal on the third day. Or the target cow may be in a bad mental state due to stress today, but may return to normal tomorrow. This may lead to excessive sensitivity in risk identification. To solve this problem, the method can be optimized, including:
[0098] If it is determined that the difference between the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow and the actual mortgage amount exceeds the preset limit, a first risk warning message indicating that the amount of the mortgage is insufficient is generated, and the preset limit can be set independently according to the actual situation. That is, when the above sum is much different from the actual mortgage amount, a risk warning is issued. If the above sum is less different from the actual mortgage amount, it means that the data fluctuates within an acceptable range and no risk warning is required.
[0099] When generating the first risk warning message, the bank may adopt existing risk control measures to handle it.
[0100] The risk identification method using dairy cows as collateral also includes:
[0101] The target physiological parameter detection data of the target cow and the second treatment data actually provided to the target cow are obtained from the target farm management system, and the photographed image of the grass fed to the target cow is identified to obtain the second grass data actually fed to the target cow; the target farm management system can be a management system used by the ranch owner who raises dairy cows. The target farm management system can obtain the data from the smart neck ring on the neck of each cow and store it in the target farm management system.
[0102] The second treatment data actually provided can be understood as the name of the drug, drug manufacturer and drug batch used in the treatment. Since the drugs in the first treatment data may be discontinued, or the target cow may suddenly become ill and there is no time to purchase the drugs in the first treatment data, the second treatment data may be different from the first treatment data. The staff who treat the target cow can obtain the second treatment data through work experience, reference materials or consultation with experts, first treat the sick target cow, and then enter the second treatment data into the target farm management system.
[0103] The camera installed in the farm is used to capture images of the grass being fed to the target cows. Then, the image recognition method can be used to perform image recognition on the images of the grass being fed. The second grass data reflecting the quality of the grass, such as the grass type and freshness, can be obtained through the image recognition result. It is understandable that different breeds of cows are raised in the farm, and different breeds of cows are fed different grass types. The grass feeders may easily feed the target cows with grass types that are not suitable for feeding due to negligence.
[0104] In addition, because the farms purchase forage in bulk, the forage that is in stock for a long time is not suitable for feeding the target cows, and it is easy for the forage feeders to feed this forage to the target cows.
[0105] According to the target physiological parameter detection data, the second forage data and the second treatment data, the actual mortgage amount of the predetermined target cow is corrected to obtain a second corrected mortgage amount; the correction method can refer to the description of the first corrected mortgage amount mentioned above. Since there is a difference between the data here and the corresponding data of the corrected first corrected mortgage amount, there will also be a difference between the second corrected mortgage amount and the first corrected mortgage amount.
[0106] If it is determined that the difference between the second revised mortgage amount and the first revised mortgage amount is greater than a preset difference threshold, a second risk warning message is generated to indicate that the risk is caused by the dairy cow breeding management. The preset difference threshold can be set independently according to actual conditions. If it is determined that the difference between the second revised mortgage amount and the first revised mortgage amount is less than or equal to the preset difference threshold, it means that there is no breeding risk caused by the dairy cow breeding management.
[0107] After the step of generating a second risk warning message indicating the risk caused by the dairy cow breeding management, the risk identification method using the dairy cow as collateral further includes:
[0108] According to the pre-established correspondence between the preset physiological parameter detection data and the preset cow health identification status result, the target cow health identification status result corresponding to the target physiological parameter detection data is determined; that is, another method is used to determine the target cow health identification status result according to the target physiological parameter detection data.
[0109] The target cow health identification result is compared with the health identification result, and if it is determined that the difference between the comparison results is outside the preset difference range, a first reminder message for manual troubleshooting of related equipment is generated. The target cow health identification result obtained by the other method is compared with the result obtained by the preset cow health identification model.
[0110] The preset difference range can be set independently according to the actual situation. Referring to the above description, if the comparison results are of different levels, it means that the difference is outside the preset difference range. For example, if the health identification status result of the target cow is good, then as long as the health identification status result is not good, the comparison results are considered to be of different levels.
[0111] It should be noted that since the preset cow health status recognition model has been trained, verified and evaluated before use, the model accuracy can be considered accurate. Therefore, if there is a problem with the health status recognition result, it can be considered that the camera that captures the cow's face image is faulty; if there is a problem with the health status recognition result of the target cow, it can be considered that the neck ring that generates physiological parameter detection data is faulty. However, which of the two types of faults it is requires manual troubleshooting of related equipment to determine.
[0112] If the difference between the comparison results is within the preset difference range, it means that the risk problem of dairy cow breeding and management is not caused by the target physiological parameter detection data and health identification status results.
[0113] After the step of generating a second risk warning message indicating the risk caused by the dairy cow breeding management, the risk identification method using the dairy cow as collateral further includes:
[0114] Obtain a first forage data item list corresponding to the first forage data, a photographed image of feeding forage when feeding the target cow corresponding to the second forage data, and feeding forage operation record information of the feeder; the first forage data item list may include relevant data items of the above-mentioned forage types and forage freshness.
[0115] Generate a second reminder message for manual verification of the forage feeding data. By manually verifying the captured image of the forage feeding when feeding the target cow and the first forage data item list, it can be determined whether the forage feeder feeds the wrong type of forage or stale forage; by manually verifying the forage feeding operation record information, it can be determined whether the forage feeder feeds the right method.
[0116] If the feeding forage data is manually verified to be correct, it means that the risk issues in dairy cow breeding and management are not caused by the first forage data and the second forage data.
[0117] After the step of generating a second risk warning message indicating the risk caused by the dairy cow breeding management, the risk identification method using the dairy cow as collateral further includes:
[0118] Obtain a first treatment data item list corresponding to the first treatment data, a photographed image of the drug feeding and / or operation process when treating the target cow corresponding to the second treatment data, and drug feeding operation record information of the drug feeding personnel; the first treatment data item list may include relevant data items such as the type of disease confirmed for the target cow, the severity, the provided treatment drugs and dosage.
[0119] Generate a third reminder message for manual verification of drug feeding data. By manually verifying the feeding drugs and the first treatment data item list when treating the target cow, it can be determined whether the drug feeder has fed the wrong drug type and dosage, etc.; by manually verifying the captured images of the operation process, it can be determined whether the feeding method of the drug feeder is appropriate.
[0120] If the drug feeding data is manually verified to be correct, it means that the risk problem in dairy cow breeding and management is not caused by the first treatment data and the second treatment data.
[0121] The risk identification method for using a cow as collateral provided in an embodiment of the present invention identifies the photographed images corresponding to the target cow face at different times of the target cow used as collateral based on a preset cow health status identification model to obtain the health identification status result of the target cow; wherein the preset cow health status identification model is obtained by training a machine learning model based on cow health status identification sample data; according to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located obtained from the blockchain, and the first treatment data provided by the medical staff to the target cow, the actual collateral amount of the target cow determined in advance is corrected to obtain a first corrected collateral amount; if it is determined that the sum of the first corrected collateral amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual collateral amount, a first risk warning message is generated to prompt that the collateral amount is insufficient; wherein the cumulative milk production is the cumulative milk production of the target cow from the date on which the target cow is used as collateral, which can accurately identify the risk of using the cow as collateral, ensure the safety and reliability of relevant data, and help to discover and avoid risks in breeding management.
[0122] Furthermore, the preset cow health status recognition model is obtained by training a machine learning model according to the cow health status recognition sample data, including:
[0123] Acquire a cow data set; the cow data set includes a plurality of captured image samples corresponding to the faces of a plurality of cows corresponding to each breed, and physiological parameter detection data of a plurality of cows corresponding to each breed; the above-mentioned embodiment can be referred to for description and will not be repeated here.
[0124] The health status analysis is performed on the physiological parameter detection data of multiple cows corresponding to each breed, and multiple image samples with normal health status are used as positive samples, and multiple image samples with abnormal health status are used as negative samples; the above embodiment can be referred to for description and will not be repeated here.
[0125] The preset cow health status recognition model is obtained by training the machine learning model according to the positive samples and the negative samples. The above-mentioned embodiments can be referred to for explanation and will not be described in detail.
[0126] Furthermore, the risk identification method using dairy cows as collateral also includes:
[0127] The target physiological parameter detection data of the target cow and the second treatment data actually provided to the target cow are obtained from the target farm management system, and the photographed image of the grass fed to the target cow when feeding the target cow is identified to obtain the second grass data actually fed to the target cow; the above-mentioned embodiment can be referred to for description and will not be repeated here.
[0128] According to the target physiological parameter detection data, the second forage data and the second treatment data, the actual mortgage amount of the predetermined target cow is corrected to obtain a second corrected mortgage amount; the above-mentioned embodiment can be referred to for description and will not be repeated here.
[0129] If it is determined that the difference between the second revised mortgage amount and the first revised mortgage amount is greater than a preset difference threshold, a second risk warning message is generated to indicate the risk caused by dairy cow breeding management.
[0130] Furthermore, after the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral also includes:
[0131] According to the pre-established correspondence between the preset physiological parameter detection data and the preset cow health identification status result, the target cow health identification status result corresponding to the target physiological parameter detection data is determined; the above embodiment can be referred to for description and will not be repeated here.
[0132] The target cow health identification result is compared with the health identification result, and if it is determined that the difference between the comparison results is outside the preset difference range, a first reminder message for manual troubleshooting of related equipment is generated. The above-mentioned embodiment can be referred to for explanation and will not be repeated.
[0133] Furthermore, after the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral also includes:
[0134] Obtain a first forage data item list corresponding to the first forage data, a photographed image of feeding forage when feeding the target cow corresponding to the second forage data, and the feeding operation record information of the feeder; refer to the above embodiment description and no further details will be given.
[0135] Generate a second reminder message for manually verifying the forage feeding data. Please refer to the above embodiment for description, and no further details will be given.
[0136] Furthermore, after the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral also includes:
[0137] Obtain a first treatment data item list corresponding to the first treatment data, captured images of the drug feeding and / or operation process when treating the target cow corresponding to the second treatment data, and drug feeding operation record information of the drug feeding personnel; refer to the above embodiment for description and no further details are given.
[0138] Generate a third reminder message for manual verification of drug feeding data. Please refer to the above embodiment for description, which will not be repeated here.
[0139] Figure 2 FIG. 1 is a schematic diagram of a risk identification device using a dairy cow as collateral provided by an embodiment of the present invention. Figure 2 As shown, the risk identification device using dairy cows as collateral provided by the embodiment of the present invention includes an acquisition unit 201, a correction unit 202 and an identification unit 203, wherein:
[0140] The acquisition unit 201 is used to identify the photographed images corresponding to the target cow face of the target cow used as collateral at different times based on a preset cow health status recognition model, and obtain the health recognition status result of the target cow; wherein the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data; the correction unit 202 is used to correct the predetermined actual mortgage amount of the target cow according to the health recognition status result, the first forage data provided by the supplier to the target farm where the target cow is located obtained from the blockchain, and the first treatment data provided by the medical staff to the target cow, to obtain a first corrected mortgage amount; the identification unit 203 is used to generate a first risk warning message indicating that the collateral amount is insufficient if it is determined that the sum of the first corrected mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount; wherein the cumulative milk production is the cumulative milk production of the target cow from the date on which it is used as collateral.
[0141] Specifically, the acquisition unit 201 in the device is used to identify the photographed images corresponding to the target cow face of the target cow used as collateral at different times based on a preset cow health status recognition model, and obtain the health recognition status result of the target cow; wherein the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data; the correction unit 202 is used to correct the predetermined actual mortgage amount of the target cow according to the health recognition status result, the first forage data provided by the supplier to the target farm where the target cow is located obtained from the blockchain, and the first treatment data provided by the medical staff to the target cow, to obtain a first corrected mortgage amount; the identification unit 203 is used to generate a first risk warning message indicating that the collateral amount is insufficient if it is determined that the sum of the first corrected mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount; wherein the cumulative milk production is the cumulative milk production of the target cow from the date on which it is used as collateral.
[0142] The risk identification device for using a cow as collateral provided by an embodiment of the present invention identifies photographed images corresponding to the target cow face at different times of the target cow used as collateral based on a preset cow health status identification model to obtain a health identification status result of the target cow; wherein the preset cow health status identification model is obtained by training a machine learning model based on cow health status identification sample data; according to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located obtained from the blockchain, and the first treatment data provided by the medical staff to the target cow, the actual collateral amount of the target cow determined in advance is corrected to obtain a first corrected collateral amount; if it is determined that the sum of the first corrected collateral amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual collateral amount, a first risk warning message is generated to prompt that the collateral amount is insufficient; wherein the cumulative milk production is the cumulative milk production of the target cow from the date on which the target cow is used as collateral, which can accurately identify the risk of using the cow as collateral, ensure the safety and reliability of relevant data, and help to discover and avoid risks in breeding management.
[0143] The embodiment of the present invention provides an embodiment of a risk identification device using dairy cows as collateral, which can be specifically used to execute the processing flow of the above-mentioned method embodiments. Its functions are not repeated here, and reference can be made to the detailed description of the above-mentioned method embodiments.
[0144] Figure 3 A schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as Figure 3 As shown, the electronic device includes: a processor (processor) 301, a memory (memory) 302 and a bus 303;
[0145] The processor 301 and the memory 302 communicate with each other via a bus 303;
[0146] The processor 301 is used to call the program instructions in the memory 302 to execute the methods provided by the above method embodiments, for example, including:
[0147] Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times of the target cow used as collateral are recognized to obtain the health status recognition result of the target cow;
[0148] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0149] According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount;
[0150] If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0151] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0152] This embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can perform the methods provided by the above method embodiments, for example, including:
[0153] Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times of the target cow used as collateral are recognized to obtain the health status recognition result of the target cow;
[0154] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0155] According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount;
[0156] If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0157] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0158] This embodiment provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program enables the computer to execute the methods provided by the above method embodiments, for example, including:
[0159] Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times of the target cow used as collateral are recognized to obtain the health status recognition result of the target cow;
[0160] Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data;
[0161] According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount;
[0162] If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated;
[0163] The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
[0164] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0165] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0166] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0168] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0169] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. A risk identification method using dairy cows as collateral, characterized in that: include: Based on a preset cow health status recognition model, the photographed images corresponding to the target cow face at different times used as collateral are recognized to obtain the health status recognition result of the target cow; Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data; According to the health identification status result, the first forage data provided by the supplier to the target farm where the target cow is located and the first treatment data provided by the medical staff to the target cow obtained from the blockchain, the predetermined actual mortgage amount of the target cow is revised to obtain a first revised mortgage amount; If it is determined that the sum of the first revised mortgage amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual mortgage amount, a first risk warning message indicating that the amount of the mortgage is insufficient is generated; The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
2. The risk identification method using dairy cows as collateral according to claim 1, characterized in that: The preset cow health status recognition model is obtained by training a machine learning model according to the cow health status recognition sample data, including: Acquire a dairy cow data set; the dairy cow data set includes a plurality of photographed image samples corresponding to the cow faces of a plurality of dairy cows corresponding to each breed, and physiological parameter detection data of a plurality of dairy cows corresponding to each breed; Perform health status analysis on physiological parameter detection data of multiple dairy cows corresponding to each breed, use multiple image samples with normal health status as positive samples, and use multiple image samples with abnormal health status as negative samples; The preset cow health status recognition model is obtained by training a machine learning model according to the positive samples and the negative samples.
3. The risk identification method using dairy cows as collateral according to claim 1 or 2, characterized in that: The risk identification method using dairy cows as collateral also includes: Acquire target physiological parameter detection data of the target cow and second treatment data actually provided to the target cow from a target farm management system, and recognize photographed images of forage fed to the target cow when feeding the target cow, to obtain second forage data actually fed to the target cow; According to the target physiological parameter detection data, the second forage data and the second treatment data, the predetermined actual mortgage amount of the target dairy cow is revised to obtain a second revised mortgage amount; If it is determined that the difference between the second revised mortgage amount and the first revised mortgage amount is greater than a preset difference threshold, a second risk warning message is generated to indicate the risk caused by dairy cow breeding management.
4. The risk identification method using dairy cows as collateral according to claim 3 is characterized in that: After the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral further includes: Determine a target cow health identification status result corresponding to the target physiological parameter detection data according to a pre-established correspondence relationship between preset physiological parameter detection data and preset cow health identification status result; The target cow health identification result is compared with the health identification result, and if it is determined that the difference between the comparison results is outside a preset difference range, a first reminder message is generated for manual troubleshooting of related equipment.
5. The risk identification method using dairy cows as collateral according to claim 3 is characterized in that: After the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral further includes: Acquire a first forage data item list corresponding to the first forage data, a photographed image of feeding forage when feeding the target cow corresponding to the second forage data, and forage feeding operation record information of a forage feeder; A second reminder message is generated for manual verification of the forage feeding data.
6. The risk identification method using dairy cows as collateral according to claim 3 is characterized in that: After the step of generating a second risk warning message indicating the risk caused by the feeding and management of dairy cows, the risk identification method using dairy cows as collateral further includes: Acquire a first treatment data item list corresponding to the first treatment data, a photographed image of the drug feeding and / or operation process when treating the target dairy cow corresponding to the second treatment data, and drug feeding operation record information of a drug feeding person; A third reminder message is generated for manual verification of drug feeding data.
7. A risk identification device using dairy cows as collateral, characterized in that: include: An acquisition unit is used to identify the photographed images corresponding to the target cow face at different times of the target cow used as collateral based on a preset cow health status recognition model to obtain a health status recognition result of the target cow; Wherein, the preset cow health status recognition model is obtained by training a machine learning model based on cow health status recognition sample data; a correction unit, configured to correct a predetermined actual mortgage amount of the target cow according to the health identification status result, first forage data provided by a supplier to a target breeding farm where the target cow is located and first treatment data provided by a medical staff to the target cow obtained from the blockchain, so as to obtain a first corrected mortgage amount; an identification unit, configured to generate a first risk warning message indicating that the amount of the collateral is insufficient if it is determined that the sum of the first revised collateral amount and the corresponding selling price of the cumulative milk production of the target cow is less than the actual collateral amount; The cumulative milk production is the cumulative milk production from the date when the target dairy cow is used as collateral.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.
