Method and device for processing yew data, storage medium and electronic equipment

Abstract

The application discloses a processing method and device of epidemic wood data, a storage medium and an electronic device, relates to the technical field of artificial intelligence, and the method comprises the steps of determining a plurality of to-be-processed regions to be processed, and acquiring a first target remote sensing image of each to-be-processed region; identifying the first target remote sensing image by means of an epidemic wood identification model to obtain first epidemic wood distribution data of each to-be-processed region; and determining an epidemic wood eradication strategy corresponding to each to-be-processed region according to the first epidemic wood distribution data of each to-be-processed region. Through the application, the problem that the efficiency of epidemic wood identification is relatively low due to the manual identification of epidemic wood in the related art is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and more specifically, to a method and apparatus for processing epidemic data, a storage medium, and an electronic device. Background Technology

[0002] Pine wilt disease is one of the most dangerous forest diseases in the forest ecosystem. It is characterized by its high pathogenicity, rapid spread, short onset time, and difficulty in control. Therefore, it is very important to effectively monitor infected trees.

[0003] Currently, most monitoring of pine wilt disease relies on manual surveys. These surveys, which involve "manually searching for hilltops," roughly estimate the number and distribution of infected trees, making it impossible to conduct a comprehensive and accurate survey of infected trees. Furthermore, the time span is very long and the update frequency is slow.

[0004] There is currently no effective solution to the problem that the manual identification of infected trees in related technologies is inefficient. Summary of the Invention

[0005] The main objective of this application is to provide a method, apparatus, storage medium, and electronic device for processing infected wood data, in order to solve the problem that the efficiency of infected wood identification is relatively low due to the manual identification method in related technologies.

[0006] To achieve the above objectives, according to one aspect of this application, a method for processing infected tree data is provided. The method includes: identifying multiple regions to be processed and acquiring a first target remote sensing image for each region; identifying the first target remote sensing image using an infected tree identification model to obtain first infected tree distribution data for each region; and determining an infected tree removal strategy corresponding to each region based on the first infected tree distribution data for each region.

[0007] Further, acquiring the first target remote sensing image for each region to be processed includes: for each region to be processed, if the region to be processed is not being identified for the first time, acquiring a historical remote sensing image of the region to be processed; if a target historical remote sensing image with a first resolution exists in the historical remote sensing image, acquiring a first remote sensing image of the region to be processed with a second resolution, and determining the target historical remote sensing image and the first remote sensing image as the first target remote sensing image, wherein the first resolution is higher than the second resolution.

[0008] Furthermore, after acquiring the historical remote sensing image, the method further includes: if the target historical remote sensing image at the first resolution does not exist in the historical remote sensing image, then acquiring a second remote sensing image at the first resolution of the area to be processed; and determining the second remote sensing image as the first target remote sensing image.

[0009] Further, if the first target remote sensing image includes the target historical remote sensing image and the first remote sensing image, the identification of the first target remote sensing image using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed includes: extracting features from the target historical remote sensing image through the first network branch of the diseased tree identification model to obtain a first feature map; extracting features from the first remote sensing image through the second network branch of the diseased tree identification model to obtain a second feature map; fusing the first feature map and the second feature map to obtain a target feature map; and identifying the target feature map using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0010] Furthermore, if the first target remote sensing image includes the second remote sensing image, the first target remote sensing image is identified using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed, which includes: extracting features from the second remote sensing image through the first network branch of the diseased tree identification model to obtain a third feature map; and identifying the third feature map using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0011] Furthermore, based on the distribution data of the first infected trees in each treatment area, the strategy for removing infected trees in each treatment area is determined by: obtaining the distribution density data of the target tree species in each area; calculating the epidemic risk score for each treatment area based on the distribution density data and the number of infected trees in the distribution data of the first infected trees in each treatment area; and determining the strategy for removing infected trees in each treatment area based on the epidemic risk score.

[0012] Furthermore, based on the epidemic risk score corresponding to each area to be treated, the strategy for removing infected trees corresponding to each area to be treated includes: obtaining the risk level corresponding to each area to be treated based on the epidemic risk score; calculating the cost data for removing infected trees based on the number of infected trees in the first infected tree distribution data of each area to be treated; and determining the strategy for removing infected trees corresponding to each area to be treated based on the risk level and the cost data for removing infected trees.

[0013] Furthermore, after determining the diseased wood removal strategy for each region based on the first diseased wood distribution data of each region to be treated, the method further includes: during the diseased wood removal process in the regions to be treated according to the diseased wood removal strategy corresponding to each region, determining the regions where removal is completed; dividing the regions where removal is completed into multiple sub-regions, and obtaining target data information for each sub-region, wherein the target data information includes at least: the first diseased wood data in the first diseased wood distribution data corresponding to each sub-region, the second diseased wood data planned for removal corresponding to each sub-region, the third diseased wood data actually removed corresponding to each sub-region, the fourth diseased wood data that has been burned corresponding to each sub-region, and the fifth diseased wood data that has been stored corresponding to each sub-region; calculating the diseased wood removal index data for each sub-region based on the target data information for each sub-region; determining the target sub-regions with abnormal diseased wood removal based on the diseased wood removal index data for each sub-region, and performing diseased wood removal again on the target sub-regions.

[0014] Furthermore, after determining the diseased tree removal strategy for each region based on the first diseased tree distribution data, the method further includes: after removing diseased trees from the regions according to the corresponding diseased tree removal strategy, acquiring a second target remote sensing image for each region, wherein the second target remote sensing image includes at least a historical remote sensing image of each region at a first resolution and a third remote sensing image of each region with completed removal at a second resolution; identifying the second target remote sensing image using a diseased tree identification model to obtain second diseased tree distribution data for each region; determining target regions with abnormal diseased tree removal based on the second diseased tree distribution data and the first diseased tree distribution data, and removing diseased trees from the target regions again.

[0015] To achieve the above objectives, according to another aspect of this application, a processing apparatus for infected tree data is provided. The apparatus includes: a first determining unit, configured to determine a plurality of regions to be processed and acquire a first target remote sensing image of each region; a first identifying unit, configured to identify the first target remote sensing image using an infected tree identification model to obtain first infected tree distribution data for each region; and a second determining unit, configured to determine an infected tree removal strategy corresponding to each region based on the first infected tree distribution data for each region.

[0016] Further, the first determining unit includes: a first acquisition module, configured to acquire historical remote sensing images of each region to be processed if the region to be processed is not being identified for the first time; and a collection module, configured to collect a first remote sensing image of the region to be processed at a second resolution if a target historical remote sensing image at a first resolution exists in the historical remote sensing images, and to determine the target historical remote sensing image and the first remote sensing image as the first target remote sensing image, wherein the first resolution is higher than the second resolution.

[0017] Furthermore, the device further includes: an acquisition unit, configured to acquire a second remote sensing image of the area to be processed at a first resolution if, after acquiring the historical remote sensing image, the target historical remote sensing image at the first resolution is not present in the historical remote sensing image; and a third determination unit, configured to determine the second remote sensing image as the first target remote sensing image.

[0018] Further, the first identification unit includes: a first extraction module, configured to extract features from the target historical remote sensing image through a first network branch of the plague-infected tree identification model to obtain a first feature map if the first target remote sensing image includes the target historical remote sensing image and the first remote sensing image; a second extraction module, configured to extract features from the first remote sensing image through a second network branch of the plague-infected tree identification model to obtain a second feature map; a fusion module, configured to fuse the first feature map and the second feature map to obtain a target feature map; and a first identification module, configured to identify the target feature map through the plague-infected tree identification model to obtain first plague-infected tree distribution data for each region to be processed.

[0019] Further, the first identification unit includes: a third extraction module, used to extract features from the second remote sensing image through the first network branch of the plague tree identification model if the first target remote sensing image includes the second remote sensing image, to obtain a third feature map; and a second identification module, used to identify the third feature map through the plague tree identification model to obtain first plague tree distribution data for each area to be processed.

[0020] Further, the second determining unit includes: a second acquisition module, used to acquire the distribution density data of the target tree species corresponding to each region; a calculation module, used to calculate the epidemic risk score corresponding to each region based on the distribution density data and the number of infected trees in the first infected tree distribution data of each region to be treated; and a determining module, used to determine the infected tree removal strategy corresponding to each region to be treated based on the epidemic risk score corresponding to each region to be treated.

[0021] Furthermore, the determining module includes: a first calculation submodule, used to obtain the risk level corresponding to each area to be treated based on the epidemic risk score corresponding to each area to be treated; a second calculation submodule, used to calculate the cost data of epidemic tree removal based on the number of epidemic trees in the first epidemic tree distribution data of each area to be treated; and a determining submodule, used to determine the epidemic tree removal strategy corresponding to each area to be treated based on the risk level corresponding to each area to be treated and the cost data of epidemic tree removal.

[0022] Furthermore, the device further includes: a third determining unit, configured to determine the diseased wood removal strategy corresponding to each region after determining the diseased wood removal strategy for each region based on the first diseased wood distribution data of each region, and to determine the region where diseased wood removal is completed during the process of removing diseased wood from the region based on the diseased wood removal strategy corresponding to each region; a dividing unit, configured to divide the region where diseased wood removal is completed into multiple sub-regions and obtain target data information for each sub-region, wherein the target data information includes at least: the first diseased wood data in the first diseased wood distribution data corresponding to each sub-region, the second diseased wood data planned for removal corresponding to each sub-region, the third diseased wood data actually removed corresponding to each sub-region, the fourth diseased wood data that has been burned corresponding to each sub-region, and the fifth diseased wood data that has been stored corresponding to each sub-region; a calculation unit, configured to calculate the diseased wood removal index data for each sub-region based on the target data information for each sub-region; and a fourth determining unit, configured to determine the target sub-region where diseased wood removal is abnormal based on the diseased wood removal index data for each sub-region, and to perform diseased wood removal again on the target sub-region.

[0023] Furthermore, the apparatus further includes: an acquisition unit, configured to, after determining the diseased tree removal strategy corresponding to each to-be-treated area based on the first diseased tree distribution data of each to-be-treated area, and after performing diseased tree removal on the to-be-treated areas according to the diseased tree removal strategy corresponding to each to-be-treated area, acquire a second target remote sensing image of each to-be-treated area, wherein the second target remote sensing image includes at least a historical remote sensing image of each to-be-treated area at a first resolution and a third remote sensing image of each to-be-treated area at a second resolution; a second identification unit, configured to identify the second target remote sensing image through a diseased tree identification model to obtain the second diseased tree distribution data of each to-be-treated area; and a fifth determination unit, configured to, based on the second diseased tree distribution data and the first diseased tree distribution data, determine the target area with abnormal diseased tree removal, and perform diseased tree removal again on the target area.

[0024] To achieve the above objectives, according to one aspect of this application, a computer-readable storage medium is provided, the storage medium storing a program, wherein, when the program is executed, the device on which the storage medium is located controls the execution of the processing method of the operating system described in any one of the above claims.

[0025] To achieve the above objectives, according to another aspect of this application, an electronic device is also provided, the electronic device including one or more processors and a memory, the memory being used to store processing methods of the one or more processors implementing the operating system described in any one of the above.

[0026] This application employs the following steps: identifying multiple areas to be processed and acquiring a first target remote sensing image for each area; identifying the first target remote sensing image using a diseased tree identification model to obtain the first diseased tree distribution data for each area; and determining the corresponding diseased tree removal strategy for each area based on the first diseased tree distribution data. This solves the problem of low efficiency in diseased tree identification caused by manual methods in related technologies. In this solution, acquiring the first target remote sensing image of the area to be processed and identifying it using a diseased tree identification model significantly improves the efficiency and accuracy of diseased tree identification in remote sensing images, avoiding manual surveys and thus improving the efficiency of diseased tree identification. Furthermore, the rapid diseased tree identification capability allows for quick and comprehensive understanding of the current distribution of diseased trees, enabling accurate determination of the corresponding diseased tree removal strategy for each area. Attached Figure Description

[0027] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0028] Figure 1 This is a flowchart of a method for processing infected tree data according to an embodiment of this application;

[0029] Figure 2 This is a schematic diagram of diseased tree identification provided according to an embodiment of this application;

[0030] Figure 3 This is a flowchart of an optional method for processing infected tree data provided in the embodiments of this application. Figure 1 ;

[0031] Figure 4 This is a schematic diagram of a decision planning system provided according to an embodiment of this application;

[0032] Figure 5 This is a schematic diagram of a device for processing epidemic tree data according to an embodiment of this application;

[0033] Figure 6 This is a schematic diagram of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0034] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0035] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present application.

[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0037] It should be noted that all information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) involved in this disclosure are information and data authorized by the user or fully authorized by all parties. For example, this system has an interface with relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving consent information from the aforementioned user or organization.

[0038] The present invention will now be described in conjunction with preferred implementation steps. Figure 1 This is a flowchart of a method for processing infected tree data according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0039] Step S101: Determine multiple regions to be processed and acquire the first target remote sensing image of each region.

[0040] Optionally, multiple areas requiring disease identification can be identified. It should be noted that these multiple areas can be identified on a township (or other administrative region) basis. After identifying these multiple areas, first target remote sensing images of these areas can be collected using a drone.

[0041] Step S102: The first target remote sensing image is identified by the plague tree identification model to obtain the first plague tree distribution data for each area to be processed.

[0042] Optionally, after obtaining the first target remote sensing image corresponding to each area to be processed, the first target remote sensing image is used to identify infected trees to obtain the aforementioned first infected tree distribution data. It should be noted that the infected tree identification model can be a deep learning network model, such as a CNN convolutional neural network. It should also be noted that the first infected tree distribution data includes, but is not limited to, the location information and quantity of infected trees.

[0043] Step S103: Based on the distribution data of the first infected trees in each area to be treated, determine the infected tree removal strategy for each area to be treated.

[0044] Optionally, the appropriate strategy for removing infected trees can be determined by using the distribution data of the first infected trees in each area to be treated. For example, if there are a large number of infected trees in a certain area to be treated, more personnel can be allocated to remove the infected trees in that area, and that area will be designated as a key monitoring area for subsequent surveys of infected trees.

[0045] In summary, acquiring the first target remote sensing image of the area to be processed and identifying the first target remote sensing image through the diseased tree identification model greatly improves the efficiency and accuracy of diseased tree identification in remote sensing images. This avoids the need for manual field surveys for diseased tree identification, thereby improving the efficiency of diseased tree identification. Furthermore, the rapid diseased tree identification capability enables a quick and comprehensive understanding of the current distribution of diseased trees, thus accurately determining the corresponding diseased tree removal strategy for the area to be processed.

[0046] To improve the accuracy of the diseased tree identification model in identifying diseased trees, the method for processing diseased tree data provided in this application embodiment includes obtaining a first target remote sensing image for each region to be processed: for each region to be processed, if the region to be processed is not being identified for the first time, then a historical remote sensing image of the region to be processed is obtained; if a target historical remote sensing image with a first resolution exists in the historical remote sensing image, then a first remote sensing image with a second resolution of the region to be processed is acquired, and the target historical remote sensing image and the first remote sensing image are determined as the first target remote sensing image, wherein the first resolution is higher than the second resolution.

[0047] Optionally, for each area to be processed, it is determined whether the area to be processed is being identified for the first time. If the area to be processed is not being identified for the first time, historical remote sensing images of the area to be processed are acquired. If a target historical remote sensing image with a first resolution exists in the historical remote sensing images, a first remote sensing image with a second resolution of the area to be processed can be acquired by a drone. Then, the target historical remote sensing image and the first remote sensing image with the second resolution are determined as the first target remote sensing image mentioned above.

[0048] It should be noted that the initial resolution only needs to be sufficient to distinguish the pine forest from background tree species (especially interfering species). For example, taking Masson pine as an example, the needles of Masson pine are approximately 12-20cm long, and the branches are approximately 24-40cm in diameter. Based on the image imaging method, the initial resolution for the first data acquisition can be set to less than or equal to 3cm. It should also be noted that the initial and second resolutions can be set according to actual needs, and are not limited to the aforementioned initial resolution of less than or equal to 3cm. For example, the initial resolution can also be set to 3.5cm. Historical remote sensing images at the initial resolution can accurately distinguish different tree species within the area to be processed.

[0049] It should be noted that, in order to improve the accuracy of subsequent tree identification, the target historical remote sensing image at the first resolution needs to be a historical remote sensing image within a certain time range. If the target historical remote sensing image was acquired a long time ago, there will be a large range of tree changes, which will lead to the problem of not being able to accurately identify the tree species in the first remote sensing image based on the target historical remote sensing image.

[0050] It should be noted that the best time to acquire high-resolution remote sensing images is in autumn. During this period, tree morphology is more complex, and broad-leaved trees will exhibit yellowing and leaf drop. Furthermore, the number of interfering tree species will increase significantly. Therefore, acquiring high-resolution images during this period can accurately identify subtle differences between infected trees and other tree species, improving the accuracy of infected tree identification.

[0051] When identifying infected trees in an area that is not being treated for the first time, and when high-resolution historical remote sensing images exist, acquiring low-resolution remote sensing images can effectively reduce image acquisition costs. When identifying infected trees, pine trees and other interfering tree species can be identified based on high spatial resolution historical remote sensing images, and then infected trees can be identified based on the low-resolution first remote sensing image.

[0052] In an optional embodiment, during the initial identification of infected trees, a remote sensing image of the area to be treated at a first resolution can be acquired by a drone. During the initial identification of infected trees in the area to be treated, the remote sensing image at the first resolution is directly identified as the aforementioned first target remote sensing image.

[0053] In conclusion, by setting the resolution of remote sensing images, we can achieve accurate identification of infected trees while effectively reducing image acquisition costs.

[0054] In an optional embodiment, after acquiring historical remote sensing images, if the target historical remote sensing image at the first resolution is not present in the historical remote sensing images, a second remote sensing image at the first resolution of the area to be processed is acquired; the second remote sensing image is then identified as the first target remote sensing image. The second remote sensing image at the first resolution can accurately identify infected trees, improving the accuracy of infected tree identification.

[0055] To improve the accuracy of the diseased tree identification model in identifying diseased trees, in the diseased tree data processing method provided in this application embodiment, if the first target remote sensing image includes a target historical remote sensing image and a first remote sensing image, the diseased tree identification model is used to identify the first target remote sensing image to obtain the first diseased tree distribution data for each area to be processed. This includes: extracting features from the target historical remote sensing image through the first network branch of the diseased tree identification model to obtain a first feature map; extracting features from the first remote sensing image through the second network branch of the diseased tree identification model to obtain a second feature map; fusing the first feature map and the second feature map to obtain a target feature map; and identifying the target feature map through the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0056] Optionally, the infected tree identification model can be a deep learning algorithm. The infected tree identification model includes at least a first network branch and a second network branch. When the first target remote sensing image includes historical remote sensing images and the first remote sensing image, the first network branch extracts features from the historical remote sensing image at a first resolution to obtain the corresponding first feature map. Then, the second network branch extracts features from the first remote sensing image at a second resolution to obtain the corresponding second feature map. The first feature map and the second feature map are then fused to obtain the target feature map. Finally, the infected tree identification model identifies the target feature map to obtain the final first infected tree distribution data for each region to be processed.

[0057] In an optional embodiment, the first resolution can be a remote sensing image of 3cm, and the second resolution can be a remote sensing image of 6cm. A schematic diagram illustrating the pest identification model's process of identifying pests from the historical remote sensing image and the first remote sensing image is shown below. Figure 2 As shown, the epidemic identification model is mainly based on CNN, with the first layer (i.e. Figure 2 The high-resolution branch model-feature extraction module in the middle is used to extract feature maps of high-resolution remote sensing images (i.e., the target historical remote sensing images mentioned above), and the second layer (i.e. Figure 2The low-resolution branch model (feature extraction module) is used to extract feature maps from the low-resolution remote sensing image (i.e., the first remote sensing image mentioned above) acquired at the current time (i.e., time t0). To fully utilize the features of the high-resolution remote sensing image, feature maps proposed by the two networks are fused. Finally, the fused feature map is used for infected tree identification to obtain the infected tree distribution data corresponding to the area to be processed (i.e., Figure 2 (The distribution of the location of the infected tree in the image at time t0). It should be noted that, as... Figure 2 As shown, the diseased tree identification model can also include a detection module, which identifies diseased trees and obtains the distribution data of diseased trees in the area to be processed.

[0058] It should be noted that the first resolution can also be set to a remote sensing image of 3.5cm, and the second resolution can also be set to a remote sensing image of 7cm. In practical applications, the resolution can be set according to actual needs. This application does not limit the first resolution and the second resolution.

[0059] In summary, this application aligns with the imaging characteristics of ground features in remote sensing images. By fully utilizing the aforementioned diseased tree identification model, it significantly improves the accuracy and efficiency of diseased tree identification and enhances data processing capabilities.

[0060] If the first target remote sensing image includes the second remote sensing image, in the method for processing infected tree data provided in this application embodiment, the first target remote sensing image is identified by an infected tree identification model to obtain the first infected tree distribution data for each area to be processed, including: extracting features from the second remote sensing image through the first network branch of the infected tree identification model to obtain a third feature map; and identifying the third feature map through the infected tree identification model to obtain the first infected tree distribution data for each area to be processed.

[0061] Optionally, if the first target remote sensing image only includes a high-resolution second remote sensing image, the third feature map is obtained by directly extracting features from the second remote sensing image through the first network branch of the plague tree identification model, and finally plague tree identification is performed based on the third feature map to obtain the aforementioned first plague tree distribution data.

[0062] In an alternative embodiment, it can be achieved through... Figure 3The flowchart shown illustrates a survey of infected trees in multiple areas, primarily involving the following steps: 1. Determining the data collection plan and area, and setting up UAV survey data collection events; 2. Determining data collection requirements: For a given area, the initial spatial resolution of the data collection should be sufficient to distinguish pine forests from background tree species (especially interfering species). Taking Masson pine as an example, the needle length of Masson pine is approximately 12-20cm, and the branch diameter is approximately 24-40cm. Based on the image imaging method, the spatial resolution of the initial data collection can be set to less than or equal to 3cm, forming a high spatial resolution base map of the area. Subsequent data collection, assuming no major forest stand alteration occurs in the area, will collect data at even lower spatial resolution to reduce image acquisition costs, forming a continuous spatiotemporal sequence of UAV remote sensing data. When identifying infected trees, the high spatial resolution base map can be used as a reference to confirm infected pine trees and other interfering tree species. 3. Set the collection frequency and cycle. Based on the transmission, spread, and outbreak patterns of pine wilt disease and operational needs, set a survey cycle. Within each survey cycle, after the initial high-precision data collection, subsequent collection intervals and image accuracy requirements can be adjusted based on operational needs. For each survey cycle, spatiotemporal distribution data of infected trees over M years (e.g., M≥3) can be generated, providing data support for subsequent control and treatment of infected trees. It should be noted that the aforementioned spatiotemporal distribution data of infected trees over M years is formed by selecting consecutive M years of spatiotemporal distribution data from each of multiple survey cycles. 4. Data processing: n1 days before eradication, the distribution of infected trees is determined based on the infected tree identification model (e.g., n1≤15). The trees are divided into high, medium, and low risk areas according to their number to assist in the formulation of eradication plans. After eradication, remote sensing image data is collected again in all reported eradication areas, with townships as the unit. n2 days after eradication, the distribution of infected trees after eradication is obtained based on the remote sensing image data after eradication (e.g., n2≤15). The comparison analysis of the data before and after eradication can help determine the current number and quality of eradication in the townships.

[0063] To improve the rationality of setting up diseased tree removal strategies, the diseased tree data processing method provided in this application embodiment, based on the first diseased tree distribution data of each to-be-treated area, determines the diseased tree removal strategy corresponding to each to-be-treated area, including: obtaining the distribution density data of the target tree species corresponding to each area; calculating the epidemic risk score corresponding to each to-be-treated area based on the distribution density data and the number of diseased trees in the first diseased tree distribution data of each to-be-treated area; and determining the diseased tree removal strategy corresponding to each to-be-treated area based on the epidemic risk score corresponding to each to-be-treated area.

[0064] Based on the epidemic risk score corresponding to each area to be treated, the strategy for removing infected trees corresponding to each area to be treated is determined as follows: based on the epidemic risk score corresponding to each area to be treated, the risk level corresponding to each area to be treated is obtained; based on the number of infected trees in the first infected tree distribution data of each area to be treated, the cost data for removing infected trees is calculated; based on the risk level and cost data for removing infected trees corresponding to each area to be treated, the strategy for removing infected trees corresponding to each area to be treated is determined.

[0065] Optionally, the distribution density data of the target tree species corresponding to each region can be obtained, such as the distribution density data of pine forests. This distribution density data can be obtained through forest resource survey data. Then, the epidemic risk score corresponding to each region to be treated can be calculated by combining the distribution density data with the number of infected trees in the first infected tree distribution data of each region to be treated.

[0066] In an optional embodiment, the epidemic risk score corresponding to each area to be processed can be calculated using formula (1):

[0067] R i =a×x i +b×ρ i (1)

[0068] Where i represents the i-th region to be processed, Ri represents the epidemic risk score of the region to be processed, and x i ρ represents the number of infected trees. i The pine forest distribution density data is represented by 'a' and 'b', which represent the weight values ​​of the number of infected trees and the pine forest distribution density data, respectively. In actual operation, the data on the distribution of infected trees is obtained from objective and real UAV imagery, while the pine forest distribution density data is calculated from forest resource survey data. Since the forest resource survey data is obtained through on-site investigation, the attribute information may deviate from the actual pine forest distribution density. Therefore, a higher empirical weight is assigned to the number of infected trees, for example, a = 0.8 and b = 0.2.

[0069] After calculating the epidemic risk score for each area to be treated, the epidemic control strategy for each area can be determined based on the epidemic risk score. For example, if an area has a high epidemic risk score, its control priority is higher, more personnel should be allocated to control the epidemic trees in that area, and it should be designated as a key monitoring area for focused supervision and verification during subsequent epidemic tree surveys.

[0070] In an optional embodiment, after obtaining the epidemic risk score corresponding to each area to be treated, the multiple areas to be treated are divided into high, medium, and low risk zones according to the epidemic risk score, i.e., the risk level corresponding to each area to be treated is determined as described above. Then, the cost data for removing infected trees is calculated based on the number of infected trees in the first infected tree distribution data of each area to be treated.

[0071] In an optional embodiment, the cost data for controlling the infested trees can be calculated using formula (2):

[0072]

[0073] Where C represents the cost of removing infected trees, d represents the cost of felling each infected tree, and N represents the number of areas to be treated.

[0074] After calculating the above-mentioned cost data for the removal of infected trees, a corresponding strategy for the removal of infected trees is obtained based on the risk level and cost data for each area to be treated.

[0075] In summary, by combining risk level and cost data for controlling infected trees, we can more rationally set the strategies for controlling infected trees in the areas to be treated.

[0076] After determining the diseased wood removal strategy for each region based on the first diseased wood distribution data of each region to be treated, the diseased wood data processing method provided in this application embodiment further includes: determining the regions to be treated during the diseased wood removal process according to the diseased wood removal strategy corresponding to each region to be treated; dividing the regions to be treated into multiple sub-regions and obtaining target data information for each sub-region, wherein the target data information includes at least: the first diseased wood data in the first diseased wood distribution data corresponding to each sub-region, the second diseased wood data to be removed according to the plan for each sub-region, the third diseased wood data to be removed according to the actual data of each sub-region, the fourth diseased wood data to be burned according to the data of each sub-region, and the fifth diseased wood data to be stored according to the data of each sub-region; calculating the diseased wood removal index data for each sub-region based on the target data information for each sub-region; determining the target sub-regions with abnormal diseased wood removal based on the diseased wood removal index data for each sub-region, and performing diseased wood removal again on the target sub-regions.

[0077] Optionally, in the process of removing infected trees in the areas to be treated according to the corresponding infected tree removal strategy for each area to be treated, in order to improve the quality of infected tree removal, infected tree supervision is required. First, the areas that have been removed are identified, then the areas that have been removed are divided into multiple sub-areas, and the target data information of each sub-area is obtained.

[0078] It should be noted that the target data includes, but is not limited to, the first infected tree data in the first infected tree distribution data for each sub-region, the second infected tree data planned for removal for each sub-region, the third infected tree data actually removed for each sub-region, the fourth infected tree data that has been burned for each sub-region, and the fifth infected tree data that has been stored for each sub-region. This data allows for an accurate assessment of whether there are any anomalies in the infected tree removal process within a sub-region.

[0079] Then, based on the target data information of each sub-region, the diseased tree removal index data of each sub-region is calculated. Finally, based on the diseased tree removal index data of each sub-region, the target sub-region with abnormal diseased tree removal is identified, and the target sub-region is subjected to diseased tree removal again.

[0080] In an optional embodiment, the above-mentioned pest control index data can be calculated using formula (3):

[0081]

[0082] Among them, m1, m2, m3, and m4 are the aforementioned indicators for the removal of infected trees. These indicators can be used to accurately determine whether there are any abnormal target sub-regions. For example, if m1 > 1.2 (it should be noted that this parameter can be set according to actual needs), it means that the number of trees felled in this area is insufficient, a large number of infected trees have not been felled, and the area needs to undergo further removal of infected trees.

[0083] For example, if m1 < 0.8, it indicates that the amount of trees felled in the area is much larger than the surveyed amount, suggesting the possibility of overreporting. The same logic applies to m2 and m1. If m3 > 2, it indicates that a large number of trees in the area were burned, which may be abnormal. If the value of m4 is not near 1, it indicates a discrepancy between the amount burned, the amount stored, and the amount felled. Theoretically, the amount burned + the amount stored = the amount felled. These indicators of diseased tree control can effectively monitor the quality of control efforts.

[0084] After determining the diseased tree removal strategy for each region based on the first diseased tree distribution data of each region to be treated, the diseased tree data processing method provided in this application embodiment further includes: after removing diseased trees from the regions to be treated according to the diseased tree removal strategy corresponding to each region to be treated, acquiring a second target remote sensing image of each region to be treated, wherein the second target remote sensing image includes at least a historical remote sensing image of each region to be treated at a first resolution and a third remote sensing image of each region to be treated at a second resolution; identifying the second target remote sensing image using a diseased tree identification model to obtain the second diseased tree distribution data of each region to be treated; determining the target region with abnormal diseased tree removal based on the second diseased tree distribution data and the first diseased tree distribution data, and removing diseased trees from the target region again.

[0085] Optionally, after the diseased trees are removed from each area according to the diseased tree removal strategy, a third remote sensing image of the second resolution of each area is acquired, as well as a historical remote sensing image of the first resolution of each area. Then, the diseased trees are identified again in the second target remote sensing image using the diseased tree identification model. Based on the second and first diseased tree distribution data, the target areas with abnormal diseased tree removal are determined, and the target areas are removed again.

[0086] In an optional embodiment, if the amount of diseased wood data after treatment is ≥10 (the parameter can be set based on business needs), then the area is the target area mentioned above.

[0087] In an optional embodiment, if the amount of data on uncut infected trees is ≥10 (the parameter can be set based on business needs), then the area is the target area mentioned above.

[0088] In an optional embodiment, the status of each area can also be determined based on the amount of data from drone surveys of infected trees before and after treatment.

[0089] The method for processing infected tree data provided in this application involves identifying multiple regions to be processed and acquiring a first target remote sensing image for each region. The first target remote sensing image is then identified using an infected tree identification model to obtain the first infected tree distribution data for each region. Based on this data, a corresponding infected tree removal strategy is determined for each region. This method solves the problem of low efficiency in related technologies where infected trees are identified manually. In this solution, acquiring the first target remote sensing image of the region to be processed and identifying it using an infected tree identification model significantly improves the efficiency and accuracy of infected tree identification in remote sensing images. It avoids the need for manual surveys, thereby improving the efficiency of infected tree identification. Furthermore, the rapid identification capability allows for quick and comprehensive understanding of the current infected tree distribution, enabling accurate determination of the corresponding infected tree removal strategy for each region.

[0090] In an optional embodiment, this application also provides a decision-making and planning system, such as... Figure 4As shown, the system includes a pest control planning module, a supervision planning module, and a pest control rectification module. The pest control planning module takes into account the distribution data of diseased trees and the density data of pine forests from the general survey and obtains the epidemic risk score corresponding to each area to be treated. Based on the proportional relationship of the epidemic risk scores, the areas to be treated are divided into high, medium, and low risk zones to assist relevant departments in planning pest control strategies such as pest control teams, personnel, and schedules. The supervision planning module calculates the diseased tree index data using the formula (3) above and formulates a supervision plan based on the diseased tree index data to effectively control the quality of pest control. The pest control rectification module performs rectification processing on each area by comparing the data volume of diseased trees before and after pest control.

[0091] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0092] This application also provides an apparatus for processing infected wood data. It should be noted that this apparatus can be used to execute the method for processing infected wood data provided in this application. The following describes the apparatus for processing infected wood data provided in this application.

[0093] Figure 5 This is a schematic diagram of a device for processing epidemic data according to an embodiment of this application. Figure 5 As shown, the device includes: a first determining unit 501, a first identifying unit 502, and a second determining unit 503.

[0094] The first determining unit 501 is used to determine multiple regions to be processed and to acquire a first target remote sensing image of each region to be processed.

[0095] The first identification unit 502 is used to identify the first target remote sensing image through the plague tree identification model to obtain the first plague tree distribution data of each area to be processed;

[0096] The second determining unit 503 is used to determine the disease removal strategy for each area to be treated based on the first distribution data of diseased trees in each area to be treated.

[0097] The diseased tree data processing apparatus provided in this application embodiment determines multiple regions to be processed by a first determining unit 501 and acquires a first target remote sensing image of each region; a first identification unit 502 identifies the first target remote sensing image using a diseased tree identification model to obtain the first diseased tree distribution data of each region; and a second determining unit 503 determines the corresponding diseased tree removal strategy for each region based on the first diseased tree distribution data. This solves the problem of low efficiency in diseased tree identification caused by manual methods in related technologies. In this solution, acquiring the first target remote sensing image of the region to be processed and identifying it using a diseased tree identification model greatly improves the efficiency and accuracy of diseased tree identification in remote sensing images, avoiding manual surveys for diseased tree identification, thereby improving the efficiency of diseased tree identification. Furthermore, the rapid diseased tree identification capability allows for quick and comprehensive understanding of the current distribution of diseased trees, thus accurately determining the corresponding diseased tree removal strategy for the region to be processed.

[0098] Optionally, in the epidemic tree data processing device provided in the embodiments of this application, the first determining unit includes: a first acquiring module, used to acquire historical remote sensing images of each area to be processed if the area to be processed is not being identified for the first time; and a collection module, used to collect a first remote sensing image of the area to be processed at a second resolution if a target historical remote sensing image at a first resolution exists in the historical remote sensing images, and to determine the target historical remote sensing image and the first remote sensing image as a first target remote sensing image, wherein the first resolution is higher than the second resolution.

[0099] Optionally, in the epidemic data processing apparatus provided in the embodiments of this application, the apparatus further includes: an acquisition unit, configured to acquire a second remote sensing image of the area to be processed at a first resolution if no target historical remote sensing image of the first resolution is present in the historical remote sensing image after acquiring historical remote sensing images; and a third determination unit, configured to determine the second remote sensing image as the first target remote sensing image.

[0100] Optionally, in the diseased wood data processing apparatus provided in this application embodiment, the first identification unit includes: a first extraction module, used to identify the first target remote sensing image through a diseased wood identification model if the first target remote sensing image includes a target historical remote sensing image and a first remote sensing image, to obtain first diseased wood distribution data for each area to be processed, and to extract features from the target historical remote sensing image through a first network branch of the diseased wood identification model to obtain a first feature map; a second extraction module, used to extract features from the first remote sensing image through a second network branch of the diseased wood identification model to obtain a second feature map; a fusion module, used to fuse the first feature map and the second feature map to obtain a target feature map; and a first identification module, used to identify the target feature map through the diseased wood identification model to obtain first diseased wood distribution data for each area to be processed.

[0101] Optionally, in the processing device for infected wood data provided in this application embodiment, the first identification unit includes: a third extraction module, used to identify the first target remote sensing image through an infected wood identification model if the first target remote sensing image includes a second remote sensing image, to obtain the first infected wood distribution data of each area to be processed, and to extract features from the second remote sensing image through the first network branch of the infected wood identification model to obtain a third feature map; and a second identification module, used to identify the third feature map through the infected wood identification model to obtain the first infected wood distribution data of each area to be processed.

[0102] Optionally, in the diseased tree data processing device provided in this application embodiment, the second determining unit includes: a second acquiring module, used to acquire the distribution density data of the target tree species corresponding to each region; a calculation module, used to calculate the disease risk score corresponding to each region based on the distribution density data and the number of diseased trees in the first diseased tree distribution data of each region to be processed; and a determining module, used to determine the diseased tree removal strategy corresponding to each region to be processed based on the disease risk score corresponding to each region to be processed.

[0103] Optionally, in the diseased wood data processing device provided in this application embodiment, the determining module includes: a first calculation submodule, used to obtain the risk level corresponding to each region to be processed based on the epidemic risk score corresponding to each region to be processed; a second calculation submodule, used to calculate the diseased wood removal cost data based on the number of diseased woods in the first diseased wood distribution data of each region to be processed; and a determining submodule, used to determine the diseased wood removal strategy corresponding to each region to be processed based on the risk level and the diseased wood removal cost data corresponding to each region to be processed.

[0104] Optionally, in the diseased wood data processing device provided in this application embodiment, the device further includes: a third determining unit, used to determine the diseased wood removal strategy corresponding to each region to be treated after determining the diseased wood removal strategy corresponding to each region to be treated based on the first diseased wood distribution data of each region to be treated, and to determine the region to be treated after removing the diseased wood in the region to be treated based on the diseased wood removal strategy corresponding to each region to be treated; a dividing unit, used to divide the region to be treated into multiple sub-regions and obtain target data information for each sub-region, wherein the target data information includes at least: the first diseased wood data in the first diseased wood distribution data corresponding to each sub-region, the second diseased wood data to be removed according to the plan for each sub-region, the third diseased wood data to be removed according to the actual data of each sub-region, the fourth diseased wood data to be burned according to the data of each sub-region, and the fifth diseased wood data to be stored according to the data of each sub-region; a calculation unit, used to calculate based on the target data information of each sub-region to obtain the diseased wood removal index data of each sub-region; and a fourth determining unit, used to determine the target sub-region with abnormal diseased wood removal based on the diseased wood removal index data of each sub-region, and to remove the diseased wood again in the target sub-region.

[0105] Optionally, in the diseased wood data processing apparatus provided in this application embodiment, the apparatus further includes: an acquisition unit, configured to, after determining the diseased wood removal strategy corresponding to each region to be processed based on the first diseased wood distribution data of each region to be processed, and after performing diseased wood removal on the regions to be processed based on the diseased wood removal strategy corresponding to each region to be processed, acquire a second target remote sensing image of each region to be processed, wherein the second target remote sensing image includes at least a historical remote sensing image of each region to be processed at a first resolution and a third remote sensing image of each region to be treated at a second resolution; a second identification unit, configured to identify the second target remote sensing image through a diseased wood identification model to obtain the second diseased wood distribution data of each region to be processed; and a fifth determination unit, configured to, based on the second diseased wood distribution data and the first diseased wood distribution data, determine the target region with abnormal diseased wood removal, and perform diseased wood removal on the target region again.

[0106] The aforementioned epidemic data processing device includes a processor and a memory. The first determining unit 501, the first identifying unit 502, and the second determining unit 503 are all stored in the memory as program units. The processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0107] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and the processing of data is achieved by adjusting kernel parameters.

[0108] The memory may include non-permanent memory in computer-readable media, such as 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.

[0109] This invention provides a computer-readable storage medium storing a program thereon, which, when executed by a processor, implements a method for processing epidemic data.

[0110] This invention provides a processor for running a program, wherein the program executes a method for processing epidemic data during runtime.

[0111] like Figure 6 As shown, this embodiment of the invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: determining multiple regions to be processed and acquiring a first target remote sensing image of each region; identifying the first target remote sensing image using a diseased tree identification model to obtain first diseased tree distribution data for each region; and determining a diseased tree removal strategy for each region based on the first diseased tree distribution data.

[0112] Optionally, acquiring the first target remote sensing image of each region to be processed includes: acquiring the first target remote sensing image of each region to be processed includes: for each region to be processed, if the region to be processed is not being identified for the first time, then acquiring historical remote sensing images of the region to be processed; if there is a target historical remote sensing image with a first resolution in the historical remote sensing images, then acquiring a first remote sensing image of the region to be processed with a second resolution, and determining the target historical remote sensing image and the first remote sensing image as the first target remote sensing image, wherein the first resolution is higher than the second resolution.

[0113] Optionally, acquiring the first target remote sensing image for each region to be processed includes: after acquiring historical remote sensing images, the method further includes: if there is no target historical remote sensing image of the first resolution in the historical remote sensing images, then acquiring a second remote sensing image of the region to be processed at the first resolution; and determining the second remote sensing image as the first target remote sensing image.

[0114] Optionally, if the first target remote sensing image includes a target historical remote sensing image and the first remote sensing image, the first target remote sensing image is identified using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed, including: extracting features from the target historical remote sensing image through the first network branch of the diseased tree identification model to obtain a first feature map; extracting features from the first remote sensing image through the second network branch of the diseased tree identification model to obtain a second feature map; fusing the first feature map and the second feature map to obtain a target feature map; and identifying the target feature map using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0115] Optionally, if the first target remote sensing image includes a second remote sensing image, the first target remote sensing image is identified using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed, including: extracting features from the second remote sensing image through the first network branch of the diseased tree identification model to obtain a third feature map; and identifying the third feature map using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0116] Optionally, determining the diseased tree removal strategy for each area to be treated based on the first diseased tree distribution data of each area to be treated includes: obtaining the distribution density data of the target tree species for each area; calculating the disease risk score for each area to be treated based on the distribution density data and the number of diseased trees in the first diseased tree distribution data of each area to be treated; and determining the diseased tree removal strategy for each area to be treated based on the disease risk score for each area to be treated.

[0117] Optionally, determining the diseased tree removal strategy for each area to be treated based on the epidemic risk score includes: obtaining the risk level of each area to be treated based on the epidemic risk score; calculating the diseased tree removal cost data based on the number of diseased trees in the first diseased tree distribution data of each area to be treated; and determining the diseased tree removal strategy for each area to be treated based on the risk level and the diseased tree removal cost data.

[0118] Optionally, after determining the diseased wood removal strategy for each region based on the first diseased wood distribution data of each region, the method further includes: determining the regions where removal is completed during the process of removing diseased wood from the regions based on the diseased wood removal strategy for each region; dividing the regions where removal is completed into multiple sub-regions and obtaining target data information for each sub-region, wherein the target data information includes at least: the first diseased wood data in the first diseased wood distribution data for each sub-region, the second diseased wood data planned for removal for each sub-region, the third diseased wood data actually removed for each sub-region, the fourth diseased wood data that has been burned for each sub-region, and the fifth diseased wood data that has been stored for each sub-region; calculating the diseased wood removal index data for each sub-region based on the target data information for each sub-region; and determining the target sub-regions with abnormal diseased wood removal based on the diseased wood removal index data for each sub-region, and performing diseased wood removal again on the target sub-regions.

[0119] Optionally, after determining the diseased tree removal strategy for each region based on the first diseased tree distribution data, the method further includes: after removing diseased trees from the regions based on the diseased tree removal strategy corresponding to each region, acquiring a second target remote sensing image for each region, wherein the second target remote sensing image includes at least a historical remote sensing image of each region at a first resolution and a third remote sensing image of each region at a second resolution; identifying the second target remote sensing image using a diseased tree identification model to obtain the second diseased tree distribution data for each region; determining the target region with abnormal diseased tree removal based on the second diseased tree distribution data and the first diseased tree distribution data, and removing diseased trees from the target region again.

[0120] The devices mentioned in this article can be servers, PCs, tablets, mobile phones, etc.

[0121] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having the following method steps: determining multiple regions to be processed and acquiring a first target remote sensing image of each region; identifying the first target remote sensing image through a diseased tree identification model to obtain first diseased tree distribution data for each region; and determining a diseased tree removal strategy corresponding to each region based on the first diseased tree distribution data for each region.

[0122] Optionally, acquiring the first target remote sensing image of each region to be processed includes: acquiring the first target remote sensing image of each region to be processed includes: for each region to be processed, if the region to be processed is not being identified for the first time, then acquiring historical remote sensing images of the region to be processed; if there is a target historical remote sensing image with a first resolution in the historical remote sensing images, then acquiring a first remote sensing image of the region to be processed with a second resolution, and determining the target historical remote sensing image and the first remote sensing image as the first target remote sensing image, wherein the first resolution is higher than the second resolution.

[0123] Optionally, acquiring the first target remote sensing image for each region to be processed includes: after acquiring historical remote sensing images, the method further includes: if there is no target historical remote sensing image of the first resolution in the historical remote sensing images, then acquiring a second remote sensing image of the region to be processed at the first resolution; and determining the second remote sensing image as the first target remote sensing image.

[0124] Optionally, if the first target remote sensing image includes a target historical remote sensing image and the first remote sensing image, the first target remote sensing image is identified using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed, including: extracting features from the target historical remote sensing image through the first network branch of the diseased tree identification model to obtain a first feature map; extracting features from the first remote sensing image through the second network branch of the diseased tree identification model to obtain a second feature map; fusing the first feature map and the second feature map to obtain a target feature map; and identifying the target feature map using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0125] Optionally, if the first target remote sensing image includes a second remote sensing image, the first target remote sensing image is identified using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed, including: extracting features from the second remote sensing image through the first network branch of the diseased tree identification model to obtain a third feature map; and identifying the third feature map using the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed.

[0126] Optionally, determining the diseased tree removal strategy for each area to be treated based on the first diseased tree distribution data of each area to be treated includes: obtaining the distribution density data of the target tree species for each area; calculating the disease risk score for each area to be treated based on the distribution density data and the number of diseased trees in the first diseased tree distribution data of each area to be treated; and determining the diseased tree removal strategy for each area to be treated based on the disease risk score for each area to be treated.

[0127] Optionally, determining the diseased tree removal strategy for each area to be treated based on the epidemic risk score includes: obtaining the risk level of each area to be treated based on the epidemic risk score; calculating the diseased tree removal cost data based on the number of diseased trees in the first diseased tree distribution data of each area to be treated; and determining the diseased tree removal strategy for each area to be treated based on the risk level and the diseased tree removal cost data.

[0128] Optionally, after determining the diseased wood removal strategy for each region based on the first diseased wood distribution data of each region, the method further includes: determining the regions where removal is completed during the process of removing diseased wood from the regions based on the diseased wood removal strategy for each region; dividing the regions where removal is completed into multiple sub-regions and obtaining target data information for each sub-region, wherein the target data information includes at least: the first diseased wood data in the first diseased wood distribution data for each sub-region, the second diseased wood data planned for removal for each sub-region, the third diseased wood data actually removed for each sub-region, the fourth diseased wood data that has been burned for each sub-region, and the fifth diseased wood data that has been stored for each sub-region; calculating the diseased wood removal index data for each sub-region based on the target data information for each sub-region; and determining the target sub-regions with abnormal diseased wood removal based on the diseased wood removal index data for each sub-region, and performing diseased wood removal again on the target sub-regions.

[0129] Optionally, after determining the diseased tree removal strategy for each region based on the first diseased tree distribution data, the method further includes: after removing diseased trees from the regions based on the diseased tree removal strategy corresponding to each region, acquiring a second target remote sensing image for each region, wherein the second target remote sensing image includes at least a historical remote sensing image of each region at a first resolution and a third remote sensing image of each region at a second resolution; identifying the second target remote sensing image using a diseased tree identification model to obtain the second diseased tree distribution data for each region; determining the target region with abnormal diseased tree removal based on the second diseased tree distribution data and the first diseased tree distribution data, and removing diseased trees from the target region again.

[0130] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0131] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0132] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0133] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0134] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0135] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0136] Computer-readable media include both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, 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 technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0137] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0138] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for processing infected tree data, characterized in that, include: Identify multiple regions to be processed and acquire the first target remote sensing image for each region; The first target remote sensing image is identified by the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed; Based on the distribution data of the first infected trees in each area to be treated, determine the infected tree removal strategy for each area to be treated; The acquisition of the first target remote sensing image for each region to be processed includes: For each region to be processed, if it is not the first time that the region to be processed has been identified as a diseased tree, then the historical remote sensing image of the region to be processed is obtained. If a target historical remote sensing image with a first resolution exists in the historical remote sensing image, then a first remote sensing image with a second resolution for the area to be processed is acquired, and the target historical remote sensing image and the first remote sensing image are identified as the first target remote sensing image, wherein the first resolution is higher than the second resolution.

2. The method according to claim 1, characterized in that, After acquiring the historical remote sensing images, the method further includes: If the target historical remote sensing image at the first resolution does not exist in the resolution of the historical remote sensing image, then a second remote sensing image at the first resolution of the area to be processed is acquired. The second remote sensing image is identified as the first target remote sensing image.

3. The method according to claim 1, characterized in that, If the first target remote sensing image includes the target historical remote sensing image and the first remote sensing image, the first target remote sensing image is identified using the infected tree identification model to obtain the first infected tree distribution data for each area to be processed, including: The first feature map is obtained by extracting features from the historical remote sensing image of the target through the first network branch of the epidemic tree identification model; The second feature map is obtained by extracting features from the first remote sensing image through the second network branch of the epidemic tree identification model. The first feature map and the second feature map are fused to obtain the target feature map; The target feature map is identified by the diseased tree identification model to obtain the first diseased tree distribution data for each region to be processed.

4. The method according to claim 2, characterized in that, If the first target remote sensing image includes the second remote sensing image, the first target remote sensing image is identified using the infected tree identification model to obtain the first infected tree distribution data for each area to be processed, including: The third feature map is obtained by extracting features from the second remote sensing image through the first network branch of the epidemic tree identification model; The third feature map is identified using the diseased tree identification model to obtain the first diseased tree distribution data for each region to be processed.

5. The method according to claim 1, characterized in that, Based on the distribution data of the first infected trees in each area to be treated, the infected tree removal strategy for each area to be treated is determined as follows: Obtain the distribution density data of the target tree species for each region; Based on the distribution density data and the number of infected trees in the first infected tree distribution data of each area to be treated, the epidemic risk score corresponding to each area to be treated is calculated. Based on the epidemic risk score corresponding to each area to be treated, the epidemic control strategy for each area to be treated is determined.

6. The method according to claim 5, characterized in that, Based on the epidemic risk score corresponding to each area to be treated, the epidemic control strategy for each area to be treated includes: Based on the epidemic risk score corresponding to each area to be processed, the risk level corresponding to each area to be processed is obtained; The cost of removing infected trees is calculated based on the number of infected trees in the first infected tree distribution data of each area to be treated. Based on the risk level of each area to be treated and the cost data of diseased tree removal, a diseased tree removal strategy is determined for each area to be treated.

7. The method according to claim 1, characterized in that, After determining the diseased tree removal strategy for each area based on the first distribution data of infected trees in each area to be treated, the method further includes: During the process of removing infected trees from the areas to be treated according to the corresponding strategy for each area to be treated, the areas where the removal has been completed are determined. The area where the treatment has been completed is divided into multiple sub-regions, and target data information for each sub-region is obtained. The target data information includes at least: the first infected tree data in the first infected tree distribution data corresponding to each sub-region, the second infected tree data planned to be treated corresponding to each sub-region, the third infected tree data actually treated corresponding to each sub-region, the fourth infected tree data that has been burned corresponding to each sub-region, and the fifth infected tree data that has been put into storage corresponding to each sub-region. Based on the target data information of each sub-region, the disease control index data of each sub-region is calculated. Based on the diseased tree removal index data of each sub-region, the target sub-region with abnormal diseased tree removal is identified, and the target sub-region is subjected to diseased tree removal again.

8. The method according to claim 1, characterized in that, After determining the diseased tree removal strategy for each area based on the first distribution data of infected trees in each area to be treated, the method further includes: After removing infected trees in the areas to be treated according to the corresponding strategy for each area, a second target remote sensing image of each area to be treated is obtained. The second target remote sensing image includes at least a first-resolution historical remote sensing image of each area to be treated and a second-resolution third remote sensing image of each area to be treated. The second target remote sensing image is identified by the diseased tree identification model to obtain the second diseased tree distribution data for each area to be processed; Based on the second and first data on the distribution of infected trees, the target area for abnormal tree removal is determined, and the target area is then subjected to another round of tree removal.

9. A device for processing epidemic wood data, characterized in that, include: The first determining unit is used to determine multiple regions to be processed and to acquire the first target remote sensing image of each region to be processed. The first identification unit is used to identify the first target remote sensing image through the diseased tree identification model to obtain the first diseased tree distribution data for each area to be processed; The second determining unit is used to determine the disease removal strategy for each area to be treated based on the first distribution data of diseased trees in each area to be treated. The first determining unit includes: a first acquisition module, configured to acquire historical remote sensing images of each region to be processed if the region to be processed is not being identified for the first time; and a collection module, configured to collect a first remote sensing image of the region to be processed at a second resolution if a target historical remote sensing image at a first resolution exists in the historical remote sensing images, and to determine the target historical remote sensing image and the first remote sensing image as a first target remote sensing image, wherein the first resolution is higher than the second resolution.

10. 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 controls the device to perform the method for processing epidemic data according to any one of claims 1 to 8.

11. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being 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 cause the one or more processors to implement the method for processing epidemic data according to any one of claims 1 to 8.

Images