Agricultural insurance platform based on block chain

Abstract

The invention relates to the technical field of block chains, in particular to an agricultural insurance platform based on a block chain, and the platform comprises a collection module which is mainly used for collecting the image, remote sensing and environment information of crops, and carrying out the storage of the block chain; the insurance policy storage module is mainly used for storing an insurance policy and generating a claim settlement threshold value and a claim payment level; the judgment module is mainly used for judging the disaster probability, distinguishing natural and non-natural factors, removing human factors and determining the natural disaster area; the loss assessment module is mainly used for assessing the actual disaster area; the claim settlement module is mainly used for calculating a claim amount and transferring the claim amount to a peasant household account; the early warning module is mainly used for sending disaster early warning information to farmers; the data analysis module is mainly used for analyzing data and displaying disaster and claim settlement trends. According to the invention, natural factors and human factors in the affected area of crops can be determined more accurately, and fairness and rationality of claim settlement are ensured.

Description

Technical Field

[0001] The present invention relates to the field of blockchain technology, and in particular to an agricultural insurance platform based on blockchain. Background Art

[0002] As an important means to ensure the stability of agricultural production, agricultural insurance plays an important role in protecting farmers' interests and promoting agricultural development. However, the traditional agricultural insurance claims process is cumbersome, data is easy to tamper with, and lacks transparency, which seriously affects the efficiency and credibility of agricultural insurance. Blockchain technology, as a distributed ledger technology, has the characteristics of data immutability, high transparency, and strong traceability, providing a new idea for solving the above problems.

[0003] However, when assessing losses, existing blockchain-based agricultural insurance platforms are affected by human factors such as false reporting and exaggeration of losses, resulting in inaccurate claims assessments, which in turn affects the fairness and sustainability of agricultural insurance.

[0004] To sum up, how to solve the problem of inaccurate damage assessment caused by human factors in agricultural insurance claims has become a difficult problem that needs to be solved urgently in this field. Therefore, it is necessary to propose an agricultural insurance platform based on blockchain. Summary of the invention

[0005] To solve the above problems, the present invention provides a blockchain-based agricultural insurance platform, which can utilize the advantages of blockchain technology, combined with advanced image recognition and big data analysis technology, to achieve accurate identification and damage assessment of crop disasters, and improve the accuracy and efficiency of agricultural insurance claims.

[0006] In order to achieve the above purpose, the technical solution of the present invention is as follows: an agricultural insurance platform based on blockchain, comprising:

[0007] The acquisition module is used to use a number of cameras and drones deployed in the crop planting area to collect images of crops and remote sensing information of crops in the covered area, generate image information of crops and remote sensing information of crops, and store and fix the image information of crops and remote sensing information of crops through blockchain.

[0008] The collection module includes a sensor collection unit, which is used to use temperature and humidity sensors, wind speed sensors, light sensors and soil pH sensors to be deployed in the crop planting area, collect environmental temperature and humidity, environmental wind speed, environmental light and soil pH of the crops in the covered area, generate crop growth environment information and crop growth soil information, and store and fix the crop growth environment information and crop growth soil information through blockchain;

[0009] The insurance policy storage module is used to store the insurance policies signed between the insurance company and the customer, extract the claim conditions and compensation amounts on the insurance policies, and generate the customer's crop damage claim threshold and compensation level amount.

[0010] The judgment module is used to set the threshold of the probability of crop disasters.

[0011] When a disaster occurs, the pre-disaster crop growth environment information, crop growth soil information, crop image information and crop remote sensing information stored in the blockchain are read and statistical analysis is performed using statistical algorithms to generate the pre-disaster crop damage probability value.

[0012] When the crop disaster probability value before the disaster is higher than the crop disaster probability threshold, the area information of crops affected by natural factors is generated.

[0013] When the crop disaster probability value before the disaster is lower than the crop disaster probability threshold, the non-natural factor disaster situation information is generated.

[0014] The judgment module includes a human factor judgment unit, which is used to utilize the disaster situation information of non-natural factors and use a behavior analysis algorithm to remove the area information affected by human factors and generate the area information of crops affected by natural factors after removing the human factors.

[0015] The damage assessment module is used to assess the actual disaster-affected area of ​​crops based on the information on the area of ​​crops affected by natural factors after removing human factors, and generate actual disaster-affected area data.

[0016] The claims module is used to compare the actual disaster-stricken area data with the crop disaster claims threshold in the insurance policy storage module. When the actual disaster-stricken area data reaches the crop disaster claims threshold, the compensation amount is calculated using the compensation level amount, and the virtual currency corresponding to the compensation amount is transferred from the insurance company's virtual account to the farmer's virtual account.

[0017] Furthermore, the acquisition module includes a disaster occurrence judgment unit, which is used to use the YOLOv8 image recognition algorithm to analyze and identify the images of crops, determine whether a disaster occurs in the crop planting area, and send the disaster occurrence information to the judgment module when a disaster occurs.

[0018] Furthermore, the judgment module includes a verification unit, which is used to verify the integrity and authenticity of the data stored in the blockchain using a hash verification algorithm. When it is found that the data stored in the blockchain has been tampered with, the policy in the policy storage module will stop the agreement.

[0019] Furthermore, the statistical algorithm in the judgment module includes the following steps:

[0020] Step 1: Data preprocessing: Clean, convert and normalize the crop growth environment information, growth soil information, image information and remote sensing information read from the blockchain to generate preprocessed data.

[0021] Step 2: Feature extraction: Based on the preprocessed data, extract the key features related to crop damage; the key features include changes in temperature and humidity, wind speed, light intensity, soil pH, and vegetation index in remote sensing images.

[0022] Step 3: Model construction: Use the extracted key features to build a crop disaster probability prediction model.

[0023] Step 4: Calculation of disaster probability: Input the crop growth environment information, crop growth soil information, crop image information and crop remote sensing information before the disaster into the disaster probability prediction model, and the disaster probability prediction model generates the crop disaster probability value before the disaster through calculation.

[0024] Furthermore, the behavior analysis algorithm in the judgment module includes the following steps:

[0025] S1, first, uses the image information of the crop area before the disaster captured by the camera to detect whether there are any signs of human activities.

[0026] S2, using cameras to capture the state of crops after disaster.

[0027] S3, combined with the crop image information recorded by the camera over a long period of time, analyzes and determines whether the crops in the disaster-stricken area are in a long-term and gradually deteriorating death state, or in a sudden death state.

[0028] S4, combining the analysis results of S1, S2 and S3, if signs of human activities are detected and the state of crops after disaster meets the characteristics caused by human factors, further spatial analysis and time series analysis are used to accurately delineate the specific areas affected by human factors, and remove the area affected by human factors from the initially determined disaster area, ultimately generating information on the area of ​​crops affected by natural factors after removing human factors.

[0029] Furthermore, the state of crops after disasters includes the direction of crop collapse and the height of crop collapse; the long-term gradually deteriorating death state includes the gradual spread of pests and diseases and the depletion of soil nutrients; the sudden death state includes the natural destruction death state and the mechanical destruction death state.

[0030] Furthermore, the behavior analysis algorithm in the human factor judgment unit also includes a verification mechanism, which is used to generate a difference threshold and a verification threshold. The image information of the crops after the disaster is used to divide the crop area after the disaster into several areas, and the post-disaster state of the crops in each area is compared with the post-disaster state of the crops in the middle area to generate the difference between the post-disaster state of the crops in each area and the middle area. When the number of sub-areas whose differences are greater than the difference threshold is lower than the verification threshold, the natural factor judgment result is output; when the number of sub-areas whose differences are greater than the difference threshold is lower than the verification threshold, the human factor judgment result is output, and the area information of the crops affected by natural factors after removing the human factors is generated based on the natural factor judgment result and the human factor judgment result.

[0031] Furthermore, the behavior analysis algorithm in the judgment module also includes a soil pH anomaly detection step, and the soil pH anomaly detection step includes:

[0032] First, the soil pH data of the crop-growing areas before and after the disaster were obtained using the soil pH sensor of the sensor collection unit, and the changes in soil pH were compared and analyzed. When the soil pH changes significantly in a short period of time and is inconsistent with the local natural climate, geological conditions and the normal growth requirements of crops, it is regarded as a sign that the soil environment has been changed by human factors.

[0033] Subsequently, combined with the image information of crops taken by the camera, observe whether there are signs of turning over and adding substances on the soil surface, and whether there is soil erosion or accumulation caused by non-natural factors; at the same time, use remote sensing images obtained by drones to analyze the spatial distribution characteristics of changes in soil pH to determine whether it presents a regular pattern of human intervention. Regular patterns of human intervention include: strip-like and block-like distribution.

[0034] Finally, if the abnormal changes in soil pH are consistent with the signs of human activities reflected in the image information, and the spatial distribution characteristics also support the soil environment changes caused by human factors, then this part of the disaster-stricken area can be regarded as an area affected by human factors and excluded from the preliminary determined disaster-stricken area, ultimately generating information on the area of ​​crops affected by natural factors after removing human factors.

[0035] Furthermore, it also includes an early warning module, which is used to send disaster warning information to farmers and insurance companies when the judgment module finds that the probability value of crop disasters exceeds the threshold value of crop disaster probability.

[0036] Furthermore, it also includes a data analysis module, which is used to analyze the information on the area of ​​crops affected by natural factors, the growing environment information of crops, the growing soil information of crops, the crop damage claim threshold, the compensation level amount and the crop damage probability value before the disaster collected by the platform, obtain the analysis results, and present the analysis results in the form of charts; the analysis results include: crop damage situation and insurance claims trend.

[0037] The above scheme has the following beneficial effects:

[0038] 1. The present invention combines blockchain technology, image recognition and big data analysis technology to achieve accurate identification and damage assessment of crop disasters, effectively solves the problem of inaccurate damage assessment caused by human factors in traditional agricultural insurance claims, improves the accuracy and efficiency of agricultural insurance claims, protects the interests of farmers, and promotes the fairness and sustainability of agricultural insurance.

[0039] 2. The present invention uses the characteristics of blockchain technology that cannot be tampered with and has high transparency to ensure the authenticity and integrity of crop disaster data and claims information, avoid the risk of data being tampered with or forged, and enhance the credibility and credibility of agricultural insurance. At the same time, the data is verified through a hash verification algorithm, which further ensures the authenticity and integrity of the data and improves the reliability and security of agricultural insurance claims.

[0040] 3. The present invention also has an early warning function. When the probability value of crop disasters exceeds the threshold value of crop disaster probability, it can send disaster warning information to farmers and insurance companies in time, helping farmers to take timely measures to reduce disaster losses. At the same time, it also provides insurance companies with timely claims preparation, thereby improving the response speed and coping ability of agricultural insurance. In addition, the data analysis module can also analyze various types of data collected by the platform, display the crop disaster situation and insurance claims trends, and provide strong data support for the development of agricultural insurance.

[0041] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a structural diagram of the agricultural insurance platform based on blockchain in an embodiment of the present invention.

[0043] Figure 2 It is a structural diagram of the judgment module in an embodiment of the present invention. DETAILED DESCRIPTION

[0044] The following is further described in detail through specific implementation methods:

[0045] As attached Figure 1-Figure 2 As shown: A blockchain-based agricultural insurance platform, mainly composed of a collection module, a policy storage module, a judgment module, a damage assessment module, a claims module, an early warning module and a data analysis module; wherein the collection module is mainly used to collect images, remote sensing and environmental information of crops, and store them on the blockchain; the policy storage module is mainly used to store policies, generate claims thresholds and compensation levels; the judgment module is mainly used to judge the probability of disasters, distinguish between natural and non-natural factors, remove human factors, and determine the natural disaster-stricken area; the damage assessment module is mainly used to assess the actual disaster-stricken area; the claims module is mainly used to calculate the amount of compensation and transfer it to the farmer's account; the early warning module is mainly used to send disaster warning information to farmers; the data analysis module is mainly used to analyze data and display disaster and claims trends.

[0046] The functions of each module are described in detail below:

[0047] The acquisition module is used to use a number of cameras and drones deployed in the crop planting area to collect images of crops and remote sensing information of crops in the covered area, generate image information of crops and remote sensing information of crops, and store and fix the image information of crops and remote sensing information of crops through blockchain.

[0048] The collection module includes a sensor collection unit, which is used to use temperature and humidity sensors, wind speed sensors, light sensors and soil pH sensors to be deployed in the crop planting area to collect the environmental temperature and humidity, environmental wind speed, environmental light and soil pH of the crops in the covered area, generate crop growth environment information and crop growth soil information, and store and fix the crop growth environment information and crop growth soil information through blockchain.

[0049] The acquisition module includes a disaster occurrence judgment unit, which is used to use the YOLOv8 image recognition algorithm to analyze and identify images of crops to determine whether a disaster has occurred in the crop planting area. When a disaster occurs, the disaster occurrence information is sent to the judgment module.

[0050] Specifically, first, multiple high-definition cameras and drones are deployed in a certain crop planting area; these devices regularly collect images and remote sensing images of crops, and generate image information of crops and remote sensing image information of crops; at the same time, the sensor acquisition unit also uses temperature and humidity sensors, wind speed sensors and light sensors to monitor and record the growth environment information of crops in real time; and uses soil pH sensors to monitor and record the growth soil information of crops in real time.

[0051] The crop growth environment information and crop growth soil information are uploaded to the blockchain for evidence storage.

[0052] The insurance policy storage module is used to store the insurance policies signed between the insurance company and the customer, extract the claim conditions and compensation amounts on the insurance policies, and generate the customer's crop damage claim threshold and compensation level amount.

[0053] The judgment module is used to set the threshold of the probability of crop disasters.

[0054] When a disaster occurs, the pre-disaster crop growth environment information, crop growth soil information, crop image information and crop remote sensing information stored in the blockchain are read and statistical analysis is performed using statistical algorithms to generate the pre-disaster crop damage probability value.

[0055] When the crop disaster probability value before the disaster is higher than the crop disaster probability threshold, the area information of crops affected by natural factors is generated.

[0056] When the crop disaster probability value before the disaster is lower than the crop disaster probability threshold, the non-natural factor disaster situation information is generated.

[0057] The statistical algorithm in the judgment module includes the following steps:

[0058] Step 1: Data preprocessing: Clean, convert and normalize the crop growth environment information, growth soil information, image information and remote sensing information read from the blockchain to generate preprocessed data;

[0059] Step 2: Feature extraction: Based on the preprocessed data, extract key features related to crop damage; key features include changes in temperature and humidity, wind speed, light intensity, soil pH, and vegetation index in remote sensing images;

[0060] Step 3: Model construction: Using the extracted key features, a crop disaster probability prediction model is constructed;

[0061] Step 4: Calculation of disaster probability: Input the crop growth environment information, crop growth soil information, crop image information and crop remote sensing information before the disaster into the disaster probability prediction model, and the disaster probability prediction model generates the crop disaster probability value before the disaster through calculation.

[0062] The judgment module includes a verification unit, which is used to verify the integrity and authenticity of the data stored in the blockchain using a hash verification algorithm. When it is found that the data stored in the blockchain has been tampered with, the policy in the policy storage module will stop the agreement.

[0063] The judgment module includes a human factor judgment unit, which is used to utilize the disaster situation information of non-natural factors and use a behavior analysis algorithm to remove the area information affected by human factors and generate the area information of crops affected by natural factors after removing the human factors.

[0064] The behavior analysis algorithm in the human factor judgment unit includes the following steps:

[0065] S1, first, uses the image information of the crop area before the disaster captured by the camera to detect whether there are any signs of human activities.

[0066] S2, using cameras to capture the state of crops after disaster.

[0067] S3, combined with the crop image information recorded by the camera over a long period of time, analyzes and determines whether the crops in the disaster-stricken area are in a long-term and gradually deteriorating death state, or in a sudden death state.

[0068] S4, combining the analysis results of S1, S2 and S3, if signs of human activities are detected and the state of crops after disaster meets the characteristics caused by human factors, further spatial analysis and time series analysis are used to accurately delineate the specific areas affected by human factors, and remove the area affected by human factors from the initially determined disaster area, ultimately generating information on the area of ​​crops affected by natural factors after removing human factors.

[0069] The state of crops after disasters includes the direction and height of crops falling; the long-term gradually deteriorating death state includes the gradual spread of pests and diseases and the depletion of soil nutrients; the sudden death state includes the natural destruction death state and the mechanical destruction death state.

[0070] The behavior analysis algorithm in the human factor judgment unit also includes a verification mechanism, which is used to generate a difference threshold and a verification threshold. The image information of the crops after the disaster is used to divide the crop area after the disaster into several areas, and the post-disaster state of the crops in each area is compared with the post-disaster state of the crops in the middle area to generate the difference between the post-disaster state of the crops in each area and the middle area. When the number of sub-areas whose differences are greater than the difference threshold is lower than the verification threshold, the natural factor judgment result is output; when the number of sub-areas whose differences are greater than the difference threshold is lower than the verification threshold, the human factor judgment result is output, and the area information of the crops affected by natural factors after removing the human factors is generated based on the natural factor judgment result and the human factor judgment result.

[0071] The behavior analysis algorithm in the human factor judgment unit also includes a soil pH anomaly detection step, which includes:

[0072] First, the soil pH data of the crop planting area before and after the disaster were obtained by using the soil pH sensor of the sensor collection unit to compare and analyze the changes in soil pH. When the soil pH changes significantly in a short period of time and is inconsistent with the local natural climate, geological conditions and the normal growth requirements of crops, it is considered a sign that the soil environment has been changed by human factors.

[0073] Then, combined with the image information of crops taken by the camera, observe whether there are traces of turning and adding materials on the soil surface, and whether there is soil erosion or accumulation caused by non-natural factors; at the same time, use the remote sensing images obtained by the drone to analyze the spatial distribution characteristics of soil pH changes and determine whether it presents a regular pattern of human intervention. The regular pattern of human intervention includes: strip-shaped and block-shaped distribution;

[0074] Finally, if the abnormal changes in soil pH are consistent with the signs of human activities reflected in the image information, and the spatial distribution characteristics also support the soil environment changes caused by human factors, then this part of the disaster-stricken area can be regarded as an area affected by human factors and excluded from the preliminary determined disaster-stricken area, ultimately generating information on the area of ​​crops affected by natural factors after removing human factors.

[0075] Specifically, for example, in a corn-growing area, high-definition cameras and drones were deployed for monitoring, and temperature and humidity sensors, wind speed sensors, light sensors, and soil pH sensors were installed for data collection. After a disaster, the camera captured large areas of corn fields falling over, and the soil surface in some areas showed signs of turning over.

[0076] The pre-disaster image information captured by the camera showed that there were signs of vehicle and human activities in the area before the disaster, such as vehicles entering and exiting and people gathering.

[0077] After the disaster, the camera captured large areas of corn fields falling down in the same direction and at similar heights, showing characteristics of mechanical damage.

[0078] Combined with the image information recorded by the camera over a long period of time, it was found that the corn in the area did not show signs of dying due to the gradual spread of pests and diseases or the long-term deterioration of soil nutrient depletion before the disaster.

[0079] Combining the analysis results of S1, S2 and S3, it is determined that the corn lodging in this area is caused by human factors. Further spatial analysis and time series analysis are used to accurately define the specific areas affected by human factors, and the areas affected by human factors are removed from the initially determined disaster-stricken areas.

[0080] In the soil pH abnormality detection step, this embodiment takes the above-mentioned corn planting area as an example for explanation:

[0081] First: Using the soil pH sensor of the sensor collection unit to obtain soil pH data before and after the disaster, it was found that the soil pH in the area dropped sharply from the original suitable range (such as pH 6.5-7.5) to the acidic range (such as pH below 4.0) in a short period of time. This change is inconsistent with the local natural climate, geological conditions and the normal growth requirements of corn, and is therefore considered a sign that the soil environment has been changed by human factors.

[0082] Afterwards: Combined with the image information of crops taken by the camera, it was observed that there were traces of turning on the soil surface and residues suspected to be fertilizers or pesticides. These signs indicate that there was human intervention in the soil environment.

[0083] At the same time: using remote sensing images obtained by drones, we analyzed the spatial distribution characteristics of soil pH changes. We found that the soil pH changes in the region showed a strip-like distribution, and this regular pattern was consistent with the characteristics of human intervention (such as irrigation, fertilization, etc.).

[0084] Finally: Based on the above information, it is determined that the abnormal change in soil pH in this area is caused by human factors. Therefore, this part of the disaster-stricken area is regarded as an area affected by human factors and is excluded from the initially determined disaster-stricken area. Finally, the area of ​​crops affected by natural factors after removing human factors is generated, providing insurance companies with accurate claims basis.

[0085] The damage assessment module is used to assess the actual damage area of ​​crops based on the information of the area of ​​crops affected by natural factors after removing human factors, and generate data on the actual affected area;

[0086] The claims module is used to compare the actual disaster-stricken area data with the crop disaster claims threshold in the insurance policy storage module. When the actual disaster-stricken area data reaches the crop disaster claims threshold, the compensation amount is calculated using the compensation level amount, and the virtual currency corresponding to the compensation amount is transferred from the insurance company's virtual account to the farmer's virtual account.

[0087] Specifically, after all human factors are removed, the damage assessment module can accurately assess the actual area of ​​crops affected by natural factors after removing human factors, based on the information on the area of ​​crops affected by natural factors provided by the judgment module.

[0088] The damage assessment module will use advanced image processing and geographic information system technology to accurately divide the disaster-stricken areas and calculate the area. By comparing image information before and after the disaster and combining it with remote sensing images obtained by drones, the damage assessment module can accurately identify the type of crops affected, the degree of damage, and the affected area. At the same time, the damage assessment module will also consider factors such as the growth cycle and yield of crops to make a scientific and reasonable assessment of the affected area.

[0089] When the damage assessment module completes the damage assessment of the actual disaster-stricken area of ​​crops, the claims settlement module will compare the actual disaster-stricken area data with the crop damage claims threshold and compensation level amount stored in the policy storage module.

[0090] If the actual disaster-stricken area data reaches the crop damage claim threshold, the claim module will initiate the payment process.

[0091] In the compensation process, the claim module will calculate the amount of compensation to be paid based on the compensation level and the actual affected area. Then, the claim module will transfer the virtual currency corresponding to the compensation amount from the insurance company's virtual account to the farmer's virtual account. The completion of this step marks the end of the entire claim process, and the farmer can also get the compensation he deserves in a timely manner.

[0092] The early warning module is used to send disaster warning information to farmers and insurance companies when the judgment module finds that the probability value of crop disasters exceeds the threshold value of crop disaster probability.

[0093] The data analysis module is used to analyze the information on the area of ​​crops affected by natural factors, the growing environment of crops, the growing soil of crops, the threshold for compensation for crop damage, the amount of compensation level, and the probability of crop damage before the disaster collected by the platform, obtain the analysis results, and present the analysis results in the form of charts; the analysis results include: crop damage situation and insurance claims trend.

[0094] By combining blockchain technology, image recognition and big data analysis technology, the present invention realizes accurate identification and damage assessment of crop disasters, effectively solves the problem of inaccurate damage assessment caused by human factors in traditional agricultural insurance claims, improves the accuracy and efficiency of agricultural insurance claims, protects the interests of farmers, and promotes the fairness and sustainability of agricultural insurance.

[0095] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. An agricultural insurance platform based on blockchain, characterized in that: include: The acquisition module is used to use a number of cameras and drones deployed in the crop planting area to collect images of crops and remote sensing information of crops in the covered area, generate image information of crops and remote sensing information of crops, and store and fix the image information of crops and remote sensing information of crops through blockchain; The collection module includes a sensor collection unit, which is used to use temperature and humidity sensors, wind speed sensors, light sensors and soil pH sensors to be deployed in the crop planting area, collect environmental temperature and humidity, environmental wind speed, environmental light and soil pH of the crops in the covered area, generate crop growth environment information and crop growth soil information, and store and fix the crop growth environment information and crop growth soil information through blockchain; The policy storage module is used to store the policies signed between the insurance company and the customer, extract the claim conditions and compensation amounts on the policies, and generate the customer's crop damage claim threshold and compensation level amount; The judgment module is used to set the threshold of the probability of crop disasters; When a disaster occurs, the crop growth environment information, crop growth soil information, crop image information and crop remote sensing information stored in the blockchain before the disaster are read and statistical analysis is performed using statistical algorithms to generate the crop disaster probability value before the disaster; When the probability value of crop damage before the disaster is higher than the threshold value of crop damage, the area information of crops damaged by natural factors is generated; When the crop disaster probability value before the disaster is lower than the crop disaster probability threshold, the non-natural factor disaster information is generated; The judgment module includes a human factor judgment unit, which is used to use the disaster situation information of non-natural factors and a behavior analysis algorithm to remove the area information affected by human factors and generate the area information of crops affected by natural factors after removing the human factors; The damage assessment module is used to assess the actual damage area of ​​crops based on the information of the area of ​​crops affected by natural factors after removing human factors, and generate data on the actual affected area; The claims module is used to compare the actual disaster-stricken area data with the crop disaster claims threshold in the insurance policy storage module. When the actual disaster-stricken area data reaches the crop disaster claims threshold, the compensation amount is calculated using the compensation level amount, and the virtual currency corresponding to the compensation amount is transferred from the insurance company's virtual account to the farmer's virtual account.

2. The blockchain-based agricultural insurance platform according to claim 1, characterized in that: The acquisition module includes a disaster occurrence judgment unit, which is used to use the YOLOv8 image recognition algorithm to analyze and identify images of crops to determine whether a disaster has occurred in the crop planting area. When a disaster occurs, the disaster occurrence information is sent to the judgment module.

3. The blockchain-based agricultural insurance platform according to claim 2 is characterized in that: The judgment module includes a verification unit, which is used to verify the integrity and authenticity of the data stored in the blockchain using a hash verification algorithm. When it is found that the data stored in the blockchain has been tampered with, the policy in the policy storage module will stop the agreement.

4. The blockchain-based agricultural insurance platform according to claim 3 is characterized in that: The statistical algorithm in the judgment module includes the following steps: Step 1: Data preprocessing: Clean, convert and normalize the crop growth environment information, growth soil information, image information and remote sensing information read from the blockchain to generate preprocessed data; Step 2: Feature extraction: Based on the preprocessed data, extract key features related to crop damage; key features include changes in temperature and humidity, wind speed, light intensity, soil pH, and vegetation index in remote sensing images; Step 3: Model construction: Using the extracted key features, a crop disaster probability prediction model is constructed; Step 4: Calculation of disaster probability: Input the crop growth environment information, crop growth soil information, crop image information and crop remote sensing information before the disaster into the disaster probability prediction model, and the disaster probability prediction model generates the crop disaster probability value before the disaster through calculation.

5. The blockchain-based agricultural insurance platform according to claim 4 is characterized in that: The behavior analysis algorithm in the human factor judgment unit includes the following steps: S1, first, uses the image information of the crop area before the disaster captured by the camera to detect whether there are signs of human activities; S2, using cameras to capture the state of crops after disasters; S3, combining the crop image information recorded by the camera for a long time, analyzing and judging whether the crops in the disaster area are in a state of long-term and gradually deteriorating death or a state of sudden death; S4, combining the analysis results of S1, S2 and S3, if signs of human activities are detected and the state of crops after disaster meets the characteristics caused by human factors, further spatial analysis and time series analysis are used to accurately delineate the specific areas affected by human factors, and remove the area affected by human factors from the initially determined disaster area, ultimately generating information on the area of ​​crops affected by natural factors after removing human factors.

6. The blockchain-based agricultural insurance platform according to claim 5 is characterized in that: The state of crops after disasters includes the direction and height of crops falling; the long-term gradually deteriorating death state includes the gradual spread of pests and diseases and the depletion of soil nutrients; the sudden death state includes the natural destruction death state and the mechanical destruction death state.

7. The blockchain-based agricultural insurance platform according to claim 6, characterized in that: The behavior analysis algorithm in the human factor judgment unit also includes a verification mechanism, which is used to generate a difference threshold and a verification threshold. The image information of the crops after the disaster is used to divide the crop area after the disaster into several areas, and the post-disaster state of the crops in each area is compared with the post-disaster state of the crops in the middle area to generate the difference between the post-disaster state of the crops in each area and the middle area. When the number of sub-areas whose differences are greater than the difference threshold is lower than the verification threshold, the natural factor judgment result is output; when the number of sub-areas whose differences are greater than the difference threshold is lower than the verification threshold, the human factor judgment result is output, and the area information of the crops affected by natural factors after removing the human factors is generated based on the natural factor judgment result and the human factor judgment result.

8. The blockchain-based agricultural insurance platform according to claim 7 is characterized in that: The behavior analysis algorithm in the human factor judgment unit also includes a soil pH anomaly detection step, which includes: First, the soil pH data of the crop planting area before and after the disaster were obtained by using the soil pH sensor of the sensor collection unit to compare and analyze the changes in soil pH. When the soil pH changes significantly in a short period of time and is inconsistent with the local natural climate, geological conditions and the normal growth requirements of crops, it is considered a sign that the soil environment has been changed by human factors. Then, combined with the image information of crops taken by the camera, observe whether there are traces of turning and adding materials on the soil surface, and whether there is soil erosion or accumulation caused by non-natural factors; at the same time, use the remote sensing images obtained by the drone to analyze the spatial distribution characteristics of soil pH changes and determine whether it presents a regular pattern of human intervention. The regular pattern of human intervention includes: strip-shaped and block-shaped distribution; Finally, if the abnormal changes in soil pH are consistent with the signs of human activities reflected in the image information, and the spatial distribution characteristics also support the soil environment changes caused by human factors, then this part of the disaster-stricken area can be regarded as an area affected by human factors and excluded from the preliminary determined disaster-stricken area, ultimately generating information on the area of ​​crops affected by natural factors after removing human factors.

9. The blockchain-based agricultural insurance platform according to claim 8, characterized in that: It also includes an early warning module, which is used to send disaster warning information to farmers and insurance companies when the judgment module finds that the probability value of crop disasters exceeds the threshold value of crop disaster probability.

10. The blockchain-based agricultural insurance platform according to claim 9, characterized in that: It also includes a data analysis module, which is used to analyze the information on the area of ​​crops affected by natural factors, the growing environment of crops, the growing soil of crops, the compensation threshold for crop damage, the amount of compensation level, and the probability of crop damage before the disaster, and obtain the analysis results, and present the analysis results in the form of charts; The analysis results include: crop damage and insurance claims trends.

Images