Loan risk assessment method and device, storage medium and electronic equipment

Abstract

The invention discloses a loan risk assessment method and device, a storage medium and electronic equipment. Relates to the field of artificial intelligence, the field of financial science and technology and other related technical fields, and the method comprises the steps: carrying out the detection and recognition of target remote sensing images of a target fishery in a plurality of time periods according to a target detection model, and obtaining the operation range information of fishing boats in the target fishery; predicting the fishing amount of the plurality of fishing boats in the target fishing ground according to the time sequence prediction model to obtain the predicted fishing amount of the plurality of fishing boats in the target time period; and according to the operation range information and the predicted fishing amount, determining a loan risk level of the target fishery, the loan risk level being used for indicating whether to handle the target loan business for the target fishery. According to the method and the device, the problem of relatively low assessment accuracy of the risk of handling the loan in the fishing ground due to the fact that the risk of handling the loan in the fishing ground is assessed by manually investigating the condition of the fishing ground on site in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the fields of artificial intelligence, financial technology and other related technical fields, and specifically to a loan risk assessment method and device, a storage medium and an electronic device. Background Art

[0002] With the development of economy and population growth, fishery has become one of the important resource industries. However, due to problems such as overfishing and environmental pollution, many fishery enterprises are facing difficulties. Therefore, it is very important to conduct credit assessment for fishing enterprises. Through credit assessment, the assets and liabilities, profitability, operating stability and development potential of fisheries can be evaluated, thereby reducing the investment risks of financial institutions and investors. At the same time, credit assessment of fisheries can also help fishery managers better understand their own business conditions and make more reasonable business plans and decisions. At present, financial institutions usually conduct manual field inspections of fisheries and manually assess the loan risks and quotas of fisheries based on the conditions of the fisheries. This method is easily affected by the subjective influence of the evaluators, resulting in a low accuracy rate of the evaluation results. At the same time, manual field inspections require a lot of time and labor costs, especially for large-scale fisheries, it is difficult to conduct comprehensive inspections, resulting in low evaluation efficiency.

[0003] Currently, no effective solution has been proposed to address the problem that the risk of fishery loans is assessed through manual on-site inspections of fishery conditions in related technologies, resulting in a relatively low accuracy rate in assessing the risk of fishery loans. Summary of the invention

[0004] The main purpose of this application is to provide a loan risk assessment method and device, storage medium and electronic device to solve the problem in the related art that the risk of fishery loans is assessed through manual on-site inspections of fishery conditions, resulting in a relatively low accuracy rate in the assessment of the risk of fishery loans.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a loan risk assessment method is provided. The method comprises: detecting and identifying target remote sensing images of a target fishing ground in multiple time periods according to a target detection model, and obtaining the operating range information of fishing boats in the target fishing ground; predicting the catch of multiple fishing boats in the target fishing ground according to a time series prediction model, and obtaining the predicted catch of the multiple fishing boats in the target time period; determining the loan risk level of the target fishing ground according to the operating range information and the predicted catch, wherein the loan risk level is used to indicate whether to process a target loan business for the target fishing ground.

[0006] Furthermore, the target remote sensing images of the target fishing ground in multiple time periods are detected and identified based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, including: performing edge detection on the target remote sensing image according to the edge detection module of the target detection model to obtain edge information of the target remote sensing image; performing feature extraction on the edge information according to the feature extraction module of the target detection model to obtain first feature information of the target remote sensing image; and identifying the first feature information according to the analysis and processing module of the target detection model to obtain the operating range information of the fishing boats in the target fishing ground.

[0007] Furthermore, predicting the catch of multiple fishing boats in the target fishing ground based on the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period includes: obtaining the fishing behavior data of the multiple fishing boats in the historical time period, wherein the fishing behavior data at least includes: the catch quantity and fishing weight of each fishing boat; performing feature extraction on the fishing behavior data based on the encoding module of the time series prediction model to obtain second feature information of the multiple fishing boats in the historical time period; and predicting the second feature information based on the decoding module of the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period.

[0008] Furthermore, feature extraction is performed on the fishing behavior data according to the encoding module of the time series prediction model to obtain second feature information of the multiple fishing boats in the historical time period, including: processing the fishing behavior data through the fully connected layer and the nonlinear activation function layer in the encoding module to obtain multiple local features of the fishing behavior data; fitting the multiple local features of the fishing behavior data through the basis function layer in the encoding module to obtain the second feature information of the multiple fishing boats in the historical time period.

[0009] Furthermore, determining the loan risk level of the target fishery based on the operating scope information and the predicted catch volume includes: performing calculations based on the operating scope information and the predicted catch volume to obtain the predicted total output value of the target fishery; calculating the ratio of the loan amount of the target loan business to the predicted total output value to obtain a target result; if the target result is greater than a preset value, determining the loan risk level to be a first risk level; if the target result is less than or equal to the preset value, determining the loan risk level to be a second risk level, wherein the second risk level is lower than the first risk level.

[0010] Furthermore, after determining that the loan risk level is the second risk level, the method also includes: when the loan risk level is the second risk level, obtaining historical loan behavior data of the target fishery; based on the historical loan behavior data, determining the loan information of the target fishery, wherein the loan information at least includes: the issuance method of the target loan business and the repayment method of the target fishery; and based on the loan information, processing the target loan business for the target fishery.

[0011] Furthermore, before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, the method also includes: acquiring the location information of the target fishing ground; based on the location information, collecting initial remote sensing images of the target fishing ground in multiple time periods; and preprocessing the initial remote sensing images to obtain the target remote sensing images.

[0012] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a loan risk assessment device is provided. The device comprises: a detection unit, which is used to detect and identify the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model, and obtain the operating range information of the fishing boats in the target fishing ground; a prediction unit, which is used to predict the catch of multiple fishing boats in the target fishing ground according to the time series prediction model, and obtain the predicted catch of the multiple fishing boats in the target time period; a first determination unit, which is used to determine the loan risk level of the target fishing ground according to the operating range information and the predicted catch, wherein the loan risk level is used to indicate whether to handle the target loan business for the target fishing ground.

[0013] Furthermore, the detection unit includes: a detection module, used to perform edge detection on the target remote sensing image according to the edge detection module of the target detection model, so as to obtain edge information of the target remote sensing image; a first extraction module, used to perform feature extraction on the edge information according to the feature extraction module of the target detection model, so as to obtain first feature information of the target remote sensing image; and an identification module, used to identify the first feature information according to the analysis and processing module of the target detection model, so as to obtain the operating range information of the fishing boat in the target fishing ground.

[0014] Furthermore, the prediction unit includes: an acquisition module, used to obtain the fishing behavior data of the multiple fishing boats in the historical time period, wherein the fishing behavior data at least includes: the fishing quantity and fishing weight of each fishing boat; a second extraction module, used to perform feature extraction on the fishing behavior data according to the encoding module of the time series prediction model, and obtain second feature information of the multiple fishing boats in the historical time period; a prediction module, used to predict the second feature information according to the decoding module of the time series prediction model, and obtain the predicted fishing volume of the multiple fishing boats in the target time period.

[0015] Furthermore, the second extraction module includes: a processing submodule, used to process the fishing behavior data through the fully connected layer and the nonlinear activation function layer in the encoding module to obtain multiple local features of the fishing behavior data; a fitting submodule, used to fit the multiple local features of the fishing behavior data through the basis function layer in the encoding module to obtain the second feature information of the multiple fishing boats in the historical time period.

[0016] Furthermore, the first determination unit includes: a first calculation module, used to perform calculations based on the operating scope information and the predicted catch to obtain the predicted total output value of the target fishing ground; a second calculation module, used to calculate the ratio of the loan amount of the target loan business to the predicted total output value to obtain a target result; a first determination module, used to determine that the loan risk level is a first risk level if the target result is greater than a preset value; a second determination module, used to determine that the loan risk level is a second risk level if the target result is less than or equal to the preset value, wherein the second risk level is lower than the first risk level.

[0017] Furthermore, the device also includes: a first acquisition unit, used to obtain historical loan behavior data of the target fishery after determining that the loan risk level is the second risk level, when the loan risk level is the second risk level; a second determination unit, used to determine the loan information of the target fishery based on the historical loan behavior data, wherein the loan information at least includes: the issuance method of the target loan business and the repayment method of the target fishery; a processing unit, used to handle the target loan business for the target fishery based on the loan information.

[0018] Furthermore, before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, the device also includes: a second acquisition unit for acquiring the location information of the target fishing ground; a collection unit for collecting initial remote sensing images of the target fishing ground in multiple time periods based on the location information; and a preprocessing unit for preprocessing the initial remote sensing image to obtain the target remote sensing image.

[0019] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a computer-readable storage medium is provided, wherein the storage medium stores a program, wherein when the program is running, the device where the storage medium is located is controlled to execute any one of the loan risk assessment methods described above.

[0020] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is also provided, the electronic device includes one or more processors and a memory, and the memory is used to store the loan risk assessment method described in any one of the above-mentioned methods implemented by one or more processors.

[0021] Through this application, the following steps are adopted: according to the target detection model, the target remote sensing images of the target fishing ground in multiple time periods are detected and identified to obtain the operating range information of the fishing boats in the target fishing ground; according to the time series prediction model, the catch of multiple fishing boats in the target fishing ground is predicted to obtain the predicted catch of the multiple fishing boats in the target time period; according to the operating range information and the predicted catch, the loan risk level of the target fishing ground is determined, wherein the loan risk level is used to indicate whether the target loan business is handled for the target fishing ground. Through this application, the problem of low accuracy in assessing the risk of handling loans for fishing grounds by manually inspecting the situation of fishing grounds in the related art is solved. In this scheme, first, the target remote sensing images of the target fishing ground in different time periods are analyzed by using the target detection model to detect and identify the operating range information of the fishing boats. Then, the catch of multiple fishing boats in the target time period is predicted by the time series prediction model to obtain the predicted catch of these fishing boats. Finally, based on the operating range information and predicted catch, the loan risk level of the target fishery is determined to evaluate whether the target fishery should be provided with loan business. The operating range information of multiple fishing boats in the fishery is determined by the target detection model, and the predicted catch of multiple fishing boats in the loan period is determined by the time series model. The loan risk level of the target fishery is comprehensively assessed based on the operating range information and the predicted catch, avoiding the traditional method of manually determining the loan risk level of the target fishery, avoiding the subjective influence of the assessor, and also improving the assessment speed of the target fishery, thereby achieving the effect of improving the assessment accuracy of the risk of applying for loans to the fishery. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0023] Figure 1 is a flow chart of a loan risk assessment method provided according to an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of a loan risk assessment method provided according to an embodiment of the present application;

[0025] Figure 3 is a schematic diagram of a loan risk assessment device provided according to an embodiment of the present application;

[0026] Figure 4 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0031] The present invention is described below in conjunction with preferred implementation steps. Figure 1 is a flowchart of a loan risk assessment method provided in accordance with an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0032] Step S101, detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model, and obtaining the operating range information of the fishing boats in the target fishing ground.

[0033] Optionally, satellite remote sensing technology is used to collect multiple target remote sensing images of the target fishing ground in different time periods, and a suitable edge detection model (i.e., target detection model), such as: DexiNed model, Canny model, Prewitt operator and Sobel operator, is used to detect and identify the above multiple target remote sensing images to obtain the operating range information of fishing boats in the target fishing ground in multiple time periods, and quantify the above operating range information by calculating statistical indicators such as the mean value and standard deviation of the fishing boat position.

[0034] Step S102, predicting the catch of multiple fishing boats in the target fishing ground based on the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period.

[0035] Optionally, historical fishing data and environmental factors (such as weather, temperature, wind speed, etc.) data of fishing vessels over a period of time are collected, and the catch volume of the above data is predicted using the N-BEATS time series prediction model to obtain the predicted catch volume of multiple fishing vessels in the target time period.

[0036] Step S103, determining the loan risk level of the target fishery based on the operation scope information and the predicted catch, wherein the loan risk level is used to indicate whether to process the target loan business for the target fishery.

[0037] Optionally, the loan risk level of the target fishery is calculated based on the operating range information of the fishing vessels in the target fishery and the predicted catch. The loan risk level can be divided into low, medium and high risks. Based on the loan risk level of the target fishery, it is determined whether to process the target loan business for the target fishery. For example, for fisheries with low and medium risk levels, the loan business can be approved; for fisheries with high risk levels, the loan business can be refused.

[0038] In summary, firstly, the target remote sensing images of the target fishing ground in different time periods are analyzed by the target detection model to detect and identify the operating range information of the fishing boats. Then, the catch of multiple fishing boats in the target time period is predicted by the time series prediction model to obtain the predicted catch of these fishing boats. Finally, based on the operating range information and the predicted catch, the loan risk level of the target fishing ground is determined to evaluate whether the loan business should be provided to the target fishing ground. The operating range information of multiple fishing boats in the fishing ground is determined by the target detection model, the predicted catch of multiple fishing boats in the loan time period is determined by the time series model, and the loan risk level of the target fishing ground is comprehensively assessed based on the operating range information and the predicted catch, which avoids the traditional method of manually determining the loan risk level of the target fishing ground, avoids the subjective influence of the evaluator, and also improves the assessment speed of the target fishing ground, thereby achieving the effect of improving the assessment accuracy of the risk of applying for loans to the fishing ground.

[0039] Optionally, in the loan risk assessment method provided in the embodiment of the present application, the target remote sensing images of the target fishing ground in multiple time periods are detected and identified based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, including: performing edge detection on the target remote sensing image according to the edge detection module of the target detection model to obtain edge information of the target remote sensing image; performing feature extraction on the edge information according to the feature extraction module of the target detection model to obtain first feature information of the target remote sensing image; and identifying the first feature information according to the analysis and processing module of the target detection model to obtain the operating range information of the fishing boats in the target fishing ground.

[0040] In an optional embodiment, the edge detection module of the edge detection model (i.e., the target detection model) such as the DexiNed model, the Canny model, the Prewitt operator, and the Sobel operator is first used to perform edge detection on the preprocessed target remote sensing image, highlighting the significant changes in the remote sensing image to obtain the edge information of the target remote sensing image. Then, the feature extraction module of the target detection model is used to extract features based on the above edge information to obtain the first feature information for distinguishing fishing boats from other objects. The above first feature information may include: information such as the shape, size, texture, and color of the fishing boats in the target fishing ground. Finally, the analysis and processing module of the target detection model is used to identify the extracted first feature information, thereby determining the operation range information such as the location, quantity, and operation type of the fishing boats.

[0041] Through the edge detection model, the edge of the target remote sensing image can be effectively extracted, and the operating range information of the fishing boats in the target fishing ground can be accurately obtained.

[0042] Optionally, in the loan risk assessment method provided in the embodiment of the present application, the catch of multiple fishing boats in the target fishing ground is predicted based on the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period, including: obtaining the fishing behavior data of the multiple fishing boats in the historical time period, wherein the fishing behavior data at least includes: the catch quantity and fishing weight of each fishing boat; extracting features from the fishing behavior data based on the encoding module of the time series prediction model to obtain second feature information of the multiple fishing boats in the historical time period; predicting the second feature information based on the decoding module of the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period.

[0043] In an optional embodiment, firstly, the fishing behavior data on the catch quantity and catch weight of multiple fishing vessels in the historical time period are obtained through relevant departments or data providers, and the collected fishing behavior data are preprocessed by missing value processing, outlier processing, data normalization and other preprocessing operations. Then, the encoding module of the N-BEATS time series prediction model is used to extract features from the preprocessed fishing behavior data, and the influence of external factors such as fluctuations, periodicity and seasonality on the catch based on the time series fishing behavior data is extracted, that is, the second feature information. Finally, the decoding module of the N-BEATS time series prediction model is used to decode and predict based on the above second feature information to obtain the predicted catch of multiple fishing vessels in the target time period.

[0044] It should be noted that N-BEATS (Neural basis expansion analysis for interpretable time series forecasting) is a deep learning model for time series forecasting, which aims to provide a time series forecasting method with strong interpretability and superior performance. The N-BEATS model combines the advantages of neural networks and Fourier basis expansion analysis, and constructs a time series model by stacking multiple "basis" structures. Each basis structure consists of a set of parameterized Fourier series, and each series contains a set of specific frequencies and periods that can capture the key patterns in time series data. The N-BEATS model stacks multiple basis structures together, and the order and number of stacking can be adjusted according to the characteristics of the time series data. Among them, each basis structure of the N-BEATS model corresponds to a specific frequency and period in the time series, and automatically learns the multi-scale and nonlinear patterns in the time series data, making it more interpretable. The N-BEATS model takes historical time series data as input and outputs the predicted value at each time point by optimizing the loss function.

[0045] Through the N-BEATS time series prediction model, the characteristic information in the historical fishing behavior data can be accurately learned, and the catch of multiple fishing vessels in the target time period can be quickly predicted based on this characteristic information.

[0046] Optionally, in the loan risk assessment method provided in the embodiment of the present application, feature extraction is performed on the fishing behavior data according to the encoding module of the time series prediction model to obtain second feature information of multiple fishing boats in the historical time period, including: processing the fishing behavior data through the fully connected layer and the nonlinear activation function layer in the encoding module to obtain multiple local features of the fishing behavior data; fitting multiple local features of the fishing behavior data through the basis function layer in the encoding module to obtain the second feature information of multiple fishing boats in the historical time period.

[0047] In an optional embodiment, the input fishing behavior data is first linearly transformed through the fully connected layer of the N-BEATS model encoding module, and the nonlinear expression ability of the model is increased through a nonlinear activation function layer (such as ReLU, tanh or sigmoid, etc.), thereby obtaining multiple local features of the fishing behavior data. Then the above multiple local features are input into the basis function layer of the encoding module to learn the key patterns in the fishing behavior data, such as trends, seasonality and other influencing information. In the N-BEATS model, the basis function layer is usually composed of a set of parameterized Fourier series that can capture periodic patterns in time series data. By adjusting the parameters of the basis function layer, the model can fit multiple local features of the fishing behavior data and integrate them into a new set of representations, namely the second feature information. The above second feature information can capture higher-level abstract concepts in the fishing behavior data, which is helpful for subsequent prediction tasks.

[0048] By processing the fishing behavior data through the fully connected layer, nonlinear activation function layer and basis function layer of the N-BEATS model, the second feature information of multiple fishing boats in the historical time period can be obtained quickly and accurately.

[0049] Optionally, in the loan risk assessment method provided in the embodiment of the present application, determining the loan risk level of the target fishery based on the operating scope information and the predicted catch includes: calculating based on the operating scope information and the predicted catch to obtain the predicted total output value of the target fishery; calculating the ratio of the loan amount of the target loan business to the predicted total output value to obtain the target result; if the target result is greater than a preset value, determining the loan risk level to be the first risk level; if the target result is less than or equal to the preset value, determining the loan risk level to be the second risk level, wherein the second risk level is lower than the first risk level.

[0050] In an optional embodiment, first, according to the predicted catch of multiple fishing boats within the operating range information and the relevant market prices, the predicted total output value of the target fishing ground is calculated by predicted total output value = predicted catch × average market price, wherein the average market price can be obtained according to historical data or market surveys. Then, the loan amount of the target loan business is obtained, and the ratio of the loan amount to the predicted total output value is calculated by target result = (loan amount / predicted total output value) × 100% to obtain the target result. Finally, according to the target result and the preset value set based on the loan policy, risk tolerance and other factors, the loan risk level is determined: if the target result is greater than the preset value, the loan risk level is determined to be the first risk level, indicating that the loan amount is higher than the predicted total output value and the risk is greater; if the target result is less than or equal to the preset value, the loan risk level is determined to be the second risk level, indicating that the loan amount is lower than the predicted total output value and the risk is smaller.

[0051] By calculating the ratio of the loan amount to the predicted total output value, the loan risk level of the target fishery can be determined quickly and accurately.

[0052] Optionally, in the loan risk assessment method provided in the embodiment of the present application, after determining that the loan risk level is the second risk level, the method also includes: when the loan risk level is the second risk level, obtaining historical loan behavior data of the target fishery; based on the historical loan behavior data, determining the loan information of the target fishery, wherein the loan information includes at least: the issuance method of the target loan business and the repayment method of the target fishery; based on the loan information, handling the target loan business for the target fishery.

[0053] In an optional embodiment, when the loan amount is relatively low relative to the predicted total output value and the risk is relatively low, the historical loan behavior data of the target fishery is collected from the database or relevant financial institutions, including information such as loan amount, issuance method, repayment method, and repayment record. The above historical loan behavior data is analyzed to understand the past loan and repayment situation of the target fishery, and then determine the loan issuance method suitable for the target fishery, such as installment issuance, one-time issuance, etc., and repayment method, such as equal principal and interest, equal principal, interest repayment on schedule and principal repayment on maturity, etc. The target loan business is processed for the target fishery based on the above loan information. At the same time, the loan amount is issued to the target fishery in accordance with the agreed loan issuance method, and the repayment situation of the target fishery is continuously followed up during the loan period to ensure that the loan is recovered on time.

[0054] Through the historical loan behavior data of the target fishery, the loan information of the target fishery can be quickly determined, and the target loan business can be accurately processed for the target fishery based on the loan information.

[0055] Optionally, in the loan risk assessment method provided in the embodiment of the present application, before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, the method also includes: acquiring the location information of the target fishing ground; based on the location information, collecting initial remote sensing images of the target fishing ground in multiple time periods; and preprocessing the initial remote sensing images to obtain the target remote sensing images.

[0056] In an optional embodiment, a geographic information system (GIS) or a map service (such as Google Maps API) is used to query the geographic coordinates (longitude and latitude) of the target fishing ground, thereby determining the scope and boundaries of the target fishing ground. Through Landsat, Sentinel-2 or other commercial satellite data providers, according to the geographic location and time range of the target fishing ground, the initial remote sensing image within the corresponding time period is downloaded from the selected data source, and the initial remote sensing image is corrected, cropped, obstruction removed, denoised, and evenly lighted, etc., to obtain the target remote sensing image.

[0057] It should be noted that the denoising operation in the preprocessing operation is based on the median filter, using an adaptive median filter to remove the high-intensity salt and pepper noise and Gaussian noise to protect the edge details of the remote sensing image. Even light processing (histogram equalization) is also called grayscale equalization, which means that an image with uneven grayscale distribution is converted into an image with uniform grayscale distribution (that is, the histogram of the image is uniform) through a certain grayscale mapping, that is, there are roughly the same number of pixels at each grayscale level. In the image after histogram equalization, the pixels will be distributed on as many grayscale levels as possible and the distribution will be uniform. Therefore, such an image will have a large grayscale range and a high contrast.

[0058] By preprocessing the initial remote sensing image, the target remote sensing image of the target fishing ground can be obtained quickly and accurately, and subsequent analysis and processing can be performed based on the target remote sensing image.

[0059] It should be noted that Figure 2 Schematic diagram of a loan risk assessment method provided according to an embodiment of the present application. Figure 2As shown, first, according to the location information of the target fishing ground, the initial remote sensing images of the target fishing ground in multiple time periods are collected, and preprocessing operations are performed on them to obtain the target remote sensing images of the target fishing ground. Then, according to the edge detection module of the target detection model, the edge detection of the target remote sensing image is performed to obtain the edge information of the target remote sensing image, and according to the feature extraction module of the target detection model, the feature extraction of the above edge information is performed to obtain the first feature information of the target remote sensing image, and according to the analysis and processing module of the target detection model, the first feature information is identified to obtain the operating range information of the fishing boats in the target fishing ground. At the same time, according to the encoding module of the time series prediction model, the feature extraction of the fishing behavior data of multiple fishing boats in the historical time period is performed to obtain the second feature information of multiple fishing boats in the historical time period, and according to the decoding module of the time series prediction model, the second feature information is predicted to obtain the predicted catch of multiple fishing boats in the target time period. Finally, the predicted total output value of the target fishery is calculated based on the operating scope information and the predicted catch, and the ratio of the loan amount of the target loan business to the predicted total output value is calculated to obtain the target result. If the target result is greater than the preset value, the loan risk level is determined to be the first risk level; if the target result is less than or equal to the preset value, the loan risk level is determined to be the second risk level. In the case where the loan risk level is the second risk level, the loan information such as the issuance method of the target loan business of the target fishery and the repayment method of the target fishery are determined based on the historical loan behavior data of the target fishery, and the loan business is processed for the target fishery based on the above loan information.

[0060] The loan risk assessment method provided by the embodiment of the present application is to detect and identify the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model, and obtain the operating range information of the fishing boats in the target fishing ground; predict the catch of multiple fishing boats in the target fishing ground according to the time series prediction model, and obtain the predicted catch of multiple fishing boats in the target time period; determine the loan risk level of the target fishing ground according to the operating range information and the predicted catch, wherein the loan risk level is used to indicate whether the target loan business is handled for the target fishing ground, which solves the problem that the risk of handling loans for the fishing ground is assessed by manually inspecting the situation of the fishing ground in the relevant technology, resulting in a relatively low assessment accuracy of the risk of handling loans for the fishing ground. In summary, in this scheme, first, the target remote sensing images of the target fishing ground in different time periods are analyzed using the target detection model to detect and identify the operating range information of the fishing boat. Then, the catch of multiple fishing boats in the target time period is predicted by the time series prediction model to obtain the predicted catch of these fishing boats. Finally, based on the operating range information and predicted catch, the loan risk level of the target fishery is determined to evaluate whether the target fishery should be provided with loan business. The operating range information of multiple fishing boats in the fishery is determined by the target detection model, and the predicted catch of multiple fishing boats in the loan period is determined by the time series model. The loan risk level of the target fishery is comprehensively assessed based on the operating range information and the predicted catch, avoiding the traditional method of manually determining the loan risk level of the target fishery, avoiding the subjective influence of the assessor, and also improving the assessment speed of the target fishery, thereby achieving the effect of improving the assessment accuracy of the risk of applying for loans to the fishery.

[0061] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0062] The present application also provides a loan risk assessment device. It should be noted that the loan risk assessment device of the present application can be used to execute the loan risk assessment method provided in the present application. The loan risk assessment device provided in the present application is introduced below.

[0063] Figure 3 Schematic diagram of a loan risk assessment device according to an embodiment of the present application. Figure 3 As shown, the device includes: a detection unit 301, a prediction unit 302, and a first determination unit 303.

[0064] The detection unit 301 is used to detect and identify the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model, and obtain the operation range information of the fishing boats in the target fishing ground;

[0065] The prediction unit 302 is used to predict the catch of multiple fishing boats in the target fishing ground according to the time series prediction model to obtain the predicted catch of multiple fishing boats in the target time period;

[0066] The first determination unit 303 is used to determine the loan risk level of the target fishery based on the operation range information and the predicted fishing volume, wherein the loan risk level is used to indicate whether to process the target loan business for the target fishery.

[0067] The loan risk assessment device provided by the embodiment of the present application comprises a detection unit 301 which detects and identifies the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model, and obtains the operating range information of the fishing boats in the target fishing ground; a prediction unit 302 predicts the catch of multiple fishing boats in the target fishing ground according to the time series prediction model, and obtains the predicted catch of multiple fishing boats in the target time period; a first determination unit 303 determines the loan risk level of the target fishing ground according to the operating range information and the predicted catch, wherein the loan risk level is used to indicate whether the target loan business is handled for the target fishing ground, and solves the problem that the risk of handling loans for the fishing ground is assessed by manually inspecting the situation of the fishing ground in the related art, resulting in a relatively low assessment accuracy of the risk of handling loans for the fishing ground. In this scheme, first, the target remote sensing images of the target fishing ground in different time periods are analyzed by using the target detection model to detect and identify the operating range information of the fishing boats. Then, the catch of multiple fishing boats in the target time period is predicted by the time series prediction model to obtain the predicted catch of these fishing boats. Finally, based on the operating range information and predicted catch, the loan risk level of the target fishery is determined to evaluate whether the target fishery should be provided with loan business. The operating range information of multiple fishing boats in the fishery is determined by the target detection model, and the predicted catch of multiple fishing boats in the loan period is determined by the time series model. The loan risk level of the target fishery is comprehensively assessed based on the operating range information and the predicted catch, avoiding the traditional method of manually determining the loan risk level of the target fishery, avoiding the subjective influence of the assessor, and also improving the assessment speed of the target fishery, thereby achieving the effect of improving the assessment accuracy of the risk of applying for loans to the fishery.

[0068] Optionally, in the loan risk assessment device provided in the embodiment of the present application, the detection unit includes: a detection module, used to perform edge detection on the target remote sensing image according to the edge detection module of the target detection model, and obtain edge information of the target remote sensing image; a first extraction module, used to perform feature extraction on the edge information according to the feature extraction module of the target detection model, and obtain first feature information of the target remote sensing image; an identification module, used to identify the first feature information according to the analysis and processing module of the target detection model, and obtain the operating range information of the fishing boat in the target fishing ground.

[0069] Optionally, in the loan risk assessment device provided in the embodiment of the present application, the prediction unit includes: an acquisition module for acquiring fishing behavior data of multiple fishing boats in a historical time period, wherein the fishing behavior data includes at least: the fishing quantity and fishing weight of each fishing boat; a second extraction module for performing feature extraction on the fishing behavior data based on the encoding module of the time series prediction model to obtain second feature information of the multiple fishing boats in the historical time period; a prediction module for predicting the second feature information based on the decoding module of the time series prediction model to obtain the predicted fishing volume of the multiple fishing boats in the target time period.

[0070] Optionally, in the loan risk assessment device provided in the embodiment of the present application, the second extraction module includes: a processing submodule, used to process the fishing behavior data through the fully connected layer and the nonlinear activation function layer in the encoding module to obtain multiple local features of the fishing behavior data; a fitting submodule, used to fit the multiple local features of the fishing behavior data through the basis function layer in the encoding module to obtain the second feature information of multiple fishing boats in the historical time period.

[0071] Optionally, in the loan risk assessment device provided in the embodiment of the present application, the first determination unit includes: a first calculation module, used to perform calculations based on the operating scope information and the predicted catch to obtain the predicted total output value of the target fishing ground; a second calculation module, used to calculate the ratio of the loan amount of the target loan business to the predicted total output value to obtain the target result; a first determination module, used to determine that the loan risk level is a first risk level if the target result is greater than a preset value; a second determination module, used to determine that the loan risk level is a second risk level if the target result is less than or equal to a preset value, wherein the second risk level is lower than the first risk level.

[0072] Optionally, in the loan risk assessment device provided in the embodiment of the present application, the device also includes: a first acquisition unit, used to obtain historical loan behavior data of the target fishery after determining that the loan risk level is the second risk level, when the loan risk level is the second risk level; a second determination unit, used to determine the loan information of the target fishery based on the historical loan behavior data, wherein the loan information includes at least: the issuance method of the target loan business and the repayment method of the target fishery; a processing unit, used to handle the target loan business for the target fishery based on the loan information.

[0073] Optionally, in the loan risk assessment device provided in the embodiment of the present application, the device also includes: a second acquisition unit, used to acquire the location information of the target fishing ground before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground; an acquisition unit, used to acquire initial remote sensing images of the target fishing ground in multiple time periods based on the location information; and a preprocessing unit, used to preprocess the initial remote sensing image to obtain the target remote sensing image.

[0074] The loan risk assessment device includes a processor and a memory. The above-mentioned detection unit 301, prediction unit 302, first determination unit 303, etc. are all stored in the memory as program units. The processor executes the above-mentioned program units stored in the memory to achieve accurate assessment of loan risks.

[0075] The processor includes a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and accurate assessment of loan risks can be achieved by adjusting kernel parameters.

[0076] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0077] An embodiment of the present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, a loan risk assessment method is implemented.

[0078] An embodiment of the present invention provides a processor, which is used to run a program, wherein the loan risk assessment method is executed when the program is run.

[0079] like Figure 4As shown, an embodiment of the present invention provides an electronic device, the device includes a processor, a memory and a program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: detecting and identifying target remote sensing images of a target fishing ground in multiple time periods based on a target detection model to obtain operating range information of fishing boats in the target fishing ground; predicting the catch of multiple fishing boats in the target fishing ground based on a time series prediction model to obtain the predicted catch of multiple fishing boats in the target time period; determining the loan risk level of the target fishing ground based on the operating range information and the predicted catch, wherein the loan risk level is used to indicate whether to process a target loan business for the target fishing ground.

[0080] Optionally, target remote sensing images of the target fishing ground in multiple time periods are detected and identified based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, including: performing edge detection on the target remote sensing image according to the edge detection module of the target detection model to obtain edge information of the target remote sensing image; performing feature extraction on the edge information according to the feature extraction module of the target detection model to obtain first feature information of the target remote sensing image; and identifying the first feature information according to the analysis and processing module of the target detection model to obtain the operating range information of the fishing boats in the target fishing ground.

[0081] Optionally, predicting the catch of multiple fishing vessels in a target fishing ground based on a time series prediction model to obtain the predicted catch of multiple fishing vessels in a target time period includes: obtaining fishing behavior data of multiple fishing vessels in a historical time period, wherein the fishing behavior data at least includes: the catch quantity and catch weight of each fishing vessel; extracting features of the fishing behavior data based on an encoding module of the time series prediction model to obtain second feature information of the multiple fishing vessels in the historical time period; predicting the second feature information based on a decoding module of the time series prediction model to obtain the predicted catch of multiple fishing vessels in the target time period.

[0082] Optionally, feature extraction is performed on the fishing behavior data according to the encoding module of the time series prediction model to obtain second feature information of multiple fishing boats in the historical time period, including: processing the fishing behavior data through the fully connected layer and the nonlinear activation function layer in the encoding module to obtain multiple local features of the fishing behavior data; fitting the multiple local features of the fishing behavior data through the basis function layer in the encoding module to obtain the second feature information of multiple fishing boats in the historical time period.

[0083] Optionally, determining the loan risk level of the target fishery based on the operating scope information and the predicted catch volume includes: performing calculations based on the operating scope information and the predicted catch volume to obtain the predicted total output value of the target fishery; calculating the ratio of the loan amount of the target loan business to the predicted total output value to obtain a target result; if the target result is greater than a preset value, determining the loan risk level to be a first risk level; if the target result is less than or equal to a preset value, determining the loan risk level to be a second risk level, wherein the second risk level is lower than the first risk level.

[0084] Optionally, after determining that the loan risk level is the second risk level, the method also includes: when the loan risk level is the second risk level, obtaining historical loan behavior data of the target fishery; determining the loan information of the target fishery based on the historical loan behavior data, wherein the loan information includes at least: the issuance method of the target loan business and the repayment method of the target fishery; and processing the target loan business for the target fishery based on the loan information.

[0085] Optionally, before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, the method also includes: acquiring the location information of the target fishing ground; based on the location information, collecting initial remote sensing images of the target fishing ground in multiple time periods; and preprocessing the initial remote sensing images to obtain the target remote sensing images.

[0086] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0087] The present application also provides a computer program product which, when executed on a data processing device, is suitable for executing a program that initializes the following method steps: detecting and identifying target remote sensing images of a target fishing ground in multiple time periods based on a target detection model to obtain operating range information of fishing boats in the target fishing ground; predicting the catch of multiple fishing boats in the target fishing ground based on a time series prediction model to obtain the predicted catch of multiple fishing boats in the target time period; determining the loan risk level of the target fishing ground based on the operating range information and the predicted catch, wherein the loan risk level is used to indicate whether to process a target loan business for the target fishing ground.

[0088] Optionally, target remote sensing images of the target fishing ground in multiple time periods are detected and identified based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, including: performing edge detection on the target remote sensing image according to the edge detection module of the target detection model to obtain edge information of the target remote sensing image; performing feature extraction on the edge information according to the feature extraction module of the target detection model to obtain first feature information of the target remote sensing image; and identifying the first feature information according to the analysis and processing module of the target detection model to obtain the operating range information of the fishing boats in the target fishing ground.

[0089] Optionally, predicting the catch of multiple fishing vessels in a target fishing ground based on a time series prediction model to obtain the predicted catch of multiple fishing vessels in a target time period includes: obtaining fishing behavior data of multiple fishing vessels in a historical time period, wherein the fishing behavior data at least includes: the catch quantity and catch weight of each fishing vessel; extracting features of the fishing behavior data based on an encoding module of the time series prediction model to obtain second feature information of the multiple fishing vessels in the historical time period; predicting the second feature information based on a decoding module of the time series prediction model to obtain the predicted catch of multiple fishing vessels in the target time period.

[0090] Optionally, feature extraction is performed on the fishing behavior data according to the encoding module of the time series prediction model to obtain second feature information of multiple fishing boats in the historical time period, including: processing the fishing behavior data through the fully connected layer and the nonlinear activation function layer in the encoding module to obtain multiple local features of the fishing behavior data; fitting the multiple local features of the fishing behavior data through the basis function layer in the encoding module to obtain the second feature information of multiple fishing boats in the historical time period.

[0091] Optionally, determining the loan risk level of the target fishery based on the operating scope information and the predicted catch volume includes: performing calculations based on the operating scope information and the predicted catch volume to obtain the predicted total output value of the target fishery; calculating the ratio of the loan amount of the target loan business to the predicted total output value to obtain a target result; if the target result is greater than a preset value, determining the loan risk level to be a first risk level; if the target result is less than or equal to a preset value, determining the loan risk level to be a second risk level, wherein the second risk level is lower than the first risk level.

[0092] Optionally, after determining that the loan risk level is the second risk level, the method also includes: when the loan risk level is the second risk level, obtaining historical loan behavior data of the target fishery; determining the loan information of the target fishery based on the historical loan behavior data, wherein the loan information includes at least: the issuance method of the target loan business and the repayment method of the target fishery; and processing the target loan business for the target fishery based on the loan information.

[0093] Optionally, before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods based on the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, the method also includes: acquiring the location information of the target fishing ground; based on the location information, collecting initial remote sensing images of the target fishing ground in multiple time periods; and preprocessing the initial remote sensing images to obtain the target remote sensing images.

[0094] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0095] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0096] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0097] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0098] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0099] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0100] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0101] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0102] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0103] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A loan risk assessment method, characterized in that: include: Detecting and identifying target remote sensing images of a target fishing ground in multiple time periods according to a target detection model to obtain operating range information of fishing boats in the target fishing ground; Predicting the catch of multiple fishing boats in the target fishing ground according to the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period; The loan risk level of the target fishery is determined based on the operating scope information and the predicted catch, wherein the loan risk level is used to indicate whether to process a target loan business for the target fishery.

2. The method according to claim 1, characterized in that According to the target detection model, the target remote sensing images of the target fishing ground in multiple time periods are detected and identified, and the operating range information of the fishing boats in the target fishing ground is obtained, including: Performing edge detection on the target remote sensing image according to the edge detection module of the target detection model to obtain edge information of the target remote sensing image; Extracting features from the edge information according to a feature extraction module of the target detection model to obtain first feature information of the target remote sensing image; The first feature information is identified according to the analysis and processing module of the target detection model to obtain the operating range information of the fishing boats in the target fishing ground.

3. The method according to claim 1, characterized in that Predicting the catch of multiple fishing boats in the target fishing ground according to the time series prediction model to obtain the predicted catch of the multiple fishing boats in the target time period includes: Acquire fishing behavior data of the plurality of fishing vessels in a historical time period, wherein the fishing behavior data at least includes: the catch quantity and catch weight of each fishing vessel; Extracting features from the fishing behavior data according to the encoding module of the time series prediction model to obtain second feature information of the multiple fishing boats in a historical time period; The second feature information is predicted according to the decoding module of the time series prediction model to obtain the predicted catch of the multiple fishing vessels in the target time period.

4. The method according to claim 3, characterized in that: The feature extraction of the fishing behavior data is performed according to the encoding module of the time series prediction model to obtain the second feature information of the plurality of fishing boats in the historical time period, including: Processing the fishing behavior data through a fully connected layer and a nonlinear activation function layer in the encoding module to obtain a plurality of local features of the fishing behavior data; The multiple local features of the fishing behavior data are fitted through the basis function layer in the encoding module to obtain the second feature information of the multiple fishing boats in the historical time period.

5. The method according to claim 1, characterized in that Determining the loan risk level of the target fishery based on the operation scope information and the predicted catch includes: Calculating based on the operation range information and the predicted catch to obtain the predicted total output value of the target fishing ground; Calculating the ratio of the loan amount of the target loan business to the predicted total output value to obtain a target result; If the target result is greater than a preset value, the loan risk level is determined to be the first risk level; If the target result is less than or equal to the preset value, the loan risk level is determined to be a second risk level, wherein the second risk level is lower than the first risk level.

6. The method according to claim 5, characterized in that After determining that the loan risk level is the second risk level, the method further includes: When the loan risk level is the second risk level, obtaining historical loan behavior data of the target fishery; Determining the loan information of the target fishery based on the historical loan behavior data, wherein the loan information at least includes: the issuance method of the target loan business and the repayment method of the target fishery; Based on the loan information, the target loan business is processed for the target fishery.

7. The method according to claim 1, characterized in that Before detecting and identifying the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model to obtain the operating range information of the fishing boats in the target fishing ground, the method further includes: Acquiring location information of the target fishing ground; Based on the location information, collecting initial remote sensing images of the target fishing ground in multiple time periods; The initial remote sensing image is preprocessed to obtain the target remote sensing image.

8. A loan risk assessment device, characterized in that: include: A detection unit, used to detect and identify the target remote sensing images of the target fishing ground in multiple time periods according to the target detection model, and obtain the operating range information of the fishing boats in the target fishing ground; A prediction unit, used to predict the catch of multiple fishing boats in the target fishing ground according to a time series prediction model, and obtain the predicted catch of the multiple fishing boats in the target time period; The first determination unit is used to determine the loan risk level of the target fishery based on the operation scope information and the predicted catch, wherein the loan risk level is used to indicate whether to process a target loan business for the target fishery.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the storage medium is controlled to execute the loan risk assessment method according to any one of claims 1 to 7 on a device.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the loan risk assessment method described in any one of claims 1 to 7.

Images