Ancient tree area monitoring method

Abstract

The invention provides an ancient tree area monitoring method and system and electronic equipment. The method comprises the steps that a first image and a second image of a monitoring area are acquired through acquisition equipment, the first image is acquired by the acquisition equipment at a first focal length, the second image is acquired by the acquisition equipment at a second focal length, and the first focal length is larger than the second focal length; processing the first image by using an ancient tree disease disaster model, and if a disease disaster exists in the first image, acquiring a central pixel of a disease disaster area in the first image; and according to the central pixel of the disease disaster area in the first image, acquiring the central pixel of the disease disaster area in the second image. The method can help the user to quickly locate the disease disaster part of the ancient tree, thereby helping to confirm the overall trend of the disease disaster of the ancient tree.

Description

Technical Field

[0001] This application relates to the field of plant monitoring technology, specifically to a monitoring method, system, and electronic equipment for ancient tree areas. Background Technology

[0002] Ancient trees, generally referring to trees that are 100 years old or older, are considered "living fossils" in nature. While ancient trees possess remarkable resilience, they become exceptionally vulnerable due to the combined effects of aging, increasingly severe human interference (environmental damage, pollution, and mismanagement), and extreme weather events caused by climate change. Pests and diseases, in particular, have become major threats to their health and even cause their death. Because ancient trees possess extremely important ecological, cultural, historical, and scientific value, the management of their diseases and pests is crucial for biodiversity conservation, ecological balance, and the protection of cultural heritage.

[0003] Monitoring diseases and disasters of ancient trees is a crucial part of disaster management for ancient trees. Traditional monitoring of diseases and disasters of ancient trees relies on a large ground-based monitoring and forecasting system and a large amount of field investigation work, which has the disadvantages of large manpower requirements, high cost and poor timeliness, resulting in the accelerated spread of diseases and disasters of ancient trees.

[0004] To address the aforementioned issues, a convenient and effective strategy for monitoring diseases and disasters affecting ancient trees is urgently needed. Summary of the Invention

[0005] In view of this, this application provides a monitoring method, system, and electronic device for ancient tree areas to solve the above problems.

[0006] According to a first aspect of this application, a monitoring method for an ancient tree area is provided, comprising: acquiring a first image and a second image of the monitoring area using a data acquisition device, wherein the first image is acquired by the data acquisition device at a first focal length, and the second image is acquired by the data acquisition device at a second focal length, wherein the first focal length is greater than the second focal length; processing the first image using an ancient tree disease disaster model; if disease disaster exists in the first image, acquiring the center pixel of the disease disaster area in the first image; and acquiring the center pixel of the disease disaster area in the second image based on the center pixel of the disease disaster area in the first image, so as to assist the user in quickly locating the disease disaster area.

[0007] According to a first aspect of this application, the method further includes: stitching together multiple second images to generate a panoramic display image; obtaining the mapping relationship between the multiple second images and the panoramic display image; and mapping the center pixels of the diseased areas in the multiple second images to the panoramic display image to obtain the center pixels of the diseased areas in the panoramic display image.

[0008] According to a second aspect of this application, a monitoring system for an ancient tree area is provided, comprising: an acquisition module, configured to acquire a first image and a second image of the monitoring area using an acquisition device, wherein the first image is acquired by the acquisition device at a first focal length, and the second image is acquired by the acquisition device at a second focal length, wherein the first focal length is greater than the second focal length; a processing module, configured to process the first image using an ancient tree disease disaster model, wherein if disease disaster exists in the first image, the center pixel of the disease disaster area in the first image is acquired; and a mapping module, configured to acquire the center pixel of the disease disaster area in the second image based on the center pixel of the disease disaster area in the first image, so as to assist the user in quickly locating the disease disaster area.

[0009] According to a second aspect of this application, the monitoring system further includes: a panoramic display module, used to stitch the second image together to generate a panoramic display image; obtain the mapping relationship between the second image and the panoramic display image; and map the center pixel of the diseased area in the second image to the panoramic display image to obtain the center pixel of the diseased area in the panoramic display image.

[0010] According to a third aspect of this application, an electronic device is provided, comprising a memory and a processor, the memory for storing a computer program, the processor for running the computer program to cause the electronic device to perform the ancient tree area monitoring method as provided in the first aspect of this application.

[0011] According to a fourth aspect of this application, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method for monitoring ancient tree areas as provided in the first aspect of this application.

[0012] According to a fifth aspect of this application, a computer program product is provided, the computer program product including instructions that, when executed by a processor of an electronic device provided in a third aspect of this application, enable the electronic device to implement the ancient tree area monitoring method provided in the first aspect of this application.

[0013] This application provides a method, system, and electronic device for monitoring ancient tree areas. The method processes a first image of an ancient tree using an ancient tree disease disaster model to obtain the center pixels of the disease disaster areas in the first image. These center pixels are then mapped to the center pixels of the disease disaster areas in a second image of the ancient tree. This allows for the determination of disease disaster characteristics in the ancient tree based on details in the first image, and helps users quickly locate the diseased areas in the second image. Furthermore, the method can stitch multiple second images together to create a panoramic image, displaying the diseased areas of the ancient tree and assisting in confirming the overall trend of disease disasters. Moreover, this method does not require personnel to visit the diseased areas of the ancient tree, thus reducing safety hazards and saving labor costs. It also reduces misjudgments caused by human error or lack of experience, thereby improving monitoring accuracy. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 The diagram shown is a flowchart illustrating an exemplary embodiment of this application for a method of detecting ancient tree areas.

[0016] Figure 2 The diagram shown is a flowchart illustrating a method for obtaining a disease disaster model of an ancient tree species, provided in an exemplary embodiment of this application.

[0017] Figure 3 The diagram shown is a flowchart illustrating a method for determining the species classification of ancient trees in a monitoring area, provided by an exemplary embodiment of this application.

[0018] Figure 4 The diagram shown is a flowchart illustrating a method for obtaining a single-species ancient tree disease disaster model based on the ancient tree species category, provided by an exemplary embodiment of this application.

[0019] Figure 5 The diagram shown is a flowchart illustrating another method for obtaining a single-species ancient tree disease disaster model based on the ancient tree species category, provided by an exemplary example of this application.

[0020] Figure 6 The diagram shown is a flowchart illustrating a method for obtaining a multi-species ancient tree disease disaster model provided in an exemplary embodiment of this application.

[0021] Figure 7The diagram shown is a flowchart illustrating another method for obtaining a multi-species ancient tree disease disaster model provided by an exemplary embodiment of this application.

[0022] Figure 8 The diagram shown is a flowchart illustrating another method for obtaining a multi-species ancient tree disease disaster model provided in an exemplary embodiment of this application.

[0023] Figure 9 The diagram shown is a schematic diagram of a monitoring system for an ancient tree area provided in an exemplary embodiment of this application.

[0024] Figure 10 The diagram shown is a block diagram of an exemplary electronic device provided in an exemplary embodiment of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0026] Traditional monitoring of diseases and disasters in ancient trees relies primarily on manual inspections and basic electronic equipment. This approach has several drawbacks: it is labor-intensive and inefficient; when identifying diseased areas, especially for tall trees, personnel or ladders are needed to reach the estimated disease location, posing safety risks; early signs of disease are less visible on the tree as a whole, making misjudgments easy; it cannot quickly pinpoint diseased areas, relying only on information such as the time of photographing to determine the approximate location; and it lacks the ability to analyze real-time data, making it impossible to grasp the overall development of disease and disaster status in ancient trees.

[0027] With the widespread application of image recognition technology in agriculture, collecting and recording detailed images of ancient trees has become an important means of protecting them. These detailed images allow for magnified observation of tree details, enabling the identification of potential diseases and damage, and thus facilitating disease monitoring. However, this method still has drawbacks: because detailed images can only pinpoint the current state of the trees, they cannot map diseases and damage to images encompassing the entire tree area. This prevents users from quickly locating affected sections, leading to the spread of diseases and damage to the ancient trees.

[0028] To address the aforementioned problems, this application proposes a monitoring method for ancient tree areas. This method can confirm the disease situation of ancient trees through detailed images and quickly and accurately locate the disease in an image of the complete ancient tree area. Therefore, it helps users locate the diseased parts of ancient trees, thereby effectively controlling the spread of disease in ancient trees. Various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0029] Figure 1 The diagram shown is a flowchart illustrating a method for detecting ancient tree areas according to an exemplary embodiment of this application. This method is executed by a monitoring system for ancient tree areas. Figure 1 As shown, the method includes the following steps S110~S130: Step S110: Acquire a first image and a second image of the monitoring area using the acquisition device.

[0030] The monitoring area is the area where the ancient tree being monitored is located. There may be one or more ancient trees in this area. The ancient trees may be of one or more species. Specifically, the ancient trees may be cypress, pine, yew, etc.

[0031] The first image is acquired by the acquisition device at a first focal length, and the second image is acquired by the acquisition device at a second focal length, wherein the first focal length is greater than the second focal length.

[0032] In one embodiment, the acquisition device may include a zoom camera, wherein a first image is acquired by the zoom camera at a first focal length, and a second image is acquired by the zoom camera at a second focal length. The acquisition device may also include a pan-tilt head, such that the zoom camera is mounted on a pan-tilt head capable of vertical and horizontal rotation. Preferably, the first focal length may be the maximum focal length of the zoom camera, and the second focal length may be the minimum focal length of the zoom camera. Specifically, the first and second images can be acquired through the following steps: Step 1: Determine the ancient tree area to be monitored. Based on the grating information of the device's pan-tilt unit, determine the start and end horizontal angles (start_x, end_x) and the start and end vertical angles (start_y, end_y) of the pre-monitoring area. Step 2: Determine the first focal length for capturing the first image of the ancient tree, that is, the first focal length for capturing the detailed image of the ancient tree; optionally, the first focal length can be the maximum focal length of the zoom camera. Step 3: Determine the horizontal angle hfov1, vertical angle vfov1, and horizontal overlap rate rh1 and vertical overlap rate vh1 of two adjacent ancient tree detail images at the first focal length; Step 4: Determine the rotation angle spacing (hfov1-rh1, vfov1-vh1) of the first focal length based on the horizontal angle hfov1, vertical angle vfov1, horizontal overlap rate rh1, and vertical overlap rate vh1. Step 5: Based on the rotation angle interval of the first focal length, capture the first image within the range of the start and end horizontal angles and the start and end vertical angles, and record the horizontal and vertical angles of the device's gimbal when capturing the first image. Step 6: Determine the second focal length for acquiring the second image of the ancient tree; optionally, the second focal length can be the minimum focal length of the zoom camera; Step 7: Determine the horizontal angle hfov2, the vertical angle vfov2, and the horizontal overlap rate rh2 and the vertical overlap rate vh2 of two adjacent second images at the second focal length; Step 8: Based on the horizontal angle hfov2, vertical angle vfov2, horizontal overlap rate rh2, and vertical overlap rate vh2 of the second focal length, determine the rotation angle spacing (hfov2-rh2, vfov2-vh2) of the second focal length. Step 9: Capture the second image based on the second angular spacing of the second focal length, and record the horizontal and vertical angles of the device's gimbal when capturing the second image; Step 10: Store the first image, the gimbal angle of the device that captured the first image, the second image, and the gimbal angle of the device that captured the second image.

[0033] In another embodiment, the acquisition device may not include a zoom camera. For example, the first image is obtained by the first camera in the acquisition device at a first focal length, and the second image is obtained by the second camera in the acquisition device at a second focal length, as long as the first focal length is greater than the second focal length.

[0034] It should be understood that the above two schemes are only examples of how to acquire the first image and the second image. The acquisition methods of the first image and the second image can also be determined based on the specific acquisition environment and equipment. This application embodiment does not specifically limit the acquisition methods of the first image and the second image.

[0035] It should be noted that, since the first focal length used when acquiring the first image is greater than the second focal length used when acquiring the second image, the first image has a narrower field of view than the second image, but the details of the ancient trees are clearer, so that users (or users using appropriate tools) can identify whether there are diseases or disasters in the ancient trees through the details of the ancient trees; the second image has a larger field of view, which can show a more complete area of ​​ancient trees so that users can observe the whole picture of the monitoring area through the second image.

[0036] It should be understood that the number of first and second images can be one or more; when the area of ​​ancient trees to be monitored is too large, the second image can also be a stitched image of multiple second images.

[0037] Furthermore, when there are higher requirements for the accuracy of the acquired images of the monitoring area, the acquisition equipment may also include a laser rangefinder, which can be mounted on the pan-tilt unit of the acquisition equipment, wherein the laser emission direction of the laser rangefinder is parallel to the optical axis of the zoom camera.

[0038] Step S120: Process the first image using the ancient tree disease disaster model. If there is disease in the first image, obtain the center pixel of the disease disaster area in the first image.

[0039] Ancient tree disease disaster model refers to a model that can identify disease disaster areas of ancient trees. This model is a neural network model or mathematical model based on deep learning. Since there are many types of ancient trees, the ancient tree disease disaster model can be a single-species ancient tree disease disaster model or a multi-species ancient tree disease disaster model.

[0040] A single-species ancient tree disease disaster model refers to an ancient tree disease disaster model that is trained only on ancient trees of a single species. For example, an ancient tree disease disaster model trained on the pine tree species can identify the disease disaster features of all pine trees in the first image and output one or more disease disaster area center pixels. Similarly, an ancient tree disease disaster model trained on the dawn redwood tree species can identify the disease disaster features of all dawn redwood trees in the first image and output one or more disease disaster area center pixels.

[0041] Since the single-species ancient tree disease disaster model is trained only on a single tree species, its advantages are simple training logic and high accuracy.

[0042] A multi-species ancient tree disease disaster model refers to combining ancient trees of multiple tree species (e.g., pine, camphor, and banyan) into a single mathematical model. For example, a multi-species ancient tree disease disaster model trained on pine, camphor, and banyan trees can simultaneously identify the disease disaster features of all three ancient trees in the first image and output one or more center pixels of disease disaster areas.

[0043] Since the multi-species ancient tree disease disaster model combines multiple types of ancient trees into a single mathematical model, its advantage is that the training cost is relatively low.

[0044] The training process for the single-species ancient tree disease disaster model and the multi-species ancient tree disease disaster model can be found in the description of the relevant embodiments in the following examples. To avoid repetition, it will not be repeated here.

[0045] Step S130: Based on the center pixel of the diseased area in the first image, obtain the center pixel of the diseased area in the second image to assist the user in quickly locating the diseased area.

[0046] Specifically, step S130 can be implemented in the following way: Obtain the first angle of the device's gimbal when capturing the first image; obtain the second angle of the device's gimbal when capturing the second image; obtain the mapping relationship between the first image and the second image based on the first angle and the second angle of the device's gimbal; based on the mapping relationship between the first image and the second image, map the center pixel of the diseased area in the first image to the second image to obtain the center pixel of the diseased area in the second image.

[0047] The first angle is the angle of the gimbal used to capture the first image, and the second angle is the angle of the gimbal used to capture the second image. The specific methods for obtaining the first and second angles can be found in the description of step S110. To avoid repetition, they will not be repeated here.

[0048] Specifically, based on the first angle and the second angle of the device's pan-tilt unit, the mapping relationship between the first image and the second image is obtained. The angle difference between the second angle of each second image and the first angle of the first image can be calculated, and the mapping relationship between the second image corresponding to the second angle with the smallest angle difference and the first image is determined.

[0049] The mapping relationship between the first image and the second image can be obtained using the OpenCV-SIFT corner detection algorithm. After determining the mapping relationship between the first image and the second image, the center pixel of the diseased area in the first image can be mapped to the second image based on the mapping relationship to obtain the center pixel of the diseased area in the second image.

[0050] Because the second image has a wider field of view, it can show a more complete area of ​​ancient trees. Therefore, after obtaining the center pixel of the diseased area in the second image, a warning mark can be placed on the second image to help users quickly locate the approximate area of ​​the disease, which will then facilitate the arrangement of personnel for further confirmation and investigation of the disease.

[0051] Therefore, this application proposes a monitoring method for ancient tree areas. This method processes a first image of an ancient tree using an ancient tree disease disaster model to obtain the center pixels of the diseased areas in the first image. Then, it maps the center pixels of the diseased areas in the first image to the center pixels of the diseased areas in a second image of the ancient tree. This allows for the determination of whether an ancient tree exhibits disease characteristics based on details in the first image, and the use of the second image to help users quickly locate the diseased parts of the ancient tree. Processing the first image using the ancient tree disease disaster model improves the efficiency of disease judgment, while the more complete second image helps users quickly locate the diseased areas, thus effectively controlling the spread of disease in ancient trees. Furthermore, this detection method does not require personnel to reach the diseased areas of the ancient tree. Therefore, it not only reduces safety hazards and saves labor costs but also reduces misjudgments caused by human error or lack of experience, thereby improving the accuracy of monitoring.

[0052] In practical applications, there may be situations where the second image cannot fully display the appearance of the ancient tree. In such cases, the center pixel of the diseased area in the panoramic image can be obtained in the following way. This can help to quickly locate the approximate area of ​​the diseased area, and at the same time, help users to confirm the proportion of the diseased area on the ancient tree as a whole, thereby inferring the overall development trend of the diseased area of ​​the ancient tree.

[0053] Specifically, the steps to obtain the center pixel of the diseased area in the panoramic display image are as follows: stitching multiple second images together to generate a panoramic display image; obtaining the mapping relationship between the multiple second images and the panoramic display image; mapping the center pixel of the diseased area in the multiple second images to the panoramic display image to obtain the center pixel of the diseased area in the panoramic display image.

[0054] The OpenCV-Stitcher algorithm can be used to stitch together multiple second images to generate a panoramic display image; the OpenCV-Sift corner detection algorithm can be used to obtain the mapping relationship between each of the multiple second images and the panoramic display image, thereby mapping the center pixel of the diseased area in each of the multiple second images to the panoramic display image.

[0055] In practical applications, users can also use the saved historical images of the first and second images to trace back the historical development of the disease and disaster areas of the ancient tree, thereby helping users to judge the cycle of disease and disaster of the ancient tree and thus realize the pre-management of disease and disaster of the ancient tree. Users can also use the first and second images to help them observe the development status of the ancient tree after the elimination of disease and disaster, thereby enabling better management of disease and disaster of the ancient tree.

[0056] The following sections will introduce the methods for obtaining single-species ancient tree disease disaster models and multi-species ancient tree disease disaster models.

[0057] Figure 2 The diagram shown is a flowchart illustrating a method for obtaining a disease disaster model of an ancient tree of a single tree species, provided in an exemplary embodiment of this application. Figure 2 As shown, the method for obtaining a single-species ancient tree disease disaster model includes the following steps S210~S220: Step S210: Determine the species of ancient trees in the monitoring area.

[0058] As mentioned earlier, the tree species categories for ancient trees can include cypress, pine, and yew, among others. The following will use pine as an example to describe the process of obtaining a single-species ancient tree disease disaster model.

[0059] In one embodiment, the species of ancient trees in the monitoring area can be manually specified; in another embodiment, the species of ancient trees in the monitoring area can also be determined by a mathematical model.

[0060] Figure 3 The diagram shown is a flowchart illustrating a method for determining the species classification of ancient trees in a monitoring area, provided in an exemplary embodiment of this application. Figure 3 As shown, the method for automatically obtaining the species classification of ancient trees in the monitoring area includes the following steps S310~S330: Step S310: Label the ancient tree species category of the training material using the labeling tool to obtain the first labeled training material.

[0061] The annotation tool is used to add structured labels to target objects in the image to generate labeled data required for training the AI ​​model. Optionally, the annotation tool can be CVAT, Labelimg, etc.; preferably, this application uses Labelimg as the annotation tool.

[0062] The training materials can be historical first images, historical second images, stitched images of historical first images, or stitched images of historical second images, etc. Users can choose according to their actual needs, and this application embodiment does not limit this.

[0063] Step S320: Train the first mathematical model based on the labeled first training material to obtain a single-species ancient tree classification model.

[0064] The first mathematical model can be a classification model, such as the ResNet model or VGG.

[0065] Step S330: Based on the first image, use the single-species ancient tree classification model to determine the species category of ancient trees in the monitoring area.

[0066] The first image can also be replaced by a second image or a spliced ​​image of the first image. Users can choose according to their actual needs, and this application embodiment does not limit this.

[0067] Specifically, taking the ancient tree species category as pine, the training material as historical first images, the annotation tool as Labelimg, and the first mathematical model as ResNet as an example, the process of determining the ancient tree species category in the monitoring area is as follows: using the Labelimg tool, historical first images containing pine trees and historical first images not containing pine trees are annotated in multiple historical first images to obtain multiple annotated historical first images. Then, a classification model is trained based on multiple annotated historical first images to obtain a pine tree species classification model. Finally, the first image is input into the pine tree species classification model, which can identify whether the ancient tree species in the first image includes pine trees.

[0068] Step S220: Obtain a single-species ancient tree disease disaster model based on the ancient tree species category in the monitoring area.

[0069] In one embodiment, the method for obtaining a single-species ancient tree disease disaster model can be found in [reference needed]. Figure 4 ,in, Figure 4 The diagram shown is a flowchart illustrating a method for obtaining a single-species ancient tree disease disaster model based on the ancient tree species category, provided by an exemplary embodiment of this application.

[0070] like Figure 4 As shown, step S220 includes the following steps S410~S420: Step S410: Based on the ancient tree species category in the monitoring area, label whether the training material has diseases or disasters to obtain the labeled second training material.

[0071] The training materials can be historical first images, historical second images, stitched images of historical first images, stitched images of historical second images, etc. Users can choose according to their actual needs, and this application embodiment does not limit this.

[0072] Since ancient tree species categories correspond to ancient tree disease and hazard categories, when labeling training materials, the ancient tree disease and hazard categories to be labeled can be determined based on the ancient tree species category, and then the training materials can be labeled based on the ancient tree disease and hazard categories. One ancient tree may correspond to one disease and hazard category, or it may correspond to multiple disease and hazard categories, and different disease and hazard categories have different image features.

[0073] For example, when the ancient tree species is pine, the corresponding disease categories for pine trees are set as: pine wilt disease, pine needle blight, pine blister rust, and pine root rot. The image characteristics of pine wilt disease are that the needles are reddish-brown and withered. The image characteristics of pine needle blight are that the needles have reddish-brown spots and fall off prematurely. The image characteristics of pine blister rust are that the bark of the branches and trunk bulges and cracks, exuding orange-yellow spores. The image characteristics of pine root rot are that the roots rot, the tree becomes weak and collapses. Therefore, when the ancient tree species is pine, if the single-species ancient tree disease disaster model defaults to recognizing only the most common pine nematode disease, then the appearance characteristics of pine nematode disease can be used to identify and label whether the pine trees in the training material have the disease (including only pine wilt disease), and this is recorded as the labeled second training material; if the single-species ancient tree disease disaster model needs to identify pine wilt disease, pine needle blight, pine blister rust, and pine root rot, then the appearance characteristics of these diseases can be used to identify and label whether the pine trees in the training material have the disease (including: pine wilt disease, pine needle blight, pine blister rust, and pine root rot), and this is recorded as the labeled second training material.

[0074] For example, when the ancient tree species is cypress, the corresponding disease categories for cypress are: cypress rot, cypress dieback, and gummosis. The visual characteristics of cypress rot are large cavities or cracks in the trunk, with the internal xylem decaying in a spongy, blocky, or flaky manner (white rot / brown rot), localized necrosis and peeling of the bark, and the possible growth of horseshoe-shaped or layered fungal fruiting bodies (such as wood-rotting fungi) on the surface. The visual characteristics of cypress dieback are that at the top of the crown, the young shoots die from top to bottom, turning reddish-brown, with a sunken boundary between diseased and healthy tissue, and dense black dots (conidiophores) on the dead branches. The visual characteristics of gummosis are that on the trunk / branches, amber or brown gum-like substances ooze from the bark cracks, which harden after drying, and the bark below the gum mass is often wet and rotten. Therefore, when the ancient tree species is cypress, if the single-species ancient tree disease disaster model defaults to recognizing only the most common cypress canker, then the appearance characteristics of cypress can be used to identify and label whether the pine trees in the training material have disease disasters (including only cypress canker), and this is recorded as the labeled second training material; if the single-species ancient tree disease disaster model needs to identify cypress canker, cypress dieback and gummosis, then the appearance characteristics of cypress can be used to identify and label whether the cypress trees in the training material have disease disasters (including: cypress canker, cypress dieback and gummosis), and this is recorded as the labeled second training material.

[0075] Step S420: Train the second mathematical model based on the labeled second training material to obtain the first single-species ancient tree disease disaster model.

[0076] The second mathematical model can be a classification model, such as the ResNet model or VGG.

[0077] Specifically, based on the labeled second training material obtained in step 410, the classification model is trained to obtain the first single-species ancient tree disease disaster model. The first single-species ancient tree disease disaster model can be used to identify whether the first image contains a certain type or several types of disease disasters.

[0078] For example, when the default tree species is pine and the default disease category is pine nematode disease, the second mathematical model is trained based on the labeled second training material obtained in step S410 to obtain a single-species ancient tree disease model for pine. At this time, the single-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of pine nematode disease. If the training material is labeled and the second mathematical model is trained for pine wilt disease, pine needle blight, pine blister rust, and pine root rot in step S410, then the single-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of pine wilt disease, pine needle blight, pine blister rust, and pine root rot.

[0079] For example, when the ancient tree species is cypress and the corresponding ancient tree disease category is cypress rot, the second mathematical model can be trained based on the labeled second training material obtained in step S410 to obtain a single-species ancient tree disease model for cypress. In this case, the single-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of cypress rot. If the training material is labeled and the second mathematical model is trained for cypress rot, cypress dieback, and gummosis in step S410, then the single-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of cypress rot, cypress dieback, and gummosis.

[0080] After obtaining the first single-species ancient tree disease disaster model in step S420, the first image is processed using the ancient tree disease disaster model in step S120. If there is disease disaster in the first image, the center pixel of the disease disaster area in the first image is obtained, including: processing the first image using the first single-species ancient tree disease disaster model to determine whether the first image has disease disaster; if the first image has disease disaster, the center pixel of the first image is taken as the center pixel of the disease disaster area in the first image.

[0081] It should be noted that since the first single-species ancient tree disease disaster model obtained in step 420 is a classification model, this model can only be used to identify whether there is disease in the first image, and cannot output the area where the disease is located. Therefore, the center pixel of the first image can be used as the center pixel of the disease area in the first image. The advantage of the first single-species ancient tree disease disaster model is that it has high training efficiency and simple logic, thus improving training efficiency and saving training time and computation costs.

[0082] In another embodiment, the method for obtaining a single-species ancient tree disease disaster model can be found in [reference needed]. Figure 5 ,in, Figure 5 The diagram shown is a flowchart illustrating another method for obtaining a single-species ancient tree disease disaster model based on the ancient tree species category, provided by an exemplary example of this application.

[0083] like Figure 5 As shown, step S220 includes the following steps S510~S520: Step S510: Using the ancient tree species categories and annotation tools in the monitoring area, the diseased and damaged areas in the training material are annotated to obtain the third training material after annotation.

[0084] The annotation tool can be CVAT, Labelimg, etc.; preferably, this application uses Labelimg as the annotation tool. The training material can be a historical first image, a historical second image, a stitched image of the historical first image, a stitched image of the historical second image, etc., and users can choose according to their actual needs. This application embodiment does not limit this.

[0085] Specifically, in step S510, the training material can be labeled using labeling tools such as Labelimg, that is, the diseased and damaged areas of the ancient trees in the training material can be labeled to obtain the labeled third training material.

[0086] For example, if the default tree species is pine and the default disease category is pine nematode disease, then if the training material contains image features of pine nematode disease, the third training material obtained after annotation in step S510 will be marked with the area where pine nematode disease is located. If the corresponding ancient tree disease category is pine wilt disease, pine needle blight, pine blister rust, and pine root rot, then if the training material contains image features of these diseases, the third training material obtained after annotation in step S510 will be marked with the areas where cypress rot, cypress dieback, and gummosis are located.

[0087] For example, when the ancient tree species is cypress and the corresponding ancient tree disease category is cypress rot, if the training material contains image features of cypress rot, then the third training material obtained after annotation in step S510 will be marked with the area where cypress rot is located; if the corresponding ancient tree disease category is cypress rot, cypress dieback, and gummosis, and if the training material contains image features of these diseases, then the third training material obtained after annotation in step S510 will be marked with the areas where cypress rot, cypress dieback, and gummosis are located.

[0088] Step S520: Train the third mathematical model based on the labeled third training material to obtain the second single-species ancient tree disease disaster model.

[0089] The third mathematical model can be an object detection model, such as the YOLO series models.

[0090] Specifically, based on the labeled third training material obtained in step S510, the target detection model is trained to obtain the second single-species ancient tree disease disaster model. The second single-species ancient tree disease disaster model can be used to identify whether the first image contains a certain disease disaster and output the area where the disease disaster is located.

[0091] For example, when the default tree species is pine and the default disease category is pine nematode disease, the third mathematical model is trained based on the labeled third training material obtained in step S510 to obtain a single-species ancient tree disease model for pine. At this time, the single-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of pine nematode disease. If so, the area where the pine nematode disease is located is output. If the corresponding ancient tree disease category is pine wilt disease, pine needle blight, pine blister rust, and pine root rot, the trained single-species ancient tree disease model for pine can be used to identify whether the first image has the disease characteristics of pine wilt disease, pine needle blight, pine blister rust, and pine root rot. If so, the area where these diseases are located is output.

[0092] For example, when the ancient tree species is cypress and the corresponding ancient tree disease category is cypress rot, the third mathematical model can be trained based on the labeled third training material obtained in step S510 to obtain a single-species ancient tree disease model for cypress. At this time, the single-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of cypress rot. If so, the area where the cypress rot is located is output. If the corresponding ancient tree disease category is cypress rot, cypress dieback, and gummosis, the trained single-species ancient tree disease model for cypress can be used to identify whether the first image has the disease characteristics of cypress rot, cypress dieback, and gummosis. If so, the area where these diseases are located is output.

[0093] After obtaining the second single-species ancient tree disease disaster model in step S520, the first image is processed using the ancient tree disease disaster model in step S120. If there is disease disaster in the first image, the center pixel of the disease disaster area in the first image is obtained, including: processing the first image using the second single-species ancient tree disease disaster model to obtain the disease disaster area in the first image; and obtaining the center pixel of the disease disaster area in the first image based on the disease disaster area in the first image.

[0094] Since the second single-species ancient tree disease disaster model obtained in step S520 is a target detection model, this model can not only identify whether there is a disease disaster area in the first image, but also output the area where the disease disaster is located. Therefore, compared with Figure 4 The method shown, Figure 5 The method shown can more accurately calculate the center pixel of the diseased area of ​​the ancient tree in the first image.

[0095] The following section will detail how to obtain disease disaster models for multiple tree species of ancient trees.

[0096] Figure 6 The diagram shown is a flowchart illustrating a method for obtaining a multi-species ancient tree disease disaster model according to an exemplary embodiment of this application. Figure 6 As shown, in one embodiment, the method for obtaining a multi-species ancient tree disease disaster model may include the following steps S610~S620: Step S610: Label whether the training material has diseases or disasters, and obtain the labeled fourth training material.

[0097] The training materials can be historical first images, historical second images, stitched images of historical first images, stitched images of historical second images, etc. Users can choose according to their actual needs, and this application embodiment does not limit this.

[0098] Specifically, the training materials can be manually labeled to determine whether they have diseased or damaged areas (diseases corresponding to one or more tree species) to obtain the labeled fourth training materials.

[0099] For example, assuming the ancient trees in the images shown in the training material are pine, camphor, and banyan trees, and the corresponding diseases and ailments of the ancient trees are pine nematode disease, twig blight, and powdery mildew, we can identify whether the ancient trees in the training material have areas with pine nematode disease, twig blight, and / or powdery mildew based on the appearance characteristics of pine nematode disease, twig blight, and / or powdery mildew, and mark them. That is, all disease and ailment areas (including pine nematode disease, twig blight, and powdery mildew) in the training material will be marked and recorded as the fourth training material after annotation.

[0100] Step S620: Train the fourth mathematical model based on the labeled fourth training material to obtain the first multi-species ancient tree disease disaster model.

[0101] The fourth mathematical model can be a classification model, such as the ResNet model or VGG.

[0102] Specifically, based on the labeled fourth training material obtained in step 610, the classification model is trained to obtain the first multi-species ancient tree disease disaster model. This first multi-species ancient tree disease disaster model can be used to identify whether the first image contains one or more specific disease disaster features.

[0103] For example, assuming the ancient trees in the images shown in the training materials are pine, camphor, and banyan trees, and the corresponding ancient tree diseases are pine nematode disease, twig blight, and powdery mildew, the fourth mathematical model is trained based on the labeled fourth training materials obtained in step S610 to obtain a multi-species ancient tree disease model for pine nematode disease, twig blight, and powdery mildew. At this time, the multi-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of pine nematode disease, twig blight, and / or powdery mildew.

[0104] After obtaining the first multi-species ancient tree disease disaster model through step S620, the first image is processed using the ancient tree disease disaster model in step S120. If there is disease disaster in the first image, the center pixel of the disease disaster area in the first image is obtained, including: processing the first image using the first multi-species ancient tree disease disaster model to determine whether the first image has a disease disaster area; if the first image has a disease disaster area, the center pixel of the area in the first image is taken as the center pixel of the disease disaster area in the first image.

[0105] Since the first multi-species ancient tree disease disaster model obtained in step 620 is a classification model, this model can only be used to identify whether the first image contains specific disease disaster features, but cannot obtain the area where the disease disaster is located. Therefore, the center pixel of the area in the first image can be used as the center pixel of the disease disaster area in the first image.

[0106] according to Figure 6 As can be seen from the process of obtaining the first multi-species ancient tree disease disaster model, the training of the first multi-species ancient tree disease disaster model does not perform ancient tree species identification, but only labels the disease disaster samples. Due to the simple training logic, the training efficiency is improved accordingly, and the training time and computation costs are saved.

[0107] Figure 7 The diagram shown is a flowchart illustrating another method for obtaining a multi-species ancient tree disease disaster model according to an exemplary embodiment of this application. Figure 7 As shown, in another embodiment, the method for obtaining a multi-species ancient tree disease disaster model may further include the following steps S710~S720: Step S710: Use the annotation tool to annotate the diseased areas in the training material to obtain the fifth annotated training material.

[0108] The annotation tool can be CVAT, Labelimg, etc.; preferably, this application uses Labelimg as the annotation tool. The training material can be a historical first image, a historical second image, a stitched image of the historical first image, a stitched image of the historical second image, etc., and users can choose according to their actual needs. This application embodiment does not limit this.

[0109] Specifically, in step S710, the training material can be labeled using labeling tools such as Labelimg, that is, the disease and disaster areas of ancient trees in the training material (the areas where the diseases and disasters of one or more tree species are located) are labeled, and the fifth training material after labeling is obtained.

[0110] For example, if the ancient trees in the images shown in the training material are pine, camphor, and banyan trees, and the corresponding diseases and ailments of the ancient trees are pine nematode disease, twig blight, and powdery mildew, then the areas where all diseases and ailments (including pine nematode disease, twig blight, and powdery mildew) are located in the training material can be marked using annotation tools such as Labelimg, and this area is recorded as the fifth training material after annotation.

[0111] Step S720: Train the fifth mathematical model based on the labeled fifth training material to obtain the second multi-species ancient tree disease disaster model.

[0112] The fifth mathematical model can be an object detection model, such as the YOLO series models.

[0113] Specifically, based on the fifth training material with annotations obtained in step S710, the target detection model is trained to obtain the second multi-species ancient tree disease disaster model. The second multi-species ancient tree disease disaster model can be used to identify whether the first image contains one or more diseases and the area where the disease is located.

[0114] For example, assuming the ancient trees in the images shown in the training materials are pine, camphor, and banyan trees, and the corresponding ancient tree diseases are pine nematode disease, twig blight, and powdery mildew, the fifth mathematical model is trained based on the fifth labeled training materials obtained in step S710 to obtain a multi-species ancient tree disease model for pine nematode disease, twig blight, and powdery mildew. At this time, the multi-species ancient tree disease model can be used to identify whether the first image has the disease characteristics of nematode disease, twig blight, and powdery mildew. If so, the regions where these diseases are located are output respectively.

[0115] After obtaining the second multi-species ancient tree disease disaster model through step S720, the first image is processed using the ancient tree disease disaster model in step S120. If there is disease disaster in the first image, the center pixel of the disease disaster area in the first image is obtained, including: processing the first image using the second multi-species ancient tree disease disaster model to obtain the disease disaster area in the first image; and obtaining the center pixel of the disease disaster area in the first image based on the disease disaster area in the first image.

[0116] Since the second multi-species ancient tree disease disaster model obtained in step S720 is a target detection model, this model can not only identify whether there is a disease disaster area in the first image, but also output the area where the disease disaster is located. Therefore, the center pixel of the disease disaster area output by the second multi-species ancient tree disease disaster model obtained in step S720 can be directly calculated based on the area where the disease disaster is located, and it can be used as the center pixel of the disease disaster area in the first image.

[0117] according to Figure 7 The process of obtaining the second multi-species ancient tree disease disaster model, as shown, demonstrates that the training of this model does not involve ancient tree species identification; only disease disaster samples are labeled. Due to the simplicity of the training logic, this improves training efficiency and saves training time and computational costs. Figure 6 Compared to the method shown, Figure 7 The method shown uses annotation tools to label the training materials, thus saving labor costs, improving the efficiency of training material annotation, and consequently improving the overall training efficiency of the model; moreover, Figure 7 The second multi-species ancient tree disease disaster model can directly output the area where the disease is located. Therefore, the center pixel of the disease area in the first image is more accurate, which is more conducive to ancient tree maintenance personnel to more accurately locate the area where the disease is located.

[0118] Figure 8 The diagram shown is a flowchart illustrating another method for obtaining a multi-species ancient tree disease disaster model according to an exemplary embodiment of this application. Figure 8 As shown, the method for obtaining a multi-species ancient tree disease disaster model may also include the following steps S810~S850: Step S810: Using the annotation tool, annotate the tree species areas and disease and disaster areas in the training materials, and obtain the annotated sixth training material and the annotated seventh training material respectively.

[0119] The annotation tool can be CVAT, Labelimg, etc.; preferably, this application uses Labelimg as the annotation tool. Training materials can be historical first images, historical second images, stitched images of historical first images, stitched images of historical second images, etc., and users can choose according to their actual needs; this application embodiment does not impose any restrictions on this. The tree species region refers to the area where a certain ancient tree species is located.

[0120] The process for annotating the sixth training material is as follows: If the training material contains any target ancient tree species, then the regions where all target ancient tree species are located are marked. If the training material does not contain any target ancient tree species, then no marking is performed. The target ancient tree species are those corresponding to the diseases and hazard identified by the multi-species ancient tree disease and hazard model. Specifically, if the diseases and hazard identified by the multi-species ancient tree disease and hazard model are pine nematode disease, twig blight, and powdery mildew, then the target ancient tree species are pine, camphor, and banyan. Therefore, the pine, camphor, and banyan trees in the images shown in the training material must be marked. That is: if the training material contains pine trees, then the region where the pine tree is located is marked; if the training material contains pine, camphor, and banyan trees simultaneously, then the regions where the pine, camphor, and banyan trees are located are marked respectively; if the training material does not contain any target ancient tree species (i.e., any one of pine, camphor, and banyan), then no marking is performed.

[0121] The process for annotating the seventh training material is as follows: If the training material contains the target ancient tree disease, then all areas where the target ancient tree disease is located are marked; if the training material does not contain the target ancient tree disease, then no marking is performed. The target ancient tree disease refers to all categories of diseases that the multi-species ancient tree disease model aims to identify. Specifically, if the diseases that the multi-species ancient tree disease model aims to identify are pine nematode disease, twig blight, and powdery mildew, then the areas of pine nematode disease, twig blight, and powdery mildew in the images shown in the training material must be marked. That is: if the training material only contains pine nematode disease, then all areas where pine nematode disease is located are marked; if the training material contains pine nematode disease, twig blight, and powdery mildew, then all areas where pine nematode disease, twig blight, and powdery mildew are located are marked; if the training material does not contain areas of the target ancient tree disease, then no marking is performed.

[0122] Step S820: Train the sixth mathematical model based on the labeled sixth training material to obtain a multi-species ancient tree species identification model.

[0123] The sixth mathematical model can be an object detection model, such as the YOLO series models.

[0124] Specifically, based on the sixth training material with annotations obtained in step S810, the target detection model is trained to obtain a multi-species ancient tree identification model. This multi-species ancient tree identification model can be used to identify whether the first image contains one or more ancient trees and output the area where the ancient tree is located.

[0125] Step S830: Based on the first image, use the multi-tree-species ancient tree species identification model to obtain tree species region images.

[0126] Specifically, the first image is input into the multi-species ancient tree identification model obtained in step S820. If the first image contains the target ancient tree species, the multi-species ancient tree identification model can output the area where the target ancient tree species is located, that is, the tree species area image.

[0127] For example, if the disease or hazard to be identified by the multi-species ancient tree disease and hazard model is pine nematode disease, twig blight and powdery mildew, then the corresponding target ancient tree species are pine, camphor and banyan. Therefore, the multi-species ancient tree species identification model obtained in step S830 can identify whether the first image contains pine, camphor and / or banyan and output the corresponding ancient tree species area (i.e., tree species area image). That is, if the first image only contains pine, then the area where all pine trees are located is output; if the first image contains pine, camphor and banyan trees at the same time, then the area where all pine, camphor and banyan trees are located is output.

[0128] Step S840: Update the tree species region image to the first image.

[0129] For example, if the first image contains only pine trees, then step S930 outputs all pine tree regions (i.e., pine tree region images), and then updates all pine tree regions to the first image; if the first image contains pine trees, camphor trees, and banyan trees, then step S930 outputs all pine tree, camphor tree, and banyan tree regions respectively, and then updates all pine tree, camphor tree, and banyan tree regions (i.e., pine tree, camphor tree, and banyan tree region images) to the first image respectively.

[0130] Step S850: Train the seventh mathematical model based on the labeled seventh training material to obtain the third multi-species ancient tree disease disaster model.

[0131] The seventh mathematical model can be a target detection model, such as the YOLO series models.

[0132] Specifically, based on the seventh training material with annotations obtained in step S810, the target detection model is trained to obtain the third multi-species ancient tree disease disaster model. The third multi-species ancient tree disease disaster model can be used to identify whether the first image contains one or more ancient tree diseases and output the area where the ancient tree disease disaster is located.

[0133] For example, if the multi-species ancient tree disease disaster model is to identify pine nematode disease, twig blight, and powdery mildew, the multi-species ancient tree disease disaster model obtained in step S850 can be used to identify whether the first image contains pine nematode disease, twig blight, and / or powdery mildew and output the areas where pine nematode disease, twig blight, and / or powdery mildew are located. Since the first image is the image updated in step S840, each first image after the update in step S840 contains only one tree species, that is: the first image after the update in step S840 may be a pine tree area image, a camphor tree area image, or a banyan tree area. The image, specifically: if the first image updated after step S840 contains only pine trees and the pine tree area contains pine nematode disease, then the multi-species ancient tree disease disaster model can identify and output all areas where pine nematode disease is located; if the first image updated after step S840 is a camphor tree and the camphor tree area contains twig dieback disease, then the multi-species ancient tree disease disaster model can identify and output all areas where twig dieback disease is located; if the first image updated after step S840 is a banyan tree and the banyan tree area contains powdery mildew, then the multi-species ancient tree disease disaster model can identify and output all areas where powdery mildew is located.

[0134] After obtaining the third multi-species ancient tree disease disaster model through step S850, the first image is processed using the ancient tree disease disaster model in step S120. If there is disease disaster in the first image, the center pixel of the disease disaster area in the first image is obtained, including: processing the first image using the third multi-species ancient tree disease disaster model to obtain the disease disaster area in the first image; and obtaining the center pixel of the disease disaster area in the first image based on the disease disaster area in the first image.

[0135] Since the third multi-species ancient tree disease disaster model obtained in step S850 is a target detection model, this model can not only identify whether there is a disease disaster area in the first image, but also output the area where the disease disaster is located. Therefore, the center pixel of the disease disaster area output by the third multi-species ancient tree disease disaster model obtained in step S850 can be directly calculated based on the area where the disease disaster is located, and it can be used as the center pixel of the disease disaster area in the first image.

[0136] according to Figure 8 The method shown first trains a multi-species ancient tree species identification model and a multi-species ancient tree disease and disaster model. Then, the multi-species ancient tree species identification model identifies and outputs the tree species regions of the ancient trees in the first image, and replaces the output tree species region image with the first image. Subsequently, the replaced first image is input into the trained multi-species ancient tree disease and disaster model to obtain the disease and disaster regions of the ancient trees. Because this method first divides the current image (i.e., the first image of the monitoring area obtained in step S110) into sections based on tree species, and then identifies the disease and disaster regions within each section, it is more efficient than traditional methods. Figure 6 and Figure 7 The method shown, Figure 8 The method shown has high recognition accuracy.

[0137] Figure 9 The diagram shown is a schematic representation of a monitoring system for an ancient tree area provided in an exemplary embodiment of this application; this monitoring system can implement the ancient tree area monitoring method provided in any of the foregoing embodiments. Figure 9 As shown, the monitoring system 900 includes an acquisition module 910, a processing module 920, and a mapping module 930. The acquisition module is used to acquire a first image and a second image of the monitoring area using an acquisition device. The first image is obtained by the acquisition device at a first focal length, and the second image is obtained by the acquisition device at a second focal length, where the first focal length is greater than the second focal length. The processing module is used to process the first image using an ancient tree disease disaster model. If disease disaster exists in the first image, the center pixel of the disease disaster area in the first image is acquired. The mapping module is used to acquire the center pixel of the disease disaster area in the second image based on the center pixel of the disease disaster area in the first image, to assist the user in quickly locating the disease disaster area.

[0138] The monitoring system 900 also includes a panoramic display module 940, which is used to stitch the second image to generate a panoramic display image; obtain the mapping relationship between the second image and the panoramic display image; and map the center pixel of the diseased area in the second image to the panoramic display image to obtain the center pixel of the diseased area in the panoramic display image.

[0139] It should be understood that, for the sake of convenience and brevity, the specific working scenarios, processes, effects, and other details of each module in the above system 900 can be referred to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0140] This application also provides an electronic device. Figure 10 The diagram shown is a block diagram of an exemplary electronic device provided in an exemplary embodiment of this application. (Refer to...) Figure 10 The electronic device 1000 includes a memory 1010 and a processor 1020. The memory 1010 stores a computer program, and the processor 1020 runs the computer program to enable the electronic device 1000 to implement the ancient tree area monitoring method provided in any of the foregoing embodiments.

[0141] The electronic device 1000 may also include a power supply component configured to perform power management of the electronic device 1000, a wired or wireless network interface configured to connect the electronic device 1000 to a network, and an input / output (I / O) interface. The electronic device 1000 can be operated based on an operating system stored in the memory 1010, such as Windows Server. TM macOS X TM Unix TM Linux TM FreeBSD TM Or similar.

[0142] This application also provides a computer-readable storage medium storing a computer program thereon. When the computer program in the storage medium is executed by the processor 1020 of the electronic device 1000, the electronic device 1000 is able to implement the ancient tree area monitoring method provided in any of the foregoing embodiments.

[0143] This application also provides a computer program product, which includes instructions that, when executed by the processor 1020 of the electronic device 1000, enable the electronic device 1000 to implement the ancient tree area monitoring method provided in any of the foregoing embodiments.

[0144] The prompting method in this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in this application are performed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, network equipment, user equipment, core network equipment, OAM, or other programmable device.

[0145] The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device, a core network device, an OAM (Operational Information Management) system, or other programmable devices.

[0146] The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video optical disc; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.

[0147] It is understood that the specific examples provided in this application are only intended to help those skilled in the art better understand the embodiments of this application, and are not intended to limit the scope of the invention.

[0148] It is understood that in the various embodiments of this application, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0149] It is understood that the various embodiments described in this application can be implemented individually or in combination, and the embodiments of this application are not limited in this respect.

[0150] Unless otherwise stated, all technical and scientific terms used in the embodiments of this application have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The term "and / or" as used in this application includes any and all combinations of one or more of the associated listed items. The singular forms "a," "the," and "the" as used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0151] It is understood that the processor in the embodiments of this application can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiments can be completed by the integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0152] It is understood that the memory in the embodiments of this application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Specifically, non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may be random access memory (RAM). It should be noted that the memory in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0153] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0154] For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be indirect couplings or communication connections between devices or units through some interfaces, and may be electrical, mechanical, or other forms.

[0155] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0156] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0157] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] The above description is merely a specific embodiment of this application, but the scope of protection of this invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for monitoring ancient tree areas, characterized in that: The acquisition device acquires a first image and a second image of the monitoring area. The first image is acquired by the acquisition device at a first focal length, and the second image is acquired by the acquisition device at a second focal length. The first focal length is greater than the second focal length. The first image is processed using an ancient tree disease disaster model. If disease disaster exists in the first image, the center pixel of the disease disaster area in the first image is obtained. Based on the center pixel of the diseased area in the first image, the center pixel of the diseased area in the second image is obtained to assist the user in quickly locating the diseased area.

2. The method according to claim 1, characterized in that, The acquisition device includes a zoom camera, the first focal length being the maximum focal length of the zoom camera, and the second focal length being the minimum focal length of the zoom camera.

3. The method according to claim 1, characterized in that, The ancient tree disease disaster model is either a single-species ancient tree disease disaster model or a multi-species ancient tree disease disaster model.

4. The method according to claim 3, characterized in that, The single-species ancient tree disease disaster model was obtained through the following method: Determine the species of ancient trees in the monitored area; Based on the ancient tree species categories in the monitored area, obtain the single-species ancient tree disease disaster model.

5. The method according to claim 4, characterized in that, The determination of the ancient tree species category in the monitoring area includes: The training materials were labeled with ancient tree species categories using a labeling tool to obtain the first labeled training materials. The first mathematical model is trained based on the labeled first training material to obtain a single-species ancient tree classification model. Based on the first image, the ancient tree species classification model for single-species ancient trees is used to determine the ancient tree species category in the monitoring area.

6. The method according to any one of claims 4 or 5, characterized in that, The step of obtaining the single-species ancient tree disease disaster model based on the ancient tree species category in the monitoring area includes: Based on the ancient tree species category in the monitoring area, the training materials are labeled to determine whether they have diseases or disasters, thus obtaining the labeled second training materials; The second mathematical model is trained based on the labeled second training material to obtain the first single-species ancient tree disease disaster model; Wherein, after obtaining the first single-species ancient tree disease disaster model, the first image is processed using the ancient tree disease disaster model. If disease exists in the first image, the center pixel of the diseased area in the first image is obtained, including: The first image is processed using the first single-species ancient tree disease disaster model to determine whether the first image has disease disasters; If the first image has a disease or disaster, the center pixel of the first image is taken as the center pixel of the disease or disaster area in the first image.

7. The method according to any one of claims 4 or 5, characterized in that, The step of obtaining the single-species ancient tree disease disaster model based on the ancient tree species category in the monitoring area further includes: Using the ancient tree species categories and annotation tools in the monitored area, the disease and disaster areas in the training material are annotated to obtain the third training material after annotation; The third mathematical model is trained based on the labeled third training material to obtain the second single-species ancient tree disease disaster model; After obtaining the second single-species ancient tree disease disaster model, the first image is processed using the ancient tree disease disaster model. If disease exists in the first image, the center pixel of the diseased area in the first image is obtained, including: The first image is processed using the second single-species ancient tree disease disaster model to obtain the area where the disease disaster is located in the first image; Based on the area where the disease is located in the first image, obtain the center pixel of the disease area in the first image.

8. The method according to claim 3, characterized in that, The multi-species ancient tree disease disaster model was obtained through the following method: The training materials are labeled to determine whether they have diseases or disasters, and the labeled fourth training materials are obtained. The fourth mathematical model is trained based on the labeled fourth training material to obtain the first multi-species ancient tree disease disaster model; After obtaining the first multi-species ancient tree disease disaster model, the first image is processed using the ancient tree disease disaster model. If disease disaster exists in the first image, the center pixel of the disease disaster area in the first image is obtained, including: The first image is processed using the first multi-species ancient tree disease disaster model to determine whether the first image has disease disasters; If the first image has a disease or disaster, the center pixel of the first image is taken as the center pixel of the disease or disaster area in the first image.

9. The method according to claim 3, characterized in that, The multi-species ancient tree disease disaster model can also be obtained through the following methods: Using annotation tools, diseased and damaged areas in the training materials are annotated to obtain the fifth set of annotated training materials; The fifth mathematical model is trained based on the labeled fifth training material to obtain the second multi-species ancient tree disease disaster model; After obtaining the second multi-species ancient tree disease disaster model, the first image is processed using the ancient tree disease disaster model. If disease exists in the first image, the center pixel of the diseased area in the first image is obtained, including: The first image is processed using the second multi-species ancient tree disease disaster model to obtain the disease disaster area in the first image; Based on the area where the disease is located in the first image, obtain the center pixel of the disease area in the first image.

10. The method according to claim 3, characterized in that, The multi-species ancient tree disease disaster model can also be obtained through the following methods: Using annotation tools, the tree species areas and disease and disaster areas in the training materials were annotated, and the sixth and seventh training materials after annotation were obtained respectively. The sixth mathematical model is trained based on the labeled sixth training material to obtain a multi-species ancient tree species identification model. Based on the first image, the tree species region image is obtained using the multi-species ancient tree species identification model; Update the image of the tree species region to the first image; The seventh mathematical model is trained based on the annotated seventh training material to obtain the third multi-species ancient tree disease disaster model; After obtaining the third multi-species ancient tree disease disaster model, the first image is processed using the ancient tree disease disaster model. If disease exists in the first image, the center pixel of the diseased area in the first image is obtained, including: The first image is processed using the third multi-species ancient tree disease disaster model to obtain the disease disaster area in the first image; Based on the area where the disease is located in the first image, obtain the center pixel of the disease area in the first image.

11. The method according to claim 1, characterized in that, The acquisition device further includes a pan-tilt unit. The step of obtaining the center pixel of the diseased area in the second image based on the center pixel of the diseased area in the first image includes: Acquire the first angle of the device's gimbal when capturing the first image; Acquire the second angle of the device's gimbal when capturing the second image; Based on the first angle and the second angle of the device's gimbal, obtain the mapping relationship between the first image and the second image; Based on the mapping relationship between the first image and the second image, the center pixel of the diseased area in the first image is mapped to the second image to obtain the center pixel of the diseased area in the second image.

12. The method according to claim 1, further comprising: Multiple second images are stitched together to generate a panoramic display image; Obtain the mapping relationship between multiple second images and the panoramic display image; The center pixels of the diseased and damaged areas in the multiple second images are mapped to the panoramic display image to obtain the center pixels of the diseased and damaged areas in the panoramic display image.

13. A monitoring system for an ancient tree area, characterized in that, include: The acquisition module is used to acquire a first image and a second image of the monitoring area through the acquisition device. The first image is acquired by the acquisition device at a first focal length, and the second image is acquired by the acquisition device at a second focal length. The first focal length is greater than the second focal length. The processing module is used to process the first image using an ancient tree disease disaster model. If there is a disease disaster in the first image, the center pixel of the disease disaster area in the first image is obtained. The mapping module is used to obtain the center pixel of the diseased area in the second image based on the center pixel of the diseased area in the first image, so as to assist the user in quickly locating the diseased area.

14. The system according to claim 13, characterized in that, The system also includes: The panoramic display module is used to stitch the second image to generate a panoramic display image; obtain the mapping relationship between the second image and the panoramic display image; and map the center pixel of the diseased area in the second image to the panoramic display image to obtain the center pixel of the diseased area in the panoramic display image.

15. An electronic device, characterized in that, include: Memory; A processor, the memory for storing a computer program, the processor running the computer program to cause the electronic device to perform the monitoring method for ancient tree areas as described in any one of claims 1 to 12.

16. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the monitoring method for ancient tree areas as described in any one of claims 1 to 12.

Images