Product risk resistance ability identification method and device, equipment and medium

Abstract

The application relates to an artificial intelligence technology and discloses a product risk resistance ability identification method, which comprises the following steps: obtaining a to-be-protected product, to-be-protected product coordinates and biological attributes of the to-be-protected product; calculating an initial risk resistance score of the to-be-protected product according to historical business records and a preset risk calculation strategy; obtaining geomorphic features in a field sampling set by using a geographical feature extraction network to obtain a geomorphic feature vector set; performing biological-environment matching on the geomorphic feature vector set, a quantized biological attribute vector and a coordinate vector by using an adaptive identification model to obtain an environment adaptation level; and performing weight calculation on the initial risk resistance score by using an environment score corresponding to the environment adaptation level to obtain a risk resistance ability value of the to-be-protected product. The application further discloses a product risk resistance ability identification device, an electronic device and a storage medium. The application can improve the accuracy of product risk control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a product risk resistance capability identification method and device, electronic equipment and a computer readable storage medium. BACKGROUND

[0002] In order to ensure the development of domestic grain industry, reduce the impact of natural disasters on agricultural production, and stabilize the income of farmers, in recent years, the area covered by agricultural insurance has rapidly expanded, resulting in a sharp increase in the number of agricultural insurance policies and an increasing number of agricultural compensation cases.

[0003] Currently, in the insurance industry, the risk resistance of the product to be protected is predicted by professional personnel on-site investigation, and the risk resistance level obtained from historical experience is corrected to formulate a reasonable insurance contract. However, through manual analysis, there are still situations such as not timely risk control and low risk detection accuracy, which increases the cost of risk control year by year. SUMMARY

[0004] The present application provides a product risk resistance capability identification method, device and computer readable storage medium, which mainly aims to solve the problem of low accuracy in identifying the risk resistance capability of the product.

[0005] To achieve the above purpose, the present application provides a product risk resistance capability identification method, which comprises:

[0006] Obtain the target coordinates of the product to be protected and the biological attributes of the product to be protected, and query the biological attributes of the product to be protected;

[0007] Obtain the historical business records of the product to be protected, and calculate the initial risk resistance score of the product to be protected according to the pre-constructed risk calculation strategy and the historical business records;

[0008] Obtain the field sampling data set of the product to be protected, and use the pre-trained geographical feature extraction network to extract the geomorphic feature of the field sampling data set to obtain a geomorphic feature vector set;

[0009] Quantify the biological attributes and the target coordinates respectively to obtain a biological attribute vector and a coordinate vector;

[0010] Use the pre-trained adaptive identification model to perform biological-environment matching on the geomorphic feature vector set, the biological attribute vector and the coordinate vector to obtain the environmental adaptation level of the product to be protected;

[0011] Use the pre-set environmental score corresponding to the environmental adaptation level to calculate the weight of the initial risk resistance score to obtain the risk resistance capability value of the product to be protected.

[0012] Optionally, the landform feature extraction network is used to extract landform features from the field sampling data set to obtain a landform feature vector set, including:

[0013] Image information is extracted from the field sampling data set for quantization to obtain a picture vector matrix;

[0014] The picture vector matrix is convoluted using a set of convolution kernels in the geographic feature extraction network to obtain a set of convolution matrices;

[0015] The set of convolution matrices is subjected to a maximum pooling operation to obtain a set of pooling matrices;

[0016] The set of pooling matrices is subjected to a flattening operation using the flatten layer of the geographic feature extraction network to obtain a set of one-dimensional feature sequences;

[0017] The set of one-dimensional feature sequences is subjected to a full connection operation to obtain a set of landform feature vectors corresponding to the field sampling data.

[0018] Optionally, the image information is extracted from the field sampling data set for quantization to obtain a picture vector matrix, including:

[0019] Image information is extracted from the field sampling data set, and the image information is subjected to noise reduction processing using a Gaussian blur algorithm to obtain a noise-reduced picture;

[0020] The noise-reduced picture is subjected to grayscale processing to obtain a grayscale image;

[0021] The grayscale values at each pixel position in the grayscale image are extracted to obtain a picture vector matrix.

[0022] Optionally, the biological attribute and the target coordinates are quantized to obtain a biological attribute vector and a coordinate vector, including:

[0023] The biological attribute is subjected to dimensionality reduction to obtain a biological main attribute;

[0024] The biological main attribute and the target coordinates of the product to be protected are quantized to obtain a biological attribute vector and a coordinate vector.

[0025] Optionally, the landform feature vector set, the biological attribute vector, and the coordinate vector are subjected to biological-environment matching using a pre-trained adaptive recognition model to obtain an environmental adaptability level of the product to be protected, including:

[0026] The adaptive recognition model is used to identify the biological type corresponding to the product to be protected and to obtain a biological-environment primary adaptation space corresponding to the biological type;

[0027] According to a target field in the biological attribute vector, the biological-environment primary adaptation space is spatially stretched to obtain a biological-environment adaptation space;

[0028] The set of geomorphic feature vectors and the coordinate vector are mapped into the biological-environment adaptation space to obtain an environment adaptation level in which a corresponding point of the set of geomorphic feature vectors is located.

[0029] Optionally, the biological-environment primary adaptation space corresponding to the biological type is obtained, including:

[0030] A set of experimental samples corresponding to the biological type is obtained, and an environment adaptability curve of the basic product type in each preset environment is constructed according to the set of experimental samples;

[0031] According to a sample adaptability result in the set of experimental samples, the environment adaptability curves in the each preset environment are dimensionally aggregated to obtain the biological-environment primary adaptation space corresponding to the biological type.

[0032] Optionally, the initial anti-risk score of the product to be protected is calculated according to the pre-constructed risk calculation strategy and the historical business record, including:

[0033] The loss occurrence rate and the claim rate of each type of product in the historical business record are queried, and the loss occurrence rates of the each type of product are sorted by level to obtain a level sorting result;

[0034] The loss level of the product to be protected in the level sorting result is queried, and the initial anti-risk score of the product to be protected is calculated by using a pre-set correlation weight coefficient, the claim rate of the product to be protected and the loss level according to the pre-constructed risk calculation strategy.

[0035] In order to solve the above problems, the application further provides a product anti-risk ability identification device, the device comprises:

[0036] An information acquisition module is configured to acquire a target coordinate of a product to be protected, query biological attributes of the product to be protected, acquire a historical business record of the product to be protected, and calculate an initial anti-risk score of the product to be protected according to a pre-constructed risk calculation strategy and the historical business record;

[0037] An information quantization module is configured to acquire a set of field sampling data of the product to be protected, perform geomorphic feature extraction on the set of field sampling data by using a pre-trained geographic feature extraction network to obtain a set of geomorphic feature vectors, and quantize the biological attributes and the target coordinate respectively to obtain a biological attribute vector and a coordinate vector;

[0038] An environment matching module is configured to perform biological-environment matching on the set of geomorphic feature vectors, biological attribute vectors and the coordinate vector by using a pre-trained adaptive recognition model to obtain an environment adaptability level of the product to be protected.

[0039] An anti-risk capability calculation module is configured to perform weight calculation on the initial anti-risk score by using a preset environment score corresponding to the environment adaptability level to obtain an anti-risk capability value of the product to be protected.

[0040] To solve the above problems, the present application further provides an electronic device, which comprises:

[0041] at least one processor; and

[0042] a memory in communication connection with the at least one processor; wherein

[0043] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the product anti-risk capability identification method described above.

[0044] To solve the above problems, the present application further provides a computer readable storage medium, which stores at least one computer program, and the at least one computer program is executed by a processor in an electronic device to implement the product anti-risk capability identification method described above.

[0045] The embodiment of the present application first obtains an initial anti-risk score of the product to be protected by querying historical business records and a pre-constructed risk calculation strategy, wherein the initial anti-risk score can be used as a historical basis for the enterprise to preliminarily warn the product to be protected; then the environmental adaptability of the product to be protected is identified by obtaining geomorphic features and biological attributes to obtain an environment adaptability level, wherein the environment adaptability level can indicate the future growth trend of the product, and the primary anti-risk score is then fine-tuned; finally, the final anti-risk capability value of the product to be protected is obtained by performing weight calculation on the dual coefficients of the environment adaptability level and the initial anti-risk score, which is more accurate than relying on a single initial anti-risk score, thereby increasing the identification accuracy of the anti-risk capability of the product to be protected. Therefore, the product anti-risk capability identification method, device, electronic device and computer readable storage medium provided by the present application can solve the problem of low accuracy in identifying the anti-risk capability of the product. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 A flowchart of a product anti-risk capability identification method according to an embodiment of the present application is shown in the figure;

[0047] Figure 2 This is a schematic diagram of an image quantization process provided in an embodiment of the present invention;

[0048] Figure 3 This is a flowchart illustrating one step of a product risk resistance identification method provided in an embodiment of the present invention;

[0049] Figure 4 A functional block diagram of a product risk resistance identification device provided in an embodiment of the present invention;

[0050] Figure 5 This is a schematic diagram of the structure of an electronic device that implements the product risk resistance identification method according to an embodiment of the present invention.

[0051] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0052] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0053] This application provides a method for identifying a product's risk resistance capability. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for identifying a product's risk resistance capability can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0054] Reference Figure 1 The diagram shown is a flowchart illustrating a product risk resistance identification method according to an embodiment of the present invention.

[0055] In this embodiment, the product risk resistance identification method includes:

[0056] S1. Obtain the target coordinates of the product to be protected and query the biological attributes of the product to be protected.

[0057] In one of the embodiments of the present application, the product to be protected can be one of the targets in agricultural insurance. The agricultural insurance is a kind of insurance specially for agricultural producers in the process of engaging in planting, forestry, animal husbandry and fishery production, which provides protection for the economic losses caused by natural disasters, accidents, diseases and other insurance accidents, wherein the planting industry is most affected by the natural environment. In the following embodiments, the planting insurance in the agricultural insurance is taken as an example to be described, that is, the product to be protected can be a crop product.

[0058] After obtaining the planting insurance business application form submitted by the customer in the embodiment of the present application, the product to be protected and the target coordinates of the product to be protected that the customer wants to protect are obtained by querying the form keywords in the business application form. For example, the planting insurance business application form records that the Lu Xiping plain with a latitude of 36° and a longitude of 115°30' wants to plant four hundred mu of high-yield machine-picked long-staple cotton new variety W54-9 cultivated by the Academy of Agricultural Sciences, wherein the product to be protected is the long-staple cotton new variety W54-9, and the target coordinates of the product to be protected are a latitude of 36° and a longitude of 115°30'.

[0059] Further, the biological attributes of the product to be protected refer to the biological habits and improved characteristics of the product to be protected. For example, it can be known by querying relevant websites or data that the light compensation point of the single leaf of the long-staple cotton new variety is 750-1000 lx, the light saturation point is 70-80 thousand lx, the physiological of cotton at each growth stage requires that the average water requirement of the field is 65-75% of the water holding capacity, and the soil water content, nutrients, temperature, air, salt content, texture and the like have a great influence on the growth of cotton, and the biological attributes of the long-staple cotton new variety include resistance to cotton bollworm, wilt, yellow wilt tolerance and the like.

[0060] S2, obtaining the historical business record of the product to be protected, and calculating the initial risk resistance score of the product to be protected according to the pre-constructed risk calculation strategy and the historical business record.

[0061] In the embodiment of the present application, the historical business record can be extracted from the pre-constructed business system. The historical business record includes the relevant business record of the product to be protected in a pre-set historical time period, such as the insurance policy record of the agricultural insurance of the insurance company in the past 5 years, wherein the common insurance policies include forest fire loss, flood disaster crop loss, pest and disease loss and the like.

[0062] In detail, in the embodiment of the present application, the initial risk resistance score of the product to be protected is calculated according to the pre-constructed risk calculation strategy and the historical business record, which includes:

[0063] Inquire the claim rate and the pay rate of each type of product in the historical business record, and rank the claim rate of each type of product by grades to obtain a grade ranking result.

[0064] Inquire the claim grade of the product to be protected in the grade ranking result, and perform risk calculation according to a pre-constructed risk calculation strategy by using a preset correlation weight coefficient, the pay rate of the product to be protected and the claim grade to obtain an initial anti-risk score of the product to be protected.

[0065] Specifically, the claim rate refers to the ratio of the number of policies that have claimed in the historical business record of a type of product to the total number of policies handled.

[0066] The embodiment of the application ranks the claim rate of each type of product by grades to obtain a grade ranking result, so that the degree of risk of each type of product can be known, and reasonable policy rules can be formulated.

[0067] In addition, the correlation weight coefficient is determined by the market form of the insurance industry, and can also be determined according to the enterprise's own strategy, and plays a fine-tuning role in the process of calculating the initial anti-risk score of the product to be protected.

[0068] Further, the pay rate refers to the proportion of the amount of money paid by the enterprise to the farm owner after the protected product is damaged, and the pay rate of agricultural insurance in China is much higher than the world average level of 70%, and different crops have different pay rates, so the pay rate is also an important indicator for considering the initial anti-risk score of the product to be protected.

[0069] S3, obtain a set of field sampling data of the product to be protected, and use a pre-trained geographical feature extraction network to extract the geomorphic feature of the set of field sampling data to obtain a set of geomorphic feature vectors.

[0070] After the customer's insurance process, the insurance enterprise needs to conduct on-site verification on the insured policy, take pictures as evidence, and obtain a set of field sampling data, wherein the set of field sampling data should be able to reflect the inspector, the inspection date, the characteristics and scale of the insured subject, and ensure that the image data is clear, complete and has not been modified.

[0071] It is not accurate enough to determine the geomorphic feature only by the target coordinates of the product to be protected, and the embodiment of the application can take more environment pictures of the set of field sampling data when conducting on-site verification, which is convenient for geomorphic feature extraction.

[0072] In the embodiment of the present application, the real sampling data set can contain text information and image information, and the embodiment of the present application classifies and processes the text information and image information. Among them, the image information such as real sampling pictures can be extracted by image recognition, such as soil, terrain; and the text information such as climate type, altitude and other information is quantified to obtain a text feature vector.

[0073] In detail, in the embodiment of the present application, the geographical feature extraction network is used to extract the topographic feature of the real sampling data set to obtain a topographic feature vector set, which includes:

[0074] The image information is extracted from the real sampling data set for quantization operation to obtain a picture vector matrix;

[0075] The convolution kernel set in the geographical feature extraction network is used to perform convolution calculation on the picture vector matrix to obtain a convolution matrix set;

[0076] The maximum pooling operation is performed on the convolution matrix set to obtain a pooling matrix set;

[0077] The flatten layer of the geographical feature extraction network is used to perform flattening operation on the pooling matrix set to obtain a one-dimensional feature sequence set;

[0078] The one-dimensional feature sequence set is subjected to full connection operation to obtain a topographic feature vector set corresponding to the real sampling data.

[0079] In the embodiment of the present application, the geographical feature extraction network is an image regression identification network based on a convolutional neural network, and is trained in advance by a pre-constructed topographic feature sample, which can identify environmental features such as terrain (hillside, flat land, etc.), soil (sandy, clay; loess, black soil, etc.) in the image.

[0080] In addition, the flatten layer is a structure in the neural network for "flattening" the input, that is, one-dimensionalizing the multi-dimensional input, which is a transition layer from the pooling layer to the full connection layer.

[0081] In detail, referring to Figure 2 In the embodiment of the present application, the image information is extracted from the real sampling data set for quantization operation to obtain a picture vector matrix, which includes:

[0082] S31, image information is extracted from the real sampling data set, and the image information is subjected to noise reduction processing by a Gaussian blur algorithm to obtain a noise reduction picture;

[0083] S32, the noise reduction picture is subjected to gray scale processing to obtain a gray scale image;

[0084] S33, extracting the gray scale value at each pixel position in the gray scale image to obtain a picture vector matrix.

[0085] Specifically, to reduce the data calculation amount, the embodiment of the present application reduces the noise interference in the image information by a Gaussian blur algorithm before performing the convolution operation, to obtain a relatively accurate denoising image, wherein the Gaussian blur algorithm is a method of averaging the points around the target pixel to obtain an average value to replace the value of the target pixel, to exclude the possibility that the target pixel is a noise point.

[0086] The present application converts the three-color pixel values into gray scale values of 0-255 through a gray scale conversion formula, and combines each gray scale value into a picture vector matrix according to the relative position of the gray scale image.

[0087] Further, in the embodiment of the present application, the convolution kernel is used to perform convolution calculation on the real sampling pictures, wherein each convolution kernel is responsible for extracting one feature, so that each real sampling picture obtains a convolution matrix set containing N convolution matrices; the convolution matrix set is imported into the pooling layer of the geographic feature extraction network, and through the maximum pooling operation, the dimensionality reduction operation is performed on the convolution matrix set while retaining the maximum feature values of each convolution matrix, to obtain a reduced pooling matrix set; then each pooling matrix is split through the flatten layer, and the split results are sequentially linked, so that each pooling matrix becomes a one-dimensional feature sequence, and the N feature sequences corresponding to each real sampling picture are subjected to the final full connection operation, to obtain the geomorphic feature vector corresponding to the real sampling picture.

[0088] The embodiment of the present application further intersects and merges the geomorphic feature vector of each image information and the text feature vector after the text information is quantized, to obtain a final geomorphic feature vector set.

[0089] S4, respectively quantizing the biological attribute and the target coordinate to obtain a biological attribute vector and a coordinate vector.

[0090] In detail, in the embodiment of the present application, the respectively quantizing the biological attribute and the target coordinate to obtain a biological attribute vector and a coordinate vector comprises:

[0091] performing dimensionality reduction operation on the biological attribute to obtain a biological main attribute;

[0092] quantizing the biological main attribute and the target coordinate of the product to be protected to obtain a biological attribute vector and a coordinate vector.

[0093] In the embodiment of the present application, the irrelevant attributes such as "long fiber, relatively thin, low strength, and original place Egypt" in the biological attributes can be reduced by a principal component analysis algorithm (PCA), and important main attributes such as "sunlight requirement, water quality requirement, and temperature requirement" are retained.

[0094] The PCA is an algorithm for reducing the loss of information in the biological attributes while reducing the indicators to be analyzed, so as to achieve the purpose of comprehensive analysis of the collected data.

[0095] In addition, the biological main attributes and the target coordinates of the product to be protected can be quantified by using a one-hot algorithm, wherein the one-hot algorithm is a form of converting a type variable into a form easy to process in machine learning, and the specific quantization and encoding process are not described in detail.

[0096] S5, using a pre-trained adaptive recognition model to perform biological-environment matching on the geomorphic feature vector set, the biological attribute vector, and the coordinate vector, to obtain an environment adaptation level of the product to be protected.

[0097] In the embodiment of the present application, the adaptive recognition model is an artificial intelligence model for viewing the position of a data point after mapping data to a specific space.

[0098] In detail, referring to Figure 3 In the embodiment of the present application, the biological-environment matching on the geomorphic feature vector set, the biological attribute vector, and the coordinate vector by using the pre-trained adaptive recognition model to obtain the environment adaptation level of the product to be protected includes:

[0099] S51, using the adaptive recognition model to identify the biological type corresponding to the product to be protected, and obtaining a biological-environment primary adaptation space corresponding to the biological type;

[0100] S52, performing spatial stretching on the biological-environment primary adaptation space according to a target field in the biological attribute vector, to obtain a biological-environment adaptation space;

[0101] S53, mapping the geomorphic feature vector set and the coordinate vector into the biological-environment adaptation space, to obtain an environment adaptation level in which a corresponding point of the geomorphic feature vector set is located.

[0102] In the embodiment of the present application, the biological-environment primary adaptation space can be a three-dimensional space constructed by sunlight, land water content, and temperature.

[0103] In detail, in the embodiment of the present application, the biological-environment primary adaptation space corresponding to the biological type is obtained by:

[0104] acquire an experimental sample set corresponding to the biological type, and construct an environmental adaptability curve of the basic product type in each preset environment according to the experimental sample set;

[0105] According to the sample adaptability result in the experimental sample set, the environmental adaptability curve in each preset environment is dimensionally aggregated to obtain a biological-environmental primary adaptation space corresponding to the biological type.

[0106] In the embodiment of the application, the biological type can be the most basic variety, such as wheat, cotton, river shrimp, sea cucumber, etc. The original type of W54-9 long-staple cotton is ordinary cotton without improvement. In the embodiment of the application, the experimental sample set of each biological type is acquired by controlling variables, for example, in the case of constant suitable light and constant suitable temperature, the environmental adaptability curve of the biological type and soil humidity is constructed. Through the environmental adaptability curve, the range of the target biological type with good growth, general growth and poor growth can be viewed. By analogy, the environmental adaptability curves of cotton and environmental temperature, and light are obtained. Finally, the biological-environmental primary adaptation space is obtained by dimension combination.

[0107] In the embodiment of the application, some characteristics of new varieties such as "drought resistance and cold resistance" can be extracted in the target field in the biological attribute vector. The two dimensions of temperature and soil humidity in the biological-environmental primary adaptation space are appropriately stretched according to the drought resistance and cold resistance to obtain the biological-environmental adaptation space.

[0108] Further, after the geomorphic feature vector set and the coordinate vector are mapped into the biological-environmental adaptation space, it can be known that the corresponding point of the geomorphic feature vector set is in an environmental adaptability level, such as a suitable sowing level, a buffer sowing level, and an unsuitable sowing level. The coordinate vector can represent information such as a query climate type.

[0109] S6, using a preset environment score corresponding to the environmental adaptability level to weight calculate the initial risk resistance score to obtain the risk resistance ability value of the product to be protected.

[0110] In the embodiment of the application, the score corresponding table of the environmental adaptability level pre-constructed in the insurance enterprise can be queried to obtain the preset environment score corresponding to the implementation environmental adaptability level, and then the initial risk resistance score is weighted calculated to obtain the risk resistance ability value of the product to be protected. The preset environment score can be customized according to the specific insurance enterprise, and is used to set the result interval of the subsequent risk resistance ability value.

[0111] Furthermore, after obtaining the risk resistance value of the product to be protected, this embodiment of the invention can limit the risk resistance value by a preset range. When the risk resistance value exceeds the preset range, it can prompt business personnel to pay attention, thereby formulating a more accurate agricultural insurance policy.

[0112] This invention first obtains an initial risk resistance score for the product to be protected by querying historical business records and a pre-constructed risk calculation strategy. This initial risk resistance score serves as a historical basis for the enterprise to issue preliminary warnings for the product. Next, by acquiring topographical features and biological attributes, the environmental adaptability of the product to be protected is identified, resulting in an environmental adaptability level. This environmental adaptability level indicates the product's future growth trend, allowing for fine-tuning of the initial risk resistance score. Finally, a weighted calculation is performed using both the environmental adaptability level and the initial risk resistance score to obtain a more accurate final risk resistance value for the product to be protected compared to relying solely on the initial risk resistance score, thus increasing the accuracy of risk resistance capability identification. Therefore, the product risk resistance capability identification method proposed in this invention can solve the problem of low accuracy in product risk resistance capability identification.

[0113] like Figure 4 The diagram shown is a functional block diagram of a product risk resistance identification device provided in an embodiment of the present invention.

[0114] The product risk resistance capability identification device 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the product risk resistance capability identification device 100 may include an information acquisition module 101, an information quantification module 102, an environment matching module 103, and a risk resistance capability calculation module 104. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0115] In this embodiment, the functions of each module / unit are as follows:

[0116] The information acquisition module 101 is used to acquire the target coordinates of the product to be protected, query the biological attributes of the product to be protected, acquire the historical business records of the product to be protected, and calculate the initial risk resistance score of the product to be protected based on the pre-constructed risk calculation strategy and the historical business records.

[0117] The information quantization module 102 is configured to obtain a field sampling data set of the product to be protected, perform geomorphological feature extraction on the field sampling data set by using a pre-trained geographic feature extraction network to obtain a geomorphological feature vector set, and quantize the biological attribute and the target coordinate respectively to obtain a biological attribute vector and a coordinate vector.

[0118] The environment matching module 103 is configured to perform biological-environment matching on the geomorphological feature vector set, the biological attribute vector and the coordinate vector by using a pre-trained adaptive recognition model to obtain an environment adaptation level of the product to be protected.

[0119] The anti-risk ability calculation module 104 is configured to perform weight calculation on the initial anti-risk score by using a preset environment score corresponding to the environment adaptation level to obtain an anti-risk ability value of the product to be protected.

[0120] In detail, each module in the product anti-risk ability recognition device 100 in the embodiment of the present application adopts the same technical means as the product anti-risk ability recognition method described in the above Figures 1 to 3 , and can produce the same technical effects, which will not be described here.

[0121] As shown in Figure 5 , it is a structural schematic diagram of an electronic device for implementing the product anti-risk ability recognition method according to an embodiment of the present application.

[0122] The electronic device 1 can include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and can further include a computer program stored in the memory 11 and executable on the processor 10, such as a product anti-risk ability recognition program.

[0123] In some embodiments, the processor 10 can be composed of an integrated circuit, for example, can be composed of a single packaged integrated circuit, or can be composed of multiple packaged integrated circuits with the same function or different functions, including one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors and combinations of various control chips, etc. The processor 10 is the control core of the electronic device, which connects all components of the electronic device by using various interfaces and lines, executes or runs programs or modules stored in the memory 11 (such as executing the product anti-risk ability recognition program), and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0124] The memory 11 includes at least one type of readable storage medium, such as a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., an SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 11 can also be an external storage device of the electronic device, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 11 can include both an internal storage unit and an external storage device of the electronic device. The memory 11 can be used to store application software installed in the electronic device and various data, such as the code of the product risk resistance ability identification program, and can also be used to temporarily store data that has been output or will be output.

[0125] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable connection and communication between the memory 11 and at least one processor 10, etc.

[0126] The communication interface 13 is used for communication between the electronic device and other devices, and includes a network interface and a user interface. Optionally, the network interface can include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), and is usually used to establish a communication connection between the electronic device and other electronic devices. The user interface can be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface can also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. The display can also be appropriately referred to as a display screen or a display unit, and is used to display information processed in the electronic device and to display a visualized user interface.

[0127] Figure 5 Only an electronic device with components is shown, and those skilled in the art can understand that, Figure 5The illustrated structure does not constitute a limitation on the electronic device 1, and can include fewer or more components than illustrated, or combine certain components, or arrange different components.

[0128] For example, although not shown, the electronic device can also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, so that functions such as charge management, discharge management, and power consumption management can be achieved through the power management device. The power supply can also include one or more direct current or alternating current power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and any other components. The electronic device can also include various sensors, Bluetooth modules, Wi-Fi modules, and the like, which are not described here.

[0129] It should be understood that the embodiments are only for illustration and are not limited in the scope of the patent application by this structure.

[0130] The product risk resistance ability identification program stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, which, when executed in the processor 10, can achieve:

[0131] Obtaining target coordinates of a product to be protected, and querying biological attributes of the product to be protected;

[0132] Obtaining historical business records of the product to be protected, and calculating an initial risk resistance score of the product to be protected according to a pre-constructed risk calculation strategy and the historical business records;

[0133] Obtaining a set of field sampling data of the product to be protected, and extracting geomorphic feature vectors from the set of field sampling data using a pre-trained geographic feature extraction network;

[0134] Quantifying the biological attributes and the target coordinates respectively to obtain a biological attribute vector and a coordinate vector;

[0135] Performing biological-environment matching on the set of geomorphic feature vectors, the biological attribute vector, and the coordinate vector using a pre-trained adaptive identification model to obtain an environmental adaptation level of the product to be protected;

[0136] Performing weight calculation on the initial risk resistance score using a preset environmental score corresponding to the environmental adaptation level to obtain a risk resistance ability value of the product to be protected.

[0137] Specifically, the specific implementation method of the processor 10 on the above instructions can refer to the description of the related steps in the corresponding embodiments of the accompanying drawings, which is not described here.

[0138] Further, the modules / units integrated in the electronic device 1 are stored in a computer readable storage medium if they are realized in the form of software function units and sold or used as independent products. The computer readable storage medium can be volatile or non-volatile. For example, the computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM).

[0139] The application further provides a computer readable storage medium, which stores a computer program, and the computer program can realize the following when executed by a processor of an electronic device:

[0140] obtaining target coordinates of a product to be protected, and querying biological attributes of the product to be protected;

[0141] obtaining historical business records of the product to be protected, and calculating an initial anti-risk score of the product to be protected according to a pre-constructed risk calculation strategy and the historical business records;

[0142] obtaining a set of field sampling data of the product to be protected, performing geomorphological feature extraction on the set of field sampling data by using a pre-trained geographical feature extraction network, and obtaining a set of geomorphological feature vectors;

[0143] quantifying the biological attributes and the target coordinates respectively to obtain a biological attribute vector and a coordinate vector;

[0144] performing biological-environment matching on the set of geomorphological feature vectors, the biological attribute vector and the coordinate vector by using a pre-trained adaptive recognition model to obtain an environmental adaptation level of the product to be protected;

[0145] performing weight calculation on the initial anti-risk score by using a preset environmental score corresponding to the environmental adaptation level to obtain an anti-risk ability value of the product to be protected.

[0146] In the several embodiments provided in the application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there can be another division manner in actual implementation.

[0147] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0148] In addition, each functional module in various embodiments of the application can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware, or in the form of hardware plus software functional module.

[0149] It is obvious for those skilled in the art that the application is not limited to the details of the above exemplary embodiments, and the application can be implemented in other specific forms without departing from the spirit or essential characteristics of the application.

[0150] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, and the scope of the application is defined by the appended claims rather than the above description, and therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the application. Any additional reference signs in the claims should not be considered as limiting the claims to which they relate.

[0151] The blockchain referred to in the application is a new application mode of distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm and other computer technologies. Blockchain, in essence, is a decentralized database, a series of data blocks associated using cryptography, each containing a batch of network transaction information for verifying the validity (anti-fake) of the information and generating the next block. The blockchain can include a blockchain underlying platform, a platform product service layer, and an application service layer.

[0152] The embodiments of the application can acquire and process related data based on artificial intelligence technology. Artificial intelligence (AI) is the use of digital computers or computer-controlled machines to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0153] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The plurality of units or devices stated in the system claims can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0154] Finally, it should be noted that the above examples are merely intended to illustrate the technical solutions of the present application and not to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application.

Claims

1. A method for identifying a product's risk resistance capability, characterized in that, The method includes: Obtain the target coordinates of the product to be protected, and query the biological habits and modification characteristics of the product to be protected as biological attributes; Obtain the historical business records of the product to be protected, and calculate the initial risk resistance score of the product to be protected based on the pre-constructed risk calculation strategy and the loss rate and payout rate of each type of product in the historical business records. Obtain a set of on-site sampling data of the product to be protected, and use a pre-trained geographic feature extraction network to extract geomorphic features from the on-site sampling data set to obtain a set of geomorphic feature vectors; The biological attributes and the target coordinates are quantified respectively to obtain biological attribute vectors and coordinate vectors; A pre-trained adaptive recognition model is used to identify the biological type corresponding to the product to be protected, and an experimental sample set corresponding to the biological type is obtained. Based on the sample adaptability results and environmental adaptability curves in the experimental sample set, a biological-environment primary adaptation space is obtained. Based on the target field in the biological attribute vector, the biological-environment primary adaptation space is spatially scaled to obtain a biological-environment adaptation space. The set of landform feature vectors and the coordinate vectors are mapped to the biological-environment adaptation space to obtain the environmental adaptability level of the point corresponding to the set of landform feature vectors. Using the preset environmental score corresponding to the environmental adaptability level, the initial risk resistance score is weighted and calculated to obtain the risk resistance value of the product to be protected.

2. The product risk resistance identification method as described in claim 1, characterized in that, The method of extracting geomorphic features from the field sampled data set using a pre-trained geographic feature extraction network to obtain a set of geomorphic feature vectors includes: Image information is extracted from the field sampling data set and quantized to obtain an image vector matrix; The set of convolution kernels in the geographic feature extraction network is used to perform convolution calculations on the image vector matrix to obtain a set of convolution matrices. Perform max pooling on the set of convolutional matrices to obtain a set of pooled matrices; The pooling matrix set is flattened using the flatten layer of the geographic feature extraction network to obtain a one-dimensional feature sequence set; Perform a fully connected operation on the set of one-dimensional feature sequences to obtain the set of geomorphic feature vectors corresponding to the field sampling data.

3. The product risk resistance identification method as described in claim 2, characterized in that, The step of extracting image information from the field sampling data set and performing quantization to obtain an image vector matrix includes: Image information is extracted from the real-world sampling data set, and the image information is denoised using a Gaussian blur algorithm to obtain a denoised image; The denoised image is then converted to grayscale to obtain a grayscale image; Extract the grayscale values ​​at each pixel location in the grayscale image to obtain the image vector matrix.

4. The product risk resistance identification method as described in claim 1, characterized in that, The process of quantifying the biological attributes and the target coordinates to obtain biological attribute vectors and coordinate vectors includes: The biological attributes are reduced in dimensionality to obtain the biological main attributes; The biological main attributes and the target coordinates of the product to be protected are quantified to obtain biological attribute vectors and coordinate vectors.

5. The product risk resistance identification method as described in claim 1, characterized in that, The biological-environment primary adaptation space is obtained based on the sample adaptability results and environmental adaptability curves in the experimental sample set, including: Based on the experimental sample set, construct environmental adaptability curves for basic product types in various preset environments; Based on the sample adaptability results in the experimental sample set, the environmental adaptability curves in each preset environment are aggregated dimensionally to obtain the biological-environment primary adaptation space corresponding to the biological type.

6. The product risk resistance identification method as described in claim 1, characterized in that, The calculation of the initial risk resistance score of the product to be protected, based on the pre-constructed risk calculation strategy and the loss rate and payout rate of each type of product in the historical business records, includes: Query the accident rate and payout rate of each type of product in the historical business records, and sort the accident rates of each type of product by level to obtain the ranking results; The risk level of the product to be protected is queried in the ranking results. Based on the pre-constructed risk calculation strategy, the risk is calculated using preset relevant weight coefficients, the loss ratio of the product to be protected, and the risk level to obtain the initial risk resistance score of the product to be protected.

7. A product risk resistance identification device, characterized in that, The device includes: The information acquisition module is used to acquire the target coordinates of the product to be protected, query the biological habits and improvement characteristics of the product to be protected as biological attributes, acquire the historical business records of the product to be protected, and calculate the initial risk resistance score of the product to be protected based on the pre-constructed risk calculation strategy and the loss rate and payout rate of each type of product in the historical business records. The information quantization module is used to acquire the on-site sampling data set of the product to be protected, extract the landform features from the on-site sampling data set using a pre-trained geographic feature extraction network to obtain a set of landform feature vectors, and quantify the biological attributes and the target coordinates respectively to obtain biological attribute vectors and coordinate vectors. The environment matching module is used to identify the biological type corresponding to the product to be protected using a pre-trained adaptive recognition model, obtain the experimental sample set corresponding to the biological type, obtain the biological-environment primary adaptation space based on the sample adaptation results and environmental adaptation curve in the experimental sample set, perform spatial scaling on the biological-environment primary adaptation space based on the target field in the biological attribute vector, obtain the biological-environment adaptation space, map the set of landform feature vectors and the coordinate vectors to the biological-environment adaptation space, and obtain the environmental adaptation level of the point corresponding to the set of landform feature vectors. The risk resistance calculation module is used to calculate the risk resistance value of the product to be protected by using a preset environmental score corresponding to the environmental adaptability level and weighting the initial risk resistance score.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the product risk resistance identification method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the product risk resistance identification method as described in any one of claims 1 to 6.

Images