Forest pest monitoring method, equipment, medium and product

Abstract

The invention relates to the technical field of environment monitoring, in particular to a forest pest monitoring method and device, a medium and a product, and the method comprises the steps: carrying out the pest recognition of a to-be-recognized pest image through a multi-scale target detection model when the pest recognition is carried out on a forest ecological system, and determining the pest recognition information, the multi-scale target detection model is obtained by performing model training on an improved YOLOv7 model by using a training set, and the multi-scale target detection model improves the detection precision of the detection model on to-be-recognized pest images with small targets, dense target shielding and multi-scale targets, so that the precision of forest pest detection is improved on the whole. Then, hazard assessment is performed based on the pest identification information corresponding to each target area in forest ecology, pest hazard assessment information is determined, prevention and control modes are selected based on the pest hazard assessment information, pest prevention and control information is determined, precise prevention and control are executed, and the pest prevention and control efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of environmental monitoring, and in particular to a forest pest monitoring method, equipment, medium and product. Background Art

[0002] As an important natural resource on the earth, forests play an irreplaceable role in maintaining ecological balance, promoting climate stability and protecting biodiversity. However, forest ecosystems face many threats, and the invasion of forest pests is one of the important issues. The rapid reproduction and spread of forest pests can cause serious damage to trees and even lead to the death of trees, thus affecting the development of the entire forest ecosystem and related industries.

[0003] In order to protect the stability and biodiversity of forest ecosystems, pests can be discovered and controlled earlier and more accurately by regularly collecting and identifying pests in forest areas. However, traditional forest pest monitoring methods mainly rely on manual inspections and sampling, and experts visually identify pest species and numbers. This method is not only labor-intensive and inefficient, but also prone to missed detection or misjudgment, resulting in low accuracy in forest pest detection.

[0004] Therefore, how to improve the accuracy of forest pest detection is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0005] The purpose of this application is to provide a forest pest monitoring method, equipment, medium and product to solve at least one of the above technical problems.

[0006] The above invention objectives of the present application are achieved through the following technical solutions: In the first aspect, the present application provides a forest pest monitoring method, which adopts the following technical solution: A forest pest monitoring method, comprising: Acquire a pest image to be identified and a multi-scale target detection model, use the multi-scale target detection model to perform pest identification on the pest image to be identified, and determine pest identification information, wherein the pest identification information includes: pest category and pest location; Performing a hazard assessment based on the pest identification information corresponding to each target area in the forest ecology to determine pest hazard assessment information, and selecting a control method based on the pest hazard assessment information to determine pest control information, wherein the pest control information is used to reduce the harm of pests to forest resources; The training method of the multi-scale object detection model includes: Acquire pest images sent by the pest monitoring lamp, perform preprocessing and data labeling based on the pest images to obtain pest sample images, and divide the pest sample images into a training set, a verification set, and a test set in proportion; An improved YOLOv7 model is obtained, and the improved YOLOv7 model is trained using the training set to obtain a multi-scale target detection model, wherein the improved YOLOv7 model is a YOLOv7 model that introduces a scale sequence feature fusion module and a triple feature encoding efficient aggregation module.

[0007] By adopting the above technical solution, when identifying pests in forest ecosystems, a multi-scale target detection model is used to identify pest images to be identified and determine pest identification information, wherein the multi-scale target detection model is obtained by training the improved YOLOv7 model using a training set. The multi-scale target detection model improves the detection accuracy of the detection model for images of pests to be identified with small targets, densely occluded targets, and multi-scale targets, thereby improving the overall accuracy of forest pest detection. Then, based on the pest identification information corresponding to each target area in the forest ecology, a hazard assessment is performed to determine the pest hazard assessment information, and based on the pest hazard assessment information, a prevention and control method is selected to determine the pest control information, and precise prevention and control is performed to improve the efficiency of pest control.

[0008] In a preferred example, the present application may be further configured as follows: the method for constructing the improved YOLOv7 model includes: Acquire multiple feature maps of different scales extracted by the backbone layer, select a first feature map based on the multiple feature maps of different scales, and determine the first feature map; Adding the scale sequence feature fusion module in the head layer, wherein the scale sequence feature fusion module performs different scale feature fusion based on the first feature map to improve the detection accuracy of the improved YOLOv7 model for small targets; The triple feature coding efficient aggregation module is added in the head layer based on the multiple feature maps of different scales, wherein the triple feature coding efficient aggregation module improves the robustness and accuracy of target detection when dense targets and multi-scale targets coexist.

[0009] In a preferred example, the present application may be further configured as follows: after the improved YOLOv7 model is trained using the training set to obtain a multi-scale target detection model, the application may further include: Using the validation set and the test set, a model performance test is performed on the multi-scale object detection model to determine the model performance; Acquiring performance standard information, and performing model performance evaluation based on the model performance and the performance standard information, and determining a model performance evaluation result; When the model performance evaluation result meets the requirements, the multi-scale target detection model is used to identify pests; when the model performance evaluation result does not meet the requirements, the pest sample images in the training set are added, and the multi-scale target detection model is iteratively trained using the added training set until the multi-scale target detection model meets the requirements.

[0010] In a preferred example, the present application may be further configured as follows: performing hazard assessment based on the pest identification information corresponding to each target area in the forest ecology to determine the pest hazard assessment information includes: Summarize pest information based on the pest identification information corresponding to each target area in the forest ecology to determine pest summary information; Acquire an ecological simulation model, and perform a hazard assessment based on the ecological simulation model and the pest summary information to determine the degree of pest hazard; Obtain historical pest monitoring data, use time series analysis technology to predict the development trend of the historical pest monitoring data and the pest summary information, and determine pest spread prediction information; A risk assessment is performed based on the pest spread prediction information to determine pest risk information, and the pest damage degree and the pest risk information are integrated to obtain the pest damage assessment information.

[0011] In a preferred example, the present application may be further configured as follows: the selection of a control method based on the pest hazard assessment information and the determination of pest control information include: Performing a control demand analysis based on the pest damage degree and the pest risk information, and selecting a target control method from multiple control methods; A pest control plan is formulated based on the target control measures and the pest summary information to determine the pest control information.

[0012] In a preferred example, the present application may be further configured as follows: after the pest control information is determined by formulating a control plan based on the target control means and the pest summary information, the following may also be included: The pest control information is sent to the pest control terminal so that pest control personnel can carry out control according to the pest control information to reduce the harm of pests to forest resources; After the pest control information is sent to the pest control terminal for a control verification period, current pest summary information is obtained, and control effect analysis is performed based on the current pest summary information to determine a control effect analysis result; When the pest control effect analysis result is not up to standard, pest control optimization is performed based on the current pest summary information and the pest control information to determine optimized pest control information.

[0013] In a second aspect, the present application provides an electronic device, which adopts the following technical solution: at least one processor; Memory; At least one application, wherein the at least one application is stored in the memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned forest pest monitoring method.

[0014] In a third aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program, which, when executed in a computer, causes the computer to execute the above-mentioned forest pest monitoring method.

[0015] In a fourth aspect, the present application provides a computer program product, which adopts the following technical solution: A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the forest pest monitoring method described above is implemented.

[0016] In summary, the present application includes at least one of the following beneficial technical effects: When identifying pests in forest ecosystems, a multi-scale target detection model is used to identify pest images to be identified and determine pest identification information. The multi-scale target detection model is obtained by training the improved YOLOv7 model using a training set. The multi-scale target detection model improves the detection accuracy of the detection model for images of pests to be identified with small targets, densely occluded targets, and multi-scale targets, thereby improving the overall accuracy of forest pest detection. Then, based on the pest identification information corresponding to each target area in the forest ecology, a hazard assessment is performed to determine the pest hazard assessment information, and a prevention and control method is selected based on the pest hazard assessment information to determine the pest control information, perform precise prevention and control, and improve the efficiency of pest control.

[0017] The first feature map is selected based on multiple feature maps of different scales, the first feature map is determined, and a scale sequence feature fusion module is added to the head layer, wherein the scale sequence feature fusion module performs different scale feature fusion based on the first feature map to improve the detection accuracy of the improved YOLOv7 model for small targets. Furthermore, a triple feature encoding efficient aggregation module is added to the head layer based on multiple feature maps of different scales. The triple feature encoding efficient aggregation module improves the robustness and accuracy of target detection when dense targets and multi-scale targets coexist. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of a forest pest monitoring method according to one embodiment of the present application; Figure 2 It is a flowchart of a training method of a multi-scale target detection model in one embodiment of the present application; Figure 3 It is a schematic diagram of the structure of an improved YOLOv7 model in one embodiment of the present application; Figure 4 It is a structural schematic diagram of a forest pest monitoring device according to one embodiment of the present application; Figure 5 It is a structural schematic diagram of an electronic device according to one embodiment of the present application. DETAILED DESCRIPTION

[0019] The following combination Figures 1 to 5 This application is described in further detail.

[0020] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, a person skilled in the art may make non-creative modifications to the present embodiment as needed, but such modifications are protected by the patent law as long as they are within the scope of the present application.

[0021] In order to make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described in conjunction with the drawings in the embodiment of the present application. Obviously, the described embodiment is a part of the embodiment 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 the field without creative work are within the scope of protection of the present application. It should be noted that in the optional embodiments of the present application, the object information and other related data involved, when the embodiments in the present application are applied to specific products or technologies, need to obtain the permission or consent of the object, and the collection, use and processing of the relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. In other words, if the data related to the object is involved in the embodiment of the present application, it needs to be obtained through the authorization and consent of the object, the authorization and consent of the relevant departments, and in accordance with the relevant laws, regulations and standards of the country and region. If personal information is involved in the embodiment, the acquisition of all personal information needs to obtain the consent of the individual. If sensitive information is involved, the separate consent of the information subject needs to be obtained, and the embodiment also needs to be implemented with the authorization and consent of the object.

[0022] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0023] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0024] The embodiment of the present application provides a forest pest monitoring method, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application. Figure 1 As shown, the method includes step S101 and step S102, wherein: Step S101: obtaining a pest image to be identified and a multi-scale target detection model, using the multi-scale target detection model to perform pest identification on the pest image to be identified, and determining pest identification information, wherein the pest identification information includes: pest category and pest location.

[0025] In the embodiments of the present application, the images of the pests to be identified are usually obtained by using advanced shooting devices to take high-definition photos of the pests successfully lured and gathered by the insect detection lamp. The insect detection lamp is a device specially designed for agricultural pest monitoring. It simulates the phototaxis or other biological characteristics of the pests to effectively attract and gather the pests in the surrounding farmland, causing them to collide with the collision screen and fall into the collection tube, and then kill them using a heating device.

[0026] In order to solve the problems of low resolution of small targets, insufficient features, and coexistence of multi-scale targets when relevant models detect forest pests, a multi-scale target detection model is proposed in the embodiment of the present application, wherein the multi-scale target detection model is based on the YOLOv7 model and is a model for identifying forest pests. The extraction of small target features is refined by introducing a scale sequence feature fusion module, and the proposed triple feature encoding efficient aggregation module solves the impact of multi-scale targets and densely occluded targets on the model recognition performance. Furthermore, the multi-scale target detection model is used to identify pests on the image of the pest to be identified, and the pest identification information is determined, that is, the image of the pest to be identified is input into the multi-scale target detection model, the multi-scale target detection model is controlled to perform multi-scale detection and non-maximum suppression operations, and the result output by the multi-scale target detection model is recorded as pest identification information, wherein the pest identification information includes: pest category and pest location. In order to improve the detection efficiency and accuracy of the multi-scale target detection model, the input pest image to be identified can be preprocessed before performing the pest identification operation. The preprocessing includes but is not limited to: image scaling, normalization, color space conversion, etc.

[0027] Step S102: Perform hazard assessment based on pest identification information corresponding to each target area in the forest ecology, determine pest hazard assessment information, and select control methods based on the pest hazard assessment information to determine pest control information, wherein the pest control information is used to reduce the harm of pests to forest resources.

[0028] For the embodiment of the present application, the pest identification information of each target area in the forest ecology is integrated to facilitate the clarification of the degree of harm of pests to forest resources, which helps to carry out precise prevention and control of pests and improve the efficiency of pest control. Therefore, based on the pest identification information corresponding to each target area in the forest ecology, a hazard assessment is performed to determine the pest hazard assessment information, and a prevention and control method is selected based on the pest hazard assessment information to determine the pest control information.

[0029] There are many ways to assess the hazard, which are not limited in the embodiments of the present application. In one feasible method, pest information is summarized based on the pest identification information corresponding to each target area in the forest ecology to determine the pest summary information; an ecological simulation model is obtained, and a hazard degree assessment is performed based on the ecological simulation model and the pest summary information to determine the degree of pest hazard; historical pest monitoring data is obtained, and the development trend of historical pest monitoring data and pest summary information is predicted using time series analysis technology to determine pest spread prediction information; risk assessment is performed based on pest spread prediction information to determine pest risk information, and pest hazard degree and pest risk information are combined to obtain pest hazard assessment information. The intuitive display of pest hazard status provides a scientific basis for subsequent pest control, helps optimize the allocation of control resources, and ensures the smooth progress and efficient completion of control work. Furthermore, there are many ways to select a control method, which will not be discussed or limited in the embodiments of the present application. In one achievable method, a control demand analysis is performed based on the degree of pest damage and pest risk information to determine the control target, and based on the control target, a target control method is selected from multiple control methods, where the control methods include but are not limited to: physical control, chemical control, and biological control; a control plan is formulated based on the target control method and pest summary information to determine pest control information. Pest control information is a targeted measure taken for specific pest problems, which improves the control effect, and accurate selection of control methods can avoid waste of resources and increased costs caused by blind control.

[0030] The training methods of multi-scale object detection models include: For the embodiments of this application, Figure 2 As shown, Figure 2 A schematic diagram of a training method for a multi-scale target detection model provided in an embodiment of the present application includes steps SA and SB, wherein: Step SA: obtain the pest image sent by the insect monitoring light, perform preprocessing and data labeling based on the pest image, obtain the pest sample image, and divide the pest sample image into a training set, a verification set and a test set in proportion.

[0031] For the embodiment of the present application, the insect activity is high from May to October, which is the optimal time period for collecting insect sample images in order to capture a variety of insect samples. The insect monitoring lamp is a device specially designed for agricultural pest monitoring. It simulates the phototaxis or other biological characteristics of pests to effectively attract and gather pests in the surrounding farmland, causing them to collide with the collision screen and fall into the collection tube, and then be killed by a heating device. An image shooting device is also provided in the insect monitoring lamp to shoot the pests that fall into the insect monitoring lamp to obtain pest images. Then, preprocessing is performed based on the pest image to obtain a pest sample image. The preprocessing operation is used to improve the quality of the pest image, reduce noise and redundant information, and enhance the useful features in the pest image to facilitate subsequent model training operations, wherein the preprocessing includes but is not limited to: image denoising, image enhancement, image cropping, color space conversion, etc. Then, the preprocessed pest images are labeled. The purpose of data labeling is to provide a basis for supervised learning for the model. Data labeling usually includes the annotation of pest categories (e.g., Scarabaeidae, Cerambycidae), the drawing of bounding boxes (used to locate the position of pests in pest images), and possible attribute labels (e.g., maturity and color of pests). That is, data labeling clarifies the pest category, location, and possible quantity information in each pest image. Finally, the images with data labeling are used as pest sample images. Then, the pest sample images are divided into training set, validation set, and test set in proportion. Preferably, the division ratio is 7:2:1, wherein the training set is used for the training process of the model; the validation set is used to evaluate the model performance and adjust the hyperparameters during the training process; and the test set is used to evaluate the final performance of the model after the training is completed.

[0032] Step SB: obtain an improved YOLOv7 model, and use the training set to train the improved YOLOv7 model to obtain a multi-scale target detection model, wherein the improved YOLOv7 model introduces a scale sequence feature fusion module and a triple feature encoding efficient aggregation module into the YOLOv7 model.

[0033] For the embodiment of the present application, in order to improve the detection accuracy of the detection model for the images of pests to be identified with small targets, densely occluded targets, and multi-scale targets, the YOLOv7 model is used as a benchmark to introduce a scale sequence feature fusion module and a triple feature encoding efficient aggregation module, wherein the scale sequence feature fusion module is used to improve the detection accuracy of small targets; the triple feature encoding efficient aggregation module is used to improve the robustness and accuracy of target detection when dense targets and multi-scale targets coexist. Furthermore, the training set is used to train the improved YOLOv7 model, that is, the loss is obtained based on the similarity of multiple pest sample images in the training set, and the loss is back-propagated to train the improved YOLOv7 model; the weighted sum of the losses of the trained improved YOLOv7 model is performed to obtain the total loss, and when the total loss meets the set loss threshold range, the trained improved YOLOv7 model is determined to be a multi-scale target detection model.

[0034] It can be seen that in the embodiment of the present application, when identifying pests in a forest ecosystem, a multi-scale target detection model is used to identify pests in the image to be identified, and the pest identification information is determined, wherein the multi-scale target detection model is obtained by training the improved YOLOv7 model using a training set, and the multi-scale target detection model improves the detection accuracy of the detection model for the image of the pest to be identified with small targets, densely occluded targets, and multi-scale targets, thereby improving the accuracy of forest pest detection as a whole. Then, based on the pest identification information corresponding to each target area in the forest ecology, a hazard assessment is performed to determine the pest hazard assessment information, and a prevention and control method is selected based on the pest hazard assessment information, the pest control information is determined, and precise prevention and control is performed, thereby improving the efficiency of pest control.

[0035] Furthermore, in order to improve the detection accuracy of the detection model for the images of pests to be identified with small targets, densely occluded targets, and multi-scale targets, in the embodiment of the present application, the construction method of the YOLOv7 model is improved, including: Acquire multiple feature maps of different scales extracted by the backbone layer, select a first feature map based on the multiple feature maps of different scales, and determine the first feature map; A scale sequence feature fusion module is added to the head layer, where the scale sequence feature fusion module fuses features of different scales based on the first feature map to improve the detection accuracy of the improved YOLOv7 model for small targets. Based on multiple feature maps of different scales, a triple feature encoding efficient aggregation module is added to the head layer. The triple feature encoding efficient aggregation module improves the robustness and accuracy of target detection when dense targets and multi-scale targets coexist.

[0036] For the embodiments of the present application, in order to improve the detection accuracy of the detection model for the images of pests to be identified with small targets, densely occluded targets, and multi-scale targets, the YOLOv7 model is used as a benchmark, and a scale sequence feature fusion module and a triple feature coding efficient aggregation module are introduced. The scale sequence feature fusion module is used to improve the detection accuracy of small targets; the triple feature coding efficient aggregation module is used to improve the robustness and accuracy of target detection when dense targets and multi-scale targets coexist.

[0037] Specifically, multiple feature maps of different scales extracted by the backbone layer are obtained, and the multiple feature maps of different scales are respectively recorded as P2-level feature maps, P3-level feature maps, P4-level feature maps, and P5-level feature maps. The first feature map is selected based on the multiple feature maps of different scales to determine the first feature map, wherein the first feature map includes: three layers of feature maps of P3, P4, and P5. Then, a scale sequence feature fusion module is added to the head layer. The execution operation of the scale sequence feature fusion module is as follows: the P3-level feature map is selected as the benchmark, and the number of channels and resolution of the P4 and P5-level feature maps are adjusted to be consistent with P3 using 1*1 convolution and nearest neighbor interpolation methods, respectively, to ensure that the feature maps have the same spatial information. Subsequently, each feature layer is increased in dimension through the Unsqueeze operation, converted into a 4D tensor "depth, height, width, channel", and then spliced ​​along the depth dimension to form a 3D feature map for 3D convolution input, which retains the feature information of different scales and lays the foundation for subsequent scale sequence feature extraction. Finally, 3D convolution, 3D batch normalization, and Leaky ReLU activation functions are used to extract scale sequence features to cope with images with spatial correlation in the dataset. Through the 3D convolution operation, the improved YOLOv7 model can capture the feature information of small target pests at different scales; 3D batch normalization helps to accelerate the training process and improve the generalization ability of the model, while the introduction of the nonlinear Leaky ReLU activation function enhances the network's expressive power. These operations work together on the 3D feature map, enabling the model to effectively learn and extract scale sequence features of small targets, thereby better coping with the challenges of low resolution and insufficient feature information of small targets in the dataset.

[0038] At the same time, two triple feature encoding efficient aggregation modules are added to the extraction path of targets of different scales to improve the model's detection performance for multi-scale targets. That is, the triple feature encoding efficient aggregation module fuses the input feature map according to large, medium and small sizes to extract local fine details of the target; the processed feature map is then input into the efficient layer aggregation network (ELAN module), and through its multi-branch, multi-level feature extraction and fusion structure, the expressive power of the feature map is further enhanced, so that the gradient can be propagated more efficiently. The triple feature encoding efficient aggregation module fully utilizes the spatial feature information in the P2 feature map that is conducive to target positioning by multiple fusions of the P2, P3, P4 and P5 level feature maps, thereby improving the detection capability of the model. For the structural diagram of the improved YOLOv7 model, as shown in the figure, Figure 3 shown.

[0039] It can be seen that in the embodiment of the present application, the first feature map is selected based on multiple feature maps of different scales, the first feature map is determined, and a scale sequence feature fusion module is added in the head layer, wherein the scale sequence feature fusion module performs different scale feature fusion based on the first feature map to improve the detection accuracy of the improved YOLOv7 model for small targets. Furthermore, a triple feature coding efficient aggregation module is added in the head layer based on multiple feature maps of different scales, and the triple feature coding efficient aggregation module improves the robustness and accuracy of target detection when dense targets and multi-scale targets coexist.

[0040] Furthermore, in order to improve the generalization ability of the model and the accuracy of pest recognition, in the embodiment of the present application, after the improved YOLOv7 model is trained using the training set to obtain the multi-scale target detection model, the following steps are also included: Use the validation set and test set to test the multi-scale object detection model and determine the model performance; Obtaining performance standard information, and performing model performance evaluation based on the model performance and the performance standard information, and determining the model performance evaluation result; When the model performance evaluation result meets the requirements, the multi-scale target detection model is used for pest identification; when the model performance evaluation result does not meet the requirements, pest sample images in the training set are added, and the multi-scale target detection model is iteratively trained using the added training set until the multi-scale target detection model meets the requirements.

[0041] For the embodiment of the present application, the training set is used for the training process of the model; the validation set is used to evaluate the model performance and adjust the hyperparameters during the training process; and the test set is used to evaluate the final performance of the model after the training is completed. Therefore, during the training process, the multi-scale target detection model is evaluated using the validation set, wherein the evaluation indicators include but are not limited to: accuracy, recall rate, F1 score, average precision mean, etc. By observing the changes in performance indicators on the validation set, it can be determined whether the model is overfitting or underfitting, and the training strategy is adjusted accordingly. Furthermore, after the model training is completed, the multi-scale target detection model is finally evaluated using the test set to determine the model performance, wherein the model performance includes but is not limited to: accuracy, recall rate, F1 score, average precision mean, precision, intersection and detection speed, etc., and the model performance is crucial for subsequent performance standard comparisons. Using the validation set and the test set to perform performance testing on the multi-scale target detection model can ensure that the multi-scale target detection model performs consistently on different data sets, thereby improving the generalization ability of the model and helping to reduce the overfitting or underfitting of the model on unknown data.

[0042] Then, the performance standard information is obtained, and the model performance is evaluated based on the model performance and the performance standard information to determine the model performance evaluation result. The performance standard information is pre-stored in the electronic device, and is the performance standard information related to pest identification found by accessing authoritative institutions, professional websites or academic databases. For the model performance evaluation, the actual model performance is compared with the pre-set performance standard information to analyze whether the multi-scale target detection model meets or exceeds the performance standard requirements in various indicators; at the same time, it is also necessary to consider the detection ability of the multi-scale target detection model on targets of different scales, the detection stability of targets of different categories, and the detection speed. Finally, based on single comparison and comprehensive evaluation, the model performance evaluation result is determined, that is, if the model meets or exceeds the performance standard requirements in various indicators, it is determined that the model performance meets the requirements; otherwise, it is determined that the model performance does not meet the requirements.

[0043] When the model performance evaluation result meets the requirements, the multi-scale target detection model is used for pest identification; when the model performance evaluation result does not meet the requirements, the pest sample images in the training set are added, and the multi-scale target detection model is iteratively trained using the added training set until the multi-scale target detection model meets the requirements. Iterative training helps the model better learn the physical signs of pests and improve the accuracy of pest identification.

[0044] It can be seen that in the embodiment of the present application, the validation set and the test set are used to perform model performance testing on the multi-scale target detection model to determine the model performance. Then, the model performance is evaluated based on the model performance and performance standard information to determine the model performance evaluation result. Using the validation set and the test set to perform performance testing on the multi-scale target detection model can ensure that the multi-scale target detection model performs consistently on different data sets, thereby improving the generalization ability of the model and helping to reduce the overfitting or underfitting phenomenon of the model on unknown data. Furthermore, when the model performance evaluation result meets the requirements, the multi-scale target detection model is used for pest identification; when the model performance evaluation result does not meet the requirements, the pest sample images in the training set are added, and the multi-scale target detection model is iteratively trained using the added training set until the multi-scale target detection model meets the requirements. Iterative training helps the model to better learn the signs of pests and improve the accuracy of pest identification.

[0045] Furthermore, in order to make the result of the hazard level assessment more objective and accurate, in the embodiment of the present application, the hazard assessment is performed based on the pest identification information corresponding to each target area in the forest ecology, and the pest hazard assessment information is determined, including: Summarize pest information based on pest identification information corresponding to each target area in forest ecology to determine pest summary information; Obtain an ecological simulation model, conduct a damage assessment based on the ecological simulation model and pest summary information, and determine the degree of pest damage; Obtain historical pest monitoring data, use time series analysis technology to predict the development trend of historical pest monitoring data and pest summary information, and determine pest spread prediction information; Risk assessment is performed based on pest spread prediction information to determine pest risk information, and pest damage degree and pest risk information are integrated to obtain pest damage assessment information.

[0046] For the embodiments of the present application, in order to quickly grasp the distribution and types of pests in the forest ecosystem, pest information is summarized based on the pest identification information corresponding to each target area in the forest ecology, and pest summary information is determined. The pest summary information is used to comprehensively reflect the types, quantities, distribution, etc. of pests in the forest ecosystem, providing key data support for the assessment of the degree of pest damage and the formulation of effective prevention and control strategies. Then, an ecological simulation model is obtained, which can be used to simulate the interaction between pests and forest vegetation. The construction method of the ecological simulation model is as follows: collect the biological characteristics, ecological behavior, historical occurrence data of pests, and obtain the types, distribution, and growth conditions of forest vegetation in forest ecology, as well as collect environmental factor data, such as climate, soil conditions, topography, etc.; according to the principles of ecological systems theory, divide the ecosystem into different components (such as producers, consumers, decomposers, environmental factors, etc.), and determine the interaction between them; define key parameters in the model, such as the reproduction rate, mortality rate, and migration rate of pests, the growth rate and mortality rate of vegetation, and the range of environmental factors; then, select a suitable modeling method, and build a model based on the defined model structure and parameters using professional modeling software or programming language to obtain an ecological simulation model. Then, the pest summary information is input into the ecological simulation model, and the ecological simulation model performs simulation operations based on the input data and the defined algorithm to simulate the growth, reproduction, migration, and other processes of pests in the forest ecosystem, as well as the interaction between them and vegetation. Furthermore, the ecological simulation model outputs the degree of harm caused by pests to the forest ecosystem, which is recorded as the degree of harm caused by pests, including but not limited to: the spread range of pests, changes in number, degree of damage to vegetation, etc. Using the ecological simulation model to evaluate the degree of harm of pest summary information can scientifically quantify the impact of pests on the forest ecosystem, which not only takes into account the number and type of pests, but also combines the complexity and dynamics of the ecosystem, making the results of the degree of harm assessment more objective and accurate.

[0047] At the same time, historical pest monitoring data is obtained, which is the summary information of pests corresponding to the preset time before the current moment in forest ecology. Then, the historical pest monitoring data is processed and analyzed by using time series analysis technology to identify the periodic laws of pest occurrence and spread; combined with the current pest monitoring data, the possible development trend of pests in the future is predicted, and the pest spread prediction information is determined, wherein the pest expansion prediction information includes but is not limited to: spread speed, spread area, quantity change, etc. Then, risk assessment is performed based on the pest spread prediction information to determine the pest risk information, that is, when performing risk assessment, the pest spread prediction information is quantitatively evaluated by using risk matrix, risk assessment and other methods, for example, calculating the economic losses and ecological impacts of pest damage, and according to the risk quantification results, the pest risks are divided into different levels and recorded as pest risk information. Risk assessment based on pest spread prediction information can comprehensively consider various risks that may be brought about by pest spread, including ecological risks, economic risks and social risks, etc. Comprehensive risk assessment helps decision makers weigh the pros and cons from multiple perspectives and make more reasonable decisions. Finally, the pest damage degree and pest risk information are integrated to obtain the pest damage assessment information.

[0048] It can be seen that in the embodiment of the present application, pest information is summarized based on the pest identification information corresponding to each target area in the forest ecology, and the pest summary information is determined. Then, the degree of harm is assessed based on the ecological simulation model and the pest summary information to determine the degree of harm of the pests. Using the ecological simulation model to assess the degree of harm of the pest summary information can scientifically quantify the impact of pests on the forest ecosystem, not only considering the number and type of pests, but also combining the complexity and dynamics of the ecosystem, so that the result of the degree of harm assessment is more objective and accurate. At the same time, historical pest monitoring data is obtained, and the development trend of historical pest monitoring data and pest summary information is predicted using time series analysis technology to determine the pest spread prediction information. Then, risk assessment is performed based on the pest spread prediction information to determine the pest risk information. Risk assessment based on pest spread prediction information can comprehensively consider the various risks that may be brought about by pest spread, including ecological risks, economic risks and social risks, etc. Comprehensive risk assessment helps decision makers weigh the pros and cons from multiple angles and make more reasonable decisions. Pest harm assessment information is obtained by combining the pest harm degree and pest risk information.

[0049] Furthermore, in order to ensure the effectiveness and efficiency of the control measures, in the embodiment of the present application, the control method is selected based on the pest damage assessment information, and the pest control information is determined, including: Analyze pest control needs based on pest damage level and pest risk information, and select target control measures from multiple control measures; Formulate prevention and control plans based on target prevention and control measures and pest summary information to determine pest control information.

[0050] For the embodiment of the present application, the degree of pest damage is used to characterize the degree of damage of pests to forest ecology, and the pest risk information is used to characterize the possible future development trend of pests and the potential damage to forest resources. Furthermore, the applicability of each control method is evaluated, and a target control method that best matches the current pest situation of forest ecology is selected from multiple control methods. Through reasonable control methods, the number of pests can be effectively controlled to ensure the effectiveness and efficiency of the control methods. Common control methods include: physical control (for example, pruning damaged branches, cleaning pest habitats), chemical control (using appropriate pesticides for spraying) and biological control (introducing natural enemies, using pathogenic microorganisms, etc.), among which physical control is suitable for local and clear pest distribution, early control and emergency treatment, and the need to protect the ecological environment; chemical control is suitable for the situation where the number of pests is large, the harm is serious, and there is an urgent and large-scale control demand; biological control is suitable for the situation where there is a long-term sustainable control demand for pests, environmental protection needs and abundant natural resources. Therefore, based on the degree of pest damage and pest risk information, it is matched with the appropriate use of each control method to determine the target control method. Furthermore, a prevention and control plan is formulated based on the target prevention and control means and pest summary information to determine the pest control information, which includes but is not limited to: prevention and control time, prevention and control area, prevention and control means and prevention and control measures, etc. Detailed prevention and control measures data items are given in the prevention and control measures. For example, for physical prevention, it includes light trapping, yellow board trapping, manual capture, etc.; for chemical prevention, it is necessary to clarify the type of pesticide, spraying concentration, spraying time and frequency, etc.; for biological control, it is necessary to determine the type of natural enemy, the number of natural enemies, the time and method of release, etc.

[0051] It can be seen that in the embodiment of the present application, the control demand analysis is performed based on the pest damage degree and pest risk information, and the target control means are selected from multiple control means. Through reasonable control means, the number of pests can be effectively controlled to ensure the effectiveness and efficiency of the control means. Then, a control plan is formulated based on the target control means and the pest summary information to determine the pest control information.

[0052] Furthermore, in order to reduce the direct harm of pests to forest resources and protect the stability and diversity of forest ecosystems, in the embodiment of the present application, a control plan is formulated based on target control measures and pest summary information. After determining the pest control information, it also includes: Send pest control information to the pest control terminal so that pest control personnel can carry out control according to the pest control information and reduce the harm of pests to forest resources; After the pest control information is sent to the pest control terminal for the control verification time, current pest summary information is obtained, control effect analysis is performed based on the current pest summary information, and the control effect analysis result is determined; When the results of the pest control effect analysis are not up to standard, pest control optimization is performed based on the current pest summary information and pest control information to determine the optimized pest control information.

[0053] For the embodiments of the present application, pest control information is sent to the pest control terminal so that pest control personnel can carry out control according to the pest control information and reduce the harm of pests to forest resources. After the pest control information is sent to the pest control terminal for the control verification time, the current pest summary information is obtained, and the control effect analysis is performed based on the current pest summary information to determine the control effect analysis result. The control verification time is a time period to ensure that the control measures are fully implemented and play a role. During this time, the pest control personnel have enough time to carry out control according to the pest control information. Therefore, after the control verification time, the current pest summary information is compared and analyzed with the pest summary information before the control. By comparing the changes in indicators such as the number of pests and distribution range at different time points, the effectiveness and applicability of the control measures are scientifically evaluated. That is, the electronic device pre-stores the pest control information expected to be achieved after the prevention and control verification time of the prevention and control measures, and compares the expected pest control information with the current pest summary information. When the standard of the expected pest control information is reached, the prevention and control effect analysis result is determined to be up to standard; otherwise, the prevention and control effect analysis result is determined to be not up to standard. Furthermore, when the prevention and control effect analysis result is not up to standard, prevention and control optimization is performed based on the current pest summary information and the pest control information, and optimization information is determined, wherein the optimization information includes but is not limited to: adjusting prevention and control measures, strengthening prevention and control efforts, introducing new technologies, etc. With regard to prevention and control optimization, it can be set according to the specific situation of pest control in forest ecology, as long as the pest control effect can be improved, and the specific implementation process is no longer limited in the embodiments of the present application. Through timely prevention and control and scientific evaluation and optimization, the growth of the number of pests can be effectively controlled, the direct harm of pests to forest resources can be reduced, and the stability and diversity of forest ecosystems can be protected.

[0054] It can be seen that in the embodiment of the present application, the pest control information is sent to the pest control terminal so that the pest control personnel can carry out management according to the pest control information and reduce the harm of pests to forest resources. Then, after the pest control information is sent to the pest control terminal for the control verification time, the control effect analysis is performed based on the current pest summary information to determine the control effect analysis result. When the control effect analysis result is not up to standard, the control optimization is performed based on the current pest summary information and the pest control information to determine the optimized pest control information. Through timely prevention and scientific evaluation and optimization, the growth of the number of pests can be effectively controlled, the direct harm of pests to forest resources can be reduced, and the stability and diversity of forest ecosystems can be protected.

[0055] The above embodiment introduces a forest pest monitoring method from the perspective of method flow, and the following embodiment introduces a forest pest monitoring device from the perspective of a virtual module or a virtual unit. For details, please refer to the following embodiment.

[0056] The present application embodiment provides a forest pest monitoring device, such as Figure 4 As shown, the forest pest monitoring device may specifically include: The pest identification module 210 is used to obtain a pest image to be identified and a multi-scale target detection model, and use the multi-scale target detection model to perform pest identification on the pest image to be identified, and determine pest identification information, wherein the pest identification information includes: pest category and pest location; The pest control module 220 is used to perform a hazard assessment based on the pest identification information corresponding to each target area in the forest ecology, determine the pest hazard assessment information, and select a control method based on the pest hazard assessment information to determine the pest control information, wherein the pest control information is used to reduce the harm of pests to forest resources; The forest pest monitoring device also includes: The model training module 230 is used to obtain the pest images sent by the pest monitoring lamp, perform preprocessing and data labeling based on the pest images, obtain pest sample images, and divide the pest sample images into a training set, a verification set, and a test set according to proportion; An improved YOLOv7 model is obtained, and the improved YOLOv7 model is trained using the training set to obtain a multi-scale target detection model, wherein the improved YOLOv7 model introduces a scale sequence feature fusion module and a triple feature encoding efficient aggregation module into the YOLOv7 model.

[0057] In a possible implementation of the embodiment of the present application, the forest pest monitoring device further includes: Improve the YOLOv7 model construction module to obtain multiple feature maps of different scales extracted by the backbone layer, select the first feature map based on the multiple feature maps of different scales, and determine the first feature map; A scale sequence feature fusion module is added to the head layer, where the scale sequence feature fusion module fuses features of different scales based on the first feature map to improve the detection accuracy of the improved YOLOv7 model for small targets. Based on multiple feature maps of different scales, a triple feature encoding efficient aggregation module is added to the head layer. The triple feature encoding efficient aggregation module improves the robustness and accuracy of target detection when dense targets and multi-scale targets coexist.

[0058] In a possible implementation of the embodiment of the present application, the forest pest monitoring device further includes: The performance evaluation module is used to perform model performance testing on the multi-scale object detection model using the validation set and the test set to determine the model performance; Obtaining performance standard information, and performing model performance evaluation based on the model performance and the performance standard information, and determining the model performance evaluation result; When the model performance evaluation result meets the requirements, the multi-scale target detection model is used for pest identification; when the model performance evaluation result does not meet the requirements, pest sample images in the training set are added, and the multi-scale target detection model is iteratively trained using the added training set until the multi-scale target detection model meets the requirements.

[0059] In a possible implementation of the embodiment of the present application, the pest control module 220, when performing a hazard assessment based on the pest identification information corresponding to each target area in the forest ecology and determining the pest hazard assessment information, is used to: Summarize pest information based on pest identification information corresponding to each target area in forest ecology to determine pest summary information; Obtain an ecological simulation model, conduct a damage assessment based on the ecological simulation model and pest summary information, and determine the degree of pest damage; Obtain historical pest monitoring data, use time series analysis technology to predict the development trend of historical pest monitoring data and pest summary information, and determine pest spread prediction information; Risk assessment is performed based on pest spread prediction information to determine pest risk information, and pest damage degree and pest risk information are integrated to obtain pest damage assessment information.

[0060] In a possible implementation of the embodiment of the present application, the pest control module 220, when performing the selection of a control method based on the pest hazard assessment information and determining the pest control information, is used to: Analyze pest control needs based on pest damage level and pest risk information, and select target control measures from multiple control measures; Formulate prevention and control plans based on target prevention and control measures and pest summary information to determine pest control information.

[0061] In a possible implementation of the embodiment of the present application, the forest pest monitoring device further includes: The pest control effect analysis module is used to send pest control information to the pest control terminal so that pest control personnel can carry out control according to the pest control information and reduce the harm of pests to forest resources; After the pest control information is sent to the pest control terminal for the control verification time, current pest summary information is obtained, control effect analysis is performed based on the current pest summary information, and the control effect analysis result is determined; When the results of the pest control effect analysis are not up to standard, pest control optimization is performed based on the current pest summary information and pest control information to determine the optimized pest control information.

[0062] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the forest pest monitoring device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0063] An electronic device is provided in an embodiment of the present application, such as Figure 5 As shown, Figure 5 The electronic device 300 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of the present application.

[0064] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0065] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but it does not mean that there is only one bus or only one type of bus.

[0066] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0067] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0068] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 5 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0069] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0070] The embodiment of the present application provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the method in any of the above embodiments is implemented. Compared with the related art, in the embodiment of the present application, when identifying pests in a forest ecosystem, a multi-scale target detection model is used to identify pests in the image to be identified, and the pest identification information is determined, wherein the multi-scale target detection model is obtained by training the improved YOLOv7 model using a training set, and the multi-scale target detection model improves the detection accuracy of the detection model for the image to be identified of pests with small targets, densely occluded targets, and multi-scale targets, thereby improving the accuracy of forest pest detection as a whole. Then, based on the pest identification information corresponding to each target area in the forest ecology, a hazard assessment is performed to determine the pest hazard assessment information, and a prevention and control method is selected based on the pest hazard assessment information, the pest control information is determined, and precise prevention and control is performed, thereby improving the efficiency of pest control.

[0071] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0072] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A forest pest monitoring method, characterized in that: include: Acquire a pest image to be identified and a multi-scale target detection model, use the multi-scale target detection model to perform pest identification on the pest image to be identified, and determine pest identification information, wherein the pest identification information includes: pest category and pest location; Performing a hazard assessment based on the pest identification information corresponding to each target area in the forest ecology to determine pest hazard assessment information, and selecting a control method based on the pest hazard assessment information to determine pest control information, wherein the pest control information is used to reduce the harm of pests to forest resources; The training method of the multi-scale object detection model includes: Acquire pest images sent by the pest monitoring lamp, perform preprocessing and data labeling based on the pest images to obtain pest sample images, and divide the pest sample images into a training set, a verification set, and a test set in proportion; An improved YOLOv7 model is obtained, and the improved YOLOv7 model is trained using the training set to obtain a multi-scale target detection model, wherein the improved YOLOv7 model is a YOLOv7 model that introduces a scale sequence feature fusion module and a triple feature encoding efficient aggregation module.

2. The forest pest monitoring method according to claim 1, characterized in that: The improved YOLOv7 model construction method includes: Acquire multiple feature maps of different scales extracted by the backbone layer, select a first feature map based on the multiple feature maps of different scales, and determine the first feature map; Adding the scale sequence feature fusion module in the head layer, wherein the scale sequence feature fusion module performs different scale feature fusion based on the first feature map to improve the detection accuracy of the improved YOLOv7 model for small targets; The triple feature coding efficient aggregation module is added in the head layer based on the multiple feature maps of different scales, wherein the triple feature coding efficient aggregation module improves the robustness and accuracy of target detection when dense targets and multi-scale targets coexist.

3. The forest pest monitoring method according to claim 1, characterized in that: After the improved YOLOv7 model is trained using the training set to obtain a multi-scale target detection model, the method further includes: Using the validation set and the test set, a model performance test is performed on the multi-scale object detection model to determine the model performance; Acquiring performance standard information, and performing model performance evaluation based on the model performance and the performance standard information, and determining a model performance evaluation result; When the model performance evaluation result meets the requirements, the multi-scale target detection model is used to identify pests; when the model performance evaluation result does not meet the requirements, the pest sample images in the training set are added, and the multi-scale target detection model is iteratively trained using the added training set until the multi-scale target detection model meets the requirements.

4. The forest pest monitoring method according to claim 1, characterized in that: The pest identification information corresponding to each target area in the forest ecology is used to perform a hazard assessment to determine the pest hazard assessment information, including: Summarize pest information based on the pest identification information corresponding to each target area in the forest ecology to determine pest summary information; Acquire an ecological simulation model, and perform a hazard assessment based on the ecological simulation model and the pest summary information to determine the degree of pest hazard; Obtain historical pest monitoring data, use time series analysis technology to predict the development trend of the historical pest monitoring data and the pest summary information, and determine pest spread prediction information; A risk assessment is performed based on the pest spread prediction information to determine pest risk information, and the pest damage degree and the pest risk information are integrated to obtain the pest damage assessment information.

5. The forest pest monitoring method according to claim 4, characterized in that: The step of selecting a control method based on the pest hazard assessment information and determining pest control information includes: Performing a control demand analysis based on the pest damage degree and the pest risk information, and selecting a target control method from multiple control methods; A pest control plan is formulated based on the target control measures and the pest summary information to determine the pest control information.

6. The forest pest monitoring method according to claim 5, characterized in that: The formulation of a control plan based on the target control means and the pest summary information, after determining the pest control information, also includes: The pest control information is sent to the pest control terminal so that pest control personnel can carry out control according to the pest control information to reduce the harm of pests to forest resources; After the pest control information is sent to the pest control terminal for a control verification period, current pest summary information is obtained, and control effect analysis is performed based on the current pest summary information to determine a control effect analysis result; When the pest control effect analysis result is not up to standard, pest control optimization is performed based on the current pest summary information and the pest control information to determine optimized pest control information.

7. An electronic device, characterized in that: include: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the forest pest monitoring method according to any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed in a computer, the computer is caused to execute the forest pest monitoring method according to any one of claims 1 to 6.

9. A computer program product, characterized in that The method comprises a computer program, wherein the computer program is executed by a processor to implement the forest pest monitoring method according to any one of claims 1 to 6.

Images